Platinum Metals ReviewE-mail: [email protected] E-ISSN 1471–0676 PLATINUM METALS REVIEW A Quarterly Survey of Research on the Platinum Metals and of Developments in their Application

PlatinumMetalsReview

www.platinummetalsreview.comE-ISSN 1471–0676

VOLUME 53 NUMBER 3 JULY 2009

© Copyright 2009 Johnson Matthey PLC

http://www.platinummetalsreview.com/

Platinum Metals Review is published by Johnson Matthey PLC, refiner and fabricator of the precious metals and sole marketing agent for the six platinumgroup metals produced by Anglo Platinum Limited, South Africa.

All rights are reserved. Material from this publication may be reproduced for personal use only but may not be offered for re-sale or incorporated into,reproduced on, or stored in any website, electronic retrieval system, or in any other publication, whether in hard copy or electronic form, without the priorwritten permission of Johnson Matthey. Any such copy shall retain all copyrights and other proprietary notices, and any disclaimer contained thereon, andmust acknowledge Platinum Metals Review and Johnson Matthey as the source.

No warranties, representations or undertakings of any kind are made in relation to any of the content of this publication including the accuracy, quality orfitness for any purpose by any person or organisation.

Editor: David Jollie; Assistant Editor: Sara Coles; Editorial Assistant: Margery Ryan; Senior Information Scientist: Keith White

Platinum Metals Review, Johnson Matthey PLC, Orchard Road, Royston, Hertfordshire SG8 5HE, U.K.E-mail: [email protected]

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. 53 JULY 2009 NO. 3

ContentsA New Palladium-Based Ethylene Scavenger to Control 112

Ethylene-Induced Ripening of Climacteric FruitBy Andrew W. J. Smith, Stephen Poulston, Liz Rowsell,

Leon A. Terry and James A. Anderson

Thermodynamic Properties of Platinum Diatomics 123By Pavitra Tandon and K. N. Uttam

“Carbons and Carbon Supported Catalysts in Hydroprocessing” 135A book review by Martyn V. Twigg

Plastic Deformation of Polycrystalline Iridium 138at Room Temperature

By Peter Panfilov, Alexander Yermakov, Olga V. Antonovaand Vitalii P. Pilyugin

Fuel Cells Science and Technology 2008 147A conference review by Donald S. Cameron

The Platinum Development Initiative: Platinum-Based Alloys for 155High Temperature and Special Applications: Part III

By L. A. Cornish, R. Süss, L. H. Chown and L. Glaner

5th International Conference on Environmental Catalysis 164A conference review by Rodney Foo and Noelia Cortes Felix

“Fuel Processing: for Fuel Cells” 172A book review by Joseph McCarney

“Platinum 2009” 174

Abstracts 175

New Patents 177

Final Analysis: The Impact of CO2 Legislation on 179PGM Demand in Autocatalysts

By Lucy Bloxham

112

IntroductionEthylene is one of the simplest plant growth

regulators and is known to play a role in manyphysiological processes in plants. Seed germinationand growth, abscission, fruit ripening and senes-cence can all be affected by ethylene (1–4). It isbelieved that the ancient Egyptians and Chinesewere aware of some of these effects and attemptedto artificially control the ripening of figs and pears.More recently, in the nineteenth century, it wasobserved that street lighting (fuelled by town gaswhich contained low levels of ethylene) was caus-ing stunting and other changes to nearby plantgrowth (2). Today, although it is well documentedthat almost all plants produce ethylene in varyingamounts, understanding how ethylene interactswith each plant type and the effect this has onplant development is a very active area of scientific research.

The ability to control ripening is very importantto the sale of many fresh produce types. Whilstconsumers often want to purchase food that canbe eaten straight away, suppliers may prefer to

market fruit in need of further ripening, to avoidlosses due to over-ripening in transit or storage.Therefore, the control of the ambient ethyleneconcentration is key to prolonging the shelf-life ofmany horticultural products.

Various methods of ethylene control areoffered commercially, including several based onethylene adsorption/oxidation. However, thesetechnologies have seen limited uptake, especially inretail packaging and fresh produce transportation.In the present paper, we report on the developmentof a new palladium (Pd)-based ethylene adsorber,which works effectively both in synthetic gasstreams and in laboratory-based trials using realfruit samples. We are now in a position to under-take ‘real-world’ testing.

Ethylene Production by Fresh ProduceFresh fruit and vegetables are generally classi-

fied in one of two ways, depending upon themechanism of ripening (5) and their capacity toproduce ethylene (Table I). Climacteric itemsrelease a burst of ethylene during ripening, accom-

Platinum Metals Rev., 2009, 53, (3), 112–122

A New Palladium-Based EthyleneScavenger to Control Ethylene-InducedRipening of Climacteric Fruit By Andrew W. J. Smith*, Stephen Poulston and Liz Rowsell Johnson Matthey Technology Centre, Blounts Court, Sonning Common, Reading RG4 9NH, U.K.; *E-mail: [email protected]

Leon A. TerryPlant Science Laboratory, Cranfield University, Cranfield, Bedfordshire MK43 0AL, U.K.

and James A. AndersonSurface Chemistry and Catalysis Group, Department of Chemistry, University of Aberdeen, Aberdeen B24 3UE, U.K.

A novel palladium-promoted zeolite material with a significant ethylene adsorption capacityat room temperature is described. It was characterised by diffuse reflectance infrared Fouriertransform spectroscopy (DRIFTS) and transmission electron microscopy (TEM) to showpalladium particles dispersed over the support. Initial measurements of the ethylene adsorptioncapacity were conducted with a synthetic gas stream at a higher ethylene concentrationthan would normally be encountered in fruit/vegetable storage, in order to obtain an acceleratedtesting protocol. Further laboratory-based trials on fruit samples show that the palladium-promoted zeolite material can be effective as an ethylene scavenger to prolong the shelf-lifeof fresh fruits.

DOI: 10.1595/147106709X462742

panied by an increase in respiration, whereas non-climacteric produce do not vary their rate ofethylene production in this fashion. Exogenousethylene (typically < 0.1–1.0 μl l–1) can initiateripening in many climacteric fruit which then canlead to autocatalytic production of ethylene by thefruit. Climacteric fruits, which include bananas,avocados, nectarines and pears, ripen after harvest,typically by softening, changing colour and becom-ing sweeter (6). The latter class, includingstrawberries, grapes and pineapples do not ripen asdramatically after harvest, but rather senesce, lead-ing to discolouration, unpleasant odour, shrinkageand general rotting. In such cases, the challenge tomaintain product quality is paramount. Other freshproduce types are sensitive to ethylene eventhough their ethylene production is very low forexample, potato tubers, bulb onions, broccoli andspring cabbage, along with some cut flowers.

To artificially reproduce the natural ripeningprocess, ethylene can be introduced during storage.This process is used in the fruit industry on freshproduce such as bananas, avocadoes and mangoes(7). As well as such deliberate exogenous introduc-tion, the level of ethylene can rise due to ethyleneemission from stored fresh produce or from acci-dental sources such as forklift truck exhaustemissions. For this reason, the industry prefers touse electric vehicles in fresh food storage areas tominimise the risk of the fresh produce coming intocontact with ethylene gas.

Excessive or uncontrolled levels of ethylene canresult in a number of problems. For example, thepremature ripening of fruits and vegetables, thefading and wilting of cut flowers and the loss ofgreen colour and increase in bitterness of vegeta-

bles are common problems when ethylene levelshave not been properly managed.

Ethylene-Sensitive Plants Plants, flowers and buds are also sensitive to

ethylene (see Table II for examples) and somerelease ethylene gas when they are cut or damaged.Some of the most detrimental effects of ethyleneon plants are:– Partial or incomplete flower abortion;– Retarded plant growth;– Growth abnormalities such as excessive leafi-

ness or the stimulated growth of daughterbulbs;

– Shortened vase lifespan of cut flowers (abscis-sion of leaves and flower petals);

– Inhibited development of immature (unopened)flower buds;

– Accelerated senescence of all types of plants;– Susceptibility to disease.

Ethylene Removal TechnologiesIn many instances, ethylene concentrations can

be controlled by ventilation of the area containing

Platinum Metals Rev., 2009, 53, (3) 113

Table I

Ethylene Production from Different Fresh Produce Types

Low Moderate High Very high (< 1.0 ml kg–1 h–1) (1–10 ml kg–1 h–1) (10–100 ml kg–1 h–1) (> 100 ml kg–1 h–1)

Pineapple, artichoke, Banana, mango, Apricot, nectarine, Apple, avocado,cauliflower, broccoli, plum, tomato pear, peach cherimoya,date, orange, rhubarb, passion fruitspinach, beetroot,green asparagus, celery, lemon, onion

Table II

Different Degrees of Ethylene Sensitivity amongstCut Flower Species

Low Moderate High Sensitivity Sensitivity Sensitivity

Tulip, Lily, freesia, Carnation,daffodil agapanthus, Geraldton

alstroemeria, waxfloweranemone,dahlia

the fresh produce. This can be aided by containerdesigns that allow for air circulation. However,ventilation is not always appropriate; for example,it cannot be used in sealed environments, such ascontrolled atmosphere storage or retail packaging.There are a number of ethylene removal technolo-gies available:– Catalysts. Often based on platinum/alumina,

these operate at elevated temperature(> 200ºC) and catalytically oxidise ethylene tocarbon dioxide (CO2) and water (8). There arealso reports of the use of photocatalytic oxida-tion of ethylene using titanium dioxide (TiO2),which can occur at room temperature (9).

– Stoichiometric oxidising agents. Mostly based onpotassium permanganate (KMnO4), whichagain oxidises ethylene and is itself reduced.Though some CO2 and water is produced,some partially oxidised species such as car-boxylic acids may also be formed.

– Sorbents. These materials work by sorption ofthe ethylene and are often based on high sur-face area materials, including activated carbon,clays and zeolites.

Ethylene Blocking TechnologiesA different approach is to inhibit ethylene

action in the produce itself, which can in turnreduce the amount of ethylene released by the pro-duce into the container or storage area. Severalchemicals have been shown to act as ethyleneinhibitors, including both volatile and aqueoustreatments:–– 1-Methylcyclopropene (1-MCP). 1-MCP is the most

widely used commercial volatile ethyleneinhibitor, which blocks ethylene binding sites.It is applied exogenously as a gas and has beenwidely applied to fruit (particularly apples) andflowers. 1-MCP is sold commercially asSmartFreshSM (10) into the fresh produceindustry and as EthylBlockTM into the floralindustry.

–– Silver thiosulfate (STS). The use of this material islargely restricted to cut flowers and it is soldcommercially under the trade name ChrysalAVB®. It is applied by putting the cut flowerstems in a solution containing the STS.

–– Aminoethoxyvinylglycine (AVG). This material issold commercially as ReTain® and acts as aplant growth regulator by blocking the produc-tion of ethylene in the plant tissue. It is sprayedonto the fruit, usually 1 to 3 weeks prior to harvesting.

New Palladium-Based EthyleneScavenging Technology

In this paper, we report on the discovery byJohnson Matthey scientists of a novel palladium-promoted material with a significant ethyleneadsorption capacity at room temperature (11). Awide range of materials were synthesised andscreened for activity. Pd gave by far the best per-formance of the promoter metals tested. Thematerial is a palladium-impregnated zeolite givingfinely dispersed palladium particles.

Initial testing was conducted with a syntheticgas stream at a higher ethylene concentration thanwould normally be encountered in fruit/vegetablestorage, in order to obtain an accelerated testingprotocol for measuring ethylene adsorptioncapacity. Ethylene adsorption capacity measure-ments were carried out at room temperature(21ºC) in a plug flow reactor using 0.1 g of activePd-based material with a gas composition of 200μl l–1 ethylene, 10% (v/v) oxygen balanced withhelium, at a flow rate of 50 ml min–1, with andwithout ca. 100% relative humidity (RH). Reactoroutlet gas concentrations were analysed using aThermo Onix ProLab mass spectrometer(Thermo Onix, Houston, Texas, U.S.A.). Amass:charge ratio of either 26 or 27 was used forethylene, as the use of nitrogen as a diluent gasleads to the presence of a large peak at m/z = 28.A value of m/z = 44 was used for CO2. Ethyleneuptake capacity was measured using a simple‘breakthrough’ measurement, in which the totalintegrated ethylene removal was determined afterthe outlet ethylene concentration from the reactorhad reached the inlet ethylene concentration,showing that the adsorber was saturated with ethylene (12).

The Pd-based material typically removed allmeasurable ethylene until breakthrough occurred.A typical example of an ethylene breakthrough


measurement under humid conditions is shown inFigure 1. Under these conditions the Pd-promotedmaterial was found to have an ethylene adsorptionof 4162 μl g–1 under ca. 100 % RH. This perfor-mance was increased to 45,600 μl g–1 under dryconditions.

Further experiments were carried out at roomtemperature in a sealed, unstirred batch reactor(0.86 l) with 0.1 g active Pd-based material and aninitial gas composition of 550 μl l–1 ethylene, 40%(v/v) air balanced with argon. Selected gas concen-trations were measured at hourly intervals with aVarian CP-4900 Micro-GC (Varian, Inc, Palo Alto,California, U.S.A.). Gas samples (40 ms duration)were taken via an automated recirculating samplingsystem. Column and injector temperatures were setat 60ºC and 70ºC, respectively. The 0.15 mm diameter, 10 m long column was packed withPoraPLOT (porous layer open tubular) Q station-ary phase. Ethylene and CO2 were calibratedagainst 10 μl l–1 ethylene balanced with air and 5%(v/v) CO2 balanced with Ar (Air Products Europe,Surrey, U.K.). A thermal conductivity detector wasused with He carrier gas at 276 kPa inlet pressure.Peak integration was carried out within the VarianSTAR software.

Under these conditions, CO2 and ethane pro-duction were observed, as plotted in Figure 2.The ethane is likely to be produced by the hydro-genation of adsorbed ethylene, the hydrogen beinggenerated by the partial dissociation of ethylene.

The selectivity to ethane varies with experimentalconditions but is typically no more than the maxi-mum value of ~ 10% observed under theconditions tested here. Ethane can be produced byplants in response to stress (13) and has not previ-ously been reported to be detrimental to plants inthe concentration range reported here. It is clear,however, that the palladium-based material is act-ing largely as an adsorber rather than as a catalyst.The mechanism of reaction is discussed further inthis article by interpretation of diffuse reflectanceinfrared Fourier transform spectroscopy (DRIFTS)data.

Characterisation of the palladium-based mater-ial was carried out to determine the Pddistribution through the support. Transmissionelectron microscopy (TEM) analysis (Figure 3)showed that the palladium particles (bright parti-cles) are dispersed over the support material. Theaverage size of the palladium particles in Figure 3was calculated to be 1.7 nm. Scanning electronmicroscopy (SEM) analysis (not shown) also iden-tified some larger palladium particles withdiameters around 20 to 40 nm. These results cor-respond well with CO chemisorption data whichgave a metal dispersion of ~ 15 %. From this dis-persion value, a slightly larger average size of thepalladium particles would be expected. This isconsistent with the observation of many small andsome larger particles by TEM (Figures 3 and 4)and SEM.


200

150

100

50

0 500 1000 1500 2000 2500 3000 3500

Eth

yle

ne c

oncentr

ation,

μl l–

1

Time, s

Fig. 1 Ethylenebreakthrough graph of thePd-based material in humidgas conditions. Ethylene wasmonitored using the mass 26signal on a massspectrometer (mass 26 and27 gave very similarbreakthrough profiles)

Ethylene–Metal Interaction viaDRIFTS Analysis

In order to gain further information regardingthe processes involved in ethylene removal, avibrational spectroscopic study was performed to probe the interactions between the ethyleneand the Pd-promoted scavenger. DRIFTS wasused to characterise the species adsorbed on theethylene scavenger after exposure to ethylene, inthe presence and absence of oxygen and watervapour.

DRIFTS allows spectra to be obtained of pow-dered samples in the presence of gaseousatmospheres. It produces infrared spectra of the

diffusely scattered radiation in the reflectancemode (Figure 5), which are analogous to those col-lected in the more conventional transmittancemode. Disadvantages of this method include theappearance of artefacts resulting from the collec-tion of specularly reflected radiation. It is alsodifficult to perform quantitative studies, since bothscattering and absorption coefficients must beconsidered. By contrast, only the latter is requiredto quantify adsorbed species when spectra are col-lected in the transmittance mode and converted toabsorbance (14). The key advantages of thismethod are that no sample preparation is neces-sary and that the cells can be operated in plug flowmode with the reactant gases being forced to trav-el through the bed of powdered scavengermaterial.

Commercial DRIFTS cells are readily availablewhich allow the collection of spectra of powderedsamples at ambient or elevated temperatures,while controlling the composition of the gaseousatmosphere. In the study performed here, the out-let port of the DRIFTS cell was coupled to aquadropole mass spectrometer (QMS) via a heated,glass-lined capillary. This permitted continualmonitoring of ethane (m/z = 30), carbon dioxide(m/z = 44) and ethylene (a value of m/z = 27 wasselected for this experiment) during exposure toethylene and during subsequent temperature-pro-grammed desorption (TPD) measurements.Experiments were performed by exposing samples


Time, h

Conce

ntr

ation o

f C

O2

and e

thyle

ne,

μl l–

1

Concentra

tion o

f eth

ane, μ

l l–

1

4 8 120

0

100

200

300

400

500 100

80

60

40

20

0

Ethylene (Pd scavenger)

Carbon dioxide

Ethane

Fig. 2 Gasconcentrationsin a batchreactor initiallycontaining 550 μl l–1

ethylene*,along with 0.1g of thePd-promotedethylenescavenger

*Note: Someethylene hasbeen removedby thescavengerprior to thefirst injection

10 nm

Fig. 3 TEM image of the Pd-promoted zeolite materialshowing nanometre size palladium particles (brightareas) on the zeolite support

with and without Pd, at ambient temperature, to aflow of ethylene in nitrogen with the addition ofeither no other gases, air or air and water. After aperiod of exposure to these gaseous atmospheres,the ethylene stream was switched off and a TPDmeasurement was performed in the presence ofthe remaining gases. Spectra were also obtained ofsamples that had been exposed to ripened fruit.

The Fourier transform infrared (FTIR) spec-trum of ethylene in nitrogen flowing through thecell, in the absence of scavenger, gave main fea-tures at 3112 cm–1 (ν9, νCH asym.), 2992 cm–1

(ν11, νCH sym.), 2048 cm–1 (2 × ν10, γCH2 rock-ing), 1889 cm–1 (2 × ν7, δCH2 out-of-plane) and1446 cm–1 (ν12, δCH2 in-plane). The absence ofbands for the IR inactive (totally symmetrical)modes at 1623 cm–1 (ν2, νC=C) and 1343 cm–1

(ν3, δCH2 in-plane) should be noted, although it isexpected that any interaction with the adsorbentmight permit these modes to be detected as thesymmetry is lifted. These interactions might alsobe expected to shift the frequencies away from thegas phase values listed above.

FTIR spectra of a sample exposed to fruit(Figure 6(a)) showed features at 1467, 1460, 1434and 1382 cm–1. FTIR spectra of the Pd-free zeoliteshowed only very weak features, including a bandat 1439 cm–1 which could be assigned to the pres-ence of adsorbates resulting from exposure toethylene (Figure 6(c)). However, these featureswere absent following heating to 50ºC in air. Thenarrow shape of the feature at 1439 cm–1, and theappearance of a single rather than a double compo-nent (as in the gas phase), confirms that thisfeature can be assigned to an adsorbed state,although the limited shift (Δν = 7 cm–1) withrespect to the gas phase feature (due to the ν12,δCH2 in-plane) would suggest a very weakly boundstate. This was confirmed by the disappearance ofbands at this frequency when the sample was heat-ed in air to 50ºC. Note that this ease of removalwas not the case for the sample containing Pd(Figure 6(b)). This, along with the detection ofother infrared bands (at 1467 and 1382 cm–1) whenPd was present, would indicate that the metalplayed a significant role in the retention of theadsorbate and in the dominant adsorbed state. Therelative intensities of the features indicate that the


Num

ber

of P

d p

art

icle

s

Pd particle size, nm

0.5 1 1.5 2 3.52.5 3 4 4.5 5 >50

40

60

80

20

100

120

140

Fig. 4 Particle size distribution ofpalladium in the palladium-promotedzeolite material, as determined from theTEM image in Figure 3

Incident beamDiffuse reflectance

Sample

Fig. 5 Scheme showing the scattering of light from apowdered sample and the collection of the diffuselyscattered component for DRIFTS analysis

presence of Pd was essential in order to achievethe 45,600 μl g–1 adsorption capacity obtainedunder dry conditions. The similarities between thefeatures resulting from exposing the sample tofruit (Figure 6(a)) and to exogenous ethylene, inparticular the features at 1467 and 1382 cm–1, arestrong evidence that similar adsorbed species existin both cases. Under these conditions, the C–Hstretching region gave two dominant bands at2969 and 2865 cm–1, which are consistent withexpectations for CH3 species (Figure 7).

The transformation from C2H4 to adsorbedspecies containing CH3 would be consistent withthe development of ethylidyne (CCH3) species (15,16). In these species, the carbon forms three bondsto the surface metal atoms (Figure 8(c)), probablyon three-fold hollow sites of Pd(111) type facets(17). Such an assignment would be consistent withthe detection of the 1467 and 1382 cm–1 band pair(Figure 6) arising from the corresponding CH3

deformation modes. The formation of a hydrogenatom for each ethylidyne species generated would


180 0 1 600 1400 1200

Refle

cta

nce/%

W a venu mb er/cm-1

20%

(a)

(b)

(b)

(a )(c)

(c)

Wavenumber, cm–1

Reflecta

nce, %

20 %

1800 1600 1400 1200

(a)

(b)

(c)

(a)

(b)

(c)

Fig. 6 DRIFT spectra of: (a) the Pd-promoted zeolitematerial following exposure tofruit, (b) the Pd-promoted zeolitematerial exposed to a flow ofethylene/nitrogen at 25ºC, (c) Pd-free zeolite exposed toethylene in nitrogen at 25ºC

3200 3 150 310 0 3050 30 00 2950 2900 28 50 2800 2 750 2700

Reflecta

nce/%

W ave num be r/cm-1

(a )

(b)

(c)

1 0%

Wavenumber, cm–1

Reflecta

nce, %

10 %

(a)

(b)

(c)

3200 3150 3100 3050 3000 2950 2900 2850 2800 2750 2700

Fig. 7 DRIFT spectra of thePd-promoted ethylene scavenger: (a) before exposure to ethylene,(b) after exposure to a flow ofethylene in nitrogen at 25ºC, (c) after exposure to a flow ofethylene in air at 25ºC

explain the formation of ethane following initialexposure of the scavenger to ethylene (Figure 2),which was observed in the mass spectrometry (MS)trace (m/z = 30) recorded during the FTIR mea-surements. The surface selection rule states thatvibrations parallel to a metal surface should not bedetected, due to the creation of an opposing imagedipole on the metal. This means that the νCH3

asymmetric and δCH3 asymmetric modes at 2969and 1467 cm–1 should be absent from the spec-trum. However, this rule will be relaxed for verysmall metal crystallites of the order of 2 nm andbelow and where extended flat facets are absentdue to the presence of steps and edges.

Such a description of the morphology is consis-tent with the TEM image presented (Figures 3 and4). This assignment cannot be confirmed from theexpected absorption at around 1130 cm–1 (νC–C),due to absorption by the support in this region.Bands due to ethylidyne were diminished in spec-tra recorded after heating the sample in nitrogen to150ºC. If air was present during this thermal treat-ment, then CO2 was detected by MS as a completeoxidation product. Below this temperature, FTIRevidence for partial oxidation included the detec-tion of carboxylate type species and a band at 2125cm–1 due to CO adsorbed at surface oxidised Pdsites. The latter was detected in spectra of the sam-ple recorded at 100ºC, indicating that the onset ofoxidation took place well below the temperature ofcomplete oxidation.

An MS trace was recorded during the collectionof spectra with exposure to ethylene. This showeda breakthrough shape which was strongly depen-dent on the composition of the gas phase, with thegreatest removal of ethylene occurring during treat-ment in nitrogen. The presence of air or air and

water reduced the adsorption capacity.Additionally, the presence of air or air and watermodified the predominant modes of adsorption,with four bands now appearing in the C–H stretch-ing region. In addition to the pair at 2957 and 2865cm–1 which have already been assigned to the CH3

modes of ethylidyne in the absence of air or water,a further pair at 2934 and 2853 cm–1 were detected,although these were very weak when water waspresent. These features are consistent with expec-tations for CH2 groups, although the vibrationalfrequencies are relatively low for vinyl species. Di-σ species of adsorbed ethylene (Figure 8(b)) givelower frequency CH modes than π-C2H4 (18)(Figure 8(a)), although the former are less favouredin the presence of adsorbed oxygen (18). The pres-ence of oxygen in the system also led to theappearance of a species giving a maximum at 1514cm–1, which was absent for the air-free system.

A similar feature was found at 1510 cm–1 inelectron energy loss spectroscopy (EELS) spectraof ethylene on oxygen covered Pd(100). This wasassigned to δCH2, so it would be tempting toassign the surface species to vinyl intermediatessuch as HCCH2 (17). However, an alternativeassignment, consistent with the known stepwisedehydrogenation of ethylene (19), is that the addi-tional maxima at 2934 and 2853 cm–1 representvibrations due to the CH3 groups of ethylidene(CHCH3), where the expected full conversion toethylidyne, observed in the absence of air andwater (18), is hindered due to the presence of theco-adsorbates which limit the activation and dissociation of the C–H bond. The lesser extent ofC–H dissociation at room temperature, and subse-quently the lesser population of the surface byadsorbed hydrogen, would explain the reduced


(a) (e)(b) (c) (d) (f)

Pd Pd Pd Pd Pd Pd

Fig. 8 Potential adsorbed species following exposure of the Pd-containing scavenger to ethylene: (a) π-bonded, (b) di-σ bonded, (c) ethylidyne, (d) ethylidene, (e) vinylidene and (f) vinyl

ability of the surface to liberate ethane resultingfrom ethylene hydrogenation during the initialexposure stages.

In summary, the surface speciation is some-what dependent on the gas mix and the moisturecontent. However, the DRIFTS study has identi-fied the key role of the palladium–ethyleneinteraction and the benefit that this has on bindingethylene to the zeolite surface.

Fresh Produce ResearchThe use of platinum group metals including pal-

ladium in different forms and on different supportsfor ethylene removal from fruit and vegetables hasbeen investigated by other authors in the past (20,21). Bailén et al. at University Miguel Hernándezhave recently published results on delaying tomatodecay using a combination of controlled atmos-phere packaging and a granular-activated carbon(GAC) or GAC impregnated with a palladium-based catalyst (22). Martínez-Romero et al. at thesame University have also reported the use of a car-tridge heater device (optimally running at 175ºC)joined to activated carbon containing palladium forethylene removal above room temperature (23).

In contrast, the new material described hereconsists of a specific combination of a preciousmetal with a zeolite and removes significantamounts of ethylene at low temperature (5ºC) androom temperature. In order to test the materialunder realistic conditions, the active material hasbeen evaluated in collaboration with the PlantScience Laboratory at Cranfield University, U.K.Bananas were included in the initial studies, as therole of ethylene in initiating ripening in these fruits

is well documented (24, 25). Initial findings, nowpublished (2), have demonstrated for the first timethat the presence of a palladium-based scavengerwas effective at removing ethylene to below phys-iologically active levels for preclimacteric greenbananas and green avocado fruits.

Reduced CO2 production and control of thecolour change from green to yellow was observedfor the preclimacteric bananas (Figure 9). The palladium-promoted ethylene scavenger was alsofound to be far superior to a KMnO4-based ethyl-ene adsorber when used in low amounts at highrelative humidity. No adverse effects on fruit qual-ity or subsequent ripening were observed afterremoval of the ethylene scavenger material.

Similar experiments were also conducted onavocados. Results showed that exogenous andendogenous ethylene concentration was reducedsignificantly with increasing amounts of the Pd-promoted material. In the presence ofPd-promoted material, ethylene was removed tobelow physiologically active levels. The effect ofethylene on the colour of avocado cv. Hass fruitsin the presence or absence of the palladium mate-rial is shown in Figure 10. Fruit held in thepresence of 100 mg or 1000 mg of the Pd-promot-ed material for three days were generally greener,and thus less ripe, than control fruit after seven toten days (Figure 10).

Furthermore, when avocados were treated withethylene and then subsequently held in the pres-ence of the Pd-promoted material (1000 mg) afterday 1, ethylene was removed to below physiologi-cally active levels. Despite the climacteric phasehaving been initiated for these fruits, the


Fig. 9 Colour of a five-day-old banana cv. Cavendish fruit previously held for three days at 16ºC in 3 l sealed jarscontaining the Pd-promoted ethylene scavenger material (0–50 mg) and previously treated with (+E) or without (–E)100 μl l–1 ethylene when at the preclimacteric stage (i.e. green) at day 0

–E

+E

0 mg 50 mg 40 mg 30 mg 20 mg 10 mg

subsequent total removal of ethylene resulted inbetter maintenance of fruit firmness as comparedto controls. This suggests that the normal andexpected climacteric respiratory rise has been dis-rupted. Therefore, for the first time an ethylenescavenger has been shown to be capable of extend-ing shelf-life even when the climacteric respiratoryrise has already been initiated.

Current work is focused on comparing the effi-cacy of the Pd-promoted ethylene scavenger to1-MCP for the control of ripening in avocado fruit,and the resultant effect on non-structural carbohy-drates and fatty acid methyl esters.

ConclusionsThe results from this study demonstrate that

the Pd-promoted ethylene scavenger described

here is effective for the control of ethylene to pro-long the shelf-life of climacteric fresh producesuch as bananas and avocados. The material hasthe potential to be used commercially, as an alter-native and/or supplemental treatment to 1-MCP.The technology does not require elevated temper-ature to remove ethylene, and the choice of zeolitesupport material makes it suitable for most freshproduce and floral applications under conditionsof high humidity and low or room temperature.Future research will elucidate further uses of thePd-promoted ethylene scavenger.

AcknowledgementsThe authors would like to thank Anglo

Platinum for sponsoring the research and develop-ment in this area.


+E

–E

0 mg 100 mg 1000 mg*1000 mg

Fig. 10 Colour of a seven-day-oldavocado cv. Hass fruit previouslyheld for three days at 12ºC in 3 lsealed jars containing thePd-promoted ethylene scavengermaterial (0, 100 or 1000 mg) andpreviously treated with (+E) orwithout (–E) 100 μl l–1 ethylenewhen at the preclimacteric stage (i.e. green) at day 0* The Pd-promoted material (1000 mg) was put into the jars after day 1, following treatment ofthe fruit with or without 100 μl l–1

ethylene

1 A. El Blindi, L. Rigal, G. Malmary, J. Molinier and L.Torres, ‘Ethylene Removal for Long TermConservation of Fruits and Vegetables’, Food Qual.Preference, 1993, 4, 119

2 “Ethylene Action in Plants”, ed. N. A. Khan, Springer-Verlag, Berlin, Heidelberg, Germany, 2006

3 F. B. Abeles, P. W. Morgan and M. E. Saltveit, Jr.,“Ethylene in Plant Biology”, 2nd Edn., AcademicPress, San Diego, California, U.S.A., 1992

4 “Fruit Quality and its Biological Basis”, ed. M. Knee,Wiley-Blackwell, Chichester, West Sussex, U.K., 2002

5 M. E. Saltveit, Postharvest Biol. Technol., 1999, 15, (3),279

6 D. V. Raghava Rao and B. S. Chundawat, ‘Extensionof Shelf Life of Lacatan Bananas Stored in Cartons’,Gujarat Agric. Univ. Res. J., 1986, 11, (2), 26

7 K. P. Sudheer and V. Indira, in “Post HarvestTechnology of Horticultural Crops”, ed. K. V. Peter,

New India Publishing, New Delhi, India, 2007, p. 858 J. Conte, A. El Blidi, L. Rigal and L. Torres, J. Food

Eng., 1992, 15, (4), 313 9 C. T. Brigden, S. Poulston, M. V. Twigg, A. P. Walker

and A. J. J. Wilkins, Appl. Catal. B: Environ., 2001, 32,(1–2), 63

10 C. B. Watkins, Biotechnol. Adv., 2006, 24, (4), 38911 T. Ilkenhans, S. Poulston and A. W. J. Smith, Johnson

Matthey PLC, World Appl. 2007/052,07412 L. A. Terry, T. Ilkenhans, S. Poulston, L. Rowsell and

A. W. J. Smith, Postharvest Biol. Technol., 2007, 45, (2),214

13 T. W. Kimmerer and T. T. Kozlowski, Plant Physiol.,1982, 69, (4), 840

14 J. A. Anderson, C. Fergusson, I. Rodríguez-Ramosand A. Guerrero-Ruiz, J. Catal., 2000, 192, (2), 344

15 M. Kaltchev, A. W. Thompson and W. T. Tysoe, Surf.Sci., 1997, 391, (1–3), 145

References


16 M. K. Ainsworth, M. R. S. McCoustra, M. A. Chesters,N. Sheppard and C. De La Cruz, Surf. Sci., 1999, 437,(1–2), 9

17 N. Sheppard and C. De La Cruz, in “Advances inCatalysis”, ed. D. D. Eley, Academic Press, San Diego,California, U.S.A., 1996, Vol. 41, p. 1

18 E. M. Stuve, R. J. Madix and C. R. Brundle, Surf. Sci.,1985, 152–153, Pt. 1, 532

19 G. A. Somorjai, A. M. Contreras, M. Montano andR. M. Rioux, PNAS, 2006, 103, (28), 10577

20 P. J. Price, Johnson Matthey PLC, British Patent Appl.2,163,637; 1986

21 K. Abe and A. E. Watada, J. Food Sci., 1991, 56, (6),1589

22 G. Bailén, F. Guillén, S. Castillo, M. Serrano, D. Valeroand D. Martínez-Romero, J. Agric. Food Chem., 2006,54, (6), 2229

23 D. Martínez-Romero, F. Guillén, S. Castillo, P. J.Zapata, M. Serrano and D. Valero, Postharvest Biol.Technol., 2009, 51, (2), 200

24 J. B. Golding, D. Shearer, W. B. McGlasson and S. G.Wyllie, J. Agric. Food Chem., 1999, 47, (4), 1646

25 J.-M. Lelièvre, A. Latchè, B. Jones, M. Bouzayen andJ.-C. Pech, Physiol. Plant., 1997, 101, (4), 727

The Authors

Dr Stephen Poulston is a PrincipalScientist at JMTC. His main interests liein heterogeneous catalysts andsorbents.

Dr Elizabeth Rowsell is a ResearchManager at JMTC. She is responsiblefor managing the Ethylene Scavengerresearch project and developing newapplications for platinum group metals.

Dr Andrew W. J. Smith is a PrincipalScientist at the Johnson MattheyTechnology Centre (JMTC), SonningCommon, U.K. He is a syntheticinorganic chemist with experience inmaterials, glass technology and catalystpreparation. He is interested indeveloping new materials and theirapplications in heterogeneous catalysis.

Dr Leon Terry is a Senior Lecturer inPlant Science and Head of the PlantScience Laboratory at CranfieldUniversity, U.K. The Plant ScienceLaboratory is one of the largestuniversity-based groups in the EUdedicated to research, consultancy andeducation in post-harvest technology offresh produce.

Professor Jim Anderson is Head of Chemistry (Research) at theUniversity of Aberdeen, U.K. His main interests lie in the areas ofsupported metal catalysis and mixed oxide-based acid catalysis andin the use of infrared spectroscopy for site quantification andinterpretation of selectivity effects in catalysis.

IntroductionThermodynamic properties of diatomic transi-

tion metal compounds are very important forinvestigating their thermal behaviour. Recently,these properties have been applied in the fabrica-tion of smart devices using intelligent materials (1)(see (2, 3) for examples using platinum). Transitionmetal compounds have a wide range of actual andpotential applications in materials science becauseof their relatively high melting points, moderatedensities, and resistance to chemical attack.Platinum-containing compounds have been used innanoscience and nanotechnology. For example,alloys such as iron-platinum (FePt), are used as nan-odots (4). Platinum itself is used in a wide range ofapplications, including catalytic converters for cars,fuel cell electrocatalysts, computer technology,optical communication, missile technology, neuro-surgery and medical science (see for example (5)).

The unique properties of platinum generate a highlevel of interest among scientists. Due to the presenceof unpaired d electrons, which have a greater angularmomentum than s and p electrons, more energy isneeded for the excitation of Pt in the molecularphase. This requires a high energy excitation devicesuch as a gas discharge laser, which leads to experi-mental difficulties. Thus the experimental study ofplatinum diatomics is very challenging and expensive.

Scientific groups such as the Scientific GroupThermodata Europe (SGTE), the Joint Institute of

High Temperatures, Russian Academy of Sciences(IVTAN) and, in the U.S.A., the Joint Army-Navy-Air Force (JANAF) Thermochemical WorkingGroup (6) and the National Aeronautics and SpaceAdministration (NASA) (7, 8), are engaged in thecritical assessment and compilation of thermody-namic data for different molecular species.

An early contribution to the development ofthermodynamic properties was made by Tolman, fordiatomic hydrogen (9). The credit for furtherdevelopment of the subject goes to Hicks andMitchell, for their work on hydrogen chloride (10).Giauque and Overstreet (11) implemented thetechnique suggested by Hicks and Mitchell for thecalculation of these properties for hydrogen, chlorineand hydrogen chloride. They modified the reportedtheory by using stretching and interaction termsfor diatomic molecules. Gordon and Barnes (12)calculated the thermodynamic properties of chlorine(Cl2), bromine (Br2), hydrogen chloride (HCl), carbonmonoxide (CO), oxygen (O2) and nitric oxide (NO)molecules. The study of thermodynamic propertiesof the phosphorous mononitride (PN) molecule wasperformed by McCallum and Liefer (13). Thethermodynamic properties of transition metal alloyswere reported by Darby (14). Calculation of thepartition function and thermodynamic properties ofthe rare gas atoms argon, krypton and xenon wasperformed by Elyutin et al. (15). In the domain oftheory of thermodynamic properties, Eu (16)

123Platinum Metals Rev., 2009, 53, (3), 123–134

Thermodynamic Properties ofPlatinum DiatomicsPROPERTIES OF PtH, PtC, PtN AND PtO CALCULATED FROM SPECTROSCOPIC DATA

By Pavitra Tandon* and K. N. Uttam** Saha’s Spectroscopy Laboratory, Department of Physics, University of Allahabad, Allahabad-211 002, India;

E-mail: *[email protected]; **[email protected]

Thermodynamic properties of diatomic molecules containing platinum (PtH, PtC, PtN andPtO) have been calculated using spectroscopic data and partition function theory. Values ofthe Gibbs energy (G), enthalpy (H), entropy (S) and specific heat capacity at constant pressure(CP) are presented for each species in the temperature range from 100 K to 3000 K. Toobtain the most accurate data, anharmonicity, nonrigidity and stretching effects have beenincorporated in the calculations. The variation of these properties with temperature is alsodiscussed in terms of different modes of molecular motion.

DOI: 10.1595/147106709X463688

reported Boltzmann entropy, relative entropy andtheir related values. Chandra and Sharma (17)calculated the partition function for carbonmonosulfide (CS) and silicon monoxide (SiO)molecules. Recently Uttam and coworkers (18, 19)estimated the thermodynamic properties ofpotassium monohalides and alkaline earth metalmonohydrides using partition function theory.

The data on thermodynamic properties havebeen reported for a large number of molecules but asurvey of the literature reveals that the values ofthermodynamic properties for some diatomic mole-cules are not yet reported accurately. Therefore wehave estimated the thermodynamic properties ofplatinum monohydride (PtH), platinum monocar-bide (PtC), platinum mononitride (PtN) andplatinum monoxide (PtO) molecules using spectro-scopic data and partition function theory. Thechoice of the temperature range from 100 K to3000 K is due to the fact that this range of temper-atures covers the applications of platinum frombiological sciences to high-temperature chemistryand astrophysics. In the present paper, we report thevalues of thermodynamic properties Gibbs energy(G ), enthalpy (H ), entropy (S ) and specific heatcapacity at constant pressure (CP) for PtH, PtC, PtNand PtO molecules at different temperatures forwhich these properties are not given in the literature.

Method of CalculationA diatomic molecule is associated with transla-

tional, rotational, vibrational and electronicmotions. Corresponding to these four types ofmotions, there are four types of energy: translation-al, rotational, vibrational and electronic energy.Translational motion is due to the three dimension-al movement of a molecule in space. Rotationalmotion is due to the rotation of the molecule as awhole about an axis passing through the centre ofgravity and perpendicular to the internuclear axis.In diatomic molecules, atoms are also able tovibrate relative to each other along the internuclearaxis and this is the origin of vibrational motion.

Motion of electrons in one atom is perturbed byelectronic and nuclear motion in the other atom.Due to this, reshuffling of orbitals takes place andthat generates molecular orbitals. This phenome-

non is responsible for electronic motion. The elec-tronic energy, ~ 1 eV to 10 eV, is very highcompared to vibrational energy, ~ 10–2 eV, rota-tional energy, ~ 10–3 eV, and translational energy,~ 10–22 eV. However, theory shows that below3000 K molecules are not excited electronically,and electronic motion only plays a significant roleabove 3000 K. Therefore electronic motion can beneglected below 3000 K.

The contributions of the different motions aresummarised on the following page.

Results and DiscussionThe calculated thermodynamic properties

namely Gibbs energy, enthalpy, entropy and heatcapacity at constant pressure, of PtH, PtC, PtNand PtO molecular gases have been estimatedfrom spectroscopic data and are collected inTables I–IV. The spectroscopic constants whichwere used for the calculation of these propertiesare displayed in Table V (20). PtH is different fromPtC, PtN and PtO since it has a 2Δ ground state.Therefore we have incorporated the ground statemultiplicity in our calculation for the improvementof the results. We have estimated the thermody-namic properties from theoretical spectroscopicdata (21) and experimental spectroscopic data (20)for the PtH molecule as shown in Table VI.

From comparison, it is clear that the Gibbsenergy has a maximum deviation of 0.12%,enthalpy has 0.11%, entropy has 0.17% and heatcapacity has 0.63% up to 2000 K. Ideally it isassumed that rotational and vibrational motions areindependent of each other, but in practice theyinteract with each other. In the present calculationwe include this effect by taking the vibrational-rota-tional partition function instead of the independentrotational and vibrational partition functions. Thisgives more accurate values of thermodynamicproperties than the values obtained from individualrotational and vibrational partition functions. Asimilar approach has been applied for the calcula-tion of thermodynamic properties of monohalidesof potassium (18), and the obtained results were inclose agreement with reported values. Accuracy ofthe data also depends on the vibrational-rotationalcoupling. If coupling is weak, the stretching con-


stant (αe) is sufficient for the calculation of thermo-dynamic properties. If coupling is strong, theincorporation of the vibration-rotation constant(γe) gives more accurate data.

It is clear that the specific heat capacity at con-stant pressure increases with temperature at lowertemperatures, but at higher temperatures thisbecomes constant while Gibbs energy, enthalpy


Contributions of the Molecular Motions to the Thermodynamic Properties

Equations for the Translational Contribution of Thermodynamic Properties

(i) Gibbs energy:

(ii) Enthalpy:

(A)

(iii) Entropy:

(iv) Heat capacity at constant pressure:

Equations for the Rotational Contribution of Thermodynamic Properties

(v) Gibbs energy:

(vi) Enthalpy: (B)

(vii) Entropy:

(viii) Heat capacity at constant pressure:

Equations for the Vibrational Contribution of Thermodynamic Properties

(ix) Gibbs energy:

(x) Enthalpy:

(C)

(xi) Entropy:

(xii) Heat capacity at constant pressure:

Equations for the Vibrational-Rotational Contribution of Thermodynamic Properties

(xiii) Gibbs energy:

(xiv) Enthalpy:

(D)

(xv) Entropy:

(xvi) Heat capacity at constant pressure:

( ) TTRTmRTHG tran 2836.7ln2

5ln

2

3

0+−−=−

( ) RTHH tran2

5

0=−

( )THG

RS trantran

0

2

5 −+=

RC tranP2

5

)(=

Note: The derivation of these equations is given in the Appendix to this paper

( ) yRTHG rot σln0

=−

( ) RTHH rot =−0

( )yRSrot σln1−=

RC rotP =)(

( ) ( )yvib eRTHG −−=− 1

0

( ) ( )10 −

=− yvib eyRTHH

( )⎥⎦

⎤⎢⎣

⎡−−⎟

⎠⎞

⎜⎝⎛

−= − y

yvib ee

yRS 1ln1

( )22

)(

1−=

y

y

vibPe

eRyC

( ) xRZRTHG vrvr lnln0

−−=−

( ) ( ) xRZT

RTHH vr lnln2

0+

∂∂

=−

( )[ ] ( )⎥⎦⎤

⎢⎣⎡

∂∂

++= vrvrvr ZT

RTZRS lnln1

( )ZT

RTC vrP ln)( ∂

∂=

( )2

2

)(

1−= y

y

vibP eeRyC

(⎢⎣

⎡−−⎟

⎠⎞

⎜⎝⎛

−= −

1ln1

yyvib e

eyRS )

( )[ ] ( )⎢⎣⎡

∂∂

++= ln ln1 vrvrvr ZT

RTZRS

and entropy continue to increase. The heat capac-ity of any system represents its capacity to containheat. At lower temperatures only translationalmotion contributes to this, while as temperature isincreased, both rotational and vibrational motionsare excited simultaneously and this increases theheat capacity. Beyond a certain temperature, thereis no further increase in degrees of molecularmotion and this explains the constant value of CP.

At the lowest temperature included in this study,100 K, the CP should be very close to the theoret-ical limit of 3.5R, or 29.101 J K–1 mol–1 (whereR = 8.31457 J K–1 mol–1). In the present case CP isless than but very close to 29.101 J K–1 mol–1. Thismight be due to the fact that quantum mechanicshas been used to describe translational motion.This means that the energy of a molecule is con-sidered to be non-zero even at absolute zero.


Table I

Calculated Thermodynamic Properties of Platinum Monohydride (PtH) Molecule at 1 Atm

Temperature, Gibbs energy, G, Enthalpy, H, Entropy, S, Specific heat T, K kJ mol–1 kJ mol–1 J mol–1 K–1 capacity at

constant pressure,CP, J mol–1 K–1

100 17.57 2.08 182.03 29.06

200 36.60 5.29 201.92 29.07

300 57.14 8.52 213.62 29.10

400 78.67 11.18 221.97 29.26

500 100.94 14.33 228.52 29.66

600 123.81 17.31 233.97 30.26

700 147.18 20.25 238.69 30.95

800 171.00 23.19 242.87 31.65

900 195.21 26.13 246.65 32.31

1000 219.78 29.06 250.10 32.89

1100 244.69 32.00 253.27 33.41

1200 269.89 34.95 256.21 33.86

1300 295.38 37.91 258.96 34.25

1400 321.13 40.87 261.53 34.58

1500 347.13 43.84 263.95 34.87

1600 373.36 46.81 266.23 35.11

1700 399.80 49.80 268.39 35.33

1800 426.47 52.80 270.43 35.52

1900 453.33 55.80 272.38 35.68

2000 480.38 58.81 274.24 35.83

2100 507.60 61.84 276.02 35.95

2200 535.00 64.87 277.72 36.06

2300 563.19 66.86 279.62 36.16

2400 590.98 69.77 281.21 36.24

2500 618.93 72.67 282.73 36.33

2600 647.02 75.58 283.63 36.40

2700 675.27 78.49 286.00 36.46

2800 703.65 81.40 287.00 36.51

2900 732.16 84.31 288.33 36.56

3000 760.81 87.21 289.61 36.61

The enthalpy of an ideal gas depends on tem-perature and its value increases with temperature.Entropy is a measure of the molecular disorder ofa system, and molecules in a system at high tem-perature are highly disorganised either in terms oftheir locations or in terms of occupation of theiravailable translational, rotational and vibrationalenergy states. In contrast, molecules at low tem-perature have less disorder, and thus have a lower

entropy. Gibbs energy similarly increases withincreasing temperature.

ConclusionsThe experimental study of platinum-containing

diatomics is expensive and difficult. For the firsttime, accurate values of thermodynamic propertiesare reported here for the diatomic molecules PtH,PtC, PtN and PtO. Gibbs energy (G ), enthalpy (H ),


Table II

Calculated Thermodynamic Properties of Platinum Monocarbide (PtC) Molecule at 1 Atm



100 17.22 2.08 201.27 29.07

200 38.47 5.57 221.44 29.33

300 61.25 8.63 233.53 30.50

400 85.07 11.55 242.51 31.96

500 109.69 14.49 249.78 33.20

600 134.98 17.41 255.93 34.14

700 160.84 20.34 261.93 34.82

800 187.20 23.26 265.95 35.32

900 214.00 26.19 270.14 35.70

1000 241.20 29.12 273.93 35.98

1100 268.76 32.05 277.38 36.20

1200 296.65 34.98 280.55 36.38

1300 324.84 37.93 283.48 36.52

1400 353.32 40.88 286.20 36.63

1500 382.05 43.83 288.75 36.72

1600 411.03 46.79 291.13 36.80

1700 440.24 49.74 293.37 36.86

1800 469.67 52.71 295.49 36.91

1900 499.31 55.70 297.50 36.96

2000 529.13 58.65 299.41 37.00

2100 559.15 61.63 301.23 37.03

2200 589.33 64.61 302.96 37.06

2300 620.11 66.86 304.80 37.09

2400 650.67 69.77 306.40 37.11

2500 681.38 72.68 307.94 37.13

2600 712.25 75.59 309.42 37.15

2700 743.27 78.50 310.84 37.16

2800 774.42 81.40 312.21 37.18

2900 885.71 84.31 313.54 37.19

3000 837.13 87.22 314.82 37.20

entropy (S ) and specific heat capacity (CP) are tabu-lated for each species over a range of temperaturesfrom 100 K to 3000 K, calculated from spectro-scopic data. The utility of this method lies in thefact that it allows the properties of chemically unsta-ble molecules to be predicted. Due to the use ofspectroscopic data, the results are highly accurate.The calculated results presented here will be usefulto experimental workers in various disciplines, from

biological applications to high-temperature chem-istry. This work brings out the usefulness of theresults of spectroscopic data in studying thermody-namic properties using statistical mechanics.

AcknowledgementsThe authors are thankful to Professor Pradip

Kumar, Head of the Department of Physics at theUniversity of Allahabad for his keen interest in this


Table III

Calculated Thermodynamic Properties of Platinum Mononitride (PtN) Molecule at 1 Atm



100 19.86 2.08 205.15 29.07

200 41.25 5.59 225.36 29.52

300 64.18 8.64 237.59 30.99

400 88.16 11.56 246.73 32.55

500 112.96 14.50 254.14 33.76

600 138.44 17.42 260.39 34.61

700 164.50 20.35 265.80 35.23

800 191.06 23.27 270.53 35.66

900 218.07 26.21 274.80 35.99

1000 245.40 29.14 278.59 36.23

1100 273.26 32.09 282.06 36.41

1200 301.36 35.03 285.25 36.56

1300 329.77 37.99 288.20 36.67

1400 358.47 40.95 290.93 36.76

1500 387.42 43.91 293.49 36.84

1600 416.62 46.88 295.88 36.90

1700 446.06 49.86 298.13 36.95

1800 475.71 52.84 300.26 37.00

1900 505.56 55.82 302.28 37.03

2000 535.61 58.81 304.19 37.06

2100 565.85 61.81 306.01 37.09

2200 596.26 64.81 307.75 37.12

2300 627.36 66.77 309.65 37.14

2400 658.16 69.77 311.26 37.16

2500 689.11 72.68 312.80 37.17

2600 720.22 75.59 314.28 37.19

2700 751.47 78.50 315.71 37.20

2800 782.86 81.40 317.09 37.21

2900 814.38 84.31 318.42 37.22

3000 846.04 87.22 319.71 37.23

work. Pavitra Tandon is thankful to the UniversityGrants Commission (UGC), New Delhi, for finan-cial assistance in the form of a fellowship.

References 1 “Materials Research: Current Scenario and Future

Projections”, eds. R. Chidambaram and S. Banerjee,Allied Publishers Pvt Ltd, New Delhi, India, 2003

2 K. Sadeghipour, R. Salomon and S. Neogi, Smart Mater.Struct., 1992, 1, (2), 172


Table IV

Calculated Thermodynamic Properties of Platinum Monoxide (PtO) Molecule at 1 Atm



100 17.52 2.08 204.23 29.08

200 39.06 5.61 224.50 29.79

300 62.17 8.65 236.91 31.55

400 86.34 11.57 246.22 33.15

500 111.35 14.50 253.76 34.28

600 137.04 17.43 260.10 35.05

700 163.32 20.35 265.56 35.59

800 190.12 23.29 270.35 35.96

900 217.36 26.22 274.62 36.23

1000 245.00 29.17 278.47 36.43

1100 273.02 32.12 281.96 36.59

1200 301.36 35.08 285.17 36.71

1300 330.01 38.04 288.13 36.80

1400 358.94 41.01 290.88 36.88

1500 388.14 43.99 293.44 36.94

1600 417.59 46.97 295.84 36.99

1700 447.26 49.96 298.10 37.03

1800 477.15 52.96 300.23 37.07

1900 507.25 55.96 302.26 37.10

2000 537.54 58.97 304.17 37.12

2100 568.03 61.98 306.00 37.14

2200 598.68 63.96 307.75 37.16

2300 630.12 66.87 309.68 37.18

2400 661.17 69.77 311.29 37.19

2500 692.38 72.59 312.84 37.21

2600 723.74 75.59 314.33 37.22

2700 755.24 78.50 315.77 37.23

2800 786.89 81.40 317.15 37.23

2900 818.67 84.31 318.49 37.24

3000 850.58 87.22 319.78 37.25

3 P. Kotzian, P. Brázdilova, K. Kalcher, K. Handlír and K.Vytras, Sens. Actuators B: Chem., 2007, 124, (2), 297

4 Z. Gai, J. Y. Howe, J. Guo, D. A. Blom, E. W. Plummerand J. Shen, Appl. Phys. Lett., 2005, 86, (2), 023107

5 “CRC Handbook of Chemistry and Physics”, 89thEdn., ed. D. R. Lide, CRC Press, Boca Raton, Florida,U.S.A., 2008

6 “NIST-JANAF Thermochemical Tables”, 4th Edn.,ed. M. W. Chase, Journal of Physical and ChemicalReference Data Monographs, Vol. 9, AmericanInstitute of Physics, Melville, New York, U.S.A., 1998,Part I and Part II


7 S. Gordon and B. J. McBride, “Computer Programfor Calculation of Complex Chemical EquilibriumCompositions and Applications I. Analysis”, NASAReference Publication 1311, NASA Glenn ResearchCenter, Cleveland, Ohio, U.S.A., October, 1994

8 B. J. McBride and S. Gordon, “Computer Programfor Calculation of Complex Chemical EquilibriumCompositions and Applications II. Users Manual and

Program Description”, NASA Reference Publication1311, NASA Glenn Research Center, Cleveland, Ohio,U.S.A., June, 1996

9 R. C. Tolman, Phys. Rev., 1923, 22, (5), 47010 H. C. Hicks and A. C. G. Mitchell, J. Am. Chem. Soc.,

1926, 48, (6), 152011 W. F. Giauque and R. Overstreet, J. Am. Chem. Soc.,

1932, 54, (5), 1731

Table V

Spectroscopic Constants of Platinum Diatomics

Molecules Molecular Vibrational Anharmonicity Rotational Stretchingweight, u constant, constant, constant, constant,

ωe, cm–1 ωexe, cm–1 Be, cm–1 αe, cm–1

PtH 196 2294.68 46.00 7.196300 0.1996000

PtC 207 1051.13 4.86 0.530440 0.0032730

PtN 209 947.00 5.00 0.455708 0.0034481

PtO 211 851.11 4.98 0.382240 0.0028300

Table VI

Comparison between Thermodynamic Properties Obtained for PtH from Theoretical Spectroscopic Dataand Experimental Spectroscopic Data at 1 Atm

From Theoretical Spectroscopic DataTemperature, Gibbs energy, G, Enthalpy, H, Entropy, S, Specific heat

T, K kJ mol–1 kJ mol–1 J mol–1 K–1 capacity at constant pressure,CP, J mol–1 K–1

100 17.59 2.08 182.20 29.06

500 101.03 14.33 228.73 29.77

1000 220.01 29.07 250.43 33.10

1500 347.55 43.87 264.37 35.01

2000 481.00 58.88 274.70 35.92

2500 618.99 74.15 282.91 36.39

3000 760.64 89.66 289.70 36.65

From Experimental Spectroscopic DataTemperature, Gibbs energy, G, Enthalpy, H, Entropy, S, Specific heat

T, K kJ mol–1 kJ mol–1 J mol–1 K–1 capacity at constant pressure,CP, J mol–1 K–1

100 17.57 2.08 182.03 29.06

500 100.94 14.33 228.52 29.66

1000 219.78 29.06 250.10 32.89

1500 347.13 43.84 263.95 34.87

2000 480.38 58.81 274.24 35.83

2500 618.93 72.67 282.73 36.33

3000 760.81 87.21 289.70 36.61


AppendixMethod of Calculation for the Individual Partition Functions

According to the non-rigid rotator model, the rotational energy of a diatomic molecule is given byEquation (xvii):

(xvii)

where Erot is the rotational energy, J is the rotational quantum number, h is Planck’s constant, c is thevelocity of light, Be is the rotational constant and De is the centrifugal constant.

The expression for vibrational energy, Evib, according to the anharmonic oscillator model is given inEquation (xviii):

(xviii)

where ωe is the vibrational constant, ωe xe is the anharmonicity constant and v is the vibrational quantumnumber.

Total energy of a diatomic molecule can be written as Equation (xix):

Et = Erot + Evib + Eele + Etran (xix)

where Et is total energy, Erot is rotational energy, Evib is vibrational energy, Eele is electronic energy and Etran

is translational energy.The partition function, Z, contains all the relevant information of any physical system. It is approxi-

mately equal to the number of quantum states having energies below the thermal energy available to themolecule in the given volume. The partition function can be expressed as Equation (xx) (22, 23):

(xx)

where gi is the degeneracy of the ith energy level, Ei is the energy of the ith level, kB is the Boltzmann con-stant, T is absolute temperature in Kelvin and i ranges over all quantum states.

Four types of energy give rise to four types of partition function. Thus the total partition function ofthe system, Zt, can be written as Equation (xxi):

Zt = ZtranZrotZvibZele (xxi)

where Ztran, Zrot, Zvib and Zele represent the partition functions for translational energy, rotational energy,vibrational energy and electronic energy respectively.

The contributions of the individual partition functions to the thermodynamic properties can be calcu-lated separately and are given in the following sections.

Translational Partition Function

Translational energy for a particle of mass m in a cuboid of sides p, q and r can be written as Equation(xxii):

( )[ ]hcJJDJJBE eerot22

)1(1 +−+=

hcvxvE eeevib⎥⎥⎦

⎤

⎢⎢⎣

⎡⎟⎠⎞

⎜⎝⎛ +−⎟

⎠⎞

⎜⎝⎛ +=

2

2

1

2

1ωω

TkE

ii B

i

egZ −∑=

12 A. R. Gordon and C. Barnes, J. Chem. Phys., 1933,1, (5), 297

13 K. J. McCallum and E. Liefer, J. Chem. Phys., 1940, 8,(6), 505

14 J. B. Darby, Ind. Eng. Chem., 1968, 60, (5), 2815 P. V. Elyutin, V. I. Baranov, E. D. Belega and D. N.

Trubnikov, J. Chem. Phys., 1994, 100, (5), 384316 B. C. Eu, J. Chem. Phys., 1995, 102, (18), 716917 S. Chandra and A. K. Sharma, Indian J. Pure Appl. Phys.,

1995, 33, (6), 34518 K. N. Uttam, S. Kumar, R. Singh and P. Tandon, Proc.

Natl. Acad. Sci., India, Sect. A: Phys. Sci., 2008, 78, (3), 24919 P. Tandon, R. Singh, S. Tiwari, S. Srivastava, S. Kumar

and K. N. Uttam, Vijnana Parishad Anusandhan Patrika,2008, 51, (3), 209

20 K. P. Huber and G. Herzberg, “Molecular Spectra andMolecular Structure. IV. Constants of Diatomic Molecules”,Van Nostrand Reinhold Co, New York, U.S.A., 1979

21 W. Liu and R. Franke, J. Comput. Chem., 2002, 23, (5), 56422 N. Davison, “Statistical Mechanics”, McGraw-Hill

Book Co, Inc, New York, U.S.A., 196223 K. S. Pitzer, “Quantum Chemistry”, Prentice-Hall,

Inc, New York, U.S.A., 1954

vxvE eeevib⎢⎢⎣

⎡⎟⎠⎞

⎜⎝⎛ +−⎟

⎠⎞

⎜⎝⎛ +=

2

2

1

2

1ωω hc


(xxii)

where nx, ny and nz are the various quantum states.

The translational partition function corresponding to this energy can be written as Equation (xxiii):

(xxiii)

Here we take an approximation and change this summation into integration to obtain Equation (xxiv):

(xxiv)

Using the standard result of integration, we get Equation (xxv):

(xxv)

Using V = pqr where V is the volume of the enclosure gives Equation (xxvi):

(xxvi)

Dividing both sides of this expression by the number of particles, N, we obtain Equation (xxvii):

(xxvii)

We consider Ztran/N instead of Ztran as the quantity Ztran/N is an intensive quantity and depends on the concentration, N/V, whereas Ztran itself is proportional to V and independent of the number of particles. Using the ideal gas equation, Equation (xxvii) can be written as Equation (xxviii):

(xxviii)

where P0 denotes one atmospheric pressure (1.01325 × 105 N m–2). Using the above partition function, the translational contribution of the various thermodynamic

properties can be estimated by the set of Equations (A) (22), where H0 represents enthalpy at a tempera-ture of 298.15 K, and R is the gas constant.

Rotational Partition Function

Partition function theory shows that the rotational partition function can be written as Equation (xxix):

(xxix)

where y is a constant given by Equation (xxx):

(xxx)

and σ is the symmetry number which has a value of 2 for a homonuclear molecule and 1 for a heteronu-clear molecule. The symmetry number is defined as the number of regions of phase space included in thepartition function calculation, which differ only by the exchange of indistinguishable particles. In quan-tum theory the symmetry number arises from the fact that only a fraction of the rotational states areallowed in a symmetrical molecule and this fraction is 1/σ.

If y is small, the summation changes into integration and we have Equations (xxxi) to (xxxiii):

(xxxi)

(xxxii)

∑∑∑⎟⎟

⎠

⎞

⎜⎜

⎝

⎛++−

=x y z

zyx

B

n n n

rn

q

n

pn

Tmkh

tran eZ2

2

2

2

2

22

8

∫∫∫⎟⎟⎠

⎞⎜⎜⎝

⎛−⎟

⎟

⎠

⎞

⎜⎜

⎝

⎛−

⎟⎟⎠

⎞⎜⎜⎝

⎛−

= zrn

Tmkh

yq

nTmk

h

xpn

Tmkh

tran dnednedneZz

B

y

B

x

B2

22

2

22

2

22

888

⎥⎥

⎦

⎤

⎢⎢

⎣

⎡⎟⎠⎞

⎜⎝⎛

⎟⎠⎞

⎜⎝⎛

⎟⎠⎞

⎜⎝⎛=

2

1

2

2

1

2

2

1

2

8

2.

8

2.

8

2 hTmkr

hTmkq

hTmkpZ BBB

tranπππ

2

3

2

2⎟⎠⎞

⎜⎝⎛=

hTmkVZ B

tranπ

NV

hTmk

NZ Btran

2

3

2

2⎟⎠⎞

⎜⎝⎛=

π

0

2

3

2

2

PTk

hTmk

NZ BBtran ⎟

⎠⎞

⎜⎝⎛=

π

( ) ( )∑ +−+=J

JyJrot eJ

yZ 1

121

σ

TBy e

7994.4=

( ) ( )dyeJZ JyJrot

112

1 +−+∫=σ

yZrot σ

1=

⎟⎟⎠

⎞⎜⎜⎝

⎛++=

2

2

2

2

2

22

8 rn

qn

pn

mhE zyx

tran

⎢⎢

⎣

⎡⎟⎠⎞

⎜⎝⎛

⎟⎠⎞

⎜⎝⎛

⎟⎠⎞

⎜⎝⎛=

2

1

2

2

1

2

2

1

2

8

2.

8

2.

8

2 hTmkr

hTmkq

hTmkpZ BBB

tranπππ

( ) ( )∑ +−+=J

JyJrot eJ

yZ 1

121

σ

( ) ( )dyeJZ JyJrot

112

1 +−+∫=σ

(xxxiii)

Using the Euler-Maclaurin formula, we get Equation (xxxiv):

(xxxiv)

Using this partition function, the rotational contribution in the thermodynamic properties may beobtained from the relations in Equation set (B) (22).

Vibrational Partition Function

For a one-dimensional harmonic oscillator, energy levels are given by Equation (xxxv):

(xxxv)

where f is the frequency of molecular vibration given by Equation (xxxvi):

(xxxvi)

in which k is the force constant and μ is the reduced mass of the molecule, and v = 0, 1, 2, 3 etc. For v = 0, the oscillator has zero point energy, E0, which from Equation (xxxv) is equivalent to hf/2.

It is convenient to subtract the zero energy state (v = 0), except for problems involving isotopic molecules. Then we have Equation (xxxvii):

Ev – E0 = hfv = hcωev (xxxvii)

Using Equation (xxxviii):

(xxxviii)

in which y is defined by Equation (xxxix):

(xxxix)

Theory shows that the vibrational partition function can then be written as Equation (xl):

(xl)

From which we obtain Equation (xli):

(xli)

The vibrational contribution of the thermodynamic properties is derived using this partition functionby the expressions in Equation set (C) (22).

Vibrational-Rotational Partition Function

If we assume that the rotational and vibrational motions do not interact with each other, the total partition function will be the multiple of the rotational, vibrational and translational partition functions.But in real cases, both types of motion do affect each other and stretching and anharmonicity also needto be considered. Then total internal energy can be given as Equation (xlii):

(xlii)

Using Equations (xliii), (xliv) and (xlv):

(xliii)

(xliv)

⎟⎟⎠

⎞⎜⎜⎝

⎛+++= .............

1531

12yy

yZrot σ

⎟⎠⎞

⎜⎝⎛ +=

2

1vhfEvib

2

1

2

1

⎟⎟⎠

⎞⎜⎜⎝

⎛=

μπkf

yvTkEE

B

v =−

0

TB

Tkhfy e

B

7994.4==

∑ −=v

yvvib eZ

( )yvvib eZ −−

=1

1


( ) ( ) ( )12

111

2

1 22

0+⎟

⎠⎞

⎜⎝⎛ +−+−++⎟

⎠⎞

⎜⎝⎛ +−= JJvJJDJJBvxE eeeee αωνω

2

eeo BB α

−=

eee xωωω 20

−=

erot B

TZσ7994.4

=


(xlv)

where x represents the multiplicity of the ground state, αe is the stretching constant, B0 is the ground staterotational constant and ω0 is the actual separation of the first two vibrational levels, and substituting intothe energy expression in Equation (xlii) this can then be written as Equation (xlvi):

(xlvi)

For simplification, we further substitute Equations (xlvii), (xlviii), (xlix) and (l):

(xlvii)

(xlviii)

(xlix)

(l)

Then we have Equation (li):

(li)

Corresponding to this energy, the vibrational-rotational partition function, Zvr, be simplified as Equation(lii) (16, 17):

(lii)

Or Equation (liii):

Zvr = ZIZC (liii)

where ZI represents the ideal partition function which is valid for the rigid rotator harmonic oscillatormodel and ZC represents the corrected partition function and gives the contribution due to centrifugalstretching, nonrigidity and anharmonicity of the diatomic molecule. Expressions for ZI and ZC can be written as Equations (liv) and (lv):

(liv)

(lv)

This partition function can be used to estimate the contribution of vibrational-rotational motion to theGibbs energy, enthalpy, entropy and heat capacity by the formulae in Equation set (D) (22).

Finally, the thermodynamic properties like Gibbs energy, enthalpy, entropy and heat capacity of platinum molecular gases are calculated by adding the translational contribution and the vibrational-rotational contribution.

])1()[1()1(000

vJJDBJJvxvvE ee αωω −+−++−−=

Tkhcu

B

0ω

=

TkhcBa

B

0=

0BDe=β

0B

eαδ =

[ ] ])1(1[)1()1(1 δβ vJJaJJxvuvE −+−++−−=

⎥⎦

⎤⎢⎣

⎡++

−+

−++

−= −

153)1(

2

1

21

)1(

12

2

aae

xueaea

Z uuuvrδβ

σ

)1(

1

uI eaZ −−

=σ

⎥⎦

⎤⎢⎣

⎡++

−+

−++=

153)1(

2

1

21

2

2

aae

xuea

Z uuCδβ

The AuthorsDr K. N. Uttam is a Lecturer in theDepartment of Physics, University ofAllahabad, Allahabad, India. His majorresearch is in the field of molecularspectroscopy of transition metals andrare earth molecules. In addition he isalso engaged in the field ofthermodynamical properties ofmolecular gases as well asenvironmental nanotechnology.

Ms Pavitra Tandon is pursuing herdoctoral degree (D.Phil.) in theDepartment of Physics, University ofAllahabad, India. Her current area ofresearch includes molecularspectroscopy and thermodynamicalproperties of transition metals and rareearth molecules.

ee xx ωω =0

⎢⎣

⎡++

−+

−++

−= −

153)1(

2

1

21

)1(

12

2

aae

xueaea

Z uuuvrδβ

σ

⎢⎣

⎡++

−+

−++=

153)1(

2

1

21

2

2

aae

xuea

Z uuCδβ

Elemental carbon is known in a variety offorms that range from diamond gemstonesthrough C60 fullerene and graphite to carbon nanotubes, but to those working with catalysts themost familiar are probably high surface area formsknown as ‘activated charcoals’. The very high surface areas of porous charcoals made from woodand peat, as well as those derived from more exot-ic materials such as coconut husks, are well suitedas catalyst supports and each have specific desirable properties. They are used widely to prepare platinum group metal (pgm) catalysts (1)that typically contain only a few per cent of pgmyet have very high activities at low temperatures.This is because, not only do pgms have high intrin-sic catalytic activity, but they can also be very finelydispersed over activated charcoals, as extremelysmall crystallites that provide high surface areas ofthe active metal. These catalysts, and especiallythose containing palladium, have been used exten-sively for many years in organic preparativechemistry (2), and today they have important rolesin hydrogenation and hydrogenolysis reactions.They even find application in some carbon–carbonbond forming processes (3) and potentially also indirect carbonylation reactions (4). In spite of theirimportance in pgm-based catalysts, carbon sup-ports are less well known in base metal catalysts.One reason for this is that, over time, there can bea tendency for metals like cobalt, iron or nickel toform carbides, especially at higher temperatures,and this leads to a loss of catalytic activity.

The present small book, as its title suggests, isabout the catalytic activity of carbons (and espe-cially activated charcoals) and metal compoundssupported on carbons, with a focus on catalytichydroprocessing reactions of crude oils. These

reactions include hydrogenation, hydrocracking,hydrodesulfurisation, hydrodenitrogenation, andhydrodemetallation processes. Huge quantities ofhydroprocessing catalysts are employed in refiner-ies, and the most common ones are sulfidedcobalt/molybdenum and nickel/molybdenum formulations, and to a much lesser extent theirtungsten analogues, supported on aluminas withvarious promoters. Reflecting the importance ofvery low sulfur hydrocarbon fuels, there is a vastliterature on hydroprocessing catalysts, and espe-cially on the nature of the active sulfide phasespresent in the operating catalysts (5, 6).

Carbon-Supported HydroprocessingCatalysts

After a short general introductory chapter outlining the basic structures of graphite, carbonblack, diamond, activated carbons, C60 fullereneand carbon nanotubes, there is another short chap-ter entitled ‘Industrial Carbons’ that providesdetails of the physical properties of each of the car-bon forms. The next chapter briefly discusses thestructure and composition of traditional hydropro-cessing catalysts and the cobalt/molybdenumphases involved in hydrodesulfurisation, and contrasts this with what is known about their carbon-supported counterparts. Here, themetal–carbon interactions are likely to be weak-ened by the presence of metal–sulfur bonds, andwhile this might be thought to inhibit deactivationvia carbide formation, it could well open up otherdeactivation paths such as loss of surface area. Theinfluence of physical properties like support poresize is also discussed.

The fourth chapter is concerned with the ability of carbons to absorb, activate and transfer

Platinum Metals Rev., 2009, 53, (3), 135–137 135

“Carbons and Carbon Supported Catalystsin Hydroprocessing”BY EDWARD FURIMSKY (IMAF Group, Ottawa, Ontario, Canada), Royal Society of Chemistry, Cambridge, U.K., 2008, 176 pages,

ISBN 978-0-85404-143-5, £95.00

Reviewed by Martyn V. TwiggJohnson Matthey PLC, Orchard Road, Royston, Hertfordshire SG8 5HE, U.K.; E-mail: [email protected]

DOI: 10.1595/147106709X465479

active hydrogen via C–H bonds at high tempera-tures. At the lower temperatures normally used forcarbon-supported pgm catalysts, these reactionsmust be much less important than the activation ofhydrogen by the supported pgms. Indeed in thepresence of platinum, palladium or rhodium, dis-sociative hydrogen chemisorption takes place onthe metal, followed by spillover onto the carbonsurface to form weakly bonded mobile H atoms.Most of the cited examples are systems based oncobalt/molybdenum formulations, although otherinteresting catalysts are referred to.

The following chapter, entitled ‘CatalyticActivity of Carbons’, begins with a comparison ofcoking tendencies of cobalt/molybdenum speciessupported on alumina and supported on activatedcharcoal. In use, activated charcoal-based catalystshave a very much smaller rate of coke formationfrom anthracene than do alumina-based catalysts,and this difference is attributed to the relative acid-ity of the supports – the alumina being acidic whilethe carbon is said to be neutral. However, it is clearthat such benefits in reactions with real feeds arenot always apparent, and this may at least in part bebecause the acid/base behaviour of activated char-coals can vary enormously depending on theirorigin and the treatments they have undergone.

The sixth chapter, ‘Carbon SupportedCatalysts’, is concerned with the preparation ofcarbon supported hydroprocessing catalysts, and itcovers most of the conventional preparative methods. Techniques used to characterise thesecarbon-supported catalysts are also discussed, andthey include a range of spectroscopic and adsorp-tion/desorption methods as well as catalyticactivity measurements. The results presented high-light that this is a complicated area of catalysis,with multiple parameters influencing the catalyst’sactivity in practice. The next chapter, ‘Kinetics andMechanism of Hydroprocessing Reactions’, dealswith hydrogenation of aromatics, hydrodesulfuri-sation, hydrodenitrogenation, hydrodeoxygenationand hydrodemetallation reactions, and reinforcesthe conclusion that this is a complex area of catal-ysis. Several tables of rate constants with variousfeeds over conventional and carbon-supportedcatalysts are provided; however, there is little

discussion about the intimate mechanisms of thesereactions. The eighth chapter, ‘CatalystDeactivation’, reminds the reader that coking reactions are an important deactivation process forconventional hydroprocessing catalysts, and thattheir carbon-supported counterparts can be lessprone to this deactivation mode.

The penultimate chapter on ‘Patent Literature’is surprisingly brief, being only one and a halfpages long and with fewer than ten patents cited –this, it is said, reflects the rather limited patent literature on carbon-supported hydroprocessingcatalysts. However, there are in fact many morerelated published patents than those listed, and it isnot clear what criteria were applied to produce thisshort list. The final two-page chapter providessome ‘Conclusions’, including the proposal thatthe activity of carbon-supported catalysts is basedon their ability to adsorb and activate hydrogen. Inthis regard, carbon blacks are said to be more ableto activate hydrogen than activated charcoals,which in turn are more active than graphite. And,of course, this is much enhanced by the presenceof metal species, especially palladium and platinum.

Commercial ApplicationsThis book contains some thought provoking

information, and it will certainly be of interest tothose involved with carbon-based catalysts. Insome instances, there appear to be benefits fromusing carbon-supported hydroprocessing catalysts,although the fundamental reasons for this are notalways delineated, nor are guide rules provided.Commercially activated charcoals often have vari-ous promoters incorporated on their surfaces, andmay have undergone an oxidative pre-treatmentthat gives a high dispersion of the active specieswhen it is applied. In the context of this book it isunclear what effects these treatments have on, forexample, carbon-supported sulfided cobalt/molybdenum hydroprocessing catalysts.

It is, however, clear that carbon supports haveattracted much more attention in areas other thanhydroprocessing reactions, and with the exceptionof promoted carbon-supported ruthenium catalystfor ammonia synthesis (7), the most successful


References1 A. J. Bird, ‘Activated Carbon Supports’, in “Catalyst

Supports and Supported Catalysts: Theoretical andApplied Concepts”, ed. A. B. Stiles, Butterworth-Heinemann, Stoneham, Massachusetts, U.S.A., 1987,pp. 107–137

2 See for example: J. C. Chaston and E. J. Sercombe,Platinum Metals Rev., 1961, 5, (4), 122

3 See for example: L. S. Liebeskind and E. Peña-Cabrera,Org. Synth., Collect. Vol., 2004, 10, 9

4 H. M. Colquhoun, D. J. Thompson and M. V. Twigg,“Carbonylation: Direct Synthesis of CarbonylCompounds”, Plenum Publishing Corp, New York,U.S.A., 1991

5 H. Topsøe, B. S. Clausen and F. E. Massoth,“Hydrotreating Catalysis”, eds. J. R. Anderson and M.Boudart, Catalysis Science and Technology, Volume11, Springer Verlag, Berlin, Germany, 1996

6 J. Scherzer and A. J. Gruia, “Hydrocracking Scienceand Technology”, Marcel Dekker, New York, U.S.A.,1996

7 S. R. Tennison, ‘Alternative Noniron Catalysts’, in“Catalytic Ammonia Synthesis: Fundamentals andPractice”, ed. J. R. Jennings, Fundamental and AppliedCatalysis Series, Plenum Press, New York, U.S.A.,1991, pp. 303–364


The ReviewerMartyn Twigg is the Chief Scientist of JohnsonMatthey PLC and was previously TechnicalDirector for the Environmental Catalysts andTechnologies Division. Following work at theUniversity of Toronto, Canada, and a fellowship atthe University of Cambridge, U.K., he joined ICIwhere he aided the development and production ofheterogeneous catalysts used in the production ofhydrogen, ammonia and methanol. Martyn hasauthored or co-authored many research papers,

written numerous chapters in encyclopedic works, and edited and contributedto several books. He edits a book series on fundamental and applied catalysis.

carbon-supported catalysts are used at low temper-atures. The hydroprocessing catalysts discussed inthis book, on the other hand, operate at relativelyhigh temperatures. Under these hydrogenatingconditions, a major practical problem could begasification of the support itself to give methane orother hydrocarbon species, and this should not beoverlooked when considering carbon-supportedcatalysts for hydrogenation applications involvinghigh hydrogen partial pressures and high tempera-tures. It is claimed in this book that carbongasification is slow under typical hydrodesulfurisa-tion conditions, although over several months’operation some irreversible structural changes maytake place. However, little information about this isavailable. Potential practical complications such asthis are not explored in any detail, although beforethese carbon-supported hydroprocessing catalystsare considered for industrial applications it is para-mount that they be better understood, thoughperhaps not to the level of detail that pgms on car-bons are understood from their widespread useand characterisation over many years.

138

IntroductionIridium is the only refractory f.c.c. metal. Its

melting point is 2446ºC, and as one of the plat-inum group metals (pgms), it exhibits excellentmechanical properties and high resistance to corro-sion at elevated temperatures (1, 2). Some featuresof this metal, such as its poor workability (inclina-tion to brittle fracture under mechanical treatment)and intrinsic fracture mode (brittle transcrystallinefracture) are unlike the deformation behaviourobserved in other f.c.c. metals and remain puzzlingfor the materials science community (3–8).Discussion of the possible mechanisms of defor-mation and fracture in iridium was begun inPlatinum Metals Review more than fifty years ago(9, 10), and continues up to the present. Becauserefining iridium is a very complicated procedure,segregation of impurities at the grain boundaries isconsidered the most likely cause of the weak

cohesive strength of grain boundaries and, hence,the poor workability of this metal (7, 8, 10, 11, 15).Indeed, it has been shown that high purity iridium can be processed like platinum (16).Polycrystalline iridium free of non-metallic impuri-ties, and its ternary alloy with rhenium andruthenium, demonstrate both transgranular cleav-age and satisfactory plasticity. Traces (~ 10 ppm)of carbon and oxygen in the matrix induce theappearance of intergranular cleavage on the fracture surfaces and, as a result, workability is cat-astrophically diminished (7, 11). The extraction ofdetrimental non-metallic impurities continues tobe a significant problem for pgm refiners (7, 8, 15).

Some physical parameters of iridium, such asits strong interatomic bonds and high elastic mod-ulus, give a basis for supposing that brittlenessmay be an intrinsic property of this metal (5, 6,20–23). For example, formal substitution of the


Plastic Deformation of PolycrystallineIridium at Room TemperatureUNIQUE PROPERTIES OF IRIDIUM, THE SOLE REFRACTORY FACE CENTRED CUBIC METAL

By Peter Panfilov* and Alexander YermakovLaboratory of Strength, Ural State University, 620083 Ekaterinburg, Russia; *E-mail: [email protected]

and Olga V. Antonova and Vitalii P. PilyuginInstitute of Metal Physics, Ural Division of the Russian Academy of Sciences, 620219 Ekaterinburg, Russia

Defect structure and its relationship with deformation behaviour at room temperature ofiridium, the sole refractory face centred cubic (f.c.c.) metal, are discussed. Small angleboundaries and pile-ups of curvilinear dislocation segments are the main features of dislocationstructure in polycrystalline iridium at room temperature, while homogeneously distributedrectilinear dislocation segments were the main element of defect structure of iridium singlecrystals at the same conditions. Small angle boundaries and pile-ups of curvilinear dislocationsegments are formed in iridium single crystals under mechanical treatment at elevatedtemperatures (≥ 800ºC) only. The evolution of defect structure in polycrystalline iridium andother f.c.c. metals under room temperature deformation occurs by the same process:accumulation of dislocations in the matrix leads to the appearance of both new sub-grainsand new grains up to the fine grain (nanocrystalline) structure. Neither single straightdislocations nor their pile-ups are observed in iridium at room temperature if small angleboundaries have been formed. This feature may be considered as the reason why polycrystallineiridium demonstrates advanced necking (high localised plasticity) and small total elongation.

DOI: 10.1595/147106709X463318

elastic modulus into cleavage criteria equationsleads to the conclusion that iridium is expected tobehave like a brittle substance (6, 20, 24). On theother hand, according to empirical theories ofplastic deformation, it should be deformed likeother f.c.c. metals (13, 20, 25). It is a paradox, butthe highest yield stress and strength of iridium sin-gle crystals under tension (~ 100 MPa and ~ 500MPa, respectively) become similar to those ofother f.c.c. metals when these parameters havebeen normalised on the elastic modulus (20). Thismay also be applied to the concept of ‘dislocationmobility’ in iridium, as understood by dislocationtheory in the field of continuum mechanics, whichdescribes the motion of single dislocations underexternal stress (normalised on the elastic modu-lus). This apparent contradiction with empiricalknowledge of the deformation and fracture off.c.c. metals merits further discussion among thematerials science community.

Behaviour of Iridium SingleCrystals

Experiments have shown that the intrinsicfracture mode of iridium single crystals is brittletransgranular cleavage. However, monocrystallineiridium is also a highly plastic material, as it frac-tures after considerable elongation (up to 80%) atroom temperature and never fails under compres-sion (19, 20, 26, 27). Hence, the brittleness ofiridium in the monocrystalline state is a uniquekind of brittle fracture, since cleavage occurs afterhuge plasticity. Therefore, single crystal iridiummay be classified as a plastic but cleavable solid(28). Study of the geometry of deformation trackson the back surfaces of deformed iridium singlecrystals has shown that octahedral slip is the soledeformation mechanism active at room tempera-ture. In contrast to other f.c.c. metals, no neckingis observed in iridium single crystal samples undertension, in spite of considerable elongation priorto failure. Analysis of the deformation track distri-bution leads to the conclusion that all plasticity ofiridium single crystals is observed during the earlystages of plastic deformation (29).

Studies by transmission electron microscopy(TEM) have shown that rectilinear dislocation

segments laid along the <110> crystallographicdirection are the main element of the defect struc-ture in iridium single crystals at room temperature(11, 26, 29). These straight dislocations aregrouped into network-like arrangements, wherethey are distributed almost equidistantly or homo-geneously. The homogeneous distribution ofdislocations takes place in thin foils of f.c.c. met-als in the early stages of plastic deformation, whendislocation density is expected to be small (30).However, the density of dislocations in iridiumsingle crystals may be so high that the foil is nolonger transparent to the electron beam. No smallangle boundaries or other types of inhomoge-neously distributed dislocations are observed insingle crystal iridium deformed at room tempera-ture. Therefore, it may be concluded that thedeformation mechanism of iridium single crystalsat room temperature is octahedral slip of the per-fect dislocation with a <110> Burgers vector (28,29). This observation distinguishes iridium singlecrystals from f.c.c. metals with low and intermedi-ate melting points.

The evolution of dislocation structure in iridium single crystals stops at the stage of homogeneous distribution of dislocations in thecrystal, which is the first stage of plastic deformation in metallic single crystals (30–32). It should be noted that motion of dislocationsnever occurs in thin foils of iridium during room temperature TEM experiments, including in situtension in the column of the microscope. Thesefeatures, together with the high yield stress of iridium single crystals, lead to the conclusion that<110> dislocations have low ability to moveunder external stress in the refractory f.c.c. metalat room temperature. As a result, the dominantdislocation arrangements in iridium single crystalscannot transform into small angle boundariesunder external stress at room temperature. Such behaviour contrasts with f.c.c. metals having lowand intermediate melting points, where thisprocess can occur at room temperature. It may beconsidered as an additional argument for the statement that all plasticity of iridium single crystals is realised during the early stages of plasticdeformation.


High Purity Polycrystalline Iridium High purity polycrystalline iridium (purity

> 99.9%, including < 0.1% metallic impuritiesand no non-metallic impurities) and iridium alloysalso exhibit brittle transcrystalline fracture as anintrinsic fracture mode under tension, but thetotal elongation of samples having a circularcross-section is less than 10% at room tempera-ture (17–19). This does not mean that iridiumwires possess poor plasticity, since advancednecking occurs in the samples even at room tem-perature, in spite of a considerable traverse rateof 1 mm min–1. The homologous temperature,THomologous, is defined as Texp (K)/Tmelt (K),where Texp = experimental temperature andTmelt = melting point. For low and intermediatehomologous temperatures (THomologous ≤ 0.5),localisation of plasticity induces a visible decreasein the total elongation of wires from 10% downto 5% at THomologous ~ 0.5 when necking to apoint takes place (17). Total elongation begins toincrease as soon as the flowing neck of a multi-neck effect appears at THomologous > 0.5. Thisbehaviour at similar traverse rates and homolo-gous temperatures is observed only in gold wiresand, hence, high purity polycrystalline iridiummay also be considered to be a plastic but cleavable substance.

It is apparent that the deformation behaviourof polycrystalline iridium on the macroscopic scalehas many features in common with iridium singlecrystals and also many differences. For example,advanced necking in polycrystalline iridium wireand the homogeneous distribution of plasticdeformation in iridium single crystals under ten-sion both point to the high plasticity of iridium.However, this is due to different effects dependingon the microstructure of the material. Currently,no information on the defect structure of poly-crystalline iridium is available in the literature.Therefore, the aim of this work is to carry out aTEM study of the dislocation structure of poly-crystalline iridium and its evolution duringdifferent stages of plastic deformation up to severedeformation. The findings will serve as a basis forfurther discussion of plastic deformation in poly-crystalline iridium at room temperature.

Preparation of Samples Samples for the research were prepared from

high purity polycrystalline iridium, free of non-metallic impurities (7). An electron-beam meltedmonocrystalline ingot was used for the manufac-ture of iridium sheets (33). Therefore, the samplescontain neither small grain boundaries nor grainboundaries in the initial state. TEM study of iridi-um single crystals agrees with this conclusion (29).The treatment procedure for work pieces includesforging of the ingot at 1500ºC to 2000ºC androlling of the sheet at ≤ 800ºC. After that, iridiumand its alloys can be deformed like platinum (7).

Recrystallisation annealing of the sheet (foriridium, the recrystallisation temperature is~ 1000ºC) is not carried out, since it induces a cat-astrophic drop in workability of iridium due tointergranular brittleness. During processing, grainscan be formed under forging only, whereas condi-tions for the appearance of small angle boundariesexist during both stages of treatment. During thefirst stage, the iridium matrix should not be conta-minated by non-metallic impurities (carbon andoxygen), because iridium oxides are volatile sub-stances at these temperatures. The environment iskept carbon free for this procedure (7). However,iridium does not interact with oxygen at tempera-tures < 1000ºC and carbon lubricant is not used inthe rolling mill facility. Hence, segregation ofharmful contaminants at the grain boundaries isalso avoided during the second stage. The fracturemode of the metal, which has been shown to be abrittle transgranular fracture, may be considered asthe proof of this supposition.

Discs with a diameter of 3 mm were stampedfrom the polycrystalline iridium sheet (thicknessof 0.3 mm; grain size ~ 0.1 mm), which was notrecrystallised after processing. Samples weredeformed under a ‘shift under pressure’ regime ina Bridgman anvil at room temperature (Figure 1).This procedure allows the highest degrees of plas-tic deformation to be reached without failure,even for brittle materials (34). The technique ofiridium thin foil preparation has been described inReferences (29, 35). The TEM study was carriedout on a conventional 200 kV microscope (JEM-200CX from JEOL).


Deformation and Defect Structureof Iridium

Under compression, high purity iridiumbehaves like a f.c.c. metal: it never fails under loadand exhibits the usual stress-strain curves expectedof a f.c.c. metal (20, 27, 36). Therefore, it may beexpected that the deformation behaviour of iridi-um would also be the same as that of other f.c.c.metals, under deformation with reduced tensile(cleaving) stresses. A shift under pressure proce-dure in a Bridgman anvil allows even brittle andalmost unworkable materials to be deformed, dueto the suppression of cleaving stresses. In metals,this technique allows severe deformations to bereached, which cannot be obtained under uniaxial(tension or compression) load (34).

Four points were chosen for examination of themicrostructure in polycrystalline iridium at differ-ent deformation degrees (initial state, 0.5 turn, 1.5turns and 3 turns), which cover the whole range ofstructural states from undeformed to severe defor-mation. The metallographic image of an iridiumdisc after 1.5 turns is shown in Figure 2. No crackson the edges or other significant deformationdefects are observed on its surface, despite consid-erable deformation of the material. Radial cracksand separation of the material close to the edgesappear after 3 turns (Figure 3). However, evensuch severe deformation does not lead to failure ofthe central part of the disc, where the level ofcleaving tensile stresses is minimal. Anotherdemonstration of the high plasticity of polycrys-

talline iridium is the fact that the thin section forTEM observations, which was obtained by elec-tropolishing in the central part of the disc after 0.5turn, has disappeared or been healed during thenext deformation (additional turn). It should benoted that polycrystalline copper and nickel exhib-it similar behaviour under the same deformationconditions (37).

TEM study of defect structure has shown thatsmall angle boundaries and dislocation pile-ups,consisting of curvilinear dislocation segments, arethe main dislocation arrangements in polycrys-talline thin foils in the undeformed state (Figure 4).No rectilinear dislocation segments or theirarrangements are observed in the material. It may


Sample

Bridgman Anvil

Fig. 1 Bridgman anvil used for the ‘shiftunder pressure’ test on the iridium samples

0.5 mm

Fig. 2 Iridium disc after ‘shift under pressure’ testing ina Bridgman anvil after 1.5 turns

be proposed that the rectilinear dislocation seg-ments become unstable under thermomechanicaltreatment of the single crystalline work piece atelevated temperatures, and the initial dislocationstructure of iridium single crystals begins to trans-form into a cellular structure. The same defectstructure is observed in polycrystalline f.c.c. metalswith low and intermediate melting points at smalldeformations, and in their single crystals at thethird stage of plastic deformation (elongation> 10%) (30, 38–40). Deformation after 0.5 turndoes not induce the appearance of new features inthe dislocation structure of polycrystalline iridium.Small angle boundaries and pile-ups of curvilineardislocation segments are the dominant attributesof the defect structure, while the dislocation

density in the material naturally increases (Figure5). The microdiffraction image confirms that localdeformation of the material is advanced. However,in contrast to the single crystalline state, rectilineardislocation segments do not appear in the materi-al under external stress.

A threshold is reached after 1.5 turns, as nocracking or separation of the sample takes place atthis stage of deformation, whereas the next stageof plastic deformation leads to the cracking of thedisc. The microstructure of the central part of thedisc looks the same as after 0.5 turn: its main mor-phological features are small angle boundaries andpile-ups of curvilinear dislocations close to them,and its microdiffraction pattern points to aseverely deformed polycrystalline state (Figure 6).This new kind of defect structure is observed inthe vicinity of the disc edge, where deformationof the material is considerably higher than in thecentre of the disc. Fine grains, which can reach afew dozen nanometres in diameter, are revealed inthe material (Figure 7). Analysis of diffraction pat-terns confirms the supposition that thenanocrystalline state begins to form in polycrys-talline iridium at this stage of plastic deformation.The microstructure of iridium in the middle partof the disc after the last deformation step (3 turns)is shown in Figure 8. This is a fine grain structure,where some grains are approximately 50 nm to100 nm in diameter, and its microdiffraction pat-tern is typical of the nanocrystalline state of f.c.c.metallic materials. It should be noted that this


0.5 mm

Fig. 3 Iridium disc after ‘shift under pressure’ testing ina Bridgman anvil after 3 turns

1 μm

Fig. 4 Microstructure of polycrystallineiridium at room temperature in the initial(undeformed) state (Inset: Electronmicrodiffraction pattern taken from thesame place)

nanocrystalline structure is stable in refractoryiridium, whereas recovery processes make thesame structural state in f.c.c. metals with low andintermediate melting points unstable at room tem-

perature (41). Taking into account the differencesbetween their melting points, it may be supposedthat the low ability of <110> dislocations to moveunder external stress in iridium is the main reason


0.5 μm

Fig. 5 Microstructure of polycrystallineiridium at room temperature after 0.5 turns(Inset: Electron microdiffraction patterntaken from the same place)

0.5 μm

Fig. 6 Microstructure of polycrystallineiridium at room temperature after 1.5 turns(centre of the disc) (Inset: Electronmicrodiffraction pattern taken from thesame place)

0.5 μm

Fig. 7 Microstructure of polycrystallineiridium at room temperature after 1.5 turns(edge of the disc) (Inset: Electronmicrodiffraction pattern taken from thesame place)

for the stability of the nanocrystalline state in irid-ium at low homologous temperatures.

Behaviour of High Purity Iridium The experiment described above has confirmed

that polycrystalline iridium, free of non-metallicimpurities such as carbon and oxygen, can behaveat room temperature like a f.c.c. metal. The ‘shiftunder pressure’ test in the Bridgman anvil waschosen simply as another deformation scheme(the first being uniaxial compression), where thelevel of cleaving stresses is minimal. Qualitativeanalysis of the defect structure of polycrystallineiridium at the different stages of plastic deforma-tion, including examination of its mainmorphological features and the character of itsevolution, does not reveal any difference com-pared to f.c.c. metals having low and intermediatemelting points. Indeed, small angle boundaries andpile-ups of curvilinear dislocation segments arecommon attributes of the dislocation structure forall f.c.c. metals (30–32). The forming of a finegrain (nanocrystalline) structure under severedeformation also agrees with the supposition thatiridium can behave like a f.c.c. metal at room tem-perature. Therefore, taking into account thatoctahedral slip of <110> dislocations is the soledeformation mechanism in iridium single crystalsat room temperature (28, 29), this mechanism ofplasticity may also be considered the dominantone for the polycrystalline metal under the sameexperimental conditions. In other words, the pres-

ence of grain boundaries in the iridium matrixdoes not provoke additional deformation mecha-nisms in polycrystalline iridium. This conclusioncorrelates with the fact that the intrinsic fracturemode of iridium free of non-metallic impurities inboth the monocrystalline and polycrystalline statesis brittle transgranular cleavage, which does notdepend on the presence of grain boundaries in thematrix (19).

In contrast to other f.c.c. metals, perhapsexcluding rhodium, which may be considered ananalogue of iridium having a melting point of1963ºC (1), the nanocrystalline state in iridium atroom temperature is stable. This means that therecovery process has been suppressed at roomtemperature. This may be explained by the lowability of <110> dislocations to move under exter-nal stress, as structural obstacles to dislocationmotion, such as second phases or dislocation bar-riers (sessile dislocations, dislocation ‘forests’ etc.),should be absent in highly plastic f.c.c. metals(30–32). This decrease in the ability of dislocationsto move in f.c.c. metals is only seen at low temper-atures. Room temperature (~ 300 K) on thehomologous temperature scale for refractory iridi-um is estimated at about 0.12. For rhodium, nickel,copper and aluminium, this homologous tempera-ture is reached at around 240 K, 150 K, 120 K and70 K, respectively. Hence, the experimentsdescribed in this paper (including thin foil prepara-tion and TEM study) must be carried out at verylow temperatures for the majority of f.c.c. metals,


0.5 μm

Fig. 8 Microstructure of polycrystallineiridium at room temperature after 3 turns(Inset: Electron microdiffraction patterntaken from the same place)

which naturally causes great technical problems.Therefore, the existence of a stable nanocrys-talline structure at this homologous temperaturecannot be considered as a unique property of irid-ium, at least until this experiment can be carriedout for a f.c.c. metal having a low or intermediatemelting point. On the microscopic scale (atomlevel), the low ability of <110> dislocations tomove under external stress in iridium may beexplained by the fact that it has the strongest inter-atomic bonds among f.c.c. metals, since it is thesole refractory metal with an f.c.c. lattice.Transition from homogeneously distributed recti-linear dislocations to small angle boundaries isstopped, which leads to the absence of necking iniridium single crystals under room temperaturetension. This may also be considered as a conse-quence of the low ability of <110> dislocations tomove under external stress.

Conclusions According to the well-known empirical scheme

of evolution of defect structure in f.c.c. metals,homogeneous distribution of single dislocationsleads to homogeneous distribution of plasticdeformation in the sample, while the transitionfrom homogeneous distribution to the appearanceof small angle boundaries or localised distributionof dislocations correlates with the start of thenecking process or the localisation of plasticity inthe neck region (31, 32). Indeed, necking alwaysoccurs after the stage of homogeneous accumula-tion of plastic deformation in both single crystal

and polycrystalline f.c.c. metals with low and inter-mediate melting points across a wide temperaturerange. However, the deformation behaviour ofrefractory iridium at room temperature depends onthe microstructure of samples (single crystal orpolycrystalline). In single crystals, necking does notoccur: straight <110> dislocations are homoge-neously distributed in the material and small angleboundaries are absent. In polycrystalline iridium(iridium wire), necking is the dominant stage ofplastic deformation: small angle boundaries andpile-ups of curvilinear dislocations are the mainfeatures of the defect structure, while no rectilineardislocation segments are observed. It should benoted that in both cases, iridium continues to be ahighly plastic substance, but this manifests in dif-ferent ways. This important difference between thedeformation behaviour of refractory iridium andother f.c.c. metals needs further experimentalstudy and discussion.

AcknowledgementsThe authors would like to thank Professor

David Lupton (W. C. Heraeus GmbH, Hanau,Germany) and Professor Easo George (TheUniversity of Tennessee, Knoxville, Tennessee,U.S.A.) for helpful discussions. This research waspartially supported by the JSC Ekaterinburg Non-Ferrous Metals Processing Plant, the Ministry ofEducation and Science of the Russian Federation(Grant No. 2.2.2.2/5579) and the U.S. CivilianResearch and Development Foundation (CRDF)(Grant No. RUXO-005-EK-06/BG7305).


1 J. W. Arblaster, Platinum Metals Rev., 1996, 40, (2), 622 R. Weiland, D. F. Lupton, B. Fischer, J. Merker, C.

Scheckenbach and J. Witte, Platinum Metals Rev., 2006,50, (4), 158

3 G. Reinacher, Metall, 1964, 18, 7314 R. W. Douglass and R. I. Jaffee, Proc. ASTM, 1962,

62, 6275 S. S. Hecker, D. L. Rohr and D. F. Stein, Metall. Trans.

A, 1978, 9, (4), 4816 C. Gandhi and M. F. Ashby, Acta Metall., 1979, 27,

(10), 15657 N. I. Timofeev, A. V. Yermakov, V. A. Dimitriev and

P. E. Panfilov, “Metallurgy and Mechanical Behaviourof Iridium”, Urals Branch of Russian Academy ofScience, Ekaterinburg, Russia, 1996

8 “Iridium”, eds. E. K. Ohriner, R. D. Lanam, P. Panfilovand H. Harada, Proceedings of the internationalsymposium held during the 129th Annual Meeting& Exhibition of The Minerals, Metals and MaterialsSociety (TMS), Nashville, Tennessee, 12th–16thMarch, 2000, TMS, Warrendale, Pennsylvania, U.S.A.,2000

9 F. D. Richardson, Platinum Metals Rev., 1958, 2, (3), 8310 B. L. Mordike and C. A. Brookes, Platinum Metals Rev.,

1960, 4, (3), 94

References

11 C. A. Brookes, J. H. Greenwood and J. L. Routbort,J. Inst. Met., 1970, 98, (1), 27

12 L. Heatherly and E. P. George, Acta Mater., 2001,49, (2), 289

13 T. J. Balk, K. J. Hemker and L. P. Kubin, Scr. Mater.,2007, 56, (5), 389

14 S. P. Lynch, Scr. Mater., 2007, 57, (2), 8515 E. K. Ohriner, Platinum Metals Rev., 2008, 52, (3), 18616 J. R. Handley, Platinum Metals Rev., 1986, 30, (1), 1217 P. Panfilov, V. Novgorodov and A. Yermakov, J.

Mater. Sci. Lett., 1994, 13, (2), 13718 P. Panfilov and A. Yermakov, J. Mater. Sci., 2004,

39, (14), 454319 P. Panfilov, J. Mater. Sci., 2005, 40, (22), 598320 C. N. Reid and J. L. Routbort, Metall. Trans., 1972, 3,

(8), 225721 R. E. MacFarlane, J. A. Rayne and C. K. Jones, Phys.

Lett., 1966, 20, (3), 23422 S. Crampin, K. Hampel, D. D. Vvedensky and J. M.

MacLaren, J. Mater. Res., 1990, 5, (10), 210723 S. P. Chen, Philos. Mag. A, 1992, 66, (1), 124 C. Gandhi and M. F. Ashby, Scr. Metall., 1979, 13, (5),

37125 P. Haasen, H. Hieber and B. L. Mordike, Z. Metallkd.,

1965, 56, (12), 83226 C. A. Brookes, J. H. Greenwood and J. L. Routbort,

J. Appl. Phys., 1968, 39, (5), 239127 A. Yermakov, P. Panfilov and R. Adamesku, J. Mater.

Sci. Lett., 1990, 9, (6), 69628 P. Panfilov, ‘Brittle Transcrystalline Fracture in Plastic

Face Centered Cubic Metal Iridium’, in “Iridium”,eds. E. K. Ohriner, R. D. Lanam, P. Panfilov and H.Harada, Proceedings of the international symposium

held during the 129th Annual Meeting & Exhibitionof The Minerals, Metals and Materials Society (TMS),Nashville, Tennessee, 12th–16th March, 2000, TMS,Warrendale, Pennsylvania, U.S.A., 2000, pp. 27–40

29 P. Panfilov, J. Mater. Sci., 2007, 42, (19), 823030 P. B. Hirsch, A. Howie, R. B. Nicholson, D. W. Pashley

and M. J. Whelan, “Electron Microscopy of ThinCrystals”, Butterworth and Co Ltd, London, 1965,549 pp

31 R. W. K. Honeycombe, “The Plastic Deformation ofMetals”, Edward Arnold, London, 1968, 477 pp,Translated into Russian, Mir, Moscow, 1972

32 J. P. Hirth and J. Lothe, “Theory of Dislocations”,McGraw-Hill, New York, 1968, 780 pp

33 P. Panfilov, A. Yermakov, V. Dmitriev and N.Timofeev, Platinum Metals Rev., 1991, 35, (4), 196

34 P. W. Bridgman “Studies in Large Plastic Flow andFracture: With Special Emphasis on the Effects ofHydrostatic Pressure”, McGraw-Hill, New York, 1952,362 pp

35 D. L. Rohr, L. E. Murr and S. S. Hecker, Metall. Trans.A, 1979, 10, (4), 399

36 H. Hieber, B. L. Mordike and P. Haasen, PlatinumMetals Rev., 1964, 8, (3), 102

37 Kh. Ya. Mulyukov, S. B. Khahizov and R. Z. Valiev,Phys. Status Solidi A, 1992, 133, (2), 447

38 P. B. Hirsch, R. W. Horne and M. J. Whelan, Philos.Mag., 1956, 1, (7), 677

39 J. E. Bailey and P. B. Hirsch, Philos. Mag., 1960, 5, (53),485

40 J. E. Bailey, Philos. Mag., 1963, 8, (86), 22341 N. A. Smirnova, V. I. Levit, V. P. Pilyugin, R. I.

Kuznetsov, L. S. Davydova and V. A. Sazonova, Fiz.Met. Metalloved., 1986, 61, (6), 1170 (in Russian)


The Authors

Peter Panfilov is a Professor ofMaterials Science at the Ural StateUniversity in Ekaterinburg, Russia. Hisscientific interests are closelyconnected with the problems of plasticdeformation, fracture and processingof the platinum group metals.

Alexander Yermakov is the TechnicalDirector of the platinum group metalsmanufacturer INTECH at Ekaterinburg,Russia. He has been a visitingresearcher at the Ural State Universitysince 1987. His research interest is thestudy of the properties and creation ofindustrial technology for the productionof noble metals and their alloys.

Olga V. Antonova is a Senior Scientistat the Institute of Metal Physics of UralDivision of the Russian Academy ofSciences in Ekaterinburg. Her area ofresearch activity is TEM study of defectstructure and phase transformations inhigh temperature alloys andintermetallics.

Vitalii P. Pilyugin is the Head of theHigh Pressure Laboratory at theInstitute of Metal Physics of UralDivision of the Russian Academy ofSciences in Ekaterinburg. His researchcovers deformation behaviour ofmaterials under high pressureincluding high temperature alloys,structural intermetallics, ceramics andbiomaterials. He is also a Professor ofMaterials Science at the Ural StateUniversity in Ekaterinburg, Russia.

Following meetings in Amsterdam, Munich andTurin (1–3), a fourth gathering was held on 8thand 9th October 2008 at the Confederation ofDanish Industry in Copenhagen, Denmark, withthe theme ‘Scientific Advances in Fuel CellSystems’ (4). These conferences alternate with theGrove Fuel Cell Symposium, with greater empha-sis on the latest technical developments in thefield. The two-day programme was compiled bythe Grove Symposium Steering Committee (5)from oral papers and posters submitted fromaround the world, and the conference was organ-ised by Elsevier. Many of the papers will bepublished in full in a special edition of Journal ofPower Sources (6).

The meeting was attended by delegates fromuniversities, research organisations and the fuelcell industry, numbers being limited to 280 by thecapacity of the venue. A total of 32 countries were represented, with 18% of delegates fromScandinavia, 60% from the rest of Europe, and16% from Asia, as well as others from the MiddleEast and Africa. Surprisingly, only 4% of delegateswere from the U.S.A.

The conference represents the state of the art inresearch and development topics on fuel cells andtheir applications and, as well as 56 oral papers,there were almost 140 high-quality poster presen-tations. There were eight oral sessions on fuels andfuel processing, modelling, systems and applica-tions, membrane science, materials science, andfinally cell and stack technology, the latter two cat-egories each occupying two sessions. Since thetopic of fuel cells covers such a wide area, for thisreview only papers involving use of the platinumgroup metals (pgms) and possible alternative mate-rials have been selected.

Delegates were welcomed to the headquartersof the Confederation of Danish Industry, situatedat the edge of the world-famous Tivoli Gardens,

by the Deputy Director General, Ole Krog. TheConfederation represents 11,000 companies inDenmark employing 500,000 workers. ThomasEgebo (Permanent Secretary of State at theMinistry of Climate and Energy, Denmark) offi-cially opened the proceedings, and spoke of theneed to improve security of energy supply, partlyby improving the efficiency of utilisation of exist-ing supplies, as well as increasing renewable energygeneration. The transport sector accounts for 25%of greenhouse gas emissions, and hydrogen andfuel cell powered vehicles have been grantedexemptions from Danish government registrationtaxes to encourage their implementation.

Egebo ended by presenting the 2008 GroveMedal to Subhash C. Singhal, who is a BattelleFellow and Director of Fuel Cells at the PacificNorthwest National Laboratory (PNNL) inRichland, Washington, U.S.A. (Figure 1). Singhal isan acknowledged world leader in solid oxide fuelcells (SOFCs), having joined PNNL in April 2000after working at Siemens Power Generation (for-merly Westinghouse Electric Corporation) fornearly thirty years, playing a key role in bringingSOFC technology from a few-watt laboratorycuriosity to fully integrated 200 kW power genera-tion systems. 100 kW systems have beendemonstrated in The Netherlands, Germany andItaly as well as the United States, while technologydemonstration programmes are in progress world-wide. The Solid State Energy Conversion Alliance(SECA) project is addressing the challenges to befaced, where cost reduction remains a major factor,together with interconnects and gas seals which arecompliant with differential thermal expansion.

Plenary PresentationOne reason for the high level of interest in fuel

cells in Denmark was explained by a plenary presen-tation entitled ‘Massive Integration of Renewable


Fuel Cells Science and Technology 2008SCIENTIFIC ADVANCES IN FUEL CELL SYSTEMS

Reviewed by Donald S. CameronThe Interact Consultancy, 11 Tredegar Road, Reading RG4 8QE, U.K.; E-mail: [email protected]

DOI: 10.1595/147106709X465460

Energy in the Power System with Fuel Cells asVirtual Power Plants’, given by Inger Pihl Byriel(Energinet.dk, Denmark). This company is ownedby the Danish state and was set up by Act ofParliament in December 2004 to operate the elec-tricity and gas grids, but has no power productioncapacity.

Twenty-five years ago, fifteen large thermalelectricity generating plants supplied all Danishpower requirements. Today, there are still fifteenlarge plants, but there are also 700 combined heatand power (CHP) units generating local powerand currently 5000 wind turbines rated at up to3000 MW. National annual demand is 35.3 TWh,which must be balanced against the 47.3 TWhactually generated by marketing the surplus. Onsome windy days, turbines and CHP can meet allof Denmark’s power demands. However, thisgenerating capacity fluctuates widely, dependingon wind strength, which cannot be accuratelyforecast.

To bridge the wide gaps between supply anddemand, ‘spinning reserve’ must be maintained inthe coal-fired thermal power stations, which haverelatively slow response times to changing loads.Excess power is sold to neighbouring Sweden andGermany, but electricity prices vary widely dependingon availability. Stores of electricity, or generatorswith fast response times, would considerably easethe difficulties in balancing supply and demand, and alternatives such as batteries, fuel cells and

electrolysers, as well as compressed air storage andheating systems, are being considered.

A distributed energy generation system consistingof large numbers of connected fuel cells forming a‘virtual grid’ is seen as a possible solution, particularlysince these generators operate at increased efficiencyunder partial loads. Danish wind turbine generatingcapacity is expected to double from 3000 MW to6000 MW by 2025, posing a considerable challenge.Denmark is one of the leading exponents ofrenewable energy generation, but this supply anddemand problem will soon be encountered by othercountries as their proportion of renewable energyincreases.

Fuels and Fuel ProcessingMost fuel cells rely on a supply of gaseous

hydrogen which may be provided by severalmeans, including storage under pressure, reform-ing hydrocarbons or electrolysis. Huge volumes ofhydrogen are generated as byproducts from indus-trial processes, one of the largest sources beingfrom chlorine or caustic soda production. A highproportion of electrolysis is still carried out usingmercury amalgam cells, which leads to contamina-tion of the hydrogen with mercury.

The effect of this has been investigated, asexplained by Karel Bouzek (Institute of ChemicalTechnology, Prague, Czech Republic) in his talk‘Influence of the Hydrogen Contamination byMercury on the PEM Type Fuel Cell Life-Time’.


Fig. 1 The 2008 Grove Medalwas presented to Dr SubhashSinghal (centre), a worldauthority on solid oxide fuelcells, by Thomas Egebo, theDanish Permanent Secretary ofState at the Ministry of Climateand Energy (right), andProfessor Lars Sjunnesson,Chairman of the GroveSymposium Steering Committee(left)

Brine electrolysis in mercury-type cells typicallyresults in hydrogen contaminated with between10 μg m–3 and 40 μg m–3 of mercury, which overlong periods may be adsorbed on the platinum cat-alysts of PEM fuel cells to form platinumamalgams. Small (50 cm2) fuel cells with electrodessupplied by E-TEK, New Jersey, U.S.A. (nowBASF Fuel Cells), with 5 mg Pt cm–2 content, andNafion® membranes were supplied with hydrogencontaining between 0 μg m–3 and 1000 μg m–3 mer-cury over extended periods. Although there wassome performance recovery during periods whenthe cells were switched off, there was a 20% powerreduction over 10,000 hours which was not recov-erable. X-Ray photoelectron spectroscopyindicates that the mercury is in a changed chemicalstate, suggesting that an amalgam is formed withthe highly dispersed platinum catalyst. This effectwas reproduced using platinum foil electrodesexposed to mercury vapour. Overall, results indi-cated that with hydrogen containing 10 μg m–3 ofmercury, polymer electrolyte membrane (PEM)fuel cells are likely to be able to operate for 10,000to 20,000 hours without catastrophic poisoning bythe mercury.

Two papers were presented on storage ofhydrogen in the form of ammonia. The first ofthese, ‘Solid Ammonia as Energy Carrier:Possibilities and Technology Development’, waspresented by Debasish Chakraborty (AmminexA/S, Denmark). Ammonia itself is more danger-ous than gasoline, since it is toxic and has a highvapour pressure at room temperature. However,in the form of a metal ammine, such as with mag-nesium, [Mg(NH3)2Cl2] or copper, [Cu(NH3)8Cl2],ammonia can be stored in solid form, from whichgaseous ammonia can be recovered and cracked torelease hydrogen.

Amminex have developed ‘HydrammineTM’which includes Mg(NH3)6Cl2 and Ca(NH3)8Cl2 andhas a volummetric hydrogen content similar to liq-uid ammonia (~ 110 kg H2 m–3), but does notnecessitate pressurised storage, and is non-haz-ardous for road transportation. The company isactively integrating this solid ammonia technologywith a SOFC, in collaboration with Topsoe FuelCell, Denmark, and Risø National Laboratory at

the Technical University of Denmark, since theSOFC operating temperature is sufficient todecompose the ammine and to generate hydrogenfrom the resulting ammonia. A thirty-cell, 450 WSOFC stack has been operated directly on ammo-nia, and the work is being extended tointermediate temperature (673 K to 773 K) fuelcells.

A prototype compact ammonia cracker capableof producing 0.7 l min–1 of ammonia with 99.9%conversion has also been developed, incorporatinga ruthenium-based catalyst to reduce the reactiontemperature. It will be operated in conjunctionwith a PEM fuel cell built by Intelligent Energy,U.K. For this purpose it is necessary to incorpo-rate an absorber to remove any residual ammoniaand prevent membrane contamination.

Further work on metal ammine complexes wasreported by Asbjørn Klerke (Technical Universityof Denmark, Lyngby). In his paper ‘IndirectHydrogen Storage in Metal Ammine Complexesfor Portable Devices’, it was emphasised thatammonia is produced for agricultural purposes ona huge scale (125 million tonnes per annum)worldwide. It can be converted to stable com-pounds such as calcium, manganese or nickelammines, with energy densities comparingfavourably with other forms of hydrogen storagesuch as liquid or compressed gas or metalhydrides. The group has investigated the effect oftemperature programmed desorption of variousammines, releasing gaseous ammonia which maybe either used directly in a SOFC or decomposedto form hydrogen and nitrogen. The calcium form[Ca(NH3)8Cl2] decomposes completely below550 K, while the manganese form [Mn(NH3)6Cl2]decomposes below 650 K, and the magnesium[Mg(NH3)6Cl2] and nickel [Ni(NH3)6Cl2] formsbelow 670 K.

On heating the nickel form, the blue solidbecomes a cream solid with a more open structureof composition [Ni(NH3)2Cl2], before finallydecomposing to yellow NiCl2. Catalysed crackingof ammonia to hydrogen and nitrogen has alsobeen investigated, with best results being obtainedusing supported ruthenium catalysts (3% Ru ontitania) at temperatures as low as 700 K.


The use of fuel cells in small-scale decentralisedenergy generation requires hydrogen to be pro-duced on site from hydrocarbons such as naturalgas, propane and liquefied petroleum gas (LPG).Although poison-tolerant pgm fuel cell catalystshave been developed, low-temperature fuel cellsstill require hydrogen with very low carbonmonoxide (CO) content to avoid impaired perfor-mance due to catalyst poisoning. Conventionalhydrogen generator systems consist of a reformeroperating at 1073 K to 1273 K, followed by a high-temperature water-gas shift reactor (HT-WGS) at570 K to 770 K, and a low-temperature shift reac-tor (LT-WGS) at 453 K to 573 K. Finally apreferential oxidation (POX) catalyst is used tooxidise CO to CO2 at 353 K to 423 K in the pres-ence of hydrogen, preferably to less than 10 ppmCO. Heat for the endothermic reformer reactioncan either be provided by heat transfer through thereactor walls, or generated in situ by oxidising aproportion of the fuel. Work is in progress to sim-plify this process by making multifunctionalreactor vessels.

Feyza Gökaliler (Bogaziçi University, Turkey)in a talk entitled ‘Oxidative Steam ReformingPerformance of Pt-Ni Catalysts’ described the useof platinum-nickel alloy catalysts on alumina sup-ports for oxidative steam reforming as the firststage, followed by the water-gas shift reaction. Byoptimising the Pt:Ni ratio of the catalyst and reac-tion parameters, it is possible to use a single reactorvessel to combine the reforming and high-temper-ature shift reactions. Parameters examinedincluded the Pt:Ni ratio of the catalyst and theoverall metal content, the hydrocarbon feed rate,and the steam:carbon and carbon:oxygen ratios inthe feed stream. These showed that the platinum-nickel system is a promising catalyst for theoxidative steam reforming (OSR) reaction. Duringthis reaction, exothermic oxidation and endother-mic steam reforming are catalysed by Pt and Nisites respectively, with the catalyst acting as a microheat exchanger. Using this catalyst, hydrogen pro-duction could be carried out at temperatures as lowas 673 K with 100% conversion of the hydrocar-bon feed. The hydrogen production rate increaseswith increasing temperature and no carbon is

deposited on the catalyst surface. Selectivity (theH2:CO ratio) is highest at low temperature andtends to decrease with increasing temperature.Similarly, the proportion of hydrogen in the prod-uct increases with decreasing residence time in thereactor.

Hydrogen can be generated by decomposingmetal hydrides, as explained by Carmen M. Rangel(Instituto Nacional de Engenharia, Tecnologia eInovação, Lisbon, Portugal) in her talk ‘HydrogenGeneration and Storage System Using SodiumBorohydride at High Pressures for Operation of a100 W-Scale PMFC Stack’. Sodium borohydride(NaBH4) contains 7.3% hydrogen by weight, corre-sponding to an energy density of 1.38 kWh kg–1.To prevent spontaneous decomposition of thematerial in aqueous solution, sodium hydroxide(NaOH) is normally added to maintain the pHabove 14, although even under these conditionsthere is a slow release of hydrogen above 333 K.Decomposition is accelerated by the presence of acatalyst, and ruthenium or nickel on a metallicfoam support is frequently used. Typical hydridefuels consist of 10% NaBH4 plus 1% NaOH, or20% NaBH4 with 3% NaOH, often with car-boxymethyl cellulose or polyacrylamide gellingagents to render the mixture thixotropic. Otherdirect oxidation types of borohydride have alsobeen developed. Applications are likely to be limit-ed to relatively low-power devices by theenergetics of manufacturing the sodium borohy-dride, but the technology appears feasible forsmall, hand-held devices.

Membrane ScienceIntense development efforts on polymer elec-

trolyte membrane fuel cells have resulted intechnology improvements on a broad front. It haslong been recognised that the limiting factor inPEM fuel cell durability was due to the membrane.Peter Gray (Johnson Matthey Fuel Cells, U.K.)outlined some of these developments in his talk‘Advances in MEA Durability in PEM Fuel Cells’.Membranes with increased chemical stability,incorporating reinforcing materials as well as addi-tives to reduce chemical attack have resulted inmembrane electrode assemblies (MEAs) able to


meet current durability targets, achieving lifetimesup to five to ten times those obtained using previ-ously available commercial materials. Emphasishas therefore moved to catalysts with improvedstability, particularly under the arduous conditionsof cell reversal. This may occasionally occur due tohydrogen starvation at cell anodes during fuel cellstart-up and shut-down, particularly in automotiveapplications. This can result in loss of electro-chemical surface area due to oxidation of thecatalyst carbon substrate, and also possible sinter-ing of the platinum catalyst. Significantimprovements in MEA durability have beenobtained by incorporating advanced membraneswith oxidation resistant catalysts, even when sub-jected to extreme voltage cycling test regimes.

Cell and Stack TechnologyDirect methanol fuel cells (DMFCs) are being

sold in significant numbers as portable electricpower sources for civilian as well as military appli-cations (7). In order to reduce parasitic powerlosses, passive-feed DMFCs, without externalpumps for feeding methanol or air to the fuel cellare being investigated. Air reaches the cathodes bydiffusion, while methanol diffuses to the anodesdue to a concentration gradient from a reservoir,and by capillary action through the electrodepores.

In his talk ‘Electrochemical Characterisation ofa Passive Monopolar DMFC Mini-Stack Operatingat Room Temperature’, Antonino S. Arico (CNR-Institute of Advanced Technologies for Energy(ITAE), Messina, Italy) described how three dif-ferent designs of passive monopolar three-cellmini-stacks of this type have been investigated.This work forms part of a programme aimed atbuilding a 500 W system for the EuropeanCommission-sponsored MOREPOWER project,in collaboration with several other organisations.Using various platinum metal loadings, optimumpower density was reached using 4 mg Pt cm–2 at294 K. Preferred catalysts are carbon supportedplatinum-ruthenium alloys for the anode and plat-inum on carbon for the cathode. Cell performancedecreased during prolonged operation, althoughthis could be restored by shaking the cells. This

was attributed to dislodging carbon dioxide gasbubbles which build up within the structure. Onechallenge still to be overcome is a high level ofmethanol crossover to the cathode, which couldbe substantially reduced or eliminated by using animproved membrane.

Under the European Commission-fundedSixth Framework research initiative, AirbusIndustrie are coordinating the CELINA project todevelop auxiliary power units for aircraft to takeover many of the functions of the presenthydraulic system. Erich Gülzow (GermanAerospace Center, Institute of TechnicalThermodynamics, Germany) in his talk‘Investigation of Low Pressure Operation ofPEFC Using CO Contaminated Gas’ describedhow PEM fuel cells can be operated at pressuresas low as 200 mbar when standard fuel cells nor-mally run at around 1.2 bar.

Since air compression is energy intensive, it isadvantageous to operate at low pressure: 200 mbarwhich corresponds to a height of 10 km. As a pre-liminary to designing a reformer system forkerosene, a special fuel cell test station has beenconstructed to operate at pressures as low as50 mbar, using single 150 cm2 format PEM cells.Typically, anodes contain 0.30 mg Pt-Ru cm–2 alloycatalysts, while cathodes contain 0.55 mg Pt cm–2.Typical gas mixtures used at the anode consist ofhydrogen plus 5 ppm, 10 ppm, and 20 ppm CO,with between 1.5 and 2.5 stoichiometric fuel flow.Conclusions to date indicate that a pressurebetween 1000 mbar and 500 mbar does not have adramatic effect on performance, with a more pro-nounced effect below 500 mbar with a minimumof 62% relative humidity in the inlet gas. The max-imum permitted carbon monoxide content was20 ppm, since pronounced performance effectswere found at concentrations above this level.

Various attempts to eliminate or minimise theamount of pgms used in PEM fuel cells are inprogress. One of these was presented by AndrewM. Creeth (ACAL Energy Ltd, U.K.) in his talkentitled ‘FlowCathTM Technology – A Route toPrecious Metal-Free Cathodes for PEM Type FuelCells’. This system is aimed at producing portablefuel cell generators, CHP systems for the home


and automotive applications. Each individual cellincorporates a three-dimensional cathode which isimmersed in a homogeneous aqueous mediator,consisting of an undisclosed transition metal redoxcatalyst. This can be regenerated in an externalreactor vessel by bubbling atmospheric oxygenthrough it. Currently about 300 ml of redox medi-um is required per 1 kW output, although this maybe reduced to 200 ml kW–1, and water managementcan be achieved within the regenerator. The cath-odes, which may be carbon based or metal meshes,are combined with conventional MEAs with plat-inum-catalysed anodes. Due to the facile nature ofthe hydrogen oxidation reaction, a very low anodepgm loading is permissible.

Peak powers of up to 440 mW cm–2 have beenobserved with durability in excess of 1500 hours.Due to the relatively large volume of catholyte, nohumidification is required for the inlet gases. As aDMFC, an open circuit voltage of 0.7 V can beobtained with power densities in excess of90 mW cm–2 and no impact is observed due tomethanol crossover at concentrations below 8 M.A ten-cell stack of 5 cm2 cells has been demon-strated, providing 60 W of power, and a 1 kWsystem is being constructed.

A second means to minimise the platinum con-tent of fuel cells was proposed by Stuart Gilby(CMR Fuel Cells, U.K.) in his talk ‘CMR FuelCells: Development of Novel DMFC Fuel CellTechnology for the Portable Market’. The compa-ny is developing a range of DMFC stack productsfor the portable electronics industry. This applica-tion requires small, compact stack designs withhigh voltage efficiency and power density to min-imise stack volume. By using alkaline solid polymerelectrolyte membranes, a wide range of catalystsmay be used, such as silver, perovskites, andpyrochlore materials. An added advantage of alka-line membranes is that electro-osmotic effects tendto move water molecules from the cathodes to theanodes, inhibiting crossover of other species.Methanol permeability therefore decreases withincreasing concentration, so that higher concentra-tions can be used at the anode. Unfortunately,alkaline membranes are less well developed thantheir acid equivalents, and also exhibit a lower

mass transfer rate for hydroxyl ions compared toprotons, so that their conductivity is roughly onetenth of those obtained for acidic membranes.This is partly offset by the high reactivity for theoxygen reduction reaction under alkaline condi-tions. In addition, soluble forms of membranematerial are not yet available. These are essential tomix with the catalysts to produce an optimum higharea three-phase interface between gas, liquid andsolid when preparing electrodes.

In a talk entitled ‘Fuel Cell Cathodes Studiedwith Density Functional Theory’, Jan Rossmeisl(Center for Atomic-scale Materials Design(CAMd), Technical University of Denmark,Lyngby) described the use of the density function-al theory (DFT) model to examine the oxygenreduction reaction on platinum and its alloys.Having developed a framework to deal with thepotential in electronic structure calculations, andby applying a quantitative version of the Sabatierprinciple, together with a database of DFT calcula-tions, they are able to predict trends in activities fordifferent electrode materials. Comparing the pre-dictions of the Sabatier model with experiments onPt and Pt3Ni {111} surfaces, most of the polarisa-tion curve features seen in experiments can beexplained in terms of the Sabatier analysis. Also,CAMd have developed a kinetic model whichallows them to study the effect of hydroxyl ioncoverage and the pressure of oxygen, based direct-ly on ab initio calculations.

Microbial fuel cells are capable of transformingchemical energy directly to electrical energy viaelectrochemical reactions. Beate Christgen(Newcastle University, U.K.) in a talk entitled‘Advances in Microbial Fuel Cells throughCathode, Anode and Membrane Development’explained that they may be used to treat materialssuch as waste water to obtain clean water and elec-trical energy. Generally a catalysed, air depolarisedcathode is used in conjunction with an anode sys-tem in which organic matter is oxidised byanaerobic bacteria acting as biocatalysts, produc-ing electrons which are transferred to a solidanode. To date, platinum-catalysed cathodes havebeen used, although less expensive alternativematerials such as metal phthalocyanines supported


on carbon, or manganese dioxide on carbon havealso been examined. Due to the low current densi-ties obtained from microbial fuel cells, it will benecessary to use cells with large surface areas toproduce a useful current output, implying thatinexpensive membrane separators will also need tobe developed.

Poster ExhibitsTwo poster sessions were held, with almost 140

posters. These included several featuring applica-tions of the pgms in fuel processing, fuel cellcatalysis and sensors.

Eini Puhakka (VTT Technical ResearchCentre of Finland) presented a poster entitled‘Comparison of Pt-Based PEMFC Cathode-Catalysts Using Modelling Approach’. Densityfunctional methods and rotating ring disc platinum electrode measurements have been usedto investigate reasons for the potential loss whichtakes place at the cathodic oxygen reduction reaction.

Julia Melke (Fraunhofer Institute for SolarEnergy Systems ISE, Germany) presented aposter entitled ‘Investigation of ReactionMechanism and Structural Changes in DirectEthanol Fuel Cell Electrodes Using X-RayAbsorption Spectroscopy’. The ethanol oxidationreaction has been investigated using carbon sup-ported platinum, platinum-ruthenium, andplatinum-tin alloys in half cells and fuel cells usingthe Δμ X-ray absorption near edge structure(Δμ XANES) technique, while structural changesto the catalysts have been studied using conven-tional extended X-ray absorption fine structure(EXAFS) spectroscopy.

Janet M. Fisher (Johnson Matthey TechnologyCentre, U.K.) presented a poster entitled ‘PdRuAlloy Nanoparticles as Anode Catalysts in DirectMethanol Fuel Cells’ aimed at developing morecost effective catalysts for small portable powerunits. The individual components of the catalystshow relatively poor activity on their own, but forthe methanol oxidation reaction, palladium-ruthe-nium anode catalysts have the same activity at353 K as commercial platinum-ruthenium anodecatalysts at 333 K.

Berker Ficicilar (Middle East TechnicalUniversity, Ankara, Turkey) reported on the‘Effect of Pd Loading in Pd-Pt Bimetallic CatalystsDoped into Hollow Core Mesoporous ShellCarbon on Performance of Proton ExchangeMembrane Fuel Cells’. Catalyst substrates weresynthesised by template replication with sub-micrometre-size core mesoporous shell silicaspheres, yielding supports with surface areas ashigh as 1290 m2 g–1 with a uniform, narrow poresize distribution centered around 3.0 nm. Amicrowave synthesis route was used to preparecarbon supported palladium-platinum alloy cata-lysts and testing was carried out in 5 cm2 single-cellPEMFC tests in conjunction with a range of phys-ical characterisation techniques.

As the range of applications for PEMFCsincreases, there is more awareness of the potentialfor poisoning. One example of poison tolerancetesting was reported by Won-Yong Lee (KoreaInstitute of Energy Research (KIER), Daejeon,South Korea) in a poster entitled ‘Influence ofNaCl Vapor on Performance and Durability of aPEFC’. Experiments involved monitoring singlecell performance while running the cathode on aircontaminated with sodium chloride vapour, andpost-test examination of the cell components. Theresults were discussed in terms of poisoning ofplatinum catalysts by Na+ ions and H+ ions in theion exchange membrane exchanged with Cl– ions.

Satoshi Ohara (Osaka University, Japan) pre-sented an interesting paper on the use of‘Palladium-Polymer Hybrid Nanoparticles forHydrogen Sensors in Fuel Cells’. DNA strandshave binding sites for metal ions and cation mole-cules on the nucleoside bases and backbone. Metal(Pd)-polymer (DNA) hybrid nanoparticles werecreated by a combination of metallisation andDNA compaction. Directly bound Pd(II) com-plexes were produced and then reduced to metalcausing a morphology change in the DNA strands,producing a spherically shaped moss-like hybrid.The palladium can still absorb hydrogen tobecome PdHx, which increases the electrical resis-tance and volume of palladium materials. Theresult is a highly sensitive hydrogen sensor orhydrogen switch device.


Poster PrizesPosters were judged by panels of members of

the Grove Symposium Steering Committee forcontent and presentation, and four awards weremade to:(a) Xiaohui Tian, Steffen Eccarius and Carsten

Agert (Fraunhofer Institute for Solar EnergySystems ISE, Germany) for their poster‘Computational Geometry Design andModeling for a Vapor-Fed Direct MethanolFuel Cell’ (Poster 1.28)

(b) Timo Kurz, Anne Grundmann and CarstenAgert (Fraunhofer Institute for Solar EnergySystems ISE, Germany) for their paper‘Modelling Heat Management in a HighTemperature PEM Fuel Cell System withAdsorption Heat Storage’ (Poster 1.54)

(c) Satoshi Ohara, Kazuyoshi Sato, MitsuoUmetsu and Tadafumi Adschiri (Osaka andTohoku Universities, Japan) for their pre-sentation ‘Palladium-Polymer HybridNanoparticles for Hydrogen Sensors in FuelCells’ (Poster 2.16)

(d) Samuel Georges, Jean-Marie Klein and YannBultel (Laboratoire d’Electrochimie et dePhysicochimie des Matériaux et desInterfaces (LEPMI), Grenoble, France) fortheir paper ‘Gradual Internal Reforming ofMethane with Electro-Catalytic Dissociationin Planar SOFC: From Model to Operation’(Poster 2.48)

Summary Findings from the Fuel Cells Science and

Technology 2008 conference were summed up ina talk by Professor Søren Linderoth (RisøNational Laboratory, Technical University ofDenmark, Roskilde). Fuel cells continue to findnew applications as auxiliary power supplies in air-craft, and as a means to utilise renewable energysources such as landfill gases and wastewater whilegenerating useful electrical power. In the area offuel processing and fuels, interest is moving tonovel fuels such as landfill gas, ammonia andethanol. Alternative methods of storing hydrogen

such as metal ammines and sodium borohydrideare also being investigated. The importance of fuelpurity has been recognised and is being intensive-ly examined. In the area of membrane science,higher operating temperatures, reduced require-ments for hydrating the input fuels, and improveddurability are being sought and achieved. DuPontNafion® remains the standard membrane bywhich others are judged.

While platinum remains the standard catalystmaterial for low-temperature fuel cells, cost reduc-tion efforts are in progress, either by replacing it(generally with a substantial performance penalty)or seeking other electrode structures or membranematerials to minimise its use. The pgms are used ina wide range of applications as well as the electro-catalyst.

The Eleventh Grove Fuel Cell Symposium willtake place at the Queen Elizabeth II ConferenceCentre in Westminster, London, U.K., from 22ndto 24th September 2009 (5), while the next confer-ence in the Science and Technology series is likelyto be held in Spain during 2010. See the Fuel CellsScience and Technology website for updated infor-mation as it becomes available (4).

References 1 D. S. Cameron, Platinum Metals Rev., 2003, 47, (1), 282 D. S. Cameron, Platinum Metals Rev., 2005, 49, (1), 163 D. S. Cameron, Platinum Metals Rev., 2007, 51, (1), 274 Fuel Cells Science and Technology:

www.fuelcelladvances.com (Accessed on 6th April2009)

5 The Grove Fuel Cell Symposium:www.grovefuelcell.com (Accessed on 6th April 2009)

6 Journal of Power Sources, 2009, Volume 193, Issue 1, pp. 1–386

7 J. Butler, ‘Portable Fuel Cell Survey 2009’, Fuel CellToday, U.K., April 2009:http://www.fuelcelltoday.com/media/pdf/surveys/2009-portable-free.pdf (Accessed on 24th June 2009)

The Reviewer

Don Cameron is an independent consultant onfuel cells and electrolysers. As well as scientificaspects, his interests include standardising andcommercialising these systems. He isSecretary of the Grove Symposium SteeringCommittee.


IntroductionThis is the third paper of four on the work

undertaken under the auspices of the PlatinumDevelopment Initiative (PDI), which was in oper-ation from April 1997 to October 2007, andcomprised Anglo Platinum, Impala Platinum,Lonmin (previously Lonrho) and Mintek. Thebackground and aims of the PDI are summarisedin (1, 2).

As well as the research work itself, the trainingof postgraduate students, and encouragement ofundergraduate and diploma students by the provi-sion of vacation work and experiential training,was considered an important output, since South

Africa has a shortage of scientists and engineers.This was seen as a great benefit, as were the collab-orations with other institutions, especially thoseoutside South Africa. During the time of the work,collaborations were started with NIMS (formerlyNRIM) in Japan, Fachhochschule Jena andBayreuth University in Germany, and Leeds,Cambridge and Oxford Universities in the U.K., aswell as four South African universities, which werethe Universities of the Witwatersrand, Cape Town,Limpopo (formerly the North) and the NelsonMandela Metropolitan University (formerly theUniversity of Port Elizabeth). The work is stillongoing, mainly as student projects, and to date


The Platinum Development Initiative:Platinum-Based Alloys for HighTemperature and Special Applications: Part IIIBy L. A. CornishDST/NRF Centre of Excellence in Strong Materials, Johannesburg 2050, South Africa,

and School of Chemical and Metallurgical 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]

L. H. ChownAdvanced Materials Division, Mintek, Private Bag X3015, Randburg 2125, South Africa,

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

and L. GlanerAdvanced 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

Under the Platinum Development Initiative, platinum-based alloys were being developedfor high-temperature and special applications for good corrosion and oxidation resistance.Work on ternary alloys had previously identified that the best of these systems for both mechanicalproperties and oxidation resistance were Pt86:Al10:Cr4 and Pt86:Al10:Ru4 (1), although themaximum precipitate volume fraction was only ~ 40% as opposed to ~ 70% achieved in nickel-based super alloys. Since Pt86:Al10:Cr4 and Pt86:Al10:Ru4 gave the best results, a range ofquaternary alloys were also made, using these compositions as a basis. The optimum compositionwas found to be around Pt80:Al14:Cr3:Ru3. Subsequently, further additions were made to thequaternary alloys to change selected properties.

DOI: 10.1595/147106709X464371

there have been eighteen postgraduate studentsinvolved, of whom fourteen have now completedtheir studies. In total, there has been involvementin seven Ph.D.s and eleven M.Sc.s, and over eightyjournal and conference papers have been published.

Development of a Quaternary AlloyUsing the platinum-aluminium system as a basis

for ordered face centred cubic (f.c.c.) precipitatesin an f.c.c. matrix (analogous to the nickel-basedsuperalloys), it was found that ternary alloying elements, and in particular chromium and rutheni-um, had additional benefits. More extensive workwas carried out on the phase relations, and Cr wasfound to stabilise the cubic form of the ~ Pt3Alphase, whereas Ru acted as a solid solutionstrengthener (3–5) However, the 2 at.% Ruamount did not stabilise the high temperature L12

form of Pt3Al, and the alloys needed Cr or another L12 stabiliser.

The composition of the quaternary alloy needed to be optimised so that the maximumproportion of the ~ Pt3Al second phase wasachieved. It was ultimately the objective to increasethe volume fraction of γ' to enhance the alloy’screep properties. Several alloys were thereforemanufactured with this objective (6), and the com-positions were selected based on the results of theternary Pt-Al-Cr and Pt-Al-Ru systems. The alloyswere prepared by arc-melting the pure elements

several times to achieve the highest possiblehomogeneity, and the samples were then heattreated at 1350ºC for 96 hours. Hardnesses of thealloys were measured using a Vickers hardnesstester with a 10 kg load. Some of the alloys weresingle-phase, and these showed cracking aroundthe indentations (6). Two alloys, Pt78:Al15.5:Cr4.5:Ru2

and Pt81.5:Al11.5:Cr4.5:Ru2.5, had large areas of~ Pt3Al, together with a fine mixture of (Pt), thesolid solution based on platinum, and ~ Pt3Al(Figure 1(a)). Another, Pt84:Al11:Cr3:Ru2, was com-posed entirely of a fine two-phase mixture, whichis the desired microstructure (Figure 1(b)).

More alloys were produced in an attempt toincrease the volume fraction of the ~ Pt3Al precip-itates further. After heat treatment (again, 96 hoursat 1350ºC in air), some oxidation took place, and,due to the small sample size, losses of aluminiumalso occurred. No improvement in the microstruc-ture was observed. The hardnesses of the alloyswere measured and the results are given in Table I.The alloys were reasonably ductile, with no crack-ing around the indentations, as was found in someof the earlier single-phase quaternary alloys (6).

In an attempt to improve the microstructure ofthe alloys, a second heat treatment was conductedfor 96 hours at 1350ºC, after which the alloyPt81.5:Al11.5:Cr4.5:Ru2.5 showed a clear, fine two-phase microstructure, possibly due to the changein its overall composition. There was no primary~ Pt3Al (so the overall composition is that of the


5 μm

(a)

5 μm

(b)

Fig. 1 SEM micrographs, in backscattered electron (BSE) mode, of the two types of two-phase alloys: (a) with primary~ Pt3Al (dark contrast) in a fine mixture of (Pt) and ~ Pt3Al; (b) fine mixture of (Pt) and ~ Pt3Al (6)

two-phase mixture: 85.2 ± 0.3 at.% Pt, 7.1 ± 0.8at.% Al, 3.1 ± 0.8 at.% Ru and 4.6 ± 0.1 at.% Cr).The loss of Al is concerning, especially after sucha short anneal, but this was the only alloy to suffersuch a large change. The precipitates inPt84:Al11:Cr3:Ru2 (Figure 1(b)) were approximatelytwice as large, but more well-defined than those ofPt85:Al7:Cr5:Ru3 alloys (6).

The Vickers hardnesses of the Pt:Al:Cr:Rualloys within the composition ranges selected wererelatively independent of both chemical contentand the number of annealing stages, and fell within the range HV10 ~ 400 to ~ 430. Hardnesseswere slightly lower after the second anneal. Thevolume fraction of ~ Pt3Al was estimated, usingimage analysis, to be approximately 25 to 30%.The highest hardness was found in the alloy with-out primary ~ Pt3Al. In the second batch ofquaternary alloys, there was no clear relationshipbetween the hardness and the composition or

microstructure. The decrease in hardness after thesecond heat treatment is likely to be due to thechanges in composition due to oxidation (6). Thebest alloy at this stage was Pt84:Al11:Cr3:Ru2, whichhad the required fine two-phase structure, with noprimary ~ Pt3Al and reasonable hardness. Similaralloys annealed at 1300ºC for 96 hours andquenched in water gave better all-round results forPt80:Al14:Cr3:Ru3 than Pt86:Al10:Cr4 because therewas more of the ~ Pt3Al phase (7–9), although itwas more coarse, as shown in Figure 2.

Oxidation ResistanceSince Pt86:Al10:Cr4 was very promising with

regard to high-temperature strength and oxidationresistance, it was decided to test a quaternary alloywith Ru as an addition to verify that the good oxidation resistance was retained. More Al wasadded in an effort to accelerate oxide scale forma-tion (10, 11). After one hour at 1350ºC, a thincontinuous oxide layer had formed. After 10hours’ exposure (Figure 3), the scale was already


Table I

Vickers Hardness of the Two-Phase QuaternaryAlloys, Using a 10 kg Load, after Annealing at1350ºC (6)

Alloy composition, Hardness Hardnessat.% after first after second

anneal, anneal, HV10 HV10

Pt85:Al11:Cr2:Ru2 430 ± 5 403 ± 20

Pt84:Al11.5:Cr2.5:Ru2 425 ± 21 403 ± 14

Pt83:Al11:Cr3.5:Ru2.5 421 ± 12 405 ± 8

Pt80.5:Al12.5:Cr4.5:Ru2.5 419 ± 22 414 ± 9

Pt81.5:Al11.5:Cr4.5:Ru2.5 423 ± 10 396 ± 6

Pt79.5:Al10.5:Cr5.5:Ru4.5 417 ± 8 415 ± 10

50 μm

Fig. 2 SEM image of Pt80:Al14:Cr3:Ru3 annealed at1400ºC for 96 hours and water quenched, showing~ Pt3Al (discrete raised) in a (Pt) matrix (8)

15 μm 15 μm 30 μm

(a) (b) (c)

Fig. 3 SEM images of the transverse sections of the Pt80:Al14:Cr3:Ru3 alloy after exposure to air at 1350ºC for anincreasing amount of time (11): (a) 1 hour; (b) 10 hours; (c) 500 hours

about three times as thick as that observed onPt86:Al10:Cr4 after the same time period. No zone ofdiscontinuous oxides, nor any other internal oxida-tion, was observed, as had been seen in some ofthe earlier ternary alloys (12, 13), Pt:Al:X where X = Re, Ta and Ti. The increased Al content of thealloys clearly accelerated the formation of a contin-uous layer, and prevented mass loss due tovolatilisation. Although this showed good proper-ties for the short test period, in the long term thealloy’s oxidation might be too severe. The alloyshould ideally form a continuous oxide layer quickly but then behave logarithmically with regardto mass increase.

Tensile TestingHigh-temperature compression and creep

results were promising, and the high-temperaturecompressive strength of Pt84:Al11:Cr3:Ru2 is signifi-cantly higher than that of Pt86:Al10:Cr4 (7–9).However, some potential manufacturers asked fortensile data, especially yield and ultimate tensilestress, since no prior data were available. Sinceresults become strain rate dependent very soon athigh temperatures, it was decided to only evaluatethe room temperature tensile properties of the bestternary alloys, compared with that of the quater-nary alloy (14).

Normal macro-scale tensile testing was notattempted because of the high material cost.Smaller specimens than the sub-size specimendescribed by the ASTM standard for tension test-ing (15) were therefore used. Small specimen testtechnology has been successfully utilised in fusionmaterials development, due to limited availabilityof effective irradiation volumes in test reactors(16). Utilising the required dimensional ratios ofthe ASTM standard (15), together with studies ofspecimen size effects (16–19) to ensure that thetest data would be comparable to those from stan-dard specimens, dimensions of 46 mm length and3 mm thickness were set for these experiments,with a gauge length of 18 mm and a gauge width of3 mm.

The specimens were prepared from 50 g ingots,manufactured by arc-melting, then aged in air in amuffle furnace at 1250ºC for 100 hours, then

quenched in water. This treatment produced ahomogeneous two-phase microstructure, withoutprimary ~ Pt3Al. The flat mini-tensile specimenswere machined from each ingot by wire spark erosion. Tensile tests were performed at a cross-head speed of 5 mm min–1. Unfortunately, themethod of testing and imperfections in some ofthe specimen shoulders rendered calibration of theextensometer impossible, resulting in the testsbeing carried out on a gauge length of 12.5 mmonly. This meant that strain could not be accurately measured, thus elongation could only beestimated by measurement of the distance betweengauge marks before and after testing, the latterbeing done after the fractured parts of the speci-mens were fitted together. Also, yield stress couldnot be determined. After testing, samples wereprepared metallographically, and then Vickershardness tests (20 kg load) were carried out. Someof the specimens failed outside the gauge length,and the results from these specimens were not carried further in the study. The average hardness-es, maximum ultimate tensile strength andestimated elongations are given in Table II. Thespread and inconsistencies in the results were disappointing. It would have been more ideal totest a wider range of specimens, but the price of Ptconstrained the number of specimens.Additionally, the results were unexpected, since itwas anticipated that Ru being a better solid solution strengthener in these alloys than Cr (20)would promote Ru alloys with a higher ultimatetensile strength. Thus, characterisation had to beundertaken to explain this discrepancy.


Table II

Tensile Testing Results for Platinum-AluminiumDerived Ternary and Quaternary Alloys at RoomTemperature (21)

Alloy Hardness, Maximum Elongation, composition, HV10 ultimate %at.% tensile strength

achieved, MPa

Pt86:Al10:Cr4 317 ± 13 836 ~ 4

Pt86:Al10:Ru4 278 ± 14 814 ~ 9

Pt84:Al11:Cr3:Ru2 361 ± 10 722 ~ 1

Using one half of the broken samples, scanningelectron microscopy (SEM) analyses underbackscattered electron (BSE) imaging using energydispersive X-ray (EDX) spectrometry were done.There were no significant differences between thetargeted and actual compositions, although thePt86:Al10:Cr4 alloy had discernable Ru content (lessthan ~ 0.1 wt.%) and the Pt86:Al10:Ru4 sample hada similar amount of Cr (21). This contaminationprobably arose from minor sputtering during melt-ing in the button-arc furnace. The fracturesurfaces of the other half of each sample were alsoexamined using SEM in secondary electron (SE)mode. Transmission electron microscopy (TEM)specimens were made and examined in a PhilipsCM200 TEM (21). X-Ray diffraction (XRD)analyses were conducted on the polished samplesusing molybdenum Kα radiation.

Microstructures were derived by SEM, XRDand TEM analyses, indicating that all samples con-tained both (Pt) and ~ Pt3Al precipitates. Thevolume fraction of the precipitates varied betweenspecimens and compositions. The Pt86:Al10:Cr4

specimens were stronger than those ofPt86:Al10:Ru4, because there was a very low volumefraction (~ 5%) of ~ Pt3Al precipitates in the alloywith Ru. Thus, it was deduced that Pt86:Al10:Ru4

had been annealed above its solvus, which wasthought to be between 1250ºC and 1300ºC. Thus,the higher ductility of this alloy was due to its near-ly single phase Pt solid solution. Pt86:Al10:Cr4 washarder and also had a higher ultimate tensilestrength than Pt86:Al10:Ru4. Having a significantvolume fraction of ~ Pt3Al, Pt84:Al11:Cr3:Ru2 wasthe hardest alloy, but had the lowest ultimate

tensile strength. The fracture surfaces (Figure 4)showed that only Pt84:Al11:Cr3:Ru2 failed intergran-ularly, with the ternary alloys failing mainly byintragranular cleavage with some localised signs ofdimpling. Thus, it is likely that the lower ultimatetensile strength was related to the intergranularfailure mode, which also correlates to its lowerelongation.

The results for these Pt-based alloys are summarised in Table III, together with values forpure Pt and selected high-temperature alloys. Thevalues of hardness and ultimate tensile strength forthe Pt-based alloys are higher than those of purePt in the soft state (i.e. not hardened). Comparedto other high-temperature alloys, such as the ferritic oxide dispersion strengthened (ODS) alloyPM2000, γ-titanium-aluminium and CMSX-4 (anickel-based superalloy), it is clearly demonstratedthat these Pt-based alloys are within the range ofthe high-temperature alloys in terms of ultimatetensile strength at room temperature. This findingis encouraging since the samples had not beenoptimised in terms of either heat treatment ormicrostructure.

Creep testing of the Pt84:Al11:Cr3:Ru2 alloy wasundertaken, and the results were worse than forPt86:Al10:Cr4. This was deduced to arise from a different atmosphere being used; initially the testswere undertaken under argon, and latterly the testswere done in air. However, the results ofPt84:Al11:Cr3:Ru2 were slightly worse than a com-mercial Pt alloy strengthened by dispersionhardening (DPH). The high-temperature com-pressive strength of Pt84:Al11:Cr3:Ru2 wassignificantly higher than that of Pt86:Al10:Cr4 (7–9).


300 μm 14 μm 80 μm

(a) (b) (c)

Fig. 4 (a) Representative secondary electron (SE)-SEM image of Pt84:Al11:Cr3:Ru2 showing intergranular fracture; (b) Representative SE-SEM image of both Pt86:Al10:Cr4 and Pt86:Al10:Ru4 showing cleavage fracture; (c) SE-SEM imageof Pt86:Al10:Ru4 showing localised dimples as a sign of ductility (21)

Other Alloying AdditionsOther additions such as cobalt or nickel have

also been tested to improve the properties of thealloys, and decrease their cost and density. For thepotential additions, phase diagram work was alsoundertaken, such as the Pt-Al-Co (26, 27) and Pt-Ni-Ru systems (28). Ni was added to improvethe solution strengthening of the matrix, althoughless solution strengthening was achieved thanhoped. Surprisingly, the melting point wasincreased by Ni additions. The work was not car-ried further because of the disappointing hardnessresults, and because work in Pt-Ni-based alloyswas ongoing in Germany (29–34).

Pt-Al-Co and Pt-Al-Co-Cr-Ru alloys were sub-jected to cold rolling on a small mill, and yieldedvery interesting results. Alloys with hardnesses

below 400 HV10 showed good cold formability(> 75% total reduction in thickness), whereas thecold formability was poor (< 40%) for hardnessesabove 450 HV10 (Figure 5).

The alloys with good formability were two-phase, with (Pt,Co) and Pt3Al phases, andcontained between 5 and 20 at.% Co and less than20 at.% Al. Excellent formability was obtained foralloys containing (Pt) and CoPt3, whereas the alloyscontaining other intermetallic phases showedextremely poor formability (< 5% total reduction).The formability (per cent total reduction) of thePt-Al-Co alloys improved sigmoidally with increas-ing per cent (Pt + Co) in the range 60 to 90 at.% (Pt + Co) (Figure 6). The cold formability of thesealloys is far superior to Pt-Al alloyed with Cr, Ruand Ni.


Table III

Hardness Range, Ultimate Tensile Strength and Elongation for Platinum-BasedAlloys, Pure Platinum and Selected High-Temperature Alloys

Alloy or metal Hardness range, Ultimate tensile Elongation, ReferencesHV strength at room %

temperature, MPa

Pt-based alloys 300–350 ~ 800 – (13, 21)

Pure Pt (soft state) ~ 40 ~ 140 – (22)

Ferritic ODS alloy – 720 14 (23)PM2000

γ-TiAl – 950 ~ 1 (24)

CMSX-4 – 870 – (25)

1000

800

600

400

200

0

Hard

ness, H

V1

0

50 60 70 80 90 100

(Pt + Co), at.%

Pt-Al-Co

Pt-Co-Al-Ru-Cr

Fig. 5 Hardness as a function of at.% (Pt + Co) for Pt-Al-Co alloys (26, 27, 35)

ConclusionsA range of platinum-based superalloys that

show very promising properties has been devel-oped by Mintek in collaboration with workers atSouth African universities. Two-phase γ/γ'microstructures, analogous to nickel-based super-alloys, consisting of cuboidal ~ Pt3Al precipitates in a (Pt) matrix, were achieved. The best alloy to date in terms of microstructure isPt84:Al11:Cr3:Ru2, since it had the required struc-ture with no primary ~ Pt3Al, and reasonable hardness. The optimum composition range isPt84:Al11:Cr3:Ru2 to Pt80:Al14:Cr3:Ru3. The oxida-tion resistance of Pt84:Al11:Cr3:Ru2 is better thaneither of the Cr- or Ru-containing Pt-Al-basedternary alloys included in the present study.However, there was concern that the alloy formedthe protective alumina film too quickly and thismight cause problems in the long term. Pt-basedalloys have hardnesses in the order of 300–350HV, with an ultimate tensile strength of ~ 800MPa. It is obvious that these Pt-based alloys have

ultimate tensile strength values within the samerange as other high-temperature alloys such asCMSX-4, which has an ultimate tensile strengthof 870 MPa. Pt-Al-Co alloys with hardnessesbelow 400 HV10 showed good cold formability(> 75% total reduction in thickness), whereas the cold formability was poor (< 40%) for hard-nesses above 450 HV10.

Part IV of the present series will appear in aforthcoming issue of Platinum Metals Review, andwill cover the corrosion of these platinum-basedalloys.

AcknowledgementsThe financial assistance of the South African

Department of Science and Technology (DST),the Platinum Development Initiative (PDI: Anglo,Impala, Lonmin and Mintek) and the DST/NRFCentre of Excellence in Strong Materials is grate-fully acknowledged.

This paper is published with the permission ofMintek.


100

80

60

40

20

0

Tota

l re

duction, %

50 60 70 80 90 100

(Pt + Co), at.%

Pt-Al-Co

Pt-Co-Al-Ru-Cr

Regression*

*Note:

%Red. = 9.04 + 88.69

1 + exp[121.36 – 1.43x]R2

= 0.89

Fig. 6 Total reduction (%Red.) as a function of at.% (Pt + Co) for Pt-Al-Co alloys (26, 27, 35)

1 L. A. Cornish, R. Süss, A. Douglas, L. H. Chown andL. Glaner, Platinum Metals Rev., 2009, 53, (1), 2

2 A. Douglas, P. J. Hill, T. Murakumo, L. A. Cornishand R. Süss, Platinum Metals Rev., 2009, 53, (2), 69

3 P. J. Hill, L. A. Cornish, M. J. Witcomb and P. Ellis,‘The (Pt)/Pt3Al Relationship in the Pt-Al-Cr and Pt-Al-Ti Systems’, in Proc. Microsc. Soc. south. Afr., Vol.30, Grahamstown, 6th–8th December, 2000, p. 13

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

5 P. J. Hill, N. Adams, T. Biggs, P. Ellis, J. Hohls, S. S. Taylor and I. M. Wolff, Mater. Sci. Eng. A., 2002,329–331, 295

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

References

7 T. Keraan and C. I. Lang, ‘High TemperatureInvestigation into Platinum-Base Superalloys’, in Proc.Microsc. Soc. south. Afr., Vol. 32, Cape Town, 3rd–5thDecember, 2003, p. 14

8 T. Keraan and C. I. Lang, ‘High TemperatureMechanical Properties and Behaviour of Platinum-Base Superalloys for Ultra-High Temperature Use’, in“African Materials Research Society Conference”,University of the Witwatersrand, Johannesburg, SouthAfrica, 8th–11th December, 2003, pp. 154–155

9 T. Keraan, “High Temperature Mechanical Propertiesand Behaviour of Platinum-Base Alloys”, M.Sc.Dissertation, University of Cape Town, South Africa,2004

10 P. J. Hill, L. A. Cornish and M. J. Witcomb,‘Microstructural Investigation of the Pt-BasedSuperalloy Pt80:Al14:Cr3:Ru3’, in Proc. Microsc. Soc.south. Afr., Vol. 31, Johannesburg, 5th–7th December,2001, p. 22

11 R. Süss, P. J. Hill, P. Ellis and L. A. Cornish, ‘TheOxidation Resistance of Pt-Base SuperalloyPt80:Al14:Cr3:Ru3 Compared to that of Pt86:Al10:Cr4’, inProc. Microsc. Soc. South. Afr., Vol. 31, Johannesburg,5th–7th December, 2001, p. 21

12 P. J. Hill, P. Ellis, L. A. Cornish, M. J. Witcomb andI. M. Wolff, ‘The Oxidation Behaviour of Pt-Al-XAlloys at Temperatures between 1473 and 1623 K’, in“Proceedings of the International Symposium on HighTemperature Corrosion and Protection 2000”,Sappora, Japan, 2000, Science Reviews, Japan, 2000,pp. 185–190

13 R. Süss, P. J. Hill, P. Ellis and I. M. Wolff, ‘TheOxidation Resistance of Pt-Base γ/γ' Analogues toNi-Base Superalloys’, in “7th European Conferenceon Advanced Materials and Processes”, Rimini, Italy,10th–14th June, 2001, Paper No. 287, CD-ROM,ISBN 8885298397

14 R. Süss, L. A. Cornish, L. H. Chown and L. Glaner,‘Tensile Properties of Pt-Based Superalloys’, presentedat “Beyond Nickel-Base Superalloys”, TMS 2004,133rd Annual Meeting and Exhibition of The Minerals,Metals and Materials Society, Charlotte, NorthCarolina, U.S.A., 14th–18th March, 2004 (Abstract)

15 ASTM Standard E8/E8M, 1993, “Standard TestMethods for Tension Testing of Metallic Metals”,ASTM International, West Conshohocken,Pennsylvania, U.S.A., 1993, pp. 130–149

16 G. E. Lucas, G. R. Odette, M. Sokolov, P. Spätig, T. Yamamoto and P. Jung, J. Nucl. Mater., 2002,307–311, Pt. 2, 1600

17 N. F. Panayotou, J. Nucl. Mater., 1982, 108–109, 45618 A. Kohyama, K. Asano and N. Igata, ‘Effects of

14-MeV Neutron Irradiations on MechanicalProperties of Ferritic Steels’, in “Influence of Radiationon Material Properties: 13th International Symposium(Part II)”, eds. F. A. Garner, C. H. Henager and N. Igata, STP 956, ASTM International, WestConshohocken, Pennsylvania, U.S.A., 1987, p. 111

19 Y. Kohno, A. Kohyama, M. L. Hamilton, T. Hirose,Y. Katoh and F. A. Garner, J. Nucl. Mater., 2000,283–287, Pt. 2, 1014

20 P. J. Hill, Y. Yamabe-Mitarai, H. Murakami, L. A.

Cornish, M. J. Witcomb, I. M. Wolff and H. Harada,‘The Precipitate Morphology and Lattice Mismatchof Ternary (Pt)/Pt3Al Alloys’, in “3rd InternationalSymposium on Structural Intermetallics”, TMS, SnowKing Resort, Jackson Hole, Wyoming, U.S.A.,September, 2001, rescheduled for 28th April–1st May,2002, pp. 527–533

21 R. Süss, A. Douglas, L. A. Cornish and B. Joja, ‘AnElectron Microscope Investigation of Tensile Samplesof Pt-Based Superalloys’, in Proc. Microsc. Soc. south.Afr., Vol. 34, Pretoria, 30th November–3rd December,2004, p. 10

22 The PGM Database, Platinum: http://www.platinummetalsreview.com/jmpgm/index.jsp

23 “Dispersion-Strengthened High-Temperature Mat-erials”, Prospectus, Plansee GmbH, Reutte, Tyrol,Austria, 2003

24 R. Pather, W. A. Mitten, P. Holdway, H. S. Ubhi andA. Wisbey, ‘Effect of High Temperature Environmenton High Strength Titanium Aluminide Alloy’, in“Advanced Materials and Processes for Gas Turbines”,eds. G. E. Fuchs, A. W. James, T. Gabb, M. McLeanand H. Harada, TMS, Warrendale, Pennsylvania,U.S.A., 2003, pp. 309–316

25 D. W. MacLachlan and D. M. Knowles, Mater. Sci.Eng. A, 2001, 302, (2), 275

26 L. H. Chown and L. A. Cornish, ‘The Influence ofCobalt Additions to Pt-Al and Pt-Al-Ru-Cr AlloySystems’, in “African Materials Research SocietyConference”, University of the Witwatersrand,Johannesburg, South Africa, 8th–11th December,2003, pp. 136–137

27 L. H. Chown, L. A. Cornish and B. Joja, ‘Structureand Properties of Pt-Al-Co Alloys’, in Proc. Microsc.Soc. south. Afr., Vol. 34, Pretoria, 30th November–3rdDecember, 2004, p. 11

28 L. Glaner and L. A. Cornish, ‘The Effect of NiAdditions to the Pt-Al-Cr-Ru System’, in Proc. Microsc.Soc. south. Afr., Vol. 33, Cape Town, 3rd–5thDecember, 2003, p. 17

29 S. Vorberg, M. Wenderoth, B. Fischer, U. Glatzel andR. Völkl, JOM, 2004, 56, (9), 40

30 S. Vorberg, M. Wenderoth, U. Glatzel, B. Fischer andR. Völkl, JOM, 2005, 57, (3), 49

31 M. Hüller, M. Wenderoth, S. Vorberg, B. Fischer, U. Glatzel and R. Völkl, Metall. Mater. Trans. A., 2005,36, (13), 681

32 M. Wenderoth, L. A. Cornish, R. Süss, S. Vorberg,B. Fischer, U. Glatzel and R. Völkl, Metall. Mater. Trans.A, 2005, 36, (3), 567

33 R. Völkl, Y. Yamabe-Mitarai, C. Huang and H. Harada,Metall. Mater. Trans. A, 2005, 36, (11), 2881

34 M. Wenderoth, R. Völkl, T. Yokokawa, Y. Yamabe-Mitarai and H. Harada, Scr. Mater., 2006, 54, (2), 275

35 R. Süss, A. Douglas, L. H. Chown, L. Glaner, N. Maledi, W. Tshawe and L. A. Cornish, ‘Pt-BasedAlloys for Aggressive Environments’, “AdvancedMetals Initiative Conference”, Gold Reef City,Johannesburg, South Africa, 18th–19th November,2008, on CD-ROM



The Authors

Lizelle Glaner is a Principal Technicianin the Advanced Materials Division atMintek, where she is in charge of theNano Characterisation Laboratory.She has worked on gold catalysts,gold jewellery alloys and platinumalloys.

Lesley Chown is a Principal Engineerin the Physical Metallurgy Group in theAdvanced Materials Division at Mintek.She has worked on continuously caststeels, platinum alloys and titaniumalloys.

Rainer Süss is a Chief Engineer in thePhysical Metallurgy Group in theAdvanced Metals Division at Mintek,as well as the co-ordinator of theStrong Metallic Alloys Focus Area inthe DST/NRF Centre of Excellence forStrong Materials. His researchinterests include phase diagrams,platinum alloys and jewellery alloys.

Lesley Cornish is a Professor at theUniversity of the Witwatersrand,South Africa, and is Director of theDST/NRF Centre of Excellence forStrong Materials, which is hosted bythe University of the Witwatersrand,and the African Materials Science andEngineering Network (AMSEN). Herresearch interests include phasediagrams, platinum alloys andintermetallic compounds.


The 5th International Conference onEnvironmental Catalysis (5ICEC) was held atQueen’s University in Belfast, Northern Ireland,from 31st August to 3rd September 2008. Theconference was split into five main categories overthe course of three days: Autocatalysis, CleanEnergy, Renewables, Green Chemistry and Air andWater. In addition to the oral presentations, therewere several plenary talks given by keynote speak-ers, together with poster sessions and vendorbooths. This review will focus on the talks relatedto the use of noble metals, including the platinumgroup metals (pgms), in autocatalyst applications.The review can be split into three sections: the firstdetails the use of silver/alumina (Ag/Al2O3) forthe selective catalytic reduction of nitrogen oxides(NOx) using hydrocarbons (HC-SCR); the seconddiscusses the use of platinum, palladium, rhodiumor iridium in lean NOx traps (LNT) and the thirdis concerned with platinum, palladium and rhodi-um used in NOx storage.

Hydrocarbon-Selective CatalyticReduction

HC-SCR is a highly desirable technology for theremoval of NOx from diesel and lean-burn gasolineexhausts, although it has proven difficult to devel-op a suitably active commercial catalyst. One of themost promising candidates is silver supported onalumina, where good SCR activity can be obtainedby the addition of hydrogen. This is referred to asthe ‘hydrogen effect’ (1, 2). The activation ofAg/Al2O3 by hydrogen has been a matter of somedebate, and no clear explanation has yet been pro-vided. It is generally thought that hydrogen causessilver to form active silver clusters; however, it hasbeen reported that the presence of silver clusters by

itself is not sufficient for good SCR activity (3–5).For example, it has been demonstrated that a simi-lar hydrogen effect was obtained through the use ofhydrogen peroxide (H2O2), which suggests that ahydroperoxy-like species is important for hydro-gen-promoted HC-SCR (6).

Atsushi Satsuma (Nagoya University, Japan)argued in his presentation ‘Role of Ag Cluster onHydrogen Effect of HC-SCR over Ag/Alumina’that the silver clusters are not necessary for goodSCR activity, but rather promote the formation ofthe active H2O2 species that is essential for the acti-vation of reactants in HC-SCR. The formation ofactive oxygen species (O2

– ) was measured by elec-tron spin resonance (ESR) to determine whethersilver clusters and hydrogen are required for thepromoted HC-SCR to take place. It was arguedthat an O2

– species is essential for the hydrogeneffect due to its high activity towards partial oxida-tion of hydrocarbons. The lack of detectable O2

–

species when silver clusters were treated in a mix-ture of carbon monoxide and oxygen followed byaddition of oxygen indicated that the presence ofhydrogen was required. Similarly, no O2

– specieswere observed on 0.5 wt.% Ag/Al2O3 catalystswhere no silver aggregation has taken place, evenafter reduction by hydrogen. It was suggested thatthis is a consequence of having the silver speciesfixed as Ag+ on the alumina surface.

The O2– species were only detected on the

2 wt.% Ag/Al2O3 that contained silver clustersafter exposure to a flow of hydrogen and oxygenfollowed by addition of oxygen (7). The reactionmechanism was evaluated by density functionaltheory (DFT) calculations using silver supportedon a MFI-type zeolite (Ag/MFI) as the model cat-alyst. The reaction is suggested to occur via the

5th International Conference onEnvironmental CatalysisPGM-BASED TECHNOLOGIES FOR NOx ABATEMENT

Reviewed by Rodney Foo* and Noelia Cortes Felix**Johnson Matthey Technology Centre, Blounts Court, Sonning Common, Reading RG4 9NH, U.K.; E-mail: *[email protected];

**[email protected]

DOI: 10.1595/147106709X465488

activation of oxygen by [HHAg4] clusters to formHOO[HAg4] clusters, which in turn leads to theformation of H2O2 species. This was to be report-ed in further detail in a paper pending submission(at the time of writing) by K. Sawabe, T. Hiro, Y. Iwata, K. Shimizu and A. Satsuma.

Similar Ag42+ clusters were detected by ultra-

violet-visible (UV-vis) spectroscopy on both theAg/MFI and Ag/Al2O3 catalysts. However, alarger contribution by Ag-O was found with thelatter catalyst. Further details on the nature of thesilver structure as analysed through the use ofextended X-ray absorption fine structure (EXAFS)and UV-vis spectroscopy can be found inReference (8).

During the discussions that followed the pre-sentation, it was mentioned that the active silverclusters may not necessarily be solely tetrahedralAg4 clusters, as there could be a range of clustersizes where the average is Ag4. The important para-meters are the agglomeration and charge of thesilver cluster. The possible differences between sil-ver supported on the MFI zeolite and supportedon alumina were also discussed, as the active silverspecies and mechanism of reaction may differbetween the two.

The next presentation, by Pyung Soon Kim(Pohang University of Science and Technology,Korea), was titled ‘Selective Catalytic Reduction ofNOx by Simulated Diesel Fuel ContainingOxygenated Hydrocarbon’. The presentation dealtwith methods to improve the low-temperatureactivity of Ag/Al2O3 catalysts through the use ofoxygenated hydrocarbons (OHCs) that have beenreported to lower the light-off temperatures, aswell as preventing catalyst deactivation by coking(9). The NOx reduction activity was investigated ina packed-bed reactor, using a mixture of simulateddiesel fuel (dodecane and m-xylene) and ethanol asa reductant. The testing was conducted at a spacevelocity of 60 h–1 in a mixture of 6% oxygen, 2.5% water vapour and helium, where the C1:NOxratio varied from 2 to 8 over a dual-bed catalystsystem with a 0.5 inch coating of Ag/Al2O3 catalystin the front followed by 0.5 inch of copper-exchanged ZSM-5 zeolite (Cu/ZSM-5) catalyst inthe rear bed.

Ammonia (NH3) was found to be a majorbyproduct, in particular when high amounts ofethanol were used in the fuel mixture. Thereforethe Cu/ZSM-5 catalyst was added as an NH3

oxidation component and to improve the high-temperature activity, as previously reported byLong and Yang (10). The influence of the catalystpreparation procedure and operating conditionswere investigated in order to optimise the dual-bedcatalyst system. The Ag/Al2O3 catalysts were char-acterised by various techniques, includingtransmission electron microscopy (TEM), UV-visspectroscopy, hydrogen temperature-programmedreduction (H2-TPR) and X-ray (excited) photoelec-tron spectroscopy (XPS).

Increasing the silver loading improved the low-temperature activity but reduced the high-temperature activity. For high-temperature use, theoptimum loading was determined to be 3.8 wt.%Ag. A similar effect was observed with the additionof ethanol, which improved low-temperature activ-ity by around 25% but negatively impacted theactivity at temperatures above 350ºC. A loss oflow-temperature activity was observed when theC1:NOx ratio was greater than 8; this is most like-ly due to the build-up of hydrocarbons coking thecatalyst.

The drying temperature used during the prepa-ration of the Ag/Al2O3 catalyst was found to havea significant impact on its catalytic activity. TheAg/Al2O3 catalyst exhibited poor activity for NOxconversion to nitrogen (N2) when dried at 65ºC,but was active for the unselective catalytic forma-tion of NH3. It was reported that the catalysts hadto be dried at temperatures above 80ºC to achievegood conversion of NOx to N2, and the best per-formance was obtained by using a dryingtemperature of 110ºC. The TEM and UV-vis char-acterisation results showed that the catalyst dried at65ºC had larger particle sizes (~ 14 nm) and awider particle size distribution, while the 110ºC-dried catalyst contained less metallic silver, and hada smaller average particle size (~ 8 nm) and nar-rower particle size distribution.

XPS results were also presented. Kim conclud-ed from these that the ionic silver species were theactive sites for NOx reduction to N2, and that the


metallic silver sites were active for NOx reductionto NH3. However, this contradicts the other char-acterisation results, and together with the largeexperimental errors reported, this would call intoquestion the accuracy and validity of the XPSresults. Further, during the discussions, it was sug-gested that the most likely influence of theCu/ZSM-5 catalysts was not NH3 oxidation, butrather its activity for NH3-SCR, which leads to animprovement in the high-temperature activity ofthe catalyst system.

Continuing with presentations based on HC-SCR studies over a Ag/Al2O3 catalytic bed, theinfluence of biodiesel fuel over a microreactor wasdiscussed next. Biodiesel is an interesting andgrowing area of research, in particular as the EUhas mandated the use of certain percentages ofbiodiesel in vehicle fuel, with set targets for 2010(11). However, there are a number of challengesinvolved in the use of biodiesels, including the factthat they are known to produce more NOx thanoil-derived diesel. The presentation given here, byJosé R. Hernández (Åbo Akademi University,Finland), was entitled ‘A Combination between aLow- and a High-Temperature Catalyst for theSCR of NOx Using Second-Generation Biodieselin Microchannels’.

The catalytic testing was conducted on amicroreactor where ten plates were coated witheither Ag/Al2O3 or Cu/ZSM-5 catalysts (fiveplates of each) in a specific arrangement, and theproducts were measured using a micro gas chro-matograph, as described in the presenter’spreviously published paper (12). The Cu/ZSM-5catalyst was found to significantly widen the tem-perature window, especially in the low-temperatureregion. The cause of the promotion effect whenboth Ag/Al2O3 and Cu/ZSM-5 catalysts are pre-sent is not fully understood, but mechanisticstudies will be carried out in the future.

Hexadecane was used as the model fuel forsimulated second-generation biodiesel, based onthe group’s previous work (13). The use of hexa-decane/biodiesel was found to improve thelow-temperature activity, but resulted in the loss ofhigh-temperature activity. The hexadecane con-centration was shown to be an important factor, as

the catalytic system could be optimised by control-ling the C1:NOx ratio. It was reported that,depending on the activity window, the C1:NOxratio should be 6 for low-temperature activity, 9for mid-range-temperature activity, and 12 forhigh-temperature activity.

This presentation invited many questions, inparticular regarding the relevance of results gener-ated using a microreactor. Hernández stated thatthe use of a microreactor improves energy effi-ciency and gives a better yield, increased safety anda better degree of process control. He also assert-ed that similar conditions are found in a carexhaust system to those in a microreactor.However, it was discussed that this may not alwaysbe the case, as the fact that the microreactor’sresults are based on changes in very small percent-ages of conversion means that they are prone tohigher levels of inaccuracy. Furthermore, addition-al factors such as mass transfer limitations andexotherms were not addressed. However, it wasrecognised that limitations in test equipment andreactants, as well as experimental convenience,favour the use of microreactors over testing onlarger monolith reactors.

Lean NOx TrapsThe next presentation, ‘A NOx Reduction

System Using Ammonia-Storage SelectiveCatalytic Reduction in Rich/Lean Excursions’, byTadao Nakatsuji (Okayama University, Japan),dealt with a novel approach to the lean NOxreduction catalyst system. One of the major disad-vantages of the lean NOx trap (LNT) system is itspoor low-temperature performance and suscepti-bility to poisoning by species such as sulfur oxides(SOx) that require high-temperature regenerationprofiles. Furthermore, the LNT only generatesnitrogen during the rich phase, as the NOx isstored during the lean phase.

To address this limitation, a transient catalyticNOx reduction catalyst was developed byNakatsuji’s group, containing a platinum-basedoxygen storage component (OSC) and a solid acidcapable of reducing NOx to N2 in both the leanand rich phases. This is an extension of the LNTconcept with an in situ NH3 formation capability in


order to carry out NH3-SCR reactions. The doublelayer catalyst consists of a solid acid top layer com-posed of 20% H-mordenite and 80% Pt/CeO2, anda bottom layer composed of either 2% Pt/Ce-Pr-ZrOx or 2% Pt/(75% Al-21% BaCO3-2% K2CO3).The ratio between the top layer for NH3 storageand the bottom layer for NOx storage was 1:3.

The reaction mechanism over the new catalystsystem is proposed to proceed through four mainreaction pathways. In the initial lean phase, theNOx species adsorb onto the OSC component.These are reduced to NH3 over the platinum cata-lyst during rich operation, and the NH3 is thenadsorbed onto the solid acid layer. Finally, thestored NH3 is used to reduce the NOx species dur-ing the lean phase, therefore allowing NOxreduction to take place during both the lean andrich phases.

The in situ Fourier transform infrared (FTIR)spectroscopy results shown during Nakatsuji’s pre-sentation supported the proposed mechanism, inwhich only nitrates are formed during the first thir-ty seconds, then these rapidly turn into ammonium(NH4

+) species after a further thirty seconds. Aftersixty seconds, the NH4

+ concentration begins todecrease, while concentration of N2 increases,together with the appearance of some NOx. Basedon the N2 profiles exhibited during the presenta-tion, there were potentially some selectivity issues.An example of one of the N2 profiles is sketchedin Figure 1.

Akira Obuchi (National Institute of AdvancedIndustrial Science and Technology (AIST), Japan)presented another interesting novel reactor design

in his talk ‘CO and NH3-Combined SCR with anInternal Heat Exchanging Reactor’. On newerengine designs, there is a trend towards emissionof exhaust gases at lower temperatures to improvefuel economy. As a consequence of this, catalyticreactions are more difficult; therefore a new reac-tor was developed to integrate a heat exchangingfunction with catalytic activity. Further details onthe reactor design can be obtained from severalpatents and papers (14–17). This is a challengingprocess, as combining a catalytic reactor with aheat exchanger system can cause a large drop inpressure and reduce overall efficiency.

The basic design of the reactor system consist-ed of different types of catalysts washcoated ontoseveral bundled stainless steel sheets at variouspositions in the reactor. The heat is recovered byuse of a counterflow system where the inlet gas isheated by the hot gases generated by the exother-mic oxidation reaction towards the centre of thereactor. See Figure 2 for a brief outline of one pos-sible configuration of the reactor scheme withthree SCR catalysts.

Obuchi used two different SCR catalysts(Cu/ZSM-5 for NH3-SCR, iridium and barium onWO3/SiO2 for CO-SCR) with Pt/Al2O3 as the oxi-dation catalyst for combustion of CO to generatethe exotherm. In general, there was a ratio of 20 gof Cu/ZSM-5 to each 5 g of IrBa/WO3-SiO2 and3 g of Pt/Al2O3 towards the outlet end of the reac-tor system. There are several different possibleconfigurations, depending on the positions of thecatalysts, whether the sheets are coated on one orboth sides, and whether one or two different


Rich phase

Time

Nitro

gen c

oncentr

ation

Lean phase

Fig. 1 Nitrogen profile over a modified lean NOxtrap

catalysts are used. The temperature in the reactionzone can be raised to 320ºC with an inlet temper-ature of 100ºC, which corresponds to an 80% heatrecovery. Under CO-SCR, the NOx conversionwas 25% (NOx inlet of 100 ppm) which can beraised to 75% when 90 ppm NH3 is added to allowNH3-SCR to take place.

The results show a potential for heat recovery,compared to the standard monolith reactor design,that will benefit the catalytic activity of the new sys-tem. However, currently between 35 W and 55 Wof electrical heating is required at the ends of thereactor to boost the activity of the system and theCO concentration has to be increased artificially.

The last pgm-related presentation of the firstday was given by Christophe Dujardin (Unité deCatalyse et de Chimie du Solide, Université desSciences et Technologie de Lille, France) and dis-cussed the ‘Nature and Evolution of PalladiumSpecies on Pd-LaCoO3 and Pd-Al2O3 on theCourse of the Reactions NO+H2+O2 Followedwith Operando EXAFS’. Perovskite-supported pgmshave been of increasing interest as potential alterna-tives to ceria-zirconia supported pgms for de-NOxapplications (18, 19). The results presented hereshowed that there are stronger interactionsbetween palladium and oxide sites on perovskitesthan on alumina, which suggests that perovskitehas a stabilising effect. The reduction of palladiumto its metallic form under hydrogen was also shown

to be delayed on perovskites compared to alumina. The catalysts were synthesised using the sol-gel

technique, with addition of 1 wt.% palladium byimpregnation, and fired at 400ºC followed by a pre-reduction in hydrogen at 250ºC. The reactionconditions were 0.1% nitric oxide (NO), 0.5%hydrogen and 3% oxygen in helium. It was interest-ing to note that pgms supported on perovskiteexhibit better catalytic activity compared to thosesupported on silica, despite their lower surface area.

It has been proposed that palladium may actu-ally be inserted into the framework of theperovskite, based on the higher XPS binding ener-gy of PdO at 500ºC. However, this is still beingdebated (20). Results of in situ XPS studies werepresented that showed preservation of theLaCoO3 structure and the conversion of Co3+ toCo2+ and Co0 when reduced and re-oxidised underthe reaction conditions.

NOx StorageThe second day of the conference focused on

one specific area of NOx control: NOx storage.The first talk of the day was by William Epling(University of Waterloo, Canada; in collaborationwith Cummins, Inc, U.S.A.) and was entitled‘Evaluating Axial Distributions of Species andTemperature on Monolith-Supported Catalysts viaSpatially-Resolved Calorimetry’. The technique ofIR thermography, which has been successfully


Outlet

Inlet

Temperature

SCR

Catalyst 1

SCR

Catalyst 2

SCR

Catalyst 3

Oxidation

Catalyst

Fig. 2 A possible layout ofthe combined CO- and NH3-

selective catalytic reductionreactor with an internal heatexchanger, using three SCRcatalysts

used previously to characterise oxidation catalysts,was used to measure the axial distribution ofnitrate species as a function of lean-phase time andtemperature, with either NO2 or NO as the NOxsource, on both model and commercial dieselNOx adsorber catalysts.

The next presentation was ‘Intermediate NH3

Generation and Utilization Inside a Lean NOxTrap Catalyst’ by Jae-Soon Choi (Oak RidgeNational Laboratory, Oak Ridge, U.S.A.). Choiexplained how NH3 is created and where and howit is used in a LNT catalyst. Intra-catalyst measure-ments were performed using spatially resolvedcapillary-inlet mass spectrometry (SpaciMS). Lean-rich cycling tests (cycles of 60 s lean, 5 s rich) werecarried out on a Pt/Ba/Al2O3 washcoated catalyst.NOx, N2, NH3 and H2 were measured over timeand it was seen that the relative amounts of andselectivity towards each species were different atdifferent points along the channel of the monolith.Working at different temperatures, it is possible tosee clearly the evolution of the different speciesduring the reaction. At low temperatures (200ºC)there is a large amount of NOx stored on the cat-alyst, and the NH3 that was formed upstream issimultaneously consumed to reduce the down-stream NOx. However, increasing the temperatureto 325ºC reduces the concentration of NH3, asNOx is less stable at this temperature and the localH2:NOx ratio is also lower. This observation sup-ports the important role of the H2:NOx ratio inNH3 formation. The results obtained suggest thattwo possible mechanisms exist, depending on theoperational temperature. At lower temperatures,NH3 is formed as an intermediate (Equation (i))that reduces the NOx stored in the catalyst(Equations (ii) and (iii)). At higher temperatures,H2 reacts with the stored NOx, reducing it directlyto N2 (Equation (iv)):

Ba(NO3)2 + 8H2 → 2NH3 + BaO + 5H2O (i)

3Ba(NO3)2 + 10NH3

→ 8N2 + 3BaO + 15H2O (ii)

3Pt-O + 2NH3 → N2 + 3Pt + 3H2O (iii)

Ba(NO3)2 + 5H2 → N2 + BaO + 5H2O (iv)

This study showed that NH3 plays an importantrole in the mechanism of a LNT.

After this interesting talk, Hiroyuki Matsubara(Toyota Motor Corporation, Japan) presented‘Novel CeO2-Al2O3 Nano Composite forInhibiting Platinum Sintering of NSR Catalyst’. Itis well known that platinum sintering by thermalageing decreases NOx storage capacity. Matsubaraand colleages have developed a new material basedon the strong interaction between platinum andceria to inhibit this effect. After the incorporationof ceria nanoparticles into the matrix of a supportwith high surface area, in this case alumina, theyobtained better platinum dispersion and smallerparticle size, which improved the NOx storageperformance of the catalyst, especially at tempera-tures lower than 300ºC.

The other major problem with the use of NOxtraps is poisoning by sulfur present in diesel fuel.‘Kinetics of Sulfur Removal from a CommercialLean NOx Trap Catalyst’ was presented byAleksey Yezerets (Cummins Inc, U.S.A.) in part-nership with the Emission Control Technologiesgroup of Johnson Matthey, U.S.A. Yezerets andcoworkers found one additional effect related tothe nature of the sulfur deposited in LNT catalysts.Different forms of sulfur can have different effectson NOx performance, as well as different kineticsfor its removal. The most important species foundwere the sulfate species, which were shown to beremovable at low temperature. The following para-meters can affect the desulfation process: (a) initial sulfur concentration (b) desulfation temperature(c) reductant concentration. This can be summarised as a mathematical expres-sion (Equation (v)):

where r is the rate, A is the frequency factor (s–1),Ea is the activation energy (137 ± 8 KJ mol–1), Tis the desulfation temperature (K),is the reductant concentration (mol l–1),the instantaneous sulfur concentration (mol l–1),n ( i) is the reaction order at that moment in timeand no is the initial reaction order.

[reductant ] n ( i)


r = A × exp(–Ea/8.314 T )× ( [reductant ] (v)) × [S ] non ( i)

[S ] ison

Todd J. Toops (Oak Ridge National Laboratory,Knoxville, U.S.A.) talked about ‘Thermal Aging ofLean NOx Trap Catalysts Using Reactor-Generated Exotherms and the Resulting MaterialEffects’. In this talk, Toops showed that there is acorrelation between ageing temperature and sur-face area losses. Four ageing temperatures (700ºC,800ºC, 900ºC and 1000ºC) were employed overthe course of several hundred ageing cycles. At700ºC, there was no impact on the performance ofthe catalyst at low temperature, at 800ºC therecould have been a slight influence, and at ageingtemperatures above 800ºC a significant decrease inNOx activity and surface area was clearly seen.This was explained by the transition of the alumi-na support from gamma to delta form at 860ºC.

The last day of the conference started with apresentation entitled ‘Model NOx StorageCatalysts: Reaction Mechanisms and Kinetics atthe Microscopic Level’ by Jörg Libuda (Universityof Erlangen-Nuremberg, Germany) and col-leagues. A catalyst based on Pd/Ba/Al2O3

was tested using a combination of scanning tunnelling microscopy, high-resolution photoelec-tron spectroscopy using synchrotron radiation,time-resolved IR reflection absorption spec-troscopy (TR-IRAS), multimolecular beammethods, and combined reactor/TR-IRAS exper-iments up to ambient pressure conditions. Libudaand his coworkers suggested two possible mecha-nisms for NOx storage after NO2 exposure,depending on the temperature: at low tempera-tures (~ 100ºC) a cooperative mechanism wasproposed, in which nitrites are formed first fol-lowed by nitrates; at high temperatures (≥ 300ºC),it was found to be a non-cooperative mechanism,where all the NOx is stored as nitrite. Palladiumparticles were found to be partially covered withBaAl2O4. Depending on the structure of theseparticles, the propensity for nitrite formation willdiffer. The adsorption properties of the palladiumparticles depend on the degree of surface oxida-tion, which changes as a function of particle sizeand reaction conditions.

The second presentation was entitled ‘NSRCatalyst Supported on an Al2O3/ZrO2-TiO2

Nano-Composite: Sulphur Resistance’ and was

presented by Naoki Takahashi (Toyota CentralR&D Labs, Inc, Japan). This study compared aNOx storage and reduction (NSR) catalyst basedon two different supports: Ba-K/Pt-Rh/AZT(AZT = 50 wt.% Al2O3, 35 wt.% ZrO2 and 15 wt.%TiO2) and Ba-K/Pt-Rh/A-ZT (A-ZT = physicalmixture of Al2O3 and ZT powders, where ZT is aZrO2-TiO2 solid solution). Both catalysts wereexposed to a sulfur dioxide-containing oxidisingatmosphere at 600ºC for 30 min and cooled to100ºC under flowing nitrogen. Then, under reduc-ing conditions, the sample was heated to 800ºC. Itwas found that the sample based on AZT had ahigher sulfur tolerance than that on A-ZT, due toa larger surface contact between Al2O3 and ZTcompared to the physical mixture.

Four-way catalysts allow diesel vehicles toreach Euro 5 legislation while at the same timedecreasing the space required by the exhaustaftertreatment system. Claire-Noelle Millet (Institut Français du Pétrole-Lyon, France) andcoworkers performed a global kinetic study offour-way catalysts: Pt/Ba/ZrO2/Al2O3. Their pre-sentation, entitled ‘Synthetic Gas Bench Study of a4 Way Catalytic Converter: Catalytic Oxidation,NOx Storage/Reduction and Impact of SootLoading and Regeneration’, followed the reactionsthat occur during lean and rich periods. It wasfound that NOx storage appeared to be inhibitedby carbon monoxide and hydrocarbon oxidationwith nitrogen dioxide (NO2) at low temperatures,possibly due to competition with carbon dioxidefor storage sites produced during hydrocarbonoxidation. Soot deposition also affected the cat-alytic NOx storage reaction. Continuous oxidationof soot by NO2 induced a slower NOx storagerate.

Sandra Capela (Instituto Superior Técnico,Lisboa, Portugal and Laboratoire de Réactivité deSurface, Université Pierre et Marie Curie, Paris,France) and coworkers presented ‘An Operando5.8 GHz Microwave-Heated FT-IR Reactor Studyof the NO2-CH4 Reaction, over a Co/Pd-HFERCatalyst’. HFER is the hydrogen form of ferrieritezeolite. The group performed two tests. The firstconsisted of a temperature-programmed surfacereaction (TPSR), in which a reaction mixture of


500 ppm NO2 and 5000 ppm methane (CH4) inargon was supplied to an IR cell-reactor whileincreasing the temperature from 110ºC to 420ºC.At ~ 180ºC, presence of formaldehyde was detect-ed by the vibration of C=O at 1745 cm–1. –NCOspecies were also detected at about 250ºC. The sec-ond test used the same reaction mixture with a5.8 Hz microwave-heated IR reactor-cell, previ-ously stabilised at either 80ºC or 200ºC byconventional means. Formaldehyde was againformed as a primary product of the reactionbetween NO2 and CH4; CO was also detected atthe surface of the catalyst, probably due to the par-tial oxidation of CH4 assisted by nitrate-likespecies. The use of a microwave field means that itis possible to heat the cell very quickly, whichallows results to be obtained faster than with aconventional operando cell.

ConclusionsThe 5th International Conference on

Environmental Catalysis covered a wide range oftopics, with autocatalysis being of particular interest.Autocatalysis currently accounts for a large propor-tion of the world’s pgm use and it is important tokeep up to date on the latest developments poten-tially leading to reduced pgm use. Several pertinenttopics were covered in the conference and it wasinteresting to compare academic research withindustry requirements and knowledge, as surprisingdifferences between the two were highlighted. Theconference is a useful forum for academics andindustrial representatives to meet and discuss theongoing research in pgm use in automotive catalysis.

The conference programme and abstracts of allthe papers presented are available on the confer-ence website (21).


1 S. Satokawa, Chem. Lett., 2000, (3), 294 2 T. Furusawa, L. Lefferts, K. Seshan and K. Aika, Appl.

Catal. B: Environ., 2003, 42, (1), 253 A. Satsuma, J. Shibata, K. Shimizu and T. Hattori,

Catal. Surv. Asia, 2005, 9, (2), 754 J. P. Breen, R. Burch, C. Hardacre and C. J. Hill, J.

Phys. Chem. B, 2005, 109, (11), 48055 P. Sazama, L. Capek, H. Drobná, Z. Sobalik, J.

Dedecek, K. Arve and B. Wichterlová, J. Catal., 2005,232, (2), 302

6 P. Sazama and B. Wichterlová, Chem. Commun., 2005,(38), 4810

7 K. Shimizu, M. Tsuzuki, K. Kato, S. Yokota, K.Okumura and A. Satsuma. J. Phys. Chem. C, 2007, 111,(2), 950

8 J. Shibata, K. Shimizu, Y. Takada, A. Shichi, H.Yoshida, S. Satokawa, A. Satsuma and T. Hattori, J.Catal., 2004, 227, (2), 367

9 K. Shimizu, M. Tsuzuki and A. Satsuma, Appl. Catal.B: Environ., 2007, 71, (1–2), 80

10 R. Q. Long and R. T. Yang, Chem. Commun., 2000, (17),1651

11 ‘The promotion of the use of biofuels or otherrenewable fuels for transport’, Directive 2003/30/ECof the European Parliament and of the Council of 8May 2003, Off. J. Eur. Union, 17th May, 2003, L 123/42:

http://ec.europa.eu/energy/res/legislation/doc/biofuels/en_final.pdf

12 J. R. Hernández Carucci, K. Arve, K. Eränen, D. Yu.Murzin and T. Salmi, Catal. Today, 2008, 133–135, 448

13 I. Kubicková, M. Snåre, K. Eränen, P. Mäki-Arvelaand D. Yu. Murzin, Catal. Today, 2005, 106, (1–4), 197

14 E. Jobson and B. Heed, Volvo AB, ‘CatalyticPurification Device’, European Appl. 1,016,777; 2000

15 G. Gaiser, ‘Reactor for Catalytically ProcessingGaseous Fluids’, World Appl. 93/22,544; 1993

16 G. Kolios, A. Gritsch, A. Morillo, U. Tuttlies, J.Bernnat, F. Opferkuch and G. Eigenberger, Appl. Catal.B: Environ., 2007, 70, (1–4), 16

17 A. Obuchi, J. Uchisawa, A. Ohi, T. Nanba and N.Nakayama, Top. Catal., 2007, 42–43, (1–4), 267

18 C. N. Costa, V. N. Stathopoulos, V. C. Belessi andA. M. Efstathiou, J. Catal., 2001, 197, (2), 350

19 M. Pirez-Engelmann, P. Granger, L. Leclercq andG. Leclercq, Top. Catal., 2004, 30–31, (1), 59

20 M. Uenishi, M. Taniguchi, H. Tanaka, M. Kimura,Y. Nishihata, J. Mizuki and T. Kobayashi, Appl. Catal.B: Environ., 2005, 57, (4), 267

21 5th International Conference on EnvironmentalCatalysis, CenTACat website:http://www.centacat.qub.ac.uk/5icec/ (Accessed on3rd June 2009)

References

Dr Rodney Foo has a Ph.D. in Chemical Engineering from theUniversity of Bath and is a Senior Scientist at the JohnsonMatthey Technology Centre, Sonning Common, U.K., working ingas phase catalysis and on the development of new automotivecatalysts. He specialises in NOx control via selective catalyticreduction.

Noelia Cortes Felix studied Chemical Science at the AutonomaUniversity of Barcelona, Spain. As a student she followed anErasmus programme at Twente University, The Netherlands,studying the impregnation of platinum and palladium on carbonnanofibres for their use as catalysts. Now at the Johnson MattheyTechnology Centre, Sonning Common, U.K., she is currentlycarrying out research into novel precious metal catalysts for thecontrol of NOx emissions.

The Reviewers


Fuel processing, where carbon-based fuels areefficiently reformed to produce hydrogen, pro-vides one route to a more extensive utilisation offuel cell technology. This multi-staged processrequires catalysis for each step. Rhodium is oftenused as a reforming catalyst, platinum for carbonmonoxide clean-up and platinum/palladium forcombustion. Base metals such as copper and zincalso find widespread application.

Gunther Kolb, in “Fuel Processing: for FuelCells”, sets out to understand the current state ofthe fragmented effort to solve the problems of fuelprocessing for niche applications. His book offersa timely and welcome overview of the expandingbody of work in which fuel processing technologyis finding application in fuel cell power (from wattto kilowatt scale).

The author is a well-known expert in this fieldand Head of Energy Technology and Catalysis atIMM, Germany. While the text is written to beinstructive towards the beginner, it is clearly direct-ed at those who wish to understand the currentissues in some detail. Kolb collates and summaris-es current information on the development of fuelprocessing technology in various areas, and high-lights the achievements that have been made todate. Throughout the book, he capitalises on hisunderstanding of both the science and the engi-neering involved in this complex interdisciplinaryfield. The early chapters serve as an excellent intro-duction to the subject, outlining the basic chemistryand engineering concepts associated with pre-reforming, partial oxidation, steam and autothermalreforming, as well as shift reactions and othermethods of cleaning up the reformed fuel.

After introducing the components, the authorencourages the reader to see the system as morethan the sum of its parts, emphasising that a fairlycomprehensive awareness of all aspects of the

chemistry involved is required for the selection ofappropriate operating conditions. The best choiceof steam:fuel and oxygen:fuel ratios and flow ratesare critical to gain optimum efficiency, but consid-eration of the physical chemistry involved is alsoimportant when designing a system for efficiency,selectivity and durability. Engineering considera-tions are also critical, and performance constraintsrequire difficult decisions to be made, for examplein the choice of reactor bed type (monolith, fixedbed, membrane etc.). The later chapters detail thespecifics of engineering, design concepts and dif-ferent types of fuel processor.

Where the book occasionally fails is in its illus-trations – some figures are difficult to read andlack sufficient annotation. Although the authoroutlines the considerable contribution that com-puter modelling continues to make towards ourunderstanding, a lack of distinction between real-world and simulated results may cause confusion.Further, the author’s strong interest in microchan-nel technology emerges in Chapter 10, which dealswith cost and production issues; however, in thecontext of the volume as a whole, this seemsacceptable.

By mid-volume, the reader cannot be in anydoubt as to the complexity of the task facing thosewho design and improve fuel processors. Kolbalso warns against common yet unsafe assump-tions, such as the idea that the engineeringtechnology of large systems can be scaled down tosmaller systems. This kind of assumption can leadto years of misdirected development work. Kolbadvocates that each situation should be consideredon its own merits, and should balance the techni-cal requirements with the economic requirementsover its entire lifetime. The inclusion of a decisiontree to map out and exemplify the types of decisionthat are required would have been very helpful to

DOI: 10.1595/147106709X465604

“Fuel Processing: for Fuel Cells”BY GUNTHER KOLB (Institut für Microtechnik Mainz GmbH, Germany), Wiley-VCH, Weinheim, Germany, 2008, 434 pages,

ISBN 978-3-527-31581-9, £130, €156, U.S.$215

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

the reader. This could have shown that, despite thehigher initial cost of precious metal catalysts, theirhigher activity, better catalyst utilisation andgreater resistance to poisoning mean that they arethe most appropriate option under some circum-stances.

One of the challenges of the book was to dealwith the difficult issue of intellectual property. Anybook such as this that describes state-of-the-arttechnology will find it difficult to be completelycurrent, as there is much knowledge that is heldoutside the public domain. However, Kolb out-lines the basic science and engineering behind thework being done, and supports it with evidencefrom the literature. This provides sufficient detailfor the educated reader to form an opinion, andsufficient referencing to help the more curious toinvestigate further. This will make the later chap-ters particularly useful to the growing numbers ofscientists and engineers who are turning theirattention and applying their skills to the technicaland commercial challenges of fuel processing.

For the general reader, this can mean that muchof the text is more detailed than they require,although the author rewards the dedicated readerwith occasional gems of insight. Throughout thebook, the author offers a balanced approach as hedeals with different theories and experimental andengineering approaches. However, he does occasionally point out incorrect assumptions ormisguided endeavours. At these points the text

comes alive, as the author adopts a cautionary toneto underline his key message.

The book’s key message throughout is thattechnological progress is being made, albeit in afragmented fashion, by experts in various disci-plines applying their knowledge and skills to thecomplex science and engineering involved. Theexamples of wasted effort may be a symptom ofthe fragmentation of the work, but Kolb’s bookmay well inspire a more coordinated approach,emphasising that much can be achieved whenmaterials scientists, chemists and engineers worktogether.

Further ReadingFor more information on policy aspects of fuel processingand fuel cells, refer to:U.S. Department of Energy, ‘On-Board Fuel ProcessingGo/No-Go Decision’, DOE Decision Team CommitteeReport, U.S.A., 6th August, 2004: http://www1.eere.energy.gov/hydrogenandfuelcells/pdfs/committee_report.pdf (Accessed on 15th June 2009) Fuel Cell Today Industry Review 2008, “Fuel Cells:Commercialisation”, Fuel Cell Today, U.K., 30th January,2008: http://www.fuelcelltoday.com/events/industry-review(Accessed on 5th May 2009)

The ReviewerJoseph McCarney is Business Development Manager for JohnsonMatthey’s Stationary Emissions Control business unit, with aparticular interest in the application of catalysis in environmentaltechnologies. Previously he was External Affairs Manager withinJohnson Matthey Fuel Cells’ Strategic Development Unit.


174

“Platinum 2009”, Johnson Matthey’s latestmarket survey of the platinum group metals(pgms), was published in May 2009. It coverssupply and demand for the calendar year 2008,together with a short-term outlook on the statusof the pgm market and pgm prices into 2009.

PlatinumThere was a market deficit of 375,000 oz plat-

inum in 2008. Global platinum demand wasdown by 5 per cent to 6.35 million oz while plat-inum supplies were also down, by 9.5 per cent, to5.97 million oz.

Supplies from South Africa fell to 4.53 millionoz in 2008. Supplies from Russia also fell, to820,000 oz, and North American supplies were flatat 325,000 oz. Supplies from other producingnations, including Zimbabwe, rose slightly to295,000 oz.

Gross demand for platinum for autocatalystswas down by 8.2 per cent to 3.81 million oz. InEurope, production of light-duty vehicles fell dueto the economic climate, while at the same time themarket share of diesel vehicles fell slightly.Therefore, despite the increasing fitment of plat-inum-containing diesel particulate filters (DPFs),platinum autocatalyst demand fell in this region. InNorth America, production of light-duty vehicleswas also down and there was continuing substitu-tion of platinum by palladium which contributedto a fall in platinum demand in the region.

Industrial demand for platinum fell to 1.76million oz in 2008. Weakening consumer demand,especially in the final quarter, led to a decline inplatinum demand from the chemical sector to395,000 oz. In the glass sector, demand fell to390,000 oz. In electronics, demand for platinumfell to 225,000 oz. Petroleum refining demand forplatinum rose to 245,000 oz. Other sources ofincreased demand for platinum included biomed-ical components and aircraft turbine blades, butdemand for dental alloys declined slightly.

Jewellery demand for platinum rose towardsthe end of 2008 as the price of platinum fell.However, demand had been affected earlier in the

year by the high price of platinum, and net platinum demand for jewellery in 2008 was downat 1.37 million oz. Physical investment demandrose to 425,000 oz.

PalladiumThe market surplus of palladium was 460,000

oz in 2008, with demand growing, despite theeconomic downturn, to 6.85 million oz. Supplieswere down to 7.31 million oz.

Supplies of palladium from South Africadeclined to 2.43 million oz in 2008. Supplies fromNorth America fell to 910,000 oz, and Russiansupplies fell to 3.66 million oz.

Demand for palladium from the autocatalystsector fell to 4.38 million oz in 2008. NorthAmerican light-duty vehicle production wasdown, causing demand for palladium to bereduced to 1.35 million oz. In Europe, palladiumdemand for autocatalysts rose to 950,000 oz, withthe increasing use of platinum-palladium catalystsfor the diesel sector outweighing the effects oflower vehicle output. In China, Japan and theRest of the World regions, combined palladiumdemand for autocatalysts rose to 2.09 million oz.

Demand for palladium from the chemicalindustry fell to 350,000 oz. Jewellery demand forpalladium was up to 855,000 oz, and physical pal-ladium investment grew to 400,000 oz in 2008.

Special FeatureThe three-page Special Feature in the book is

entitled ‘Palladium Use in Diesel OxidationCatalysts’.

Availability of “Platinum 2009”The book can be downloaded, free of charge, as

a PDF file in English, Chinese or Russian by visiting the Platinum Today website at:http://www.platinum.matthey.com/publications/Pt2009.html. Alternatively the English version can beordered in hard copy, by filling in the form at:http://www.platinum.matthey.com/publications/Request_Hard_Copies_of_PGM_Review.html orby emailing a request to: [email protected].

Platinum Metals Rev., 2009, 53, (3), 174

Platinum 2009

DOI: 10.1595/147106709X465172

CATALYSIS – REACTIONSDevelopment of an Amphiphilic Resin-Dispersionof Nanopalladium and Nanoplatinum Catalysts:Design, Preparation, and Their Use in GreenOrganic TransformationsY. UOZUMI and Y. M. A. YAMADA, Chem. Rec., 2009, 9, (1),51–65

An amphiphilic polystyrene-poly(ethylene glycol)resin-dispersion of Pd nanoparticles exhibited highcatalytic performance in the hydrodechlorination ofchloroarenes under aqueous conditions. Amphiphilicresin-supported Pd and Pt nanoparticle catalysts wereactive for the aerobic oxidation of alcohols in H2Ounder an atmospheric pressure of O2(g).

Preparation of Nano-Pd/SiO2 by One-Step FlameSpray Pyrolysis and Its Hydrogenation Activities:Comparison to the Conventional ImpregnationMethodO. MEKASUWANDUMRONG, S. SOMBOONTHANAKIJ, P.PRASERTHDAM and J. PANPRANOT, Ind. Eng. Chem. Res., 2009,48, (6), 2819–2825

Nano-Pd/SiO2 catalysts synthesised in one-stepflame spray pyrolysis (FSP) were compared to thoseon flame-made SiO2 supports by conventionalimpregnation. Metallic Pd particles < 3 nm in sizewere obtained directly by one-step FSP, whileimpregnation gave PdO with crystallite sizes 5–12nm. TOF values for 1-heptyne hydrogenation on theone-step FSP catalysts decreased from 66.2 s–1 to 4.3s–1 as Pd loading increased from 0.5 wt.% to 10 wt.%.

Steam Reforming of Methane, Ethane, Propane,Butane, and Natural Gas over a Rhodium-BasedCatalystB. T. SCHÄDEL, M. DUISBERG and O. DEUTSCHMANN, Catal.Today, 2009, 142, (1–2), 42–51

Steam reforming (SR) was investigated over a Rh-based cordierite monolithic honeycomb catalyst. Theproduct distribution was analysed as a function oftemperature (250–900ºC) and steam-to-carbon ratio(2.2–4) for two honeycomb channel densities (600 and900 cpsi) and an uncoated monolith. Ethane, propane,and butane are converted at much lower temperaturethan methane, also in the natural gas mixtures.

EMISSIONS CONTROLReduction of NOx by H2 on Pt/WO3/ZrO2 Catalystsin Oxygen-Rich ExhaustF. J. P. SCHOTT, P. BALLE, J. ADLER and S. KURETI, Appl. Catal.B: Environ., 2009, 87, (1–2), 18–29

Pt/WO3/ZrO2 (1) with a Pt loading of 0.3 wt.% anda W content of 11 wt.% had high deNOx activitybelow 200ºC and high overall N2 selectivity.Additionally, (1) exhibited outstanding hydrothermalstability and resistance against SOx.

Synthesis and Characterization of Rh and RhOx

Particles Supported on Zeolites with High Activityof Lean NOx–CO–H2 ReactionT. NAKATSUJI, T. YAMAGUCHI, J. LI, N. SATO and Y.MATSUZONO, Catal. Commun., 2009, 10, (6), 763–767

Rh and RhOx nanoparticles supported on zeoliteswere synthesised in a hydrothermal process using agel mixture of zeolite and Rh precursors. Thenanoparticles loaded onto Na-β zeolite and Na-ZSM-5 as dispersed particles and agglomerated particles,respectively. The nanoparticles supported on Na-ZSM-5 are more efficient in catalysing NOxreduction in lean conditions using H2 and CO.

FUEL CELLSQuantitative Characterization of Catalyst LayerDegradation in PEM Fuel Cells by X-RayPhotoelectron SpectroscopyF.-Y. ZHANG, S. G. ADVANI, A. K. PRASAD, M. E. BOGGS, S. P.SULLIVAN and T. P. BEEBE Jr., Electrochim. Acta, 2009, 54,(16), 4025–4030

The elemental concentrations and chemical states ofC, F, Pt, O and S on a PEM fuel cell catalyst layerwere determined by XPS. XPS signals characteristicof the ionomer decrease after ~ 300 h of fuel celloperation. Ionomer degradation was characterised bya decrease of CF3 and CF2 species and an increase inoxidised forms of C. The surface concentrations of Fand Pt decreased from 50.1% to 38.9%, and from0.4% to 0.3%, respectively. The oxidized states of Ptand C were substantially higher for the used samples.

Kinetics of Oxygen Reduction Reaction on Corich core-Ptrich shell/C ElectrocatalystsM. H. LEE and J. S. DO, J. Power Sources, 2009, 188, (2), 353–358

Prepared nanosized Pt/C (1) and Corich core-Ptrichshell/C (2) were shown to have the same f.c.c. crystalstructure by XRD, and their mean particle sizes(TEM) were 3.58 and 4.12 nm, respectively. The massactivity and specific activity of the ORR on (2) were10.22 A g–1 and 2.73 × 10–5 A cm–2, which were 1.5and 1.8 times those of the ORR on (1). The kineticsof the ORR on (1) and (2) in 0.5 M HClO4 wereexamined using film-type electrocatalysts on a RDE.

Deactivation/Reactivation of a Pd/C Catalyst in aDirect Formic Acid Fuel Cell (DFAFC): Use ofArray Membrane Electrode AssembliesX. YU and P. G. PICKUP, J. Power Sources, 2009, 187, (2), 493–499

A multi-anode, liquid-fed fuel cell was developedfor the study of DFAFC anode catalysts.Deactivation of Pd/C is caused by the electrooxida-tion of the formic acid, and does not occursignificantly at open circuit. Deactivated anodes wereshown to only be electrochemically reactivated byreversing the cell voltage. Reactivation is slow (> 1min) if the voltage is less negative than –0.2 V.


ABSTRACTS

DOI: 10.1595/147106709X465505


METALLURGY AND MATERIALSHalogen-Induced Corrosion of PlatinumE. DONÁ, M. CORDIN, C. DEISL, E. BERTEL, C. FRANCHINI, R.ZUCCA and J. REDINGER, J. Am. Chem. Soc., 2009, 131, (8),2827–2829

The interaction of Cl with Pt(110) was studied in anultrahigh vacuum environment. Up to half a mono-layer of Cl formed an adsorbate structure. Compressionto higher local coverages led to erosion of Pt atomsfrom the top layer and formation of PtCl4 pentamers.The Pt defects healed after annealing and a long-range-ordered PtCl4/Cl/Pt(110) adlayer was formed.Coadsorption of the Cl layer with CO caused forma-tion of PtCl4 but no volatile compounds.

The Role of Destabilization of Palladium Hydridein the Hydrogen Uptake of Pd-ContainingActivated CarbonsV. V. BHAT, C. I. CONTESCU and N. C. GALLEGO, Nanotechnology,2009, 20, (20), 204011 (10 pages)

A sample containing Pd embedded in activated Cfibre (2 wt.% Pd) was compared with commercial Pdnanoparticles deposited on microporous activated C(3 wt.% Pd) and with nanocrystalline Pd. The phasetransformations were analysed over 0.003–10 bar H2

partial pressures and at several temperatures using insitu XRD. Volumetric H2 uptake measurements veri-fied these results. Higher degrees of Pd–C contactsfor Pd particles embedded in the microporous Cmatrix induce efficient ‘pumping’ of H out of β-PdHx. Thermal cleaning of C surface groups prior toH2 exposure further enhances the H pumping powerof the microporous C.

Rietveld Analysis of Neutron Powder Diffraction ofMg6Pd Alloy at Various Hydriding StagesJ. HUOT, A. YONKEU and J. DUFOUR, J. Alloys Compd., 2009,475, (1–2), 168–172

Mg6Pd alloy was obtained by ball milling fromrolled Mg and small pieces of Pd foil. The evolutionof the crystal structure of Mg6Pd was studied bysimultaneous Rietveld refinement of the neutron andX-ray powder diffraction data. Samples with differentD contents were measured, corresponding to reactionend-products of the proposed hydrogenation step.After full hydrogenation, Mg6Pd transforms to MgPdand MgD2. Increases in lattice parameters of MgPdalloy agree well with measured H capacities.

New Hard and Superhard Materials: RhB1.1 and IrB1.35

J. V. RAU and A. LATINI, Chem. Mater., 2009, 21, (8), 1407–1409Phase-pure RhB1.1 and IrB1.35 bulk materials were

prepared using an electron beam apparatus. PowderXRD patterns for both borides were obtained; theRietveld analysis results are presented. The Vickersmicrohardness data revealed the superhard nature ofIrB1.35: under 0.49 N of applied load, it exhibits a max-imum hardness of 49.8 ± 6.0 GPa. RhB1.1 bulk wasfound to be hard: under 0.49 N of applied load, itexhibits a hardness of 22.6 ± 1.5 GPa.

CHEMISTRYAn Old Reaction in New Media: Kinetic Study of aPlatinum(II) Substitution Reaction in Ionic LiquidsI. CORREIA and T. WELTON, Dalton Trans., 2009, (21),4115–4121

A kinetic study was carried out on the substitutionreaction of [Pt(dpma)Cl]+ with thioacetate in roomtemperature ionic liquids and molecular solvents. Thereaction follows an associative mechanism with atwo-term rate law and is the same in all the solventsstudied. The reaction rate followed the order: H2O >ionic liquids and DMSO > methanol. No ‘ionic liquideffect’ was found.

Studies on the Complexation between Generation-4.5 Methyl Ester-Terminated Poly(amidoamine)Dendrimer and Pd2+ Ions in MethanolW. LU, G. LI, Y. LUO and Y. JIN, J. Appl. Polym. Sci., 2009, 112,(5), 2854–2858

The interactions between Pd2+ ions and G4.5-COOCH3 PAMAM dendrimers were investigated byUV-vis and FTIR. The addition of K2PdCl4 results incovalent attachment of the PdCl3– alcoholysis productof this complex to tertiary amines within the den-drimers under the appropriate conditions. XPS dataindicated a 1:3 Pd:Cl ratio. The maximum loading ofPd2+ ions within the G4.5-COOCH3 dendrimers was80. The best pH value for complexation was 8.3.

Synthesis and Characterization of Water-SolublePalladium(II)-Functionalized Diphosphine ComplexesM. SUNJUK, M. AL-NOAIMI, G. ABU SHEIKHA, E. LINDNER, B.EL-ESWED and K. SWEIDAN, Polyhedron, 2009, 28, (8),1393–1398

H2O-soluble cis-[Pd(L)(OAc)2] (1) (L = CH2(CH2PR2)2

and R = (CH2)6OH; (CH2)nP(O)(OEt)2, n = 2–6 or 8;or (CH2)3NH2) were obtained by reacting Pd(OAc)2

with L in a 1:1 mixture of CH2Cl2:CH3CN. L wereprepared photochemically by hydrophosphination ofthe corresponding 1-alkenes with H2P(CH2)3PH2.

PHOTOCONVERSIONA Versatile Color Tuning Strategy for Iridium(III) andPlatinum(II) Electrophosphors by Shifting the Charge-Transfer States with an Electron-Deficient CoreG.-J. ZHOU, Q. WANG, W.-Y. WONG, D. MA, L. WANG and Z. LIN,J. Mater. Chem., 2009, 19, (13), 1872–1883

Red cyclometallated Ir(III) and Pt(II) electrophos-phors with enhanced electron-injection/electron-transporting features were prepared by using an elec-tron-trapping fluoren-9-one chromophore. There is aswitch of the MLCT character of the transition fromthe pyridyl groups in the traditional Ir(III) and Pt(II)ppy-type complexes to the electron-deficient ringcore. The electron-withdrawing character of the fusedring results in more stable MLCT states, inducing asubstantial red-shift of the triplet emission energyfrom yellow to red for the Ir(III) complex and evengreen to red for the Pt(II) complex.

CATALYSIS – APPLIED ANDPHYSICAL ASPECTSPalladium-Zinc Nanocolloidal ParticlesMITSUBISHI CHEM. CORP. Japanese Appl. 2008-264,761

A method for manufacturing PdZn nanocolloidalparticles of formula Pd(100 – x)Znx where 30 ≤ x ≤ 80is described. Compounds of Pd and Zn are mixedwith a ligating organic compound, a reducing agentand a solvent and heated. Particles 1–50 nm in diam-eter are formed with a thin organic layer on thesurface preventing coalescence, and can be used as-isin a liquid carrier or dispersed on a support such as silica gel to catalyse the dehydrogenation of alcohols.

CATALYSIS – INDUSTRIAL PROCESSPalladium on Bacterial Cellulose SupportINDIAN INST. TECHNOL. World Appl. 2008/122,987

A novel design for a reactor using immobilised Pd isproposed. Metallic Pd is deposited on a support ofbacterial cellulose coated on a series of acrylic discs.These are rotated, alternately allowing H2(g) to beadsorbed from a gaseous feed and to reduce chlorinat-ed pollutants or nitroaromatic compounds in theliquid feed. The method can be used for treatment ofwastewater and decolourisation of textile dyes.

CATALYSIS – REACTIONSProduction of Chiral ββ-Amino Acid DerivativesSOLVIAS AG U.S. Patent 7,495,123

The claimed process bypasses the need for synthe-sis of a protected β-amino acrylic acid substrate as anintermediate. It is catalysed by a Rh complex with chi-ral phosphine ligands, which may be either preformedor generated in situ through the addition of the Rhmetal precursor and ligand to the reaction mixture.

Palladium-Catalysed Aryl Cross-CouplingA. S. IONKIN U.S. Appl. 2009/0,054,650

A new catalytic method for the coupling of aryl moi-eties is claimed. A hetero aryl halide is reacted with anarylboronic acid in the presence of a Pd compound incombination with a compound containing a dialkyl-phosphine moiety. The Pd compound may includePdCl2(dppf), Pd(OAc)2, Pd(PPh3)4 or Pd2(dba)3. Thedialkylphosphine may be di-tert-butylphosphine ordiadamantylphosphine.

Ruthenium-Catalysed Direct Synthesis of AmidesD. MILSTEIN et al. U.S. Appl. 2009/0,112,005

A novel process for preparation of amides fromalcohols and amines is claimed, catalysed by a Ru com-plex based either on a dearomatised PNN-type ligand(where PNN = 2-(di-tert-butylphosphinomethyl)-6-(diethylaminomethyl)pyridine), or its precursors. Aprimary amine is directly acylated by an equimolaramount of a primary alcohol to produce the desiredamide with H2 as the only byproduct. High yields andturnover numbers are claimed.

EMISSIONS CONTROL NOx Reduction Catalyst for FCCG. YALURIS et al. U.S. Appl. 2009/0,045,101

A catalytic additive for reduction of NOx producedin the catalyst regeneration zone in full-burn fluid cat-alytic cracking (FCC) processes is described, based ona mixture of: (a) a zeolite-free acidic metal oxide, (b) analkali/ alkaline-earth metal, (c) an oxygen storage com-ponent, (d) Pd (preferably 40–1500 ppm) and (e) apgm, preferably Pt, Rh and/or Ir (preferably 25–1500ppm). Amounts added to the circulating FCC mixtureare preferably 0.1–20 wt.% of the cracking catalyst.

Porous Platinum-Alumina Cryogel CatalystNATL. INST. ADV. IND. SCI. TECHNOL.

Japanese Appl. 2009-018,225A manufacturing method for a highly-porous and

durable catalyst for the treatment of VOCs in indus-trial exhaust gas is described. A chelated solution ofPt with ammonium oxalate is added to a sol preparedfrom AlO(OH). Gelling is induced by addition ofurea and the gel is freeze-dried and calcined. Pt con-stitutes 0.5–5 wt.% of the catalyst body in the form ofhighly-dispersed ultrafine particles.

FUEL CELLSPlatinum Nanostructures for PEMFC CatalystsTOYOTA ENG. MANUF. NORTH AMER. INC

World Appl. 2008/051,284Dendritic nanostructures containing Pt or Pt alloys

are claimed for use in fuel cell catalysts. The dendrimers may be spherical with a diameter of 1–1000 nm (preferably 5–100 nm), or disc-shapedwith a thickness of ~ 1–10 nm and may be partiallyfused. Preparation occurs in a fluid medium contain-ing a reducing agent, a precursor such as a Pt complexand a matrix on which the structures are grown andwhich can be used to manipulate their shape.

RuTe2 Catalyst for Fuel CellMITSUBISHI CHEM. CORP Japanese Appl. 2008-287,927

A RuTe2-based catalyst which is stable enough foruse in both the cathode and anode in a PEMFC isclaimed. The stability is obtained through use of a N-containing Ru precursor, preferably Ru(NO)(NO3)3.The catalyst can be used as a coating on a C-based sup-port and can form part of a MEA or MEA stack.

METALLURGY AND MATERIALSNi-Ti-Pt Shape Memory AlloyU.S. ADM. NASA U.S. Patent 7,501,032

A Ni-Ti-Pt shape memory alloy with a maximumwork output of ≥ 5 J cm–3 (preferably 10–15 J cm–3 ) isdisclosed. The transition temperature is 100–400ºC,preferably 200–350ºC, and the hysteresis is < 50ºC,preferably 10–20ºC. The composition is (in at.%): 50–52Ti (preferably 50–50.5), 10–25 Pt (preferably 15–23)and the balance Ni, which may be partially substitutedby 0.5–2 C and < 5 of one or more of Pd, Au and Cu.

177

DOI: 10.1595/147106709X465163


NEW PATENTS

Platinum Metals Rev., 2009, 53, (3)

Device for Manufacturing Glass FibreTANAKA KIKINZOKU KOGYO KK

Japanese Appl. 2008-266,057Molten glass flows from a circulating stream

through a cylindrical bushing block (1), is filteredthrough a screen (2) and is then extruded through anumber of nozzles mounted in a base plate (3). (2) and(3) are composed of Pt or a RhPt (Rh = 5–20 wt.%)alloy and the interior of (1) is coated in a protectivelayer of the same material. (1) and (3) are electricallyheated at 10 A mm–2 to prevent spot cooling of themolten glass.

APPARATUS AND TECHNIQUEPlatinum-Modified Pollen FilterA. J. OSINGA World Appl. 2009/048,324

A Si-based coating modified by adding ~ 5 ppm Ptis claimed to be effective to filter pollen and otherallergens from the air. An emulsion of a siloxanecomplex of Pt catalyst with an average particle size of~ 0.2 μm is added to a water-based silicone emulsionand held at quiescence for one day to formcrosslinked silicone particles. The coating is appliedto a gauze screen to form the filter.

Iridium Single-Atom TipACADEMIA SINICA U.S. Appl. 2009/0,110,951

A method for preparing an Ir tip of atomic sharp-ness is described. An Ir rod with a polycrystalline or<210> monocrystalline structure is sharpened by amethod such as electrochemical etching to a taperedend of radius 5–200 nm and heated in a vacuumchamber under O2 pressure 10–8–10–5 Torr in twostages, to form a tip terminating in a single atom. Thetip is stable, regenerable, capable of sustaining electricfields up to ~ 53 V nm–1 and a perfect point electronsource.

Palladium-Based Thin FilmSUMITOMO METAL MINING CO LTD

Japanese Appl. 2008-279,329A H2-permeable membrane is obtained by sputter

deposition of Pd or Pd alloy on a glass-type substratepossessing a thermal expansion coefficient of8–15 × 10–6 ºC–1. A metal substrate layer which maybe Cu can be deposited between the substrate andfilm and dissolved by acid to exfoliate the film. A verythin membrane (0.1–5 μm thick) is obtained which isrelatively defect-free and exhibits minimal curling.

BIOMEDICAL AND DENTALPlatinum Complexes for Treatment of TumoursUNIV. WARSZAWSKI World Appl. 2009/041,841

Novel peptide-Pt complexes are described, whichtypically have the formula (OP-AA)-PtX2, where OPis an opioid peptide, AA is an amino acid residue ofmethionine, cysteine, histidine, 1,3-diaminebutanoicacid or 1,4-diaminepentanoic acid and X is a halogen,preferably Cl. The complexes are claimed to combinethe anticancer properties of Pt with the analgesicactivity of the peptide.

Dental Alloy with High Palladium ContentARGEN CORP World Appl. 2009/046,260

A Ni-based dental alloy with a high Pd content isdisclosed. The composition is (in wt.%): 25–45 Pd(preferably ~ 25), 15–30 Cr (preferably ~ 25) and atleast 5 wt.% Mo and/or W (preferably 12), with thebalance being Ni. Optionally the Ni may be partiallysubstituted by up to 1.5 wt.% Si and up to 10 wt.%Re, Nb and/or Ta. The addition of Pd makes for analloy which can be easily cast, ground and bonded toporcelain, and where the thermal expansion is well-matched to that of porcelain.

Iridium Complex for Detection of CancerGUNMA UNIV. Japanese Appl. 2008-281,467

The novel application of electroluminescent com-pounds of Ir to the measurement of O2 concentrationin living cells is described. Complexes of Ir(III) witharomatic ligands, preferably containing N, S or O, andspecifically BTP2Ir(acac) (1) are indicated. Whenapplied to tissue, and using a suitable detectionmethod, (1) will emit red phosphoresence in theabsence of dissolved O2. (1) can be used for thedetection and imaging of cancer and has the advan-tage of being non-invasive.

Iridium Oxide Conductive Coating for Medical DeviceMEDTRONIC INC U.S. Appl. 2009/0,047,413

Disclosed is a coating applied to the housing ofimplantable medical devices, especially those housingpowered components. It includes a carrier and0.1–90 wt.% of a therapeutic agent, specifically Agnanoparticles. The conductive carrier can be Ir oxide,Pt, Pt black, graphite or other forms of C, etc. Suchcoatings containing Ir oxide are described to be effec-tive at inhibiting bacterial growth.

ELECTRICAL AND ELECTRONICSFeRh AFM-FM Phase Change Material for PMRHEADWAY TECHNOL. INC U.S. Appl. 2009/0,052,092

A perpendicular magnetic recording (PMR) head isdescribed which contains an antiferromagnetic-ferro-magnetic (AFM-FM) phase change material in themain pole layer, which may be FeRh or FeRhX whereX = Pd, Pt or Ir and where Rh is > 35 at.% of thetotal. During non-write operations the material is in anAFM state to minimise remanence, and during writeoperations it is switched to an FM state by heating.Minimal pole erasure during non-writing and highwritability compared to single pole writers is claimed.

SURFACE COATINGSRhAl Overlay Coatings for Gas Turbine ComponentsGENERAL ELECTRIC U.S. Appl. 2009/0,061,086

A RhAl coating system is claimed, consisting of:25–90 at.% Rh, 10–60 at.% Al and forming predom-inately a B2-phase RhAl intermetallic. It may alsocontain up to 25 at.% of one or more of Pt, Pd, Ruand Ir and up to 20 at.% of the base metal and alloy-ing elements of the substrate. It may be applied as anenvironmental coating or a diffusion barrier coating.

178


FINAL ANALYSISThe Impact of CO2 Legislation on PGMDemand in Autocatalysts

It is estimated that emissions from road trans-port contribute 17% of total anthropogenic carbondioxide (CO2) emissions (1). Regulations cominginto force in Europe and in the U.S.A. are seekingto change this. In December 2008, the EuropeanParliament approved a directive to reduce the aver-age CO2 emissions of new passenger cars to130 g km–1 by 2015, with a three-year phase-inperiod. Emissions will be reduced by a further10 g km–1 through external measures including theincreased use of biofuels, lower rolling resistancetyres and efficiency improvements to auxiliarydevices such as air conditioning units. Overall, the120 g km–1 target represents an improvement infuel economy of about 25% from current levels(2). In the U.S.A., where vehicles have traditional-ly been much larger and less fuel efficient than inEuropean markets, President Obama has unveiledan ambitious plan to improve the fuel economy ofpassenger cars and light trucks in the country to anaverage of 35.5 U.S. miles per gallon (mpg) by2016, from the current 25 U.S. mpg average (3).

While there are longer-term clean transportoptions in development, such as electric and fuelcell vehicles, three different engine technologiesare seen to be crucial in improving fuel consump-tion, and thus lowering CO2 output to meet these2015 and 2016 targets. These technologies arediesel engines, downsized turbocharged gasolineengines combined with direct injection, and hybridgasoline- or diesel-electric vehicles. Each of thesethree technologies will have a different impact onplatinum group metals (pgms) use.

Diesel vehicles already account for over half ofnew vehicles sold in Europe, and this high share isexpected to be maintained over the next decade.However, with improvements in gasoline directinjection (GDi) engines and greater numbers ofvehicle manufacturers offering this technology, itis likely that we will see more of these vehicles onEuropean roads in future. Some estimates putpenetration of GDi engines as high as 28% of the

gasoline vehicle market in Europe by 2010 (4). Inthe U.S.A., increasing uptake of GDi vehicles isalso expected. However, diesel sales are currentlylow, at around 5% of overall light vehicle sales.While there is potential for some growth in thismarket share, the main focus for automakers in theregion is on gasoline-electric hybrid powertraindevelopment.

Diesel EnginesDiesel engines are 20% to 30% more fuel effi-

cient than similar sized conventional gasolineengines, and therefore produce less CO2. On aver-age, diesel vehicles use more pgm than theirgasoline counterparts since they typically operateat lower temperatures, boosting the need for pgmuse in the catalytic aftertreatment. Hydrocarbon(HC) and carbon monoxide (CO) emissions aremanaged through the use of a diesel oxidation cat-alyst (DOC), and many new diesel vehicles arefitted with a diesel particulate filter (DPF) to con-trol PM emissions. Some vehicles use a catalysedsoot filter (CSF) to control the CO, HC and PMemissions. Traditionally, platinum has been usedas the main catalytic component in dieselaftertreatment due to its excellent oxidation activ-ity at low temperature and resistance to ‘poisons’in the exhaust stream, particularly sulfur.However, the greater availability of cleaner (lowersulfur) diesel fuel in Europe has resulted in theintroduction of some palladium into diesel emis-sion control systems (5).

Stricter legislation entering into force inEurope in 2014 will focus on reducing NOx emis-sions from diesel vehicles, thus requiring the useof additional catalytic aftertreatment, althoughsome cars sold now already incorporate NOxaftertreatment. Two forms of aftertreatment canbe used to reduce NOx emissions: a NOx trap(containing pgm), or selective catalytic reduction(SCR) using urea as the reductant where the SCRcatalyst itself does not use pgm, although the

DOI: 10.1595/147106709X465901

aftertreatment system as a whole may still containpgm. Both technologies are expected to be used tomeet Euro 6 legislation, with vehicle size being akey consideration in the choice between the use ofa NOx trap and an SCR catalyst.

Given the considerably higher amount of pgmused in diesel aftertreatment as compared withgasoline, any increase in the share of diesel engineswill lead to greater demand for pgms.

Downsized Gasoline EnginesDownsized gasoline engines (such as turbo-

and super-charged engines), particularly whencombined with GDi, provide similar, or improvedperformance at reduced engine size and hence thepotential for greater fuel economy. In recentlydeveloped GDi vehicles, the catalyst size tends tobe smaller, but the pgm loading may be higher.While the net effect on pgm content is not yetclear, at present the catalyst loadings for GDiengines are broadly similar to naturally aspirated(conventional) gasoline engines.

Hybrid Powertrain Hybrid vehicles offer fuel economy benefits

over similar sized gasoline or diesel engines bycombining the internal combustion engine with anelectric motor. Because of the relatively small num-bers of hybrid vehicles produced today, thecomparison between similar sized hybrid and con-ventional gasoline or diesel vehicles in terms ofpgm loading is not well defined and is complicatedby two opposing factors. A hybrid vehicle general-ly has a smaller engine than its conventionalgasoline or diesel counterpart, as the electric motorand battery assist during acceleration, and couldtherefore be expected to require less pgm for thecatalyst. However, in the U.S.A., where the major-ity of hybrid vehicles are sold today, these vehiclesare typically manufactured to meet more stringentCalifornian SULEV emissions standards (6), andtherefore require higher pgm loadings. In the com-ing years, assuming that vehicles are manufacturedto meet the same regional emissions standards, thecurrent view is that a switch to hybrid vehiclesfrom conventional gasoline or diesel vehicles willhave little impact on overall uptake of pgm.

Vehicle DownsizingMoving away from the subject of engine tech-

noogies, a shift to smaller, lower-cost and morefuel efficient vehicles could reduce the averagepgm loading per vehicle. Recently, there is evidenceof a move to smaller vehicles in both the Europeanand U.S. markets, but it is not yet clear whether thiswill be a long-term trend. In Europe, the growth inthe small car segment is seen as a temporary effectcaused by the scrappage incentives which are inplace in key markets (7). These provide a one-offpayment towards the cost of a new car to con-sumers scrapping an older vehicle. In the U.S.A.,some consumers downsized from passenger trucksand sports utility vehicles (SUVs) to smaller vehi-cles in response to the higher fuel prices in 2008and, despite falling gasoline prices this year, ana-lysts are anticipating that the share of passengertrucks will continue to drop slowly over time.

Conclusion Precise trends in regional uptake of pgms

depend on both the extent to which each of theabove strategies is pursued and consumer prefer-ences, and remain difficult to predict. Overall,however, it is unlikely that tighter CO2 emissionslimits will strongly affect pgm demand from theautomotive sector in either direction within theforeseeable future.

LUCY BLOXHAM

References1 International Energy Agency, “CO2 Emissions from

Fuel Combustion 1971–2005”, OECD Publishing,Paris, France, 2007, 576 pp

2 ‘Reducing CO2 emissions from light-duty vehicles’,European Commission initiatives on the Europawebsite: http://ec.europa.eu/environment/air/transport/co2/co2_home.htm (Accessed on 1st June 2009)

3 ‘Obama gets tough on fuel economy’, CNNMoney.com Special Report, 19th May, 2009: http://money.cnn.com/2009/05/18/autos/new_fuel_economy_standards/ (Accessed on 9th June 2009)

4 T. Lewin, ‘New emissions rules will boost GDIsales’, Automotive News Europe, 28th May, 2007

5 ‘Palladium Use in Diesel Oxidation Catalysts’, SpecialFeature, in “Platinum 2009”, Johnson Matthey, 2009:http://www.platinum.matthey.com/uploaded_files/int_2008/09special_featurepdindieselcatalysts.pdf

6 Emission Standards: U.S.A.: Cars and Light-Duty Trucks:California: http://www.dieselnet.com/standards/us/ld_ca.php (Accessed on 29th June 2009)

7 ‘Germans wary of car scrapping scheme’, BBC Newsreport, 16th April, 2009: http://news.bbc.co.uk/2/hi/business/8003508.stm (Accessed on 1st June 2009)

180Platinum Metals Rev., 2009, 53, (3)

EDITORIAL TEAM

EditorDavid Jollie

Assistant EditorSara Coles

Editorial AssistantMargery Ryan

Senior Information ScientistKeith White

E-mail: [email protected]

Platinum Metals Review is the quarterly E-journal supporting research on the science and technology of the platinum group metals and developments in their application in industry

http://www.platinummetalsreview.com/

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

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

Documents

Platinum Metals ReviewE-mail: [email protected] E-ISSN 1471–0676 PLATINUM METALS REVIEW A Quarterly Survey of Research on the Platinum Metals and of Developments in their Application