PLATINUM METALS REVIEW · Platinum Group Metals and Their Oxides in Semiconductor Photo...

UK ISSN 0032-1400

PLATINUM METALS REVIEW A Quarterly Survey of Research on the Platinum Metals and

of Developments in their Application in Industry www.matthey.com and www.platinum.matthey.com

VOL. 47 JANUARY 2003 NO. 1

Contents

Platinum Group Metals and Their Oxides in Semiconductor Photosensitisation By A. Mills and S.-K. Lee

Catalyst Life Cycle Conference By Stephen Poulston and Stephen Pollington

Rhodium Dendrimer Catalysts

Vehicle Emissions Control Technologies By Marlyn K Twigg

Ruthenium Light-Switch Effects

Palladium(+l) Carbonyl Clusters in the Catalytic Oxidation of Unsaturated Compounds

By Tatiana A. Stromnova

Alcohol Oxidation by Ruthenium Catalyst

Fuel Cells - Science and Technology 2002 By Donald S. Cameron

Eighth Grove Fuel Cell Symposium

Uphill Effects on Hydrogen Diffusion Coefficients in Pd,,Ag,, Alloy Membranes

By X. Q. Tong, ,F A. Lewis, S. E. J. Bell and J. fermak

Platinum Group Metals Technology in Ekaterinburg

Noble and Rare Metals Conference

Abstracts

New Patents

Find Analysis: Catalysts - Myths and Realities By D. E. Grove

2

13

14

15

19

20

27

28

31

32

36

36

37

41

44

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

Platinum Group Metals and Their Oxides in Semiconductor Photo sensitisation BASIC PRINCIPLES, METAL PHOTODEPOSITION AND WATER PHOTOSPLITTING REACTIONS

By A. Mills' and S.-K. Lee Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow G1 lXL, Scotland 'E-mail: a.miIIsQstrath.ac.uk

The hasic principles of semiconductor photochemistry, purticularl~v using titania as a semiconductor photocatalyst, are discussed. When a platintrm group metal or its oxide is deposited onto the surface of a sensitised semiconductor the overall ejiciency of the reactions it takes part in are ofien improved, especial1.v when the deposits are used as hvdrogen and ox,vgen catalysts, respectively. Methods of depositing metal or metal oxide are examined, and a particular focus is given to a photodeposition process that uses a sacrificial electron donor. Platinum group metal and platinum group metal oxide coated semiconductor photocatalysts are prominent in heterogeneous systems that are capable of'the photoreduction, oxidation and cleavage of water. There is a recent renaissance in M'ork on watei--splitting semiconductor-sensitised photosystems, but there are continued concerns over their irreproducibility, longevity and photosynthetic nu ture.

Photosensitisation of a semiconductor involves its electronic excitation by absorbed photons which promote its ability to mediate chemical reactions. At the end of such a reaction, the photo- sensitised semiconductor remains chemically (and usually physically) unchanged (l), as, generally speaking, does a conventional catalyst. A semiconductor has a manifold of electron energy levels Wled with electrons - the valence band (W) and also a manifold of higher energy levels that are largely vacant - the conduction band (CB). The energy difference between these two bands is called the bandgap energy (Ed. A general semiconductor photosensitisation reaction can be summarised by:

semiconductor

hv 2 Ebg A + D- A' + D- (9

where the change in the Gibbs free energy for this reaction may be negative (the usual reported case) when photocatalysis occurs, or positive when photosynthesis occurs (24). The usual form of a semiconductor photosensitiser in Reaction (i) is as particles of micrometre to nanometre diameter, which are aggregates of nanocrystals. These parti-

des are used either as a powder dispersion or layered to form thin films (typically, 100-10,000 nm thick).

Each semiconductor particle can be considered to act as a micro-photoelectrochemical cell and the basic features of such cells for promoting a general chemical reaction, say Reaction (i), are shown in Figure 1 (U). The left hand side of the diagram shows a rough aggregated semiconductor particle comprised of one or more nanocrystals absorbing a photon of light of bandgap energy, hv 2 E,. This generates an electron-hole pair.

The fate of the components of the electron- hole pairs determines the overall photoactivity of a semiconductor. For instance, the photogenerated electron and hole can recombine in the bulk or at the surface, probably via trap states. Electron-hole recombination usually dominates semiconductor photosensitisation so the overall process is often not very efficient (typically < 1%) with respect to photons. If, however, the photogenerated electron and hole can make their separate ways to the surface of the semiconductor particle, they may react with species, such as A and D, adsorbed at the surface and drive Reaction (i) forward. The greater the

Pkatinum Metah Rm, 2003,47, (I), 2-12 2

Fig. I The kfi hand side shows the major deactivation processes that can occw on a particle comprising maiiv nanociptals, ufter excitation with light of ultra-handgap enetw has crcated an electron-hole paic The processes include: (a) electron-hole recombination - in the hulk, (6) electron-hole recombination - at the surface.

A D

raA +

D

E

(c) reduction of an electron acceptor: A , - at the surfiice by a photogenerated electron, and (d) oxidation of’ an electron donor, D, ~ at the surjace by a photogenerated hole. The right hand side shows the electron energetics associated with Reaction (i). sensitised by a semiconductor particle

overall rate of this process, compared with the bulk and surface back reactions, the greater the photo-efficiency of the system.

Charge Transport In these nanocrystalline semiconductor photo-

systems the semiconductor is not deliberately doped and the combination of low or negligible doping and small size prevents the formation of any appreciable space charge layer at the surface of the semiconductor particles (5-7). This situation is different from that found in macro-photoelectrochemical cells where the semiconductors are usually deliberately doped to improve function, and as a consequence, possess a fully-formed space-charge layer that allows for the efficient separation of the photogenerated electron-hole pairs in the layer via migration (6, 7). Most nanocrystalline photosensitiser systems lack a space-charge region and the photogenerated electrons and holes move from the bulk to the surface mainly via diffusion.

Despite this apparent drawback, in nanocrystalline materials at least, transport of the charge carriers (electrons and holes) appears much faster (10 ps) than bulk recombination (100 ns) and these materials can thus act as efficient photosensitisers of reactions at their surface. Obviously, diffusion is only possible if a concentration gradient is set up for both holes and electrons and this is thought to be achieved by trapping states at the surface. For

most oxide semiconductor photosensitisers, such as titania and zinc oxide, it is generally believed that the photogenerated holes are readily trapped by surface-adsorbed water to form adsorbed hydroxyl radicals, which can diffuse, or ‘spill over’, across the photocaralyst surface (24). Recent work indicates such species can desorb from the surface and then diffuse into the reaction medium over distances exceeding tens of microns (8). Due to the apparent facile hole trapping action exhibited by many semiconductor photosensitisers, the photogenerated electrons tend to accumulate at the surface of the semiconductor. For titania these trap sites are T i 0 surface species, becoming Ti(IIIT) on trapping the electrons.

Illuminating this system can thus cause a build- up of negative charge on the surface of the semiconductor particles if an easily reducible species, such as oxygen, is absent. Electrophoretic mobility measurements show that on illumination titania particles are negatively charged (9). Illuminating ti& under anaerobic conditions in the presence of an easily oxidised species, such as EDTA, causes the photosensitiser to turn blue, indicaang the presence of Ti(III) (10). The blue colour quickly disappears when oxygen is admitted as Ti(I1l) is converted to T i m .

One way to prevent the accumulation of photogenerated electrons on semiconductor particles is to add easily reducible species, such as ceric (Ce4+), ferric $e3+) and PtCL” ions; for example

Phtimm Met& Rev., 2003, 47, (1) 3

H l c t a l p t

T i 0 2 ( a )

SED products

i rrev. decornp. 7 products

SEA-

SEA

(b)

Fig. 2 Schematic illustrations of the electron energetics associated with: (a) the photoreduction of water by a sacrificial electron donor (SED), sensitised by semiconductor particles which huve surface deposits of u hydrogen cata!v.st, such as Pt. Products are formed bv irreversible decomposition. (b) the photooxidation of water by a sacrificial electron acceplor (SEA), sensitised by semiconductor purticles with surjhce deposits of an oxygen catal.vst, such as RuO?. Products are formed by irreversible decomposition. and (c) the photocleavage of watec sensitised by semiconductor particles with su@e deposits of a hydrogen cuta1v.st and an oxygen catalvst

PtCL,’- is readily reduced to Pt metal. Thus, if a solution containing titania and HZPtCL, is illum- nated, Pt is usually photodeposited onto the titania surface (11). This is a common way to deposit a metal on semiconductors.

In semiconductor photochemistry, platinum group metal (pgm) deposits on the surface of semiconductor particles are often used to act as traps or

wells for any photogenerated electrons that may accumulate. The pgm deposits are assumed to reduce the overall probability of electron-hole recombination and so increase the overall effiden- cy of the photosystem. Such electron wells must be continuously ‘drained’ during illumination if they are to function efficiently.

In the absence of oxygen but in the presence of water, most pgms, if made sufficiently reducing by accumulated negative charge, will readily reduce water to Hz. In this process, most pgms stabilise the intermediate hydrogen atoms and catalyse their combination to form Hz. For increased efficiency a sacrificial electron donor (SED), such as EDTA or methanol, must be added to remove irreversibly any photogenerated holes or oxidising species, such as hydroxyl radicals, from the semiconductor surface. Many systems that overall photoreduce water to H:! have been reported, but most utilise an SED and a UV-absorbing semiconductor photocatalyst.

In water photooxidation by semiconductor photocatalysis, a sacrificial electron acceptor (SEA), such as Fe3’ or silver (Ag’) ions, is usually added to the system to prevent accumulation of any photogenerated electrons. Oxides of the pgms, such as RuO, or IrOz, which are recognised 0 2

evolution catalysts (12), are often deposited on the surface of the semiconductor photocatalyst to improve the efficiency of water oxidation.

Finally, in studies of water photodissociation by semiconductor photosynthesis (AG“ > 0), a HZ and an 0 2 catalyst, both pp-based, are usually deposited on the surface of the semiconductor particles to help effect the overall process.

The general electronic features of semiconductor-sensitised photosystems for the reduction, oxidation and cleavage of water are illustrated in Figure 2.

Semiconductor Photosensitisers There are many semiconductor materials but

few that are sufficiently robust photochemically and chemically to be described as a ‘photocatalyst’ or ‘photosensitiser’. Most semiconductors, such as CdS, ZnO or tungsten oxide (WO,), are either photochemically or chemically unstable. For

P h t w m Metub h., 2003,47, (1) 4

Phase Ebg, eV

An at as e 3.23 Rutile 3.02

instance, CdS can be photoanodically corroded to form an inert layer of surface sulfur and dissolves in acid, i.e. it is both photochemically and chemically unstable. ZnO is soluble in alkali and in complexing agents, such as EDTA. WO, is soluble in alkali and is mechanically ‘soft’, making production of robust film difficult. Of the very few photochemically and chemically stable semiconductor photosensitisers, one compound dominates: titania (titanium dioxide), that is TiOz.

Titania exists as anatase, rude and brookite crystalline forms. The latter is not common, does not possess good photocatalytic properties and is rarely used as a photocatalyst. Anatase is generated by the usual low temperature production methods, such as alkaline hydrolysis of titan- i u m o compounds followed by calcination at moderate temperatures (400-500°C). Anatase readily converts to rude at elevated temperatures (> 700°C) although this phase change is often accompanied by extensive sintering (13). As a consequence, rude usually has a much lower specific surface area (by a factor of 10 or more) than the anatase from which it was derived.

Titania is chemically and biologically inert, pho- tostable, photoactive and cheap (2-4). In Table I are key electronic features of anatase and rude. Their high bandgap energies show that the major drawback in using them as photosensitisers is that they only strongly absorb UV hght (rather than visible light). Titania only absorbs %33% of the solar spectrum so is of limited use as a photosensitiser in any solat-driven system. Despite this, much research has been carried out on titania- based systems for water reduction, oxidation and splitting, as the photogenerated electrons and holes on titania have favourable redox potentials

E ~ E ’ (vs. NHE), V a t pH 0 EVE’ (Vs. NHE), V a t pH 0

-0.32 2.91 -0.1 1 2.91

( E C B -= E”(H’/Hz) and and Em >> Eo(O2/H20)). Thus, the photogenerated electrons on both rude and anatase are sufficiently reducing to be able to reduce water to Hz (E”(H+/Hz) = 0 V). The photogenerated holes are more oxidising than fluorine (E”(FJq = 2.85 V) and can oxidise water to form hydroxyl radicals (E”(’OH/HzO) = 2.31 V) or oxygen (E”(Oz/H20) = 1.23 V). Titania is the most used semiconductor in photosystems for water reduction, oxidation or cleavage.

Deposition of PGMs and PGM Oxides onto Titania

In many semiconductor-sensitised water reduction, oxidation or cleavage photosystems a hydrogen catalyst, usually a pgm, and/or an oxygen catalyst, usually a pgm oxide, is/are required, or used (not quite the same thing), see Figure 2. The Hz catalysts and 0 2 catalysts described will be limited to platinum (Pt) and ruthenium dioxide (RuOz), respectively, the most popular used. Titania is generally the semiconductor photosensitiser.

An early, now classic, approach to depositing Pt onto a titania surface is to impregnate the semiconductor powder particles with a pgm salt (for example HZPtCh) followed by reduction in a stream of Hz, at - 480°C for 15 hours (14).

Another popular method requires the initial preparation of a citrate colloid of the pgm, followed by destabdisation in the presence of the semiconductor. Destabilisation of a pgm sol is achieved by stripping the colloid particles of their electrostatically protective citrate outer layer (1 5) or by coagulating the colloid by addition of an excess of salt, such as NaCl(l6). Both approaches produce effective photocatalysts for reducing

Phfimm MetuLr Rev., 2003, 47, (1) 5

* Both Ecn and Eva v a 9 with pH: ECE = Ec&H = 0) - 0.059 x pH; E m = Eia(pH = 0) - 0.059 x pH

Table I

Electronic Properties of Rutile and Anatase

was reported in 1978 by Kraeutler and Bard, with titania as the semiconductor and water as the SED (18). To ensure complete reduction of the photodeposited pgm salt to metal, the system was irradiated under anaerobic conditions for 4 hours; further, the solution was neutralised, had an excess of hydroquinone added to it and then maintained at 50°C for 12 hours under nitrogen. Acetic acid was soon substituted as the SED, for Reaction (3) (19). Since then many different SEDs have been used, although for Pt photodeposition, methanol or ethanol is preferred.

The key processes behind pgm photodeposition onto the semiconductor particle surface (Reaction (ii)) are shown in Figure 4. Ultra- bandgap irradiation generates an electron-hole pair that can migrate to the surface, where the photogenerated electron is sufficiently reducing (see Table I) to reduce most pgm salts to their zero- valence metal form. In titania, the photogenerated hole is sufficiently oxidising to oxidise water and most organics.

The organic chosen to remove the photogenerated hole is an SED (easily and irreversibly oxidised). If methanol or ethanol is the SED, the initial product of their oxidation via the photogenerated hole or the hyclroxyl radical is an a-radical that can reduce most pgm salts. Using TiOz pho- toelectrodes in a study of this system, pgms can be deposited onto the titania surface by the oxidative route alone, without involving pgm reduction by photogenerated conductance band electrons (20). Obviously, when both pgm reduction routes (oxidative and reductive) operate (Reaction (ii) with methanol or ethanol as the SED) for each photon absorbed two reducing equivalents are produced. This is often referred to as a 'current- doubling' effect, as it was first observed as a d o u b l q in reduction current in macro-photoelectrochemical cells (20).

When pgms are photodeposited via Reaction (ii), the oxidation state of the deposited metal is assumed to be zero. (The white titania starting material is certainly grey at the end of the process!) However, if the SED is acetic acid or water, Pt is photodeposited on titania in the Pt (0) and QI), or Pt QI) and (rv) oxidation states, respectively, and

Fig. 3 Tvpicul tran.smi.s.sion electron micrograph of a Degussa P25 titania photocatalyst having Pt island deposits on its surfuce. generuted by a P t colloid coagirlatioti method (16). The black arrowhead points to one of the Pt islands. ?picallv - 4 nm in diameter

water. A typical titania photocatalyst (Degussa P25) with a polka-dot coating of - 4 nm Pt part- cles, created by the colloid coagulation method is shown in Figufe 3.

Another deposition method involves the room temperature chemical reduction of a pgm salt in the presence of the semiconductor in aqueous solution. The reducing agent is usually sodium borohydride but, more recently, zinc powder has been used (17).

However, the most common method of depositing metals, especially pgms, onto titania and other semiconductor photocatalysts, is by photodeposition:

semiconductor

hv L Ebg M"' + SED A Md + products (ii)

where M"' is the pgm salt and the SED is methanol or ethanol, but can be almost any easily oxidised organic solvent, or dissolved species (such as EDTA or cysteine), or even water. Before irradiation the photosystem must be rendered anaerobic (usually by sparging the reaction solution with nitrogen or argon) since 0 2 interferes with pgm reduction. The first example of Reaction (ii)

P&tinum Metah Rev., 2003, 47, (1) 6

kM‘

irrev decornp.

products

irrev decornp.

products

Fig. 4 Schematic ilhrstrution showing, under anaerobic conditions, the electron energetics associated with the photoreduction of a metal salt (M”’) to.form metul deposits (M). The sucriJicial rlectron donor (SED). is .sensitised bv u particle of senziconductor photocutulyst

not Pt(0) as expected and often presumed (21). If the pgm/semiconductor photocatalyst produced is used to promote a reduction reaction, such as water reduction, any pgm deposit not in the (0) oxidation state will quickly become so upon illumination, due to the reducing nature of the environment. The need to reduce the pgm deposit to the (0) oxidation state may be the cause of the initial ‘induction’ period sometimes observed in such systems when hydrogen photogeneration is monitored.

Physical and Chemical Properties Any semiconductor with a pgm (usually Pt)

deposited on its surface has very different physical and chemical properties to the original uncoated semiconductor material. For example, plaiinised titania samples form poorer dispersions than unplatinised titania and do not adhere as well to borosilicate glass (22,23). The pzc (point of zero charge) and pzzp (point of zero zeta potential) of a platinised titania photocatalyst decrease with increasing Pt loading. Thus, at a 5% Pt loading, the pzc and pzzp for Pt/Ti02 are 5.6 and 5.1, respectively, whereas for naked Ti02 (Degussa P25 Ti02) they are 7.6 and 6.3, respectively (24). The amount of oxygen photoadsorbed by a titania photocatalyst decreased with increasing Pt loading (25). As factors which determine the overall rate of the process, all these characteristic differences may be as important (if not more so) as the assumed

electron-trapping and catalytic actions of the pgm, especially when taken in combination.

Photodeposition of a pgm oxide (Ru02) onto the surface of semiconductor particles is usually achieved by either alkaline aerobic hydrolysis of a pgm salt, such as RuC?, to RUOZ (26), or thermal decomposition of the pgm in a high oxidation state, such as RuO, to RUOZ (27). When an aqueous dispersion of the semiconductor powder is stirred at room temperature with the pgm chloride or high oxidation state pgm oxide, the product is likely to be a highly hydrated form of the oxide: Ru02.xHz0 or ‘RuOq, which is not a good oxygen catalyst as it is prone to oxidative corrosion (12). Ru02.xH20 is converted into a more stable and active oxygen catalyst, such as Ru02 by heat treatment. This ‘thermal activation’ step rarely seems to be applied and its omission may explain the often less-than-beneficial effects of ‘RuOZ’ deposits on sacrificial semiconductor photocatalytic systems for water oxidation. Indeed, today, most 02-evolving semiconductor-sensitised photosystems do not use a pgm oxygen catalyst, as it appears superfluous.

Photoreduction and Oxidation of Water

In the late 1970s and early 1980s research into artificial photosynthetic systems, such as water splitting, reached a peak. Work on macro-photoelectrochemical cells incorporaung single-crystal semiconductor photoanodes had advanced to micro-photoelectrochemical cells based on semiconductor particles (28). Water photocleavage using such cells was problematic, but the use of semiconductor particles as micro-photoelectrochemical cells to mediate water reduction (and concomitant oxidation of an SED, such as EDTA or methanol), or water oxidation (and concomitant reduction of an SEA, such as Fe” ions) proved easy to achieve and highly reproducible.

Most photocatalysts were able to mediate water reduction to Hz by SEDs only if a suitable Hz catalyst, such as a pgm, was present. The pgm was usually deposited onto the semiconductor by one of the techniques described earlier. The system also worked well if the pgm was simply mixed in with the semiconductor in a finely divided form,

Phfinum Metub Rev., 2003,47, (1) 7

Table II

Systems for the Semiconductor-Sensitised Photoreduction of Water to Hydrogen

Semiconductor

Ti02

TiOz TiO2 Ti02 TiO2 8, TiOdSi02 TiO2 Ti02 TiO2

Electron donor

biomass (including grass, wood, algae, seaweed, cockroach and urine)

glucose MeOH, tPrOH

i-PrOH MeOH-f-BuOH

EDTA MeOH EtOH

Hydrogen catalyst

Pt (4%)

Pt (1 2-4%) Pt (10-0.05%)

Pt, Pd, Rh, Au (0.5%) Pt (0.6%)

Pt (2-0.1 Yo) Rh(bW33'

Pt

for example Pt black. The basic overall process can be summarised as follows:

semiconductor/ pgm

hv 2 Ebg SED + 2H' products + Hz? (iii)

The semiconductor is invariably anatase. Figure 2(a) shows the electron transfer processes associated with Reaction (iii).

Table I1 contains some of the (now hundreds) of TiOJpgm sacrificial systems for water reduction (29-36). The most intriguing is the first entry, where the SED is various types of biomass (29). Such micro-photoelectrochemical systems were

0.02-0.04

0.085 - - -

0.03 - -

References

then being promoted as potential solar-driven biomass conversion systems to deal with unwanted biological waste material, such as sugar cane/maize waste from alcohol production. Nothing came of this idea, probably due to difficulties of scale-up and the low conversion efficiencies of such systems. Certainly the quantum yields for hydrogen evolution, @(H2) , are very low (typically 2-4"/0), and as noted earlier, sunlight contains little W.

An example of Reaction (iii) is given in a study of the photoreduction of an ethano1:water mix

(5050 V/V) by titania with various levels of deposited Pt (36), see Figure 5. The rate of hydrogen

E 0 91f W

29

29,30 31 32 33 34 35 36

c

z I

I? -""I

m j 250

W ^. ^ ^ 0.3 0.4 0.5 n u. I U.L

r Pt I

i5 I 1

10 20 30 P t I T i 0 2 . w t .I.

4 0

Fig. 5 The meusured vuriatioii of'tiie rate of'liydrogen evoiirtioti u s u,firnction of )!'I. !% Pt 011 u Pl/TiO> photocutu!vst wheti illuminnted. The SED is ethanol (50:50 v h MVth water). [Pt/TiO2] = (10 g dni') . The light source M'US (I

500 W Xe lump. The irwert diugrum is an enlurgemerit of' dutu in the niuiri diugrani over the range 0-0.4 wt.% Pt

Phfinum Metdr h., 2003,47, (1) 8

Table HI

Systems for the Semiconductor-Sensitised Oxidation of Water to Oxygen

Semiconductor

Ti02 Ti02 Wos Wos WOa and Ce02 Ti02 Ti02-WO3

Electron acceptor

Fe3'

Fe3' or none

Fe3' and Ce4' Fe3' Fe3'

[PtC16]*-

Ag'

Oxygen catalyst

Rh, Ru, Ir, Au, Co or none

Pt, Rh, Ru, Ru02 or none

none Ru

none

-

RUOZ.XHZO

photoevolution is shown as a function of the per- centage of Pt (w/w) deposited onto the titania powder. The Pt is essential for the overall process to occur, but hgh (> 20 wt.%) Pt levels seem to have a detrimental effect on the overall kinetics of the process, probably due to surface screening by the Pt particles as well as an increased likelihood of electron-hole recombination. This feature is common and often more clearly seen in this area of semiconductor photocatalysis.

The semiconductor-sensitised photocatalytic oxidation of water by an SEA can be expressed by:

semiconductor/pgrn oxide

hv 2 E b g

SEA + 2H20 > products + OZ? (iv)

Figure 2@) shows the electron transfer processes associated with Reaction (iv). Curiously, there are far fewer studies of Reaction (iv) than of Reaction (iii). This may be because the H2-evolving system, nominally providing a route to generate a useful fuel, would have attracted fund- ing via the 'alternative energy' initiatives available then. Oxygen photogeneration from water appears to be a less 'useful' process to study!

Table I11 contains examples of Reaction (iv) (37-42). These show the dominance of titania as

the semiconductor photocatalyst. Many of the early examples illustrate the assumption made in the 1980s, that for efficient water oxidation, an oxygen catalyst, such as RuOz, should be added to the system, usually as h e particles on the semiconductor photocatalyst. In fact, the photo-

@(OZ) I References

0.0031

- - - 42

generated hole on titania and wo3 (the two most commonly used semiconductor sensitisers) is so oxidising that an oxygen catalyst seems not to be needed. Work has also shown that water photooxidation is possible without a pgm oxide oxygen catalyst, if wo3 is the photosensitiser and Fe3+ ions are the SEA (38). Research shows that the addition of an oxygen catalyst, such as RuO,, has little or no effect on the ability of Ti02 or WOS to oxidise water to O2 (37,38).

The variation in concentration of the photogenerated oxygen (measured using a Clark cell) in a W03/Fe3+ photosystem was monitored as a function of irradiation time for a series of repeat irradiations, see Figure 6. The rate of photogenerated oxygen decreases with repeated use. This is due to inhibition by ferrous ions (Fez+) photogenerated during the process (38). Thus, not surprisingly, Fe3+ ions are not a particularly good example of an SEA material, since the product of electron scavenging, Fez+, interferes with the hole scavenging process - water oxidation to oxygen.

The Photocleavage of Water The semiconductor-sensitised photocleavage

of water into hydrogen and oxygen can be summarised as follows:

H, catalyst/ semiconductor/O, catalyst

hv 2 E b g

2H20 > 2H2f + Oz? (v)

where the H2 catalyst is usually Pt, and the O2 catalyst is RuO2 (or more usually RuOZ.XHZO) or

Phtinum Metah Rcv., 2003,41, (1) 9

37 11 38 39 40 41

6.

m

‘ E 4 - -

E ‘0 0

- 2- 6 u

nothing. The semiconductor photosensitiser is invariably titania or SrTi03. Reaction (v) has been and is the subject of controversy. The early claims of ‘bifunctional’ photosynthetic systems, based on titania, capable of splitting water with high quantum efficiencies (up to 30”/.) under ambient conditions appear irreprodudble, and now, with hindsight, unjustified (2, 27). However, interest in water splitting has recently increased and those involved suggest that forcing conditions, such as reduced pressures and elevated temperatures, are likely to be needed, to help strip out the photogenerated 0 2 and H2 from the system before they have time to back react (2,43).

Table IV (4442) lists most of the reported titania- and SrTiO3-based water splitting photosystems. The new interest stems from work by Sayama and Arakawa in the early 1990s showing that while Reaction (v) is difficult or impossible in water at any pH, it can be readily achieved in concentrated (> 2 M) sodium carbonate solution, using platinised titania (50). The role of the carbonate is not as a pH effect, instead, it is proposed that peroxycarbonates form in solution and these somehow enhance oxygen evolution (43).

While it is now accepted that a pgm oxide is not needed to act as an oxygen catalyst for Reaction (v), it might be thought that Pt or another pgm will

always be needed in any system where water is reduced to Hz. This may not be the case. Recent work has successfully achieved Reaction (v) using

Fig. 6 Psqfiles of the observed dissolved [OZ] ivr.siis illiimiricition time /?IV ( I WOJ disyeraiori (6.5 g dr~ii’) iri 0.005 r d d d HrSOj. coritaining 0.01 rnol cIrii-3 ,/i.rric cliloride. Pl~o/;les ( U - ( d ) were rc~cordecl coriseciitivrlr r i r i t l sho\c. that the rote o f wnter o.\-iricrtiori tiecrcwses wfth rrperrted irrotlititiori. This is dire to irihihitioii hj,,fi.rroirs ior i s ,/oi.riied duririg the oiwcrll light driveti p‘ocrss. The light source \ i v s t i 900 W ltimp,fitttd u.itli (I liecrt arid UV,/ilter.

a layered perovskite, such as K2LazTi,0io, with - 3 wt.% deposit of nickel (Ni) as a photosensitiser material but without a pgm H2 catalyst. The Ni may be acting as a H2 catalyst (61).

Visible Light Splitting All these water-splitting systems require W-

light to drive them forward, and although the visible light photocleavage of water via Reaction (v) has been claimed, notably using CdS as the photocatalyst, these fmdings are even more controversial and irreproducible than those based on titania (2). Reaction (v) has recently been driven by visible light alone, using delafossite (CuFeO2) as the photosensitiser, without a separate Hz or O2 catalyst, pgm-based or otherwise (43). Others have reported that Inl-xNixTa04 (x = M . 2 ) coated with NiO, or ‘RuOZ’, is a visible-light activated water- splitting photocatalyst (62). However, all claims of water splitting are still controversial and require reproduction by other groups.

Even if the systems in Table IV can be repro- duced, albeit inefficiently, questions of longevity and their photosynthetic nature remain unan- swered. The long search for the ‘Holy Grail’ of photochemistry: a semiconductor-sensitised system for water splitting that is efficient, long-lasting, reproducible and driven by visible light, is still far away. If such a system is discovered it is hard not to believe that at least one component will be a pgm, and most probably Pt. Although there is

Phtinnm Mefah h., 2003, 47, (1) 10

Table IV

Systems for the Semiconductor-Sensitised Photocleavage of Water into Hydrogen and Oxygen

Semiconductor

TiO?

Ti02

Ti02 doped with Cr

SrTiOa

SrTiOs

Ti02 and B/TiOn

NiO-Ti02

Zr02

BaTi409

K2La2Ti301 o

Ni,Ta04

CuFeOz

Hydrogen catalyst

Pt

Pt

Pt

Pt, Ir, Pd, Os, Ru, Re or Co

Pt

Pt

none

none

none

Ni

none

none

much that still needs to be achieved in developing semiconductor-sensitised systems for water splitting, another area of research that involves semiconductor photochemistry and pgms appears to offer great commercial promise; namely, water and air purification by semiconductor photocatalysis. This will be discussed in a later issue of this Journal, and the role of pgms, most notably Pt and Pd, will be described.

Conclusions Semiconductors can photosensitise several dif-

ferent reactions, usually with titania as the semiconductor because of its excellent photochemical and chemical stability and high activity as a sensitiser. In many cases the presence of a pgm, or pgm oxide, can improve the overall efficiency of the photochemical reaction. Of the many methods of depositing a pgm onto a semiconductor surface, the most popular involves photodeposi- don, with ethanol or methanol as an SED. Depositing a pgm oxide, such as RuOz, can be achieved by oxidation of a pgm chloride or reduction of a volatile pgm oxide. The physical characteristics of a pgm-coated semiconductor are

Oxygen catalyst

none

‘ R u O ~ ’

‘RuOn’

none

none

none

none

none

‘ R ~ 0 2 ’

none

‘RuOn’ or NiO

none

References

44-50

26,51, 52

53

54

55.56

57

58

59

60

61

62

43

quite different from those of the original semiconductor material. Semiconductor photochemistry has focused on systems capable of the photoreduction, oxidation and cleavage of water.

In water photoreduction systems, a pgm as a H2 catalyst is essential, either deposited on the semiconductor surface or intimately mixed with it. Many systems have been reported and may be precursors of a water-splitting system. The semiconductor-sensitised photooxidation of water has been less well studied and using a pgm oxide as an 0 2 catalyst does not appear essential, especially if a stable oxide semiconductor is used as the sensitiser. The latest systems resulung from the renewed interest in heterogeneous water-splitting photosystems appear to work under visible light illumination without a pgm-based HZ and/or 0 2

catalyst. However, all reported water-splitting systems are controversial and require confirmation. Further work is certainly required to create a reproducible, stable, efficient photosystem for water splitting probably requiring one or more pgms.

Acknowledgement The authors wish to thank the EPSRC for f inand support.

Phfinwn Metals Rev., 2003, 47, (1) 11

0.0066

1

2

3

4

5 6 7 8

9

10 11

12 13 14

15

16 17

18

19

20

21

22

23

24

25

26

27

28 29 30

31

32

33 34

References N. Serpone and A. V. Emeline, Int. J. Pbotoenqy, 2002, 4, 91 A. Mills and S. Le Hunte, J. Pbotocbem. Pbotobiol. A: Cbem., 1997, 108, 1 M. R Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann, Cbem. Rey., 1995,95,69 A. L. Linsebigler, G. Lu and J. T. Yates, Cbem. Rev., 1995,95,735 J. W. M. Jacobs, J. Pby. Cbem., 1986,90,6507 D. E. h P n a and A. H d a , J PbS. cb., l983,87, 4919 H. Gerischer, J. Pbs. Cbem., 1984,88, 6096 T. Tatswa, %-I. Tachibana, T. Wwa, D. A. TTk and A. Fujishima,]. Pbs. Cbem., 1999, 103, 8033 w. w. D m , y- h w a and A. J. Bard,]. c h m . Soc., 1981, 103,3456

J. S. Curran, J. Domenech, N. Jaffrezic-Renault and R Philippe, J. Pbs. Chem., 1985, 89,957 A. Mills, Cbem. SOC. Rey., 1989,18,285 A. Mills and S. Moffls, q. flt., (Ref. 2), 1993,71,285 P. Pichat, J.-M. Herrmann, J. Disdier, H. Courbon and M.-N. Mozzanegay J. 1981, 5* 627 D' w. Bahnemann, J' Monig and Chapman1 1. P ~ J . Cbem., 1987,91, 3782 A. Mills, J. Cbem. SOL, Chem. Commun., 1982,367 J. C. Crittenden, J. Liu, D. W. Hand and D. L. Perram, Wuter Res., 1997,31,429 B. Kraeutler and A. J. Bard, J. Am. Cbem. Soc., 1978, 100,2239 Commun., 1992,150

B. Kraeutler and A. J. Bard, J. h. Cbem. SOC., 1978, 100,5985 R Baba, R Konda, A. Fujishima and K. Honda, Cbem. Lett., 1986,1307 C. Sungbom, M. Kawai and K. Tanaka, BuU Cbem. SOC. Jpn., 1984, 57, 871 G. A-Sayyed, J.-C. DOliveira and P. Pichat, op. cit., (Ref. 2), 1991, 58, 99 D. Hufschmidt, D. Bahnemann, J. J. Testa, C. A. Emilio and M. I. Litter, op. cit., (Ref. 2), 2002, 148, 223 N. Jaffrezic-Renault, P. Pichat, A. Foissy and R. Mercier, J. Pbys. Cbem., 1986,90, 2733 H. Courbon, J. M. Herrmann and P. Pichat, J. Pbys. Cbem., 1984, 88, 5210 E. Borgarello, J. Kiwi, E. Pelizzetti, M. Visca and M. Graetzel, J. h. Cbem. SOC., 1981,103,6324 D. Duonghong, E. Borgarello and M. Graetzel, ]. Am. Chem. SOC., 1981,103,4685 A. J. Bard, J. Pbotocbem., 1979,10, 59 T. Sakata and T. Kawai, Now. J. Cbem., 1981,5,279 M. R St. John, A. J. Furgala and A. F. Sammells,J. P l y . Cbem., 1983,87,801 p. M.-N' Mozzme@l J' Disdier and J'-M' Herrmann, Now. J. Chem., 1982,6,559 F. H. Hussien and R Rudham, J. Cbem. Soc., Furu& Truns. I, 1984, 80, 2817 O. Enea and A. Ali, A. Mills and G. Porter, op. it., (Ref. 32), 1982, 78, 3659

35 P. Cruendet, K. K. Rao, M. Gratzel and D. 0. Hall, Biocbemie, 1986, 68, 217

36 T. Sakata, T. Kawai and K. Hashimoto, Cbem. Pbys. Lett.., 1982, 88,50

37 A. Mills and G. Porter, op. flt., (Ref. 32), 1982,78,3659 38 J. R. Darwent and A. Mills, J. Cbem. Soc., Fur+

Truns. 11, 1982, 78, 359 39 W. Erbs, J. Desilvestro, E. Borgarello and M.

Gratzel, J. Phs. Cbem., 1984,88,5827 40 G. R. Bamwenda, T. Uesigi, Y. Abe, K. Sayama and

H. Arakawa, &pl. Cutd A: Gen., 2001, 205, 117 41 T. Ohno, F. Tanigawa, K. Fujihara, S. Izumi and M.

Matsumara, q. it., (Ref. 2), 1999,127, 107 42 T. Ohno, F. Tanigawa, K. Fujihara, S. Izumi and M.

Matsumara, op. cit., (Ref. 2), 1998,118, 41

Domen, cutuL T ~ & ~ , 1998,44,17 44 G. N. Schrauzer and T. D. Guth, J. Am. Cbem. SOC.,

1977,99,7189 45 H. van D m e and W. K. Hall,]. &. Cbem. Soc.,

1979,101,4373 46 S. Sat0 and J. M. White, Chem. P l y . Lett.., 1980,72,83 47 T. Kawai and T. Sakata, Cbem. Pbys. Let., 1980,72,87 48 F. T. Domen and G. A. Somojai, Nufure (London),

1980,285,559 49 S. Sato, New]. Cbem., 1988,12, 859 50 K.

51 M. Gramel, 52 E. Borgarello, J. Kiwi, E. Pelizzetti, M. Visca and M.

Gratzel, Nuhm (London), 1981,289,158 53 E. Borgarello, J. Kiwi, M. Gratzel, E. Pelizzetti and

M. Vista,]. Am. cbem. 3% 1982,104,2996 54 J. M. Lehn, J. P. Sauvage and R. Ziessel, Now. J.

Cbim., 1980,4,623 55 K. Yamaguti and S. Sato, Nouu. J. Cbim., 1986,1,217 56 K. Domen, A. Kudo, T. Onishi, N. Kosugi and H.

Kuroda, J. Pbs. Cbem., 1986, 90,292 57 S.-C. Moon, H. Mametsuka, S. Tabata and E.

suz&, cutul ~ ~ 4 , 2 0 0 0 , 58, 125 58 A. Kudo, K. Domen, K. Maruya and T. Onish,

Cbem. Pbys. Let., 1987,133,517 59 K Sayama and H. Acakawa, J. Pbs. Chi.., 1993,97,531 60 Y. Inoue, Y. Asai and K. Sato, J. Cbem. SOC., Far+

Trans., 1994, 90, 797 61 T. Takata, K. Shinohara, A. Tanaka, M. Hara, J. N.

Kondo and K. Domen, op. it., (Ref. 2), 1997,106,45 62 2. Zou, J. Ye, K. Sayama and H. Arakawa, Nature,

2o01, 414, 625

F. Howe and M- =tzdJ. PbS. Cbem., 1987391, 3906 43 T. Takata, A. Tanah, M, Hma, J, N, Kondo and K.

and H' Arakawa, 1'

1981, 14, 376

The Authors Andrew Mills is a Professor of Physical Chemistry at the University of Strathclyde. His interests include dye and semiconductor photochemistry, electrochemical sensor development for gases, sensors for use in clinical analysis and catalysis of redox reactions.

Soo-Keun Lee is a Postdoctoral Research Fellow at the University of Strathclyde. His research interests include semiconductor photochemistry, optical sensors and laser photochemistry.

1988, 12, 853

Phtinum Metub Rm, 2003,47, (1) 12

Catalvst Life Cvcle Conference J J

By Stephen Poulston and Stephen Pollington Johnson Matthey Technology Centre, BlOUntS Court, Sonning Common, Reading RG4 9NH, U.K.

The Catalyst Life Cycle conference, held at the University of Bath on 1G17th September, 2002, was a joint meeting of the U.K.’s Royal Society of Chemistry and Institution of Chemical Engineers under their subject groups of SURCAT, Process Technology and Applied Catalysis. This is a biannual thematic meeting, with a previous one in Edinburgh being on kinetics and in silu methods in heterogeneous catalysis. Approximately 75 partid- pants, mostly from U.K. universities and industry, attended the conference. There were 16 oral presentations - half of which were concerned with the platinum group metals (pgms) - and a number of poster presentations. As expected from the spon- soring groups, the talks provided an interesung combinadon of applied catalysis and more fundamental catalytic chemistry on the microreactor scale. With the conference subtitle of ‘Preparation, Activation, Deactivation and Regeneration’, most talks focused on catalyst deactivation/regeneration and covered such diverse areas as large scale industrial reactions (for example, catalytic combustion and partial oxidation), automotive catalysis, fuel cells and line chemical synthesis.

The opening invited lecture by John Birtill (consultant, U.K.) was entitled ‘But will it last until the shutdown?’. This presentation introduced techniques for interpreting and predicting the effects of catalyst decay in industrial process cad - ysis. Kinetic expressions for describing catalyst decay and possible approaches for determining the mechanism of decay were discussed.

From the industrial plant scale the talks moved to the nanoscale with Mike Bowker (University of Reading, U.K.) describing scanning tunnelling microscopy studies of model supported-catalysts. Atomic scale processes can be observed for pgms (Rh and Pd) deposited in situ on TiOz single crys- tals in ultra hgh vacuum conditions. Effects such as spillover of chemisorbed oxygen, encapsulation under oxidising conditions and compound forma-

tion between metal and the reduced support were demonstrated.

The second invited lecture, given by Pierre Gallezot (CNRS, Villeurbanne, France), was on the stability of platinum metal catalysts in the liquid phase hydrogenation and oxidation of organic compounds. He began by outlung the main causes of catalyst deactivation: sintering, leaching and poisoning by strongly adsorbed species. One example he described was the hydrogenation of glucose to sorbitol which is used in the food and pharmaceutical industries. This reaction is usually performed with a Raney Ni catalyst which deactivates by leaching of the metal promoter. An alternative catalyst that can be used is ruthenium (Ru) on activated carbon. This has &her activity, especially when platinum (Pt) is added, and gives extremely high conversion/selecuvity and &her stability than the Ni catalyst. A second example described was the total oxidation of organic pollutants in water. This uses a Ru/Ti02 catalyst which provides advantages over homogeneous catalyst alternatives such as Cu salts, which are efficient but have issues associated with catalyst recovery.

The third and final invited lecture was by Malcolm Green (University of Oxford, U.K.) who discussed new catalysts for the conversion of methane (CI-L) to liquid hydrocarbons. Besides looking at catalysts for syngas production from CH4, he demonstrated the performance of a new Co-based catalyst for Fischer Tropsch synthesis suggesting that gas to liquid technology would become increasingly important in the near future as oil reserves are depleted.

David Jackson (University of Glasgow, U.K.) presented work on butane dehydrogenation over a Pt/ALO, catalyst using a pulsed flow microreactor. Changes in product formation and carbonaceous residue were observed for sequential pulses of butane for different catalyst pretreatments. The first pulse produced no butene but only CI-L and

P h t i n m ~ Mefais Rm, 2003,47, (l), 1%14 13

carbonaceous residue. With subsequent pulses of butane the build up of the carbonaceous deposit increased and the selectivity to butene also increased.

Jenny Jones (University of Leeds, U.K.) presented work on sulfur poisoning of CH, combustion catalysed by Pt, Pd and Rh. Sulfur compounds, such as mercaptans, are used as odorants in domestic supplies of natural gas, so the effects of a range of sulfur compounds on catalyst deactivation were studied. Catalysts used were 2 wt.?!o

Pd, Pt and Rh on alumina with a gas feed of 4% CH4 in air. Activity was found to increase in the order: Pt < Rh < Pd. TEM helped to distinguish between sulfur-induced agglomeration and simple site blocking effects in catalyst deactivation.

Automotive Catalysis The final session contained three presentations

on automotive catalysis. The first two were concerned with NOx storage/reduction (NSR) catalysts for NOx control in lean-burn engines. These catalysts consist of a Pt/Ba component which stores NOx as nitrate during lean engine operation. The engine periodically runs rich and under this condition the nitrate is decomposed and the released NOx is reduced with the help of a Rh component in the catalyst. David James (University of Reading, U.K.) presented pulsed flow microreactor work on the effect of Pt loading on the decomposition of Ba(N03)z and the difference between using Hz or CO as the reductant. Pt was shown to promote nitrate decomposition with the extent of the promotion increasing with Pt loading. H2 was shown to be superior to CO in NOx reduction.

The second presentation on NSR catalysts was by Stephen Poulston (Johnson Matthey) who described the effects of gas composition on the regeneration of nitrated or sulfated model NOx storage catalysts consisting of Pt/Ba. Again the advantage of Hz over CO as a reductant was highlighted. The influence of COZ in the simulated gas feed was also described. COZ is often omitted from synthetic exhaust feed compositions but has an effect on catalyst regeneration by lowering the temperature of Ba nitrate and sulfate decomposition.

The final talk by James Anderson (University of Dundee, U.K.) was on the use of an oxychlorination treatment to regenerate engine-aged three-way catalysts. Oxychlorination partially recovered cab- lyst activity and, in the case of Pd-based catalysts, also the oxygen storage capacity; there was some evidence for redispersion of Pt but not of Rh.

Conclusions This short highly-directed conference provided

an insight into many of the current themes of interest to catalyst scientists. Selected papers will

be published in a special issue of Cata&is Tohy. The next conference is expected to be a meet-

ing to mark the retirement of Professor Geoff Webb on 1618th July, 2003, at the University of Glasgow (http://www.chem.gla.ac.uk/colloquia/ catalyst/Catalysis-Symposium.html). For further information visit the websites at www.rsc.org and www.icheme.org.

The Authors Stephen Poulston is a Senior Scientist at the Johnson Matthey Technology Centre. His interests lie in automotive catalyst technology.

Steve Pollington is a Senior Scientist at the Johnson Matthey Technology Centre. He is interested in process catalysis and automotive catalyst technology.

Rhodium Dendrimer Catalvsts Rhodium (Rh) complexes have high reactivity

and selectivity and could be used in the hydroformylation of long chain alkenes to form aldehydes - an important industrial process. However, due to their difficult and costly recovery this particular use of Rh has been restricted.

Now, researchers at the University of St. Andrews, Scotland, have successfully used diphenylphosphine functionalised polyhedral oligomeric silsesquioxane (POSS) dendrimers (with 16 or 48 diphenylphosphine end groups) as ligands for the Rh catalysed hydroformylation of oct-1-ene (L. Ropartz, K. J. Haxton, D. F. Foster, R E. Moms, A. M. 2. Slawin and D. J. Cole-Hamilton,J. Cbem. SOC., DaLton Trans., 2002, (23), 432M334).

Unexpectedly high regioselectivity to the linear aldehyde (86Yo) was obtained with a POSS dendritic ligand (with a spacer of five atoms between the P atoms, and C-Si linkage). Small molecule analogues and other dendritic ligands had lower selectivity.

Phtimm Metub Rev., 2003, 47, (1) 14

Vehicle Emissions Control Technologies By Martyn V. Twigg Catalytic Systems Division, Johnson Matthey, Orchard Road, Royston, Hertfordshire SG8 5HE, U.K.

The world’s largest and most important automotive congress is held yearly in Detroit by the Society of Automotive Engineers (SAE), and regu- larly attracts over fifty thousand delegates. There are many technical sessions, with those devoted to describq advances in the control of exhaust g a s

pollutant emissions always being well attended (1). In this paper, key trends in catalyst emissions control systems which depend upon platinum group metal @gm) catalysts are illustrated by research described in a small number of the papers presented at the 2002 World Congress (2). The reference numbers of the papers are cited in parentheses (3).

Three-Way Catalysts for Conventional Gasoline Engines

California is the home of the demandmg legislation that is driving emissions control standards towards ever lower levels. The California Super Ultra Low Emissions Vehicle (SULEV) standards will reduce hydrocarbon (HC) emissions from cars in the American Federal Test Procedure (FIT’) to 0.01 g/mile after 120,000 miles. The corresponding engineering targets required for series vehicle production are, of course, even lower than this. Such low levels can only be achieved with a combination of precise engine fuelling, efficient in-cylinder combustion and a highly efficient three-way catalyst (IWC) system. To achieve the SULEV hydrocarbon levels, the

catalytic conversion of HC in the catalyst system must be fully functioning within a few seconds of starting an engine. To facilitate rapid hght-off a catalyst is mounted as close as possible to, or on, the exhaust manifold, but in this position it experi- ences higher operating temperatures than when located in the traditional underfloor location. Occasional engine malfunction could also expose this catalyst to very high temperatures because of its proximity to the engine. For example, engine misfires with concomitant substantial HC oxida-

tion exotherms could take place over the catalyst. Therefore, when mounted on or near the manifold, catalyst longevity has often been a problem. Johnson Matthey and Ford have reported results (2002-01-0351) from a new generation of high performance TWCs of exceptionally lvgh thermal durability which alleviate this problem. During evaluation of these catalysts the target catalyst ageing temperature was increased from the usual 960 to 1050°C in an ageing cycle that had occasional temperature excursions up to 1080°C. Even with this very harsh ageing both the new platinum/ rhodium (Pt/Rh) and palladium/rhodium (Pd/Rh) catalysts, with pgm loadings less than a quarter of what was originally used, achieved European Stage IV emissions limits on a 1.6 litre engine.

The optimisation of pgm levels is a recurring theme for TWCs. For example, OMG have results (2002-01-0345) from a computer model calibrated using emissions data from catalyst on one car. The model was then applied to another vehicle equipped with front and underfloor catalysts. Increasing the catalyst cell-density to above 600 cpsi (cells per square inch) gave little improvement, and, when compared to the original manufacturer’s system, the volume of the front catalyst could be reduced by some 40% without excessively increasing HC emissions. In addition, the pgm loadmg of the underfloor catalyst could be reduced by as much as 500/0. This work highhghts how effective new catalyst formulations are, and that the performance of the underfloor catalyst depends strongly on its interaction with the air/fuel control system (especially when there is rear sensor control) and on the response characteristics of the oxygen storage component in the catalyst.

Catalyuc Solutions (2002-01-0344) have been working on a sport utility vehicle (Sw> with a 4.6 litre V8 engine that has advanced LEV-I1 calibration, and also on a car equipped with a 2.4 litre four-cylinder engine having a SULEV calibration.

Phiinurn Metals h., 2003, 47, (l), 15-19 15

Hgh cell-density, low thermal-mass ceramic substrates (900 cpsi for the S U V , 600 and 900 cpsi for the car) were used to facilitate fast hght-off. California reformulated gasoline with a low sulfur content (40 ppm S) was used in the vehicle tests. The TWCs were aged at 950°C for the SUV, and at

900°C for the car, and not surprisingly LEV-I1 emissions standards were met on the SUV with less catalyst volume and less pgms than ollginally. The car with the SULEV calibration achieved ULEWII limits with relatively low pgm loadings. There are benefits from having a short zone of relatively hgh pgm loading on the front part of a catalyst and such ‘zoned’ or ‘striped‘ catalysts are already manufactured by other catalyst companies.

During hard accelerations and hgh-speed cruises nitrogen oxides @Ox) are emitted at the hghest levels. In cars having two catalyst systems, the larg- er underfloor catalyst is normally responsible for controlling NOx emissions. N.E. ChemCat (2002- 01-0348) have shown that metallic-state rhodium is the most effective active phase for reduction of NOx to nitrogen @2), according to Equations (i) and (ii):

2 N 0 + 2C0 -b Nz + 2C02 (i) 2 N 0 + 2H2 + N2 + 2H20 (4

While NOx can be reduced by carbon monoxide (CO), hydrogen (H2) is particularly effective. Hydrogen can be obtained by HC steam reforming, and also from CO via the water gas shift reaction, see Equations (iii) and fiv), respectively, both being efficiently catalysed by rhodium:

‘CH2’ + H2O + CO + 2Hz CO + H2O + C02 + H2

(iii) 6 4

An enhancement in the steam reforming reaction, which both removes HC and provides H2 for facil- itating NOx reduction, appears to occur in advanced TWCs that have maximised rhodium function. Under appropriate conditions the catalyst described by N.E. ChemCat enabled some 50% reduction in pgm usage.

A zeolite-based catalysed hydrocarbon-trap (CHTTM) can help to control the initially formed HCs by retaining them until the catalyst light-off temperature is reached. The effectiveness of such a

system depends on a number of often application- dependent factors, such as the nature of the zeolite used to retain the HC, and the types of HC involved at cold start. Older engines produce large amounts of unburned fuel-derived long chain HCs, whereas the cleaner engines in new cars produce lower levels of these HCs. Johnson Matthey (2002-01-0730) showed results obtained when the front catalyst on an SUV engine (that had both close-coupled and underfloor catalysts) was replaced by a CHTTM. In all cases the CHTm improved the HC conversion. Such systems could be used in series production when HC absorbents, of the same high thermal durability as the advanced, exceptionally stable state-of-the-art TWCs, become available.

The advantages of combining advanced TWC formulations and ultra-&-wall ceramic substrate have been explored by, for example, NGK, DaimlerChrysler and OMG (2002-01-0349). They have concluded that there is little to be gained from using cell densities hgher than 900 cpsi. Catalyst canning concerns and pressure-drop con- siderations associated with high cell-density catalyst are seen as important and, in any case, the strategy for cold-start warm-up usually has more importance than catalyst cell-density effects.

NOx-Storing Catalysts for Lean-Burn Gasoline Engines

Operating a gasoline engine under lean conditions can improve fuel economy as fuel is burnt only when needed, particularly under part-load conditions, during idling and on low-speed cruises when full engine power is not required. Running the engine in this way is similar to the operation of a diesel engine, but it is more complex to obtain smooth combustion with gasoline spark-ignition engines, and optimal performance requires stratification of the &/fuel mixture in the cylinder. The most common technology being developed to achieve the stratification and thus optimal performance involves the direct injection (Dn of fuel into each cylinder. Normal TWCs are not good at removing NOx from the exhaust of lean-bum engines, because NOx cannot be efficiently reduced to nitrogen in the presence of excess

Phtinum Metah Rey., 2003,41, (1) 16

oxygen. Therefore, an alternative approach is being used which stores NOx as nitrate phases under lean conditions, in a so-called “Ox-trap’. Periodically (every few minutes) the nitrate is converted to nitrogen by fuel-rich pulses in the exhaust gas; the pulses are produced by suitable adjustment of the engine management system. These reactions are illustrated by the idealised Equations (v)-(viii). Here, MO is a basic metal oxide, and under exhaust-gas operating conditions the stable species will be the carbonate rather than the oxide. Equations (v) and (vi) describe the NOx-trapping process, and Equations (vii) and (viii) describe the regeneration process under rich conditions. Equation (v) is catalysed by Pt and Equation (viii) by Rh.

NO + %Oz + N 0 z (v) NO2 + MO + MN03 (4

mo3 -9 MO 4- %oz NO (vii) 2NO + 2CO + Nz + 2COz (viii)

The nitrate phase in a NOx-trap is usually derived from an alkaline earth compound (such as barium or strontium). These are basic and are gradually converted to very stable sulfates during prolonged use via reaction with sulfur dioxide (SO,) in the exhaust gas. The SO, is derived from sulfur compounds originally present in the fuel, see Equations (ix) and (x). Oxidation of SO2 to so3,

Equation (ix), is catalysed by Pt.

soz %oz + so3

MO + so3 + MS04 (4

As a result the capacity of the NOx-trap decreases over time and so the fuel-rich regeneration pulses have to be made more frequently. This has a detr- mental impact on fuel economy. IdeaUy it is desirable not to have any sulfur compounds in the fuel as, even with low sulfur fuel, sulfate will accumulate in the NOx-trap. To overcome this problem it is necessary to periodically ‘desulfate’ the NOx-trap by exposure to very lugh-temperature reducing conditions to recover the o q p d NOx-trapping capacity.

One means of achieving high catalyst temperature to desulfate the NOx-trap has been described by Ford (2002-01-0733). They alternate the

&/fuel ratio so that the oxygen stored in the NOx-trap during lean periods is used to oxidise HC, CO and Hz in the high temperature, enriched periods. By continuing this process high desulfation temperatures can be achieved even during low-load operation. Ford have characterised the sulfate species present in the sulfated NOx-traps by infrared spectroscopy, and have showed that surface sulfate species decompose more easily than the bulk compounds.

Ford have also (2002-01-0731) described an online method for estimating the sulfate levels in NOx-traps. It involves closed loop NOx-trap purging, with one heated exhaust gas oxygen (HEGO) sensor positioned before the NOx-trap and one after, to provide information about the oxygen and NOx storage capacities. If the oxygen storage capacity (OSC) does not change significantly over time the NOx capacity changes can be seen in the observed values. This could form the basis of a control system for a desulfation strategy. It is possible that if the nitrate and OSC capacities have different chemical reactivities, they could be accessed under different temperature conditions.

Volkswagen and OMG (2002-01 -0346) have described in detail the engine management and exhaust gas emissions control systems on the lean- bum FSI (fuel stratified injection) engine in the VW Lupo. Here a NOx sensor (4), downstream of the NOx-trap, is used to monitor NOx-trap activity and to initiate regeneration. The system is adaptive and only undergoes rich regeneration when the trap is sufficiently full to warrant it. This feature further improves fuel economy. However, overall fuel economy depends on the sulfur content of the fuel since this determines how frequently the fuel-consuming desulfations take place. When the desulfation regeneration does take place there is a danger that hydrogen sulfide might form under the high-temperature rich conditions. This is eliminated by periodically switching between rich and lean conditions. As noted above this will also tend to maintain the lugh temperature because the oxygen stored in the NOx-trap and on other catalysts in the system burns the com- bustible components in the rich gas.

OMG (2002-01-0057) are also workmg on the

Phtinrcm Metah Rex, 2003, 41, (1) 17

dynamic chemical processes taking place in a NOx-trap during NOx absorption and subsequent regeneration. They have developed concepts relat- ing to NOx storage sites at the surface and in the bulk. The former are readily available, are initially occupied and readily regenerated, while access to the latter is diffusion limited and proceeds via a shrinking core-type process.

Toyota (2002-01-0732) have used potassium compounds to trap NOx at temperatures higher than are possible with alkaline earth species, and their formulations contain materials such as titania and zirconia to provide improved sulfur tolerance. However, the very low sulh-containing gasoline that is needed to obtain the required levels of fuel economy is not yet widely available.

Catalytic Particulate Control Systems for Diesel Engines

Soot or particulate matter (I'M) from a diesel engine may be considered as a hgh-surface carbon core onto which are adsorbed a variety of HC species and other partially oxidised organic compounds, together with water and sulfuric and nitric acids. Controlling diesel exhaust gas PM emissions attracts attention because of potential adverse health effects, particularly in the urban environment.

Several kinds of filter could be used to trap the PM, but the trapped PM must be removed to prevent the build-up of an excess pressure drop across the flter that would prevent the engine functioning properly. The trapped PM has to be removed by oxidation to harmless CO, and water. The temperature at which diesel PM bums in air (typically > 550°C) is significantly higher than the normal temperature of the exhaust gas from a diesel engine. Several approaches have been used to remove PM and prevent a filled filter from causing excessive backpressure. These include fuel additives to lower the combustion temperature of the PM, and various heater devices to increase the gas temperature. In the past, the use of electrical heaters or burners resulted in serious problems as the temperature rise caused by the exothermic combustion above the ignition temperature had sometimes melted the filter material!

The most successful approach, however, has been to combust trapped PM with nitrogen dioxide (NO,) rather than oxygen. The reaction with NO, takes place at relatively low temperatures, and can be achieved on a heavy-duty diesel engine, in a truck or bus, during most operating conditions. The required NOz is obtained by oxidation of the already present NO in the exhaust gas, by passing it over a Pt oxidation catalyst. Thus the device requires no attention during normal use. However, catalytic oxidation of NO is inhibited by the presence of SO, in the exhaust gas, so low sulfur fuel is necessary. This device, comprising an oxidation catalyst upstream of a particulate filter, can function continuously at appropriate temperatures. It is called a continuously regenerating trap (CRTm).

In California, a large collaborative programme involving ARCO, National Renewable Energy Laboratory, Johnson Matthey, West Virginia University, Engelhard, Battelle, and Abilene Christian University (2002-01 -0433) has monitored heavy-duty diesel test fleets operating on low sulfur diesel fuel and retrofitted with either a CRTTM system or a catalysed particulate filter. The vehicles included grocery trucks, tanker trucks, refuse haulers, school buses and transit buses. After operating for a year the emissions were evaluated. The huge volume of results confirmed that retrofitted systems performed well, with little or no significant loss in their ability to reduce PM emissions - to over 90% in some instances after h o s t 200,000 miles.

Another field trial, this time with New York City buses fitted with a CRTT'\', has been reported by Johnson Matthey, Environment Canada, New York City Transit, New York State Department of Environmental Conversation, Equilon Enterprises, Coming and Sprague Energy (2002- 01-0430). Again, after about a year in service, reductions in particulate emissions to more than 90% were found, demonstrating the long term durability of the CRT"" in appropriate applications.

Johnson Matthey have also developed (2002- 01-0428) a modified CRTW system which has a Pt-catalysed PM filter as well as an upstream oxidation catalyst - the CCRT"'. Its performance has

Phfimm Metals Rev., 2003,47, (1) 18

been compared with that of a conventional CRTm and with a catalysed filter. The regeneration efficiency of the CCRTm has been found to be better than that of a CRTm, which in turn is significantly better than only a catalysed filter. The CCRTTM operates well in problematic situations where CRT'm performance is marginal. This is probably due to reoxidation in the filter of NO, formed from PM oxidation by NOZ. Each NO molecule is therefore used several times in PM oxidation via reaction with NOZ.

In situations where the exhaust gas temperature is too low for the PM/N02 reaction to be effective or where the N0x:PM ratio is too low for the reaction involving NO2 to remove all of the PM, it is necessary to provide a means of increasing the temperature to 55OOC or higher to initiate PM combustion with oxygen. In cars with diesel engines this can be achieved by having a flexible fuelling system that enables injection of fuel into the cylinders during exhaust strokes, or perhaps injection of fuel directly into the exhaust gas. Partially burnt fuel in the exhaust gas is then oxidised over the Pt catalyst in front of the filter, and the exotherm produced is sufficient to raise the gas temperature to the point where the PM bums. However, it is important to control the rate of PM combustion in the filter to limit excessively hlgh exotherm temperatures, particularly when the exhaust gas flow rate is low. Ford (2002-01-0427) have highlighted the practical control parameters that could be used to do this and have concluded that it is best to resmct the amount of oxygen present during combustion. This can be done reliably by combining exhaust gas recirculation (EGR) with an inlet air throttle. A strategy that can deal with transient response needs combined with as much forward control as possible is deemed to be necessary for overall successful operation.

Conclusions Significant advances in the exhaust emissions

control areas are taking place, and in many of them pp-based catalysts play vital roles. Over recent months in Europe the demand for diesel powered cars has been growing. Some of the emissions control systems described here will help towards

achieving future emissions legislation require- ments. If these systems show good in-field durability this could fuaher increase the interest in diesel light-duty applications in North America, which would result in lower COz emissions and significant fuel savings. At the next Detroit SAE World Congress we can confidently expect that further innovative and exciting developments in emissions control technology will be reported.

References 1 For a review of previous years' emissions control

papers at the Detroit SAE see: Platnnm Metuh Rey., 2001,45, (2), 71; &id, 2000,44, (2), 67 Cob0 Center, Detroit, Michigan, 4-7th March, 2002 Most papers are available in electronic format on two CD-ROMs (mew Emission Technology from the SAE 2002 World Congress', SP-l703CD, and 'Direct Injection SI Engine Technology 2002', SP- 1693CD). Copies of these md individual papers are available from: SAE, 400 Commonwealth Drive, Warrendale, PA 15096, U.S.A. See also www.sae.org For a description of the piinuples of operation of the NOx sensor see: M. V. Twigg, P/atnnm Met& Rm., 1996, 40, (3), 111

2 3

4

The Author Martyn Twigg is the European Technical Director of Johnson Matthey Catalytic Systems Division. His main interests are in applying advanced chemical concepts to highly efficient emissions control systems. He is the author of numerous research papers in this area and is the editor of the book series "Fundamental and Applied Catalysis".

Ruthenium Light-Switch Effects Scientists at the University of North Carolina at

Chapel W, U.S.A., report temperature-dependent excited-state lifetime measurements, in protic and aprotic solvents (MeOH, BuCN, MeCN) on [R~@py)~dppz]*' which suggest that the light-switch effect is competitive (M. K. Brennaman, J. H. Alstrum-Acevedo, C. N. Fleming, P. Jang, T. J. Meyer and J. M. Papanikolas, J. Am. Cbem. SOC., 2002,124, (50), 15094-15098).

The dppz ligand has bpy-like and phz-like states. The bpy is associated with the bright state and phz with the dark state. The bpy-like state is similar in size to the corresponding orbital in the 'MLCT state in [R~(bpy),]~+; it is entropically favoured and populated at high temperatures. The dark state is lowest in energy and is populated at low temperatures. The switch effect results from competition between the energetic and entropic factors, not h m state reversal.

Pkahnum Metah h., 2003,47, (1) 19

Palladium(+ 1) Carbonyl Clusters in the Catalyac Oxidation of Unsaturated Compounds By Tatiana A. Strornnova N. S. Kurnakov Institute of General and Inorganic Chemistry, Russian Academy of Sciences, 119991 Moscow, Russia

Some palladium carbonyl clusters containing the palladium in the (+I) formal oxidation state, and derivatives of these clusters investigated in our laboratories are examined in this paper: In purticular. tetranuclear carbonyl carboxylates ofpalladium are discussed as a model of key intermediates in various processes, including CO oxidation by a palladium catalyst with molecular oxygen as the oxidant. A second type o f cluster, a palladium cationic cluster which contains N-donor ligands acts as a catalyst precursor for processes such as exhaust gas purification. A third type of palladium carbonyl cluster includes bimetallic comp0und.v which catalyse new reactions, such as the low-temperature dehydration of alcohols.

Palladium (Pd) carbonyl complexes are of interest to chemists due to their role in industrially important catalytic processes involving carbon monoxide (CO), such as CO oxidation, the synthesis of dialkyl oxalates and dialkyl carbonates, and the carbonylation of alcohols, unsaturated hydrocarbons, nitroaromatic compounds and others (1-5). To understand the mechanism of Pd-catalysed reactions involving CO, the properties and reactivities of palladium carbonyl complexes have to be studied.

Complexes that contain palladium atoms in the (+1) formal oxidation state are not as abundant and stable as complexes contamng palladium in the (+2) and (0) oxidation states (6). Pd(+l) complexes have often been postulated as the active intermediates in reactions, such as the Pd-catalysed isomerisation and carbonylation of alkenes. The first evidence of the participation of Pd(+l) complexes in the catalytic and stoichiomettic transformation of unsaturated compounds came from the work of Moiseev and coworkers as early as the 1960s (7).

A group at the N. S. Kurnakov Institute in Moscow has developed methods for the synthesis of Pd(+l) complexes containing carbonyl ligands. These methods (8-11) have allowed a range of new complexes in platinum metals chemistry to be produced, such as, principally:

tetranuclear neutral carbonyl carboxylate clusters

of composition P&(CO).+(OCOR)4 0, with a closed structure, where R = Me, Et, Ph, CMe3, CF3, CCb, CHKl(8,9). Figure 1 shows the planar metal framework (rectangle, square or rhombus), with bridging carbonyl and carboxylate ligands coordinated on the sides of the metal hame.

cationic carbonyl clusters of composition p&(CO)&]& @I), where L = 1,lO-phenanthroline @hen) or 2,2'-bipyridyl (bipy); X = OAc, C104, CF3COO. F w e 2 shows the tetrahedral metal core (L = phen; X = OAc). In these clusters the tetrahedral metal core forms from two mutua- ly perpendicular fragments, Pd(p-CO)Pd, which lie in parallel planes and are bonded by a direct metal- metal bond (lo), and

0

I 0 I

I

, CH3

Fig. I Structure vf cluster Pd4(,-CO)r(p-OCOR)r (I), where R = CH3. Carbonyl and carbvxylate ligands bridge the four metal atoms

PLdinnm Metah Rey., 2003,47, (l), 2 6 2 7 20

F i g 2 Structure of the palladium cation cluster [ P ~ ~ ( ~ - C O ) ~ p h e n 4 ] ( O C O C H ~ ) ~ (11). The four metul utoms fbrm a tetrahedral core

a heterobimetallic palladim-molybdenum (Pd- Mo) duster of composition Na2{P&[CpMo(C0)3].+} (III) where the Pd atoms are in the +0.5 formal oxidation state and the four Pd atoms in the duster anion form a square with a bridging CpMo(CO)3 group on each side, see Figure 3. The main features of the cluster anion are that all the eight metal atoms (of Pd and Mo) lie in the same plane (1 l), and that the duster does not show any tendency to exist in the doso-polyhedral form characteristic of the majority of clusters. Using CpMo allowed the creation of bimetallic compounds with metal-metal bonds.

Fig. 3 Structure ?fun anion of'thc Pd-Mo cluster (Ill) Nuz{Pd4[CpMo(CO) r J d . The sodirrm utorns are not pre.Yent

The main question to be considered here is: what role do these clusters, with the palladium atoms in unstable formal oxidation states, play in catalytic reactions (mainly oxidations)? There are three interesting aspects to this question:

The first is what is the behaviour of the synthesised palladium clusters under conditions that are very similar to those of the key stage of the catalytic process?

The second aspect is the search for new reactions where the synthesised clusters could be a catalyst (or a precursor that would easily transform into a catalyst).

The third is what are the clusters that are involved in the catalysis of known reactions, including industrially important processes?

Palladium Carbonyl Clusters as a Model of Key Intermediates

The oxidation of the coordinated CO group is a key stage in redox reactions involving CO that are catalysed by palladium and its compounds. The most likely intermediates of such processes are complexes containing palladium in the intermediate oxidation state between (+2) and (0). It is known that the thermolysis of transition metal car- bony1 complexes includes CO elimination or the disproportionation of two molecules of CO to give COZ and carbide cluster formation (12). However, in contrast to this, the thermolysis of palladim carbonyl carboxylate clusters (I) proceeds principally by a different route (13). We have found that the thermolysis of clusters (I) at 11Ck120"C in an inert atmosphere (argon) gives Pd metal, COz and a diacyl RC(=O)C(=O)R, according to the equation:

P&(C0)4(OCOR)4 + 4Pd + 4co2 + 2RC(=O)C(=O)R

The process includes oxygen atom transfer from the carboxylate group to the carbonyl, that is, the carboxylate group is an oxygen atom donor. It is proposed that this oxidation process includes the rupture of the Pd-0 bond accompanied by an increase of negative charge on the oxygen atom and an attack by the oxygen atom on the carbon atom of the CO group (it can be formally shown

Phtinum Metalr Rev., 2003, 47, (1) 21

as CO insertion into the Pd-0 bond) to give the unstable intermediate:

R R

Decomposition of this unstable intermediate leads to CO, elimination and the formation of the coordinated acyl group:

R \

0 /" - o\cpo I I

WPd-Pd'A - Pdz[RCO] + COz

Recombination of two of the coordinated acyl radicals gives diacyk

2Pd@CO] + 4Pd(0) + RC(=O)C(=O)R

The thermolysis of cluster (I) where (OCOR) = (OAc), in benzene or toluene in an argon atmosphere produces COz of less than 15 per cent from the stoichiometric amount. The main product of oxidation of the coordinated CO groups was found to be the correspondmg aryl carboxylic acid (benzoic acid in the case of benzene and tolylic acid in the case of toluene):

P&(CO)~(OAC)~ + AI-H -+ Pd + COz + AKOOH + ......

- 15% 7&75%

where OAc = acetate; Ac = CH,CO; Ar = aryl. It is proposed that the formation of the unsta-

ble intermediate in this reaction is accompanied by the oxidative addition of an arene molecule proceeding via Ar-H bond rupture. The next transformation includes COz insertion into the Ar- Pd bond and the reductive elimination of a molecule of aryl carboxylic acid.

The above mentioned process of inner-sphere oxidation of coordinated CO groups proceeds at 11CL12O"C. However, the oxidation of coordinated carbonyl groups by nucleophiles containing an oxygen atom, such as water molecules, acetate ions, aliphatic alcohols or phenol, proceeds at

room temperature (14, 15). The carbonyl carboxy-

late complexes (I) are decomposed in the presence of OAc anions as:

NaOAc P&(CO)~(OAC)~ + 4Pd(0) + 2CO + 2COz + 2AczO

A kinetic investigation of this reaction showed that the mechanism includes a repeated step of CO insertion into the Pd-0 bond, COZ elimination, followed by AczO formation. It can be assumed that the interaction of clusters (I) with water proceeds via the same mechanism, according to:

P&(CO)~(OAC)~ + 2H20 + 4Pd(0) + 2CO + 2COz + 4AcOH

The reaction of clusters of (I) with CI-C2 aliphatic alcohols (methanol, ethanol, i-propanol) proceeds by a few paths (16) to give the products of oxidation of coordinated CO (COZ and diakylcarbonates) and organic products of alcohol oxidation (acetaldehyde in the case of ethanol). Thus, the reaction of clusters (I) with nucleophiles containing an oxygen atom (both inner-sphere cai- boxylate groups coordinated in clusters and outer-sphere molecules of alcohols, water and acetate ions) results in the reduction of Pd(+l) to Pd(0) metal and the oxidation of the coordinated CO to COZ or diakylcarbonates. This reaction may be considered as a model of the key stage of processes that proceed with the participation of CO and the substrates mentioned above.

The interaction of carbonyl carboxylate clusters 0 with a nitrosoarene molecule, ArNO, which contains two potential donor atoms, oxygen and nitrogen, proceeds via an oxygen atom transfer from the nitrosoarene molecule to the coordinated CO to give COZ and the very active arylnitrene species [ArN:]. Transformation of [ArN:] then depends on the composition of the aryl radical.

In the case of nitrosobenzene ArNO (Ar = Ph), a number of products containing the PhN:] group were obtained. These were: azoxybenzene PhN=N(O)Ph, azobenzene PhN=NPh, aniline PhNHz and a palladium complex of composition Pd~(p-OCOR)z(PhNC6H~O)z (IV) which contains an amide ligand, see Figure 4 (17). The formation of the amide ligand proceeded with the participation of the phenylnitrene species and may

Phdmm Metalr Rev., 2003, 47, (1) 22

Fig. 4 Structure of the palladium comp[ex Pd2(p-OCOCH3)2(q '-PhNC6H4NO)2, which contains aniide 1igand.s

include the next steps: [PhN:] species formation and simultaneous

COz elimination, followed by coordination of a second molecule of nitroso-

benzene, and PhN:] species insertion into the C-H bond of

the phenyl ring according to the Scheme I below. Deprotonation of the resulting amine leads to the formation of an amide ligand.

In the case of the reaction of carbonyl carboxylate clusters (I) with c-nitrosotoluene ArNO (Ar = MeC6H4) (18), the formation of tolylnitrene species is also confirmed by COZ elimination and the formation of organic products such as:

6fNC0 and tolylisocyanate

Thus it can be concluded that the interaction of carbonyl carboxylate clusters (I) with nitrosobenzene and o-nitrosotoluene includes the oxidation of the coordinated CO goup to give COZ and q l - nitrene species. The arylnitrene species can be easily transformed into arylisocyanate (at Pco = 1 atm, 20°C). The oxidation of CO by the oxygen of nitrosoarenes and coordination to the nitroso-

arenes is one of the main stages of the reductive carbonylation of nitroarenes:

ArNOz + 3CO + ArNCO + 2C02 (9 This process allows a wide range of nitrogen-containing organic products to be obtained by an environmentally friendly route. It is postulated in the literature that the mechanism of Reaction (i) includes consecutive reductions of the nitro-

benzene molecule:

+co +co +co

- coz - coz ArNOz + A N 0 + ArN: + ArNCO

(4 @) (4

Palladium and its compounds are the most active catalysts for Reaction (i). The generation of the arylnitrene species [AN] (Stage (b)) and the formation of arylisocyanates ArNCO (Stage (c)) is one possible mechanism.

Reactions Catalysed by Palladium Carbonyl Clusters and Derivatives

Some of the compounds synthesised were found to be efficient catalysts for redox reactions. Among these compounds, the Pd-Mo cluster Naz {P&[CpMo(CO)3]4} QIl) is particularly interesting, for the formation of bimetallic clusters in which the Pd and Mo are connected by metal- metal bonds. This cluster is highly soluble and

Scheme I Ph'

Pkatinum Metub h., 2003,47, (1) 23

azoxytoluene

ditolylamine

azotoluene

CP -

CP I

\ 0

0 Mo I CP

RCHz-OH 7 R Y

- CP

‘o-H R H

HXC‘ I

.-- .-Mow]

The mechuiiisiii of dehydrutioii of olcohols cota!wed b). the Pd-Mo cluster ([If)

fairly stable in non-aqueous solvents. It does not contain readily oxidisable phosphine or related ligands. As two metals of different natures, palladium and molybdenum, constitute the cluster metal core, the cluster may simultaneously activate two different molecules or two different functional groups of the same substrate molecule.

Low-Temperature Dehydration of Alcohols via the Carbene Mechanism

It was shown (19) that the Pd-Mo cluster 011) is a good catalyst for the selective oxidation of alcohols in air. In the presence of cluster (III), methanol (MeOH) is transformed into methylfor- mate MeOCHO, ethanol (EtOH) into acetal MeCH(OEt)Z, and benzyl alcohol (PhCHzOH) is changed into benzaldehyde PhCHO.

In the absence of oxygen, the Pd-Mo cluster (III) catalyses an unusual reaction, the low-temperature (60-SOT) dehydration of alcohols (20). In the presence of the Pd-Mo cluster both aliphatic (MeOH, EtOH, i-PrOH, MesCCH20H) and

aromatic (PhCHzOH, PhZCHOH) alcohols are dehydrated. Dehydration of benzyl alcohol yields trans-stilbene and water. In the case of aliphatic alcohols the dehydration results in the formation of water and a wax-like hydrocarbon. The rate of dehydration of aromatic alcohols exceeds by twice that for aliphatic ones.

What IJ the Reaction? Acid catalysts are known to promote an inter-

molecular dehydration of alcohols to form ethers, according to:

R-OH + HO-R + H20 + R-0-R

Ethers were not found in the reaction mixture. A more unusual peculiarity of the reaction is the dehydration of alcohols that do not have a hydrogen atom in the p position to the hydroxyl group in their molecules, for example, methanol, neopentyl alcohol, benzyl alcohol and diphenyl- carbinol. In these cases the P-elimination of water:

P a R-CR‘-CH* + H20 + R-CR’=CHz

I 1

H O H

is not possible. Hence this reaction might be assumed to proceed by the mechanism of a-elimination, which includes the removal of the hydroxyl group and a hydrogen atom from the same carbon atom. This assumption was confirmed by the fact that tert-butanol did not undergo dehydration because of the absence of a-hydrogen atoms in its molecule. AU these facts suggest that the reaction proceeds via a carbene mechanism according to Scheme 11.

Probably, according to Scheme 11, the first stage of the reaction includes the oxidative addition of an alcohol molecule to the Pd-Mo bond. This is followed by the formation of an intermediate which contains the alkyl group o-bound to a Pd atom. Proton transfer from the akyl group to the hydroxyl produces a complex containing the coordinated water and carbene ligands. Further transformation of the complex includes the elimination of a water molecule and the dimerisation or oligomerisation of the carbene species into the hydrocarbon reaction products.

Phtinmn Met& Rev., 2003,47, (1) 24

Scheme II

Additional evidence for the formation of coordinated carbene species was obtained from the experiment on reaction mixture hydrogenolysis. A solution of the Pd-Mo cluster was exposed to an alcohol for some days at 60”C, and after cooling to

20°C was reacted with molecular hydrogen (1 am, 20°C). After the last procedure, the corresponding hydrocarbons (methane from methanol and toluene from benzyl alcohol) were found in the reaction mixture.

Palladium Catbonyl Clusters as Catalysts (or Precursors) of Known Processes Homogeneous Catahsts

Palladium carbonyl carboxylate complexes have no potential as homogeneous catalysts in oxidation reactions because they have low solubility in organic solvents and poor stability in the presence of oxygen-con- nucleophiles. However, Pd carbonyl complexes with additional hgands that have N- and P-donor atoms have more prospects in homogeneous catalysis.

(a) Oxihtiw Acekmykation and O&ve AiXoykation OfAlkenes

The oxidation of alkenes proceeds via two different routes in the presence of Pd and its compounds: (a) A ‘classical’ mechanism consisting of altemat- ing stages in which P d O is reduced to Pd(0) by oletin, and the Pd(0) is reoxidised by an oxidant. (b) A ‘cluster’ mechanism in which olefin oxidation occurs at the surface of a low-valence palladium cluster, such as a giant Pd561 cluster. This mechanism does not suggest alternating oxidation and reduction stages but the nanocluster acts as an ‘electron mediator’, transferring electrons from the olefin molecule to the oxidant molecule.

Moisew and coworkers have studied Pd-catalysed olefin oxidation proceeding via the ‘classical’ and ‘cluster’ mechanisms (21-25). It was shown that the complexes obtained by interaction of (I) with L (L = phen or bipy, Pd/L > 1) are suitable catalysts for olefin oxidation. In alcohol solutions of these complexes, selective oxidative alkoxyla- tion of alkenes proceeds under mild conditions (60-10O0C, Pdkene = 0.7 atm, Po2 = 0.3 atm) at a

good rate. Under these conditions propylene is transformed into ally1 methyl ether:

CH2=CHCH3 + MeOH + 1/202 + CHZ=CHCHzOMe + HzO

and dbutylene is transformed to the ether:

CHz=C(CH3)2 + MeOH + 1/202 + CHz=C(CH3)CHzOMe + HzO

Thus, in both cases only alkyl groups of substrates will be oxidising which is evidence of catalysis by Pd in a lower oxidation state.

In the oxidation of propylene in alcohol media, the presence of PdPI) could be considered to lead to a mixture of ethers including the products of the different groups of the alkene molecule. The nature of the products formed in these reactions is evidence for the ‘cluster’ mechanism of catalysis that includes the molecular clusters or nanoparticles formed in these systems (26).

Acetic acid solutions of the above complexes are good catalysts for CO oxidation by air:

co + 1/202 + co2

The velocity of CO oxidation is lo-’ mol COJl of complex/hour at 20T, where pdl = 0.2 g-atom/l.

This is evidence that the Pd complexes obtained by the interaction of carbonyl carboxylate complexes with N- and P-containing arenes can be used as precursors for the catalyst used for the oxidation of alkenes and CO under mild conditions. In the case of alkenes only their alkyl groups will be oxidising. These complexes can serve as models of the active centres of the catalysts for the oxidation of unsaturated compounds and help search for potential catalpc systems.

Hetemgeneous Catabsts

All the catalytic reactions mentioned above occur in the presence of homogeneous catalysts. We would now also like to obtain data about the activity of the synthesised clusters in conditions that are very close to industrial ones. The complexes obtained from cluster (I) and phen or bipy were immobilised on oxide supports. We used commercial supports: silica and a-Alz03 and also Ti02, where the Ti surface was oxidised electro-

Pkztinum Met& Rm, 2003,47, (1) 25

I I

400 500 TEMPERATURE: * C

6oo I chemically. The catalysts all contained 0.1 wt.% Pd. The PdO clusters immobilised on the oxide carriers were found to have high activity in the following reactions: CO oxidation in air:

co + 1/202 -+ co* CO oxidation by nitrogen oxides:

CO + NOX -+ COz + Nz Alkane oxidation in air.

ZC&IIO + 1 3 0 2 + 8COz + 10Hz0

(4

(b)

(4 Reduction of nitrogen oxides by methane:

2NOx + CH, -+ Nz + COz + 2Hz0 (d) Using the data obtained, effective heteroge-

neous catalysts for exhaust gas purification were developed (27, 28). These catalysts can also be used to help with ecological problems, such as the purification of off-gas from metallurgical or petrochemical plants, or engines.

Catalysts prepared from palladium clusters immobilised on a support of spherical silica part- cles showed high activity for the oxidation of CO and hydrocarbons. Complete conversion of CO was achieved at 16CL18O"C (V,,,, = 4000 h-', 4% CO). This temperature is the lowest for known catalysts. The high activity of the prepared catalyst was achieved by using palladium cluster compounds; the catalyst contains 5 to 10 times less noble metal than the usual industrial catalysts (for example, the Russian catalyst APK-2 contains 2 wt.% Pd).

The catalytic activity of the cluster catalyst immobilised on a Sibunit carrier (activated carbon sorbent prepared from coal) was investigated at the Institute of Catalysis, Siberian Branch of the RAS

Fig. 5 Coiiiparison ofcatullfic activity iri N0.x conversion o f three cata!vsts iiiipregrioted onto ciii intiirstriul a-AlrOj support: Vp = velocitv ofgas mixture I catalvst prepured.fi.oiii clirster ( I ) trntl pheri. contoiiiirig 0. I % Pd 2 HzPtCI6. coritai:iing 0. I !% Pt 3 indirstriol catcrlj.st APK-2; this

cutui\.st hosed on the plotiiiuni salt.

catnlvst corltc1i:ls 2% Pd

(Novosibirsk). The complete oxidation of methane and butane in air occurred at a significantly lower temperature than on regular catalysts. For instance, C4 hydrocarbons were oxidised at - 300"C, instead of at > 350°C for the usual industrial catalysts.

Purifying the polluting gases from nitric acid plants may be achieved by using a palladium cluster catalyst. Similarly, decomposition of NOx is an environmental challenge for the chemical industry. Around 50 UKI-7 plants of capacity 120,000 tonnes per year of nitric acid are now operating in the former Soviet Union and each of them emits 60,000 m3 h ' of NOx-containing pollution. All these plants presently use deNOx catalytic systems to reduce the concentration of NOx from - 0.1% to less then 50 ppm. The reduction process is based on the reaction of NOx with natural gas, see Reaction (d).

Therefore, to compare the activity of the palladium cluster catalyst proposed by us with the industrial APK-2 catalyst, both were impregnated on the same industrial a-Alz03 carrier. The catalp- ic activities were comparable although the amount of palladium in our cluster catalyst (0.1Yo) was twenty times less then that in APK-2 (20/0), see Figure 5. Moreover, the working temperatures of the cluster catalyst appeared to be at least 100°C lower then that of the industrial one.

Conclusions The current work shows that Pd(+l) complex-

es can play a role in catalysis. While our work in this area has ended, we believe further work could produce further interesting results.

Platnwi Meiuh Rev., 2003,47, (1) 26

Acknowledgement This work was supported by the Russian Foundation for

Basic Research, project No. 99-03-32519.

1

2

3

4

5

6

7

8

9

10

11

12 13

14

15

16

17

18

19

20

21

References T. A. Stromnova and I. I. Moiseev, Rms. Cbem. Rev., 1998, 67, (6), 485 0. N. Temkin, G. M. Shulakovsky and L. G. Bruk, Kbim. Pmm. (R...), 1983,5, 22 L. G. Bruk, I. V. Oshanina, 0. N. Temkin ef d., J, Mot. Cbem., 1995, 104, 9 L. G. Bruk, I. V. Oshanina, 0. N. Ternkin ef d., I?. Akad Nauk @us. Cbem. Bull), 1998,6,1104 L. G. Bruk, 0. V. Marshaha, I. V. Oshanina, 0. N. Ternkin et al, I?. &. Nauk, 1999,10,1899 0. N. Temkin and L. G. Bruk, Rnss. Cbem. Rev., 1983,52,207 I. I. Moiseev and S. V. Pestrikov, DoM Akad. Nank SSSR, 1966,171,151 I. I. Moiseev, T. A. Stromnova, M. N. Vargaftik, G. Ja. Mazo, L. G. Kuzmina and Yu. T. Smchkov, J. Cbem. Soc., Cbem. Commun., 1978,28 T. A. Stromnova, N. Yu. Tihonova, D. I. Kochubey and I. I. Moiseev, DOH. Akd Nauk, 1994,335, (5), 602 T. A. Stromnova, M. N. Vargaftik, T. S. Khodashova, M. A. Porai-Koshits and I. I. Moiseev, fiodnats. Kbim. @us.), 1982, (6), 1254 T. A. Stromnova, I. N. Busygina, S. B. Katser, A. S. Antsyshkina, M. A. Porai-Koshits and I. I. Moiseev, J. Cbem. SOC., Cbem. Commun., 1988, 114 J. S. Bradley, Ah. Organomet. Cbem., 1983,22, 1 (a) T. A. Stromnova, I. N. Busygina, N. Yu. Tihonova and I. I. Moiseev, Menakhm Commun., 1991, 1, (2), 58; @) T. A. Stromnova, N. Yu. Tihonova, L. K Shubochkin and I. I. Moiseev, Koodnats. Kbim. (Russ.), 1993,19, (6), 460 T. A. Stromnova, M. N. Vargaftik and I. I. Moiseev, ]. Organornet. Cbem., 1983,252,113 I. I. Moiseev, T. A. Stromnova and M. N. Vargaftik, J. Mol CafaL, 1994, 86, 71 T. V. Chernyshova, T. A. Stromnova, M. N. Vargaftik and I. I. Moiseev, DOH. Akad Nauk (Rms.), 1996, 348, 780 T. A. Stromnova, S. T. Orlova, I. P. Stolyarov, S. B. Katser and I. I. Moiseev, Do& &. Nank (Russ.), 1997,352, (l), 68 T. A. Stromnova, S. T. Orlova, D. N. Kazjulkin, I. P. Stolyarov and I. L. Eremenko, I?. &. N a k , Ser. Khim., 2000, 147 T. A. Stromnova, 1. N. Busygina, S. B. Katser, A. S. Antsyshkina, M. A. Porai-Koshits and I. I. Moiseev, All-Union Conf. Cad . Reaction in Liquid Phase, Alma-Am, 1988, Abstracts, Vol. 3, p. 42, in Russian T. A. Stromnova, I. N. Busygina, S. B. Katser, A. S. Antsyshkina, M. A. Porai-Koshits and I. I. Moiseev, I?. Akad Nank SSSR, Ser. Dim.., 1987,1435 I. I. Moiseev, 0. G. Levanda and M. N. Vargafak, J. Am. Cbem. SOC., 1966,88,3491

22

23

24

25

26

27

28

I. I. Moiseev, ‘%Complexes in Liquid-Phase Olefin Oxidation”, Nauka, Moscow, 1970, in Russian I. P. Stolarov, M. N. Vargaftik and I. I. Moiseev, Kinet. W. (R...), 1987,28, 1359 M. N. Vargaftik, V. P. Zagorodnikov, I. P. Stolarov, I. I. Moiseev, D. I. Kochubey, V. A. Likholobov, A. L. Chuvilin and K. I. Zamaraev,]. Mol Cafal., 1989, 53,315 I. I. Moiseev and M. N. Vargafak, ‘Catalysis with palladium clusters’, in “Catalysis by Di- and Polynuclear Metal Complexes”, eds. R D. Adams and F. A. Cotton, Wiley-VCH, New York, 1998,

T. A. Stromnova and M. N. Vargaftik, I?. Akad Nauk SSSR, Ser. k%im,, 1980,236 V. I. Toshynsky, V. A. Borovaja, T. A. Stromnova, V. I. Atroschenko, I. I. Moiseev and M. N. Vargaftik, Avtorskoe Svid. SSSR N 1363588 ot 1.09.1 987 V. I. Toshynsky, V. A. Borovaja, V. I. Atroschenko, I. I. Moiseev, T. A. Stromnova and M. N. Vargafak, Avtorskoe Svid. SSSR N 12693222 ot 8.07.1986

p. 395

The Author Tatiana A. Stromnova is a Senior Scientist at the N. S. Kurnakov Institute of General and Inorganic Chemistry, Russian Academy of Sciences, in Moscow. Her interests are platinum group metal chemistry and the application of platinum group metals and their compounds in catalysis.

Alcohol Oxidation by Ruthenium Catalyst Traditional oxidations of alcohols use heavy

metal oxidants, resulting in noxious waste and unwanted coproducts (P&finm Metah Rev., 2002, 46, (l), 26). Alternative homogeneous catalytic systems have been developed, but are limited mostly to oxidation of activated benzylic and dyl ic alcohols, or large amounts of additives (NaOAc, N a O H and K2C03) are required.

Now, researchers at the University of Tokyo have developed an efficient heterogeneous Ru catalyst that can oxidise alcohols with 1 atm of O2 or air, without additives 6 Yamaguchi and N. Mizuno, hp. Cbem. Ink Ed, 2002, 41, (23), 45384542). The 1.4 WL% RU/&o3 catalyst was prepared by stirring y- ALO, with an aqueous solution of RuCb at room temperature, and then treated with aqueous NaOH.

The RU/Alzo3 system had high catalytic activity for oxidising activated and non-activated alcohols with 1 atm of 0 2 ; reaction selectivity was over 97%. Even less reactive primary alcohols (1-octanol and 1-decanol) could be oxidised. The Ru/AlZO3 system is reusable, recoverable and no Ru leaching occurs, which make it a possible system for indusmal use.

P/atntm Metah Rev., 2003,47, (1) 27

Fuel Cells - Science and Technology 2002 SCIENTIFIC ADVANCES IN FUEL CELL SYSTEMS REPORTED IN AMSTERDAM

By Donald S. Cameron The interact Consultancy, Reading, U.K

The first of a new series of biannual European fuel cell conferences was held on the 25th and 26th September, 2002, in Amsterdam. These conferences are to be alternated with the Grove Symposium, giving a balance between the scientific and commercial aspects of the technology. The conference theme: ‘Scientific Advances in Fuel Cell Systems’, was chosen to focus on the scientific challenges posed by the introduction of these novel power generators. Authors from around the world were invited to submit oral papers and posters d e d m g their progress in the field.

The meeting was fully subscribed with 247 participants from 33 countiies, representing un- versities, research organisations, and fuel cell and system manufacturers worldwide. Forty oral papers, chosen from the original submitted abstracts, and over a hundred high quality poster presentations were made. Most of the oral papers and many of the posters are to be published in a special edition of the JoumaL ofpower Soumes.

The meeting was organised as a series of six sessions entitled: Materials’ (2 sessions), ‘Stack and cell engineering’, ‘Balance of plant and advanced fuel cell systems’, ‘Electrocatalysis and fuel processing’ and Modelhg and engineering’. The importance of materials to the advancement of the technology was emphasised by their featuring in no less than three of the sessions. The first of the two Materials sessions concentrated on developments in hydrogen storage, membrane technology and membrane electrode assemblies. The second Materials session was mainly concerned with corrosion-resistant materials for high and low temperature fuel cells. From such a wide range of topics, this brief review focuses on a selection of presentations that make use of the platinum group metals (pgms).

In his keynote talk, Professor Robert Selman (Illinois Institute of Technology, U.S.A.) empha-

sised the role of science in fuel cells, and enumer- ated some of the most important developments since the principle of the fuel cell was discovered in 1839 (1). A wide range of approaches are being used to reduce the cost of fuel cells and to make them commercially attractive, including performance improvements in terms of power density, durability and component costs. Improved techniques to exploit the pgms are playing an important part in enabling more compact and less expensive fuel cells and hydrogen generator systems to be constructed. They are also making direct oxidation of organic fuels, such as methanol, feasible.

Hydrogen Generation An efficient and compact source of hydrogen

or a suitable means of storing the gas are critical to

the success of most fuel cell systems. The development of fuel cells with a wide range of power outputs has led to a number of hydrogen generation routes being explored. It has become evident that for many applications, such as polymer electrolyte membrane (PEW fuel cells, high purity hydrogen is required, since impurities, such as carbon monoxide (CO) can lead to rapid performance deterioration.

Reformers In the paper ‘High temperature hybrid steam-

reforming for hydrogen generation without catalyst’, Phillipe Marty (l‘Ecole des Mines d’Albi- Carmaux, France) proposed the use of hybrid steam reforming. This incorporates partial combustion of the fuel with injected air to generate heat for the endothermic reforming reaction. Experiments are at a relatively early stage, mainly with fundamental reaction studies. Possibly - because of the high temperatures needed in the absence of catalysts - the typical reaction product is only 30% hydrogen, with the balance being

Phhn~m Metah h., 2003, 47, (l), 28-31 28

made up of nitrogen from the atmosphere, 15% CO and carbon dioxide (C02).

In his talk entitled ‘A new generation of water- gas shift catalysts for fuel cell applications’, Wolfgang Ruettinger (Engelhard Corporation, U.S.A.) described the limitations of the existing copper/zinc oxide-based reformer catalysts, and Engelhard’s attempts to develop replacement materials that are more thermally stable and less pyrophoric. In their current reformer system a desulfuiiser (to remove sulfur from natural gas) is followed by an autothermal reformer. Here, oxidation of a small proportion of the fuel produces heat to reform the remainder - to a gas containing hydrogen plus &lo% CO. This passes to a water gas shift reactor (which reduces the CO content to 1-5./,>. By then injecting 1-2% oxygen and passing the gas through a partial oxidation catalyst, the CO content can be reduced to less than 10 ppm. Engelhard‘s newly developed Selectram Shift catalyst has somewhat lower activity than the mixed oxide-based reformer catalysts but greatly reduced sensitivity to oxygen ingress. The water gas shift reaction demands a large reactor vessel, but to make the system more compact, it is likely to be replaced by a monolith-supported pgm catalyst; no details of this were given. Similarly, in order to reduce the CO content of the product, a two-stage partial oxidation system that contains pgm catalysts which incorporate inhibitors to prevent methane formation, is being introduced. This is likely to enable space velocities in excess of 30,000 volumes per hour in the future.

In her talk ‘Advances in fuel processor catalysts’, Jessica Re-h (Johnson Matthey Fuel Cells, U.S.A.) described the development of reformers mainly for PEM fuel cell systems. These are intended for mobile applications, with the capability of operating on a variety of fuels including natural gas, propane and hgher hydrocarbons, although for small portable systems, hydrogen stores or methanol reformers are likely to be used. However, for mobile systems, an autothermal-type reformer system has been adopted to minimise volume and weight. There is a fuel pretreatment reactor mainly to remove sulfur, and reforming followed by a water gas shift reactor. Finally, the

addition of 1-2% oxygen before a last reactor selectively oxidises CO to COZ.

For mobile systems, which are weight and volume critical, in order to produce sufficient hydrogen for a 75100 kW system from a reactor of volume less than 2 litres, space velocities of over 100,000 volumes per hour are necessary. A reformer system fuelled by gasoline has already been demonstrated, functioning at 140,000 per hour space velocity for over 500 hours; after that period it still retained 99.8% of its efficiency. In order to achieve this, it is necessary to use advanced monolith support materials with very high surface areas and to reduce sulfur impurities in the fuel feed to low levels. For the water gas shift reactor, operaang at a space velocity of over 50,000 per hour, Johnson Matthey have developed lugh performance pgm-based catalysts. These exhibit high activity at less than 300°C and avoid the pos- sibility of methane formation.

The Johnson Matthey FP05 prototype reformer, weighing 75 kg and of volume 470 litres, has operated on natural gas for over 2000 hours with over 100 start-up and shut-down cycles. Start up currently takes 20 minutes, although it should be possible to reduce this to around 5 minutes.

Hydrogen Storage Most of the fuel cell powered passenger vehi-

cles and buses currently under development rely on hydrogen stored under pressure for their operation. In his talk ‘FUCHSIA: Fuel cell and hydrogen store for integration into automobiles’, Professor Rex Harris (University of Birmingham, U.K.) described an international cooperative project going on in Germany, Switzerland and the U.K. He reviewed the various types of hydrogen store currently being developed under the EU Fifth Framework Programme. Hydrogen stores which reversibly retain over 7% of their weight of gas are now available. Liquid hydrogen has a very high energy density, but thermodynamics dictate that over 30% of the heating value is used to liq- uefy the gas. Compressed gases are currently being stored at extremely high pressures in advanced composite cylinders, in the solid state as hydrides, and adsorbed on carbon or glass microspheres.

Ph?nam Met& h., 2003,47, (1) 29

Work at the University of Birmingham, U.K., indicates that the kinetics of absorption and desorption of hydride systems, based on magnesium- nickel alloys, can be substantially improved using dispersions as low as 0.1% by weight of ruthenium (Ru) and palladium (Pd).

Electrocatalysis and Direct Fuel Cells In a talk entitled 'New fuel cell electrocatalysts

based on mesoporous precious metals', Anthony Kucernak (Imperial College, London, U.K.) described their techniques to produce pgm and pgm alloy catalysts in the aqueous phase in the form of small, nanopardculate spheres. Metal areas of 35 m2 g-' (Pt) and 40 m2 g-' (Pt-Ru) can be produced for unsupported catalysts, while metal areas of 5&90 m2 g-' can be produced on mesoporous carbon support materials. When used to oxidise carbonaceous fuels, the type of catalyst structure appears to affect CO absorption. The catalysts are extremely effective for the direct anodic oxidation of methanol, formic acid and formaldehyde in fuel cells. Catalysts of Pt-Ru alloy have been found to be more active for this purpose since they are less affected by strongly chemisorbed carbonaceous species.

Graham Hards (Johnson Matthey Fuel Cells, U.K.) in a paper entitled 'New catalyst and MEA developments for high performance PEM fuel cells', gave details of the latest Johnson Matthey oxygen reduction and hydrogen oxidation catalysts that have tolerance to CO and COZ impurities. In order to obtain high catalytic activity, the pgms are highly subdivided, yielding surface areas of up to

150 m2 g-', compared to the surface area of atomic dispersions of around 200 m2 g-'. However, the nature of the PEM fuel cell demands that the catalytic layer is kept as thin as possible to provide a narrow boundary between reactant gases and electrolyte. To obtain reasonable metal loadings in these thin layers means that the catalyst must be applied, either as the pure metal or as high concentrations, on conducting substrates such as high surface area carbon blacks. Typically, 4&50 wt.%

Pt by weight of catalyst is applied to a cathode (oxygen reduction) catalyst. With new deposition routes, even at 70 wt.% Pt on carbon, metal areas

of over 90 mz g-' can be obtained. The PEM fuel cell performance of MEAs employing these new catalysts is significantly improved over the current materials. Alloys of Pt and Ru, at an atomic ratio of 1:1, have been used for some time to provide resistance to CO poisoning as anode catalysts. These catalyst provide good tolerance to levels of CO of only 100 ppm and below. More recently a new anode combination of Pt-Ru and Pt-Mo has demonstrated extraordinary levels of fuel cell performance tolerance with reformate fuel containing up to 2000 ppm CO.

Further improvements in catalyst performance can be obtained by increasing the intrinsic activity of the catalyst metal particles. These can be achieved by alloying the pgm with elements such as chromium, iron and manganese, using thermal treatments to form true alloys. Even though the resulting alloys tend to have lower metal areas after heat treatment, higher intrinsic activities (that is, activity per unit of surface area) more than com- pensate for surface area losses. The use of alloy catalysts has been well established in phosphoric acid fuel cells over many years.

In his talk 'Improving the tolerance of PEM fuel cells for reformate gas: results and perspec- tives', Gaby Janssen (Energy Research Centre of the Netherlands (ECN)) enlarged on the variety of techniques available for reducing the effects of impurities in hydrogen on cell performance.

Reformate gas typically contains nitrogen, COz, CO and water, as well as hydrogen. One reaction that can occur at the anode is the reduction of COz by hydrogen to form more CO. By incorporating a small proportion of oxygen (1-29'0) in the reformate gas and passing it over a suitable catalyst, much of the CO in the fuel can be oxidised to C02, although the reverse reaction must be pre- vented in the cells. A second option is to periodically starve the fuel cell of fuel, so that the anode compartments reach oxidising potentials sufficient to oxidise any CO chemisorbed on the surface. However, this technique is less desirable since it entails temporarily closing down the fuel cell. ECN have examined a series of carbon- supported Pt-Pd alloy anode catalysts in PEM cells for oxidation of CO in contaminated hydrogen

PIotinum Metafr Rcv., 2003, 47, (1) 30

fuel at 80°C. Those with higher Pd concentrations have reduced coverage by CO. ECN have demonstrated 20 cell stacks with up to 500 hours of stable operation in reformate gas without an air bleed.

Considerable work has been carried out to develop direct oxidation fuel cells. In his talk Direct 2-propanol fuel cells', Zhgang Qi (H-Power Corporation, U.S.A.) described their work on fuel cells that operate on various fuels, including methanol, ethanol, 1 -propanol, 2-propanol and ethylene glycol. Fuel concentrations of 0.5-2.0 molar and temperatures of 7CLlOO"C are typically used. The use of 2-propanol in conjunction with a polymer electrolyte membrane reduces some of the problems due to organic crossover to the air cathode encountered with more volatile organic fuels. In small (25 cm') cell tests, current densities up to 200 mA cm-' have been obtained at 0.5 V, compared to the 0.3 V typical of direct methanol cells. Voltage decay, caused by impurities adsorbed on the catalysts during operation on the organic fuel, recovers on standing at open circuit voltage or on reversing the polarity of the cells.

A different approach that is being adopted at the University of Newcastle, U.K., was described by Professor Keith Scott in his talk 'Direct methanol fuel cells: solutions to the problem of crossover'. When using 2 mg un-' pgm (Pt-Ru) anode catalysts and 2 mg cm-' Pt cathode catalysts in conjunction with Nafion polymer membrane (6CL120 pm thick) at 100°C, substantial methanol oxidation occurs at the air cathode. Various cathode catalysts which exhibit selective oxidation properties towards methanol have been assessed. These include cathode catalysts of iron tetrameth- oxyphenylporphyrins, (Fe TMTP) and Ru-based compounds, such as Ru-Se. To reduce methanol crossover, techniques such as using low methanol concentrations can be used combined with lower- permeability radiation-grafted polymer materials.

Modelling and Engineering The importance of optimising fuel cell systems

in areas such as thermal balance was emphasised in the final session. Advanced studies are being carried out on hybrids of high temperature fuel cells (such as solid oxide and molten carbonate cells) in

combination with gas turbines to produce systems having almost 70% overall efficiency. This is achieved by using the fuel cell as the combustion stage of gas turbines, with the compressor supply- ing high pressure oxidant, and utilising the waste heat from the fuel cells to drive the compressor. The main applications for these are as stationary generators providing combined heat and power.

The conference was concluded by the award of four prizes. The best submitted oral presentation was won by Klaus-Dieter Kreuer (Max-Planck- Institut, Germany) and the best poster prize by Adi Aharon p e l Aviv University, Israel). The prize for the most original contribution was won by Eric Middleman (Nedstack, The Netherlands) for a talk on self-organising nanostructures in PEM fuel cell electrodes that incorporated strings of catalyst and wetproohg agent. Whitney Collella (University of Oxford, U.K.) won the best student contribution.

Conclusion The variety of presentations, diverse range of

science and hgh quality of work, demonstrate the wide support that is helping to bring fuel cell technology to the consumer.

Reference 1 W. R. Grove, Phil. Mag., 1839,14,127; ibid., 1842,21,

417

The Author Don Cameron is an independent consultant on the technology of fuel cells and electrolysers. As well as the scientific aspects, his interests include the standardisation and commercialisation of these systems.

Eighth Grove Fuel Cell Symposium The Eighth Grove Fuel Cell Symposium will be

held at ExCeL in London, from 24th to 26th September, 2003. The theme of the meeting will be 'Building Fuel Cell Industries' to reflect the rapidly growing infrastructure needed to support the technology as commercialisation proceeds. The venue was chosen so that displays of stationary generators, vehicles and the refuelling techniques being developed could be viewed in comfort.

Details of the symposium can be obtained from http://www.grovefuelcell.com or from Sarah Wilkinson: Tel: +44 (0)1865 843691.

Pkztimm Metals Rey., 2003,47, (1) 31

Uphill Effects on Hydrogen Diffusion Coefficients in Pd77Ag23 Alloy Membranes INFLUENCES DUE TO GORSKY EFFECT AND LATTICE STRAIN GRADIENT FACTORS

By X. Q. Tong, F. A. Lewis and S. E. J. Bell School of Chemistry, Queen's University, Belfast BT9 5AG, N. Ireland

and J. Cermak Institute of Physics, Academy of Sciences of the Czech Republic, Na Slovance 2, CZ-182 21 Praha 8, Czech Republic

The choice of palladium (Pd) and Pd alloys selected for use in hydrogen permeation membranes is determined by the values for hydrogen solubility and hydrogen diffusion coefficients, D H ,

with high values being preferred (1-5). Using pure Pd membranes at temperatures I 300°C can be complicated by the possible involvement of the a ++ p phase hydride transition regions, with the likely consequent formation of irreversible distor- tions (1-10). However, the problems of phase transition and related hysteresis effects can be much reduced by using membranes made of care- fully chosen Pd alloys.

For example, Pd-Ag (palladium-silver) alloys in the composition range Pd7,Ag2, to P d 7 h 5 have been successfully used as hydrogen purification membranes over a wide range of equilibration conditions with respect to hydrogen pressure, p, hydrogen content, n (n = H/M atomic ratio, where M is metal) and temperature, T, (from p-c(n)-T relations). Relationships between n and D H , have been derived by using both electrochemical and gas-phase equilibration techniques (610).

Representations of DH-n Relations Figure 1 compares various forms of the DH-n

relationship, for catalytically preactivated surfaces, at 50°C (410). It shows satisfactory agreement between results obtained either by gas-phase or electrolytic techniques over the higher, P-phase, range, where DH increa~es with inma~ing n. However, over the lower, a-phase ranges of n, studies using the electrochemical technique, have shown an initial opposing trend of &mmng values of D H with inmmng values of n (%lo).

Tubular Membrane Studies Tubular membranes have been used in a more

recent series of studies on electrochemical hydrogen permeation in 0.02 N H2S04 at 25 or 50°C (15-23, 25, 26). Parameters measured were the inner-tube hydrogen-gas pressures and internal surface electrode potentials. In electrolytic experiments, essentially analogous to those of Kussner (8), further progressive increases in the hydrogen content (n) of membrane surfaces were introduced by a stepwise series of electrolytic cathodisations.

Uphill Hydrogen Transfer Effects After interruptions made to each additional

outer-surface hydrogen-charging process, open- circuit conditions were maintained for both

0.1 0.2 0.3 0.4 n = H / M

Fig. 1 Representation of the dependence of liydrogen dlfficsion coefficients in Pd77Agzj ul1o.v~ on initial hvdrogen contents ,fioni work by: Kiissner (8): Hicknian (9): Kuhalla and Baranowski referred to in (4)

Pkztinum Metah Rev., 2003,47, (l), 32-36 32

Fig. 2a Time dependent incremental changes of internal surface electrode potentials at 50*C,,for Pd77Ag~3 tubular membranes (dia. 8.0 mm, wall 0.4 mm thick) with Pd black coats after cathodisation in 0.02N H z S O ~ at 15 mA cm-’, and ajier establishment of steady state values,for En. Corresponding values for hydrogen content, no, were derived from p(E)-n-T (15, 25, 30) relationships. Breakthrough times, tL, are indicated interpolations

> E W‘

Eo, mV tl, min

0 867.9 4.8 A 50.40 22.0 0 40.40 32.0 X 38.10 39.5 0 33.75 30.5 W 22.40 7.8

Fig. 2b Complementary incremental changes of thermodynamically equivalent values of hydrogen gas pressures within the tubular Pd77Ag~j membranes, calculated via the amended 2FE = RT In p relationships, F and R are the Faraday and gas law constants, respectively

po,to= noP/M) t L , m i n 0 0.00 0.000 18.3 a 0.42 0.023 23.9 0 0.91 0.044 32.3 X 1.09 0.049 39.9 o 1.53 0.062 29.0 W 3.70 0.195 6.8

10 20 30 40 50 TIME, min

surface and internal hydrogen equilibration processes, during periods of gradual decay to new steady state interim values of internal hydrogen pressure, p, and electrode potential, E. The p and E interim open-circuit values were then adopted as new initial values, PO and &, together with the next initial value of hydrogen content, no, determined from available p@)-c(n)-T relationships (4-10).

Figure 2a shows an example of time-dependent measurements of inner-surface electrode potential plots for a membrane of 8.0 mm inner diameter, 0.4 mm wall thickness, with inner and outer Pd

black coats. This followed the resumption (at 50°C) of cathodisation at 15 mA cm-’. Earlier established equilibration conditions ace at the EO values (23).

In Figure 2b, the values of p have been calculated from the values of E given in Figure 2% using the thermodynamically equivalent, when corrected, correlation:

2F = RT In 0 Eo PO

F is the Faraday constant and R is the g a s constant. The time dependent paths of the AE-t and the

Pkztinum MetaLr Rm, 2003, 47, (1) 33

:! 2

0.0 0.1 0.2

20 0

10.0 6 E

X

00 ,E

I n r H / M

derived Ap-t plots (Figures 2a and 2b, respectively) are typical of similar results that have been interpreted in forms of the uphill Gorsky Effect (15, 18, 21). These involve hydrogen interstitial transfer processes which operate in a direction opposite to the hydrogen permeation flux.

The values of the breakthrough times, tL, in Figures 2a and 2b correspond to intersection points between the AE-(Ap)-t axes and the later, more linear, stages of the time plots. Values of DH were then calculated (15, 20) using Relation (ii):

D H = - 12 (i)

6 t ~

1 is the thickness of the membrane wall (0.4 mm).

Comparison of DH-n Relations Figure 3 compares results obtained using alter-

native ways of deriving the relationship between DH and n at 50°C. The sources used to determine the DH values are: [a] a set of measurements obtained from direct hydrogen pressure records (26) @J] calculations using Relation (ii) with ti. values from Figure 2b [c] replotting the corresponding diffusion data of Kiissner (Fig. 9 in (8))

From Figure 3 it can be seen that there are overall similarities between the results presented by Kiissner (8) and more recent analogous data

Fig. 5 Coiiiparison ofcatu(~~tic activity iri N0.x conversion o f three cata!vsts iiiipregriored onto ciii intiirstriul a-Alr03 sllpport: Vp = velocitv ofgas mixture I catalvst prepured.fkoiii cluster ( I ) tint1 pheri. contaiiiirig 0. I % Pd 2 HzPtCI6. cwiifaining 0.1% Pt 3 indirstriol catci(\.st APK-2; this

cutu!\'st hosed on the plotiiiuni salt.

catolvsr cor l ths 2% Pd

(23, 26, 27). In particular, each DH-n plot has regions of apparent &muse of DH with inmmng n over an initial range from n = 0 to - 0.14.2.

For the two more recent determinations, [a] and [b], the results are again similar to earlier analogous Pd77Ag2?Hn reports (25, 26, 29). In these cases the temporary sign reversal of the incremental changes of permeation rate, when hydriding is restarted, has been identified with periods of elas- tic strain gradient-induced uphill Gorsky Effect (opposing the permeation direction) (1 5-26, 2S34) on internal hydrogen transfer.

Kiissner had not considered this explanation when he described the reversed sign (Fig. 7 in (8)) in terms of a transition to a type of plastic vis- coelastic state (8, 10). He did not however cite or present physical evidence of any associated struc- tural or defect changes.

Figures 2 and 3 also show that Kiissner's results (8) over the lower range of hydrogen contents: n - 0.0-0.2 in Pd-i?Agz,, could have been interpreted differently in terms of concurrent uphill hydrogen interstitial diffusion effects (Atmax in Figure 3) without significantly altering the overall elasticity characteristics.

Summary For each example in Figure 3, the apparent

&muse in DH with inmuse in the a-phase hydrogen content, n, can be equated with longer time inter-

Phtinutn Metub Kev., 2003,47, (1) 34

vals and corresponding longer periods of uphill opposing-direction hydrogen permeation flux. This leads to longer times for attaining the break-

through times, tL, and so, through Relation (ii), causes misleading apparently decreasing values of DIr with increasing n. In a broader context, this

Acknowledgements are due in regard to collaborative sup- survey to supPoit Of a port from Johnson Matthey PLC, The Royal Society, London non-Fickian classification of Pd alloy-hydrogen

difusion systems, with possibilities for easy control of concentration gradients and boundary conditions (4, 12-15,29,30).

Acknowledgements

and the Czech Academy of Sciences.

References 1 T. Graham, Pbihs. Trans. R Sac. Lonhn, 1866, 156,

415; Pmc. R Sac. London, 1868,16,422; ibid., 1869,17, 212,506 1991,171,213

18 K. Kandasamy, F. A. Lewis, J. P. Magennis, S. G. McKee and X. Q. Tong, Z. P4s. Cbem. Neue Fake,

19 K. Kandasamy, X. Q. Tong and F. A. Lewis, J. P4s.: Conhs. Matter, 1992,4, LA39

2o K. Kandasamy, F. A. Lewis and x. Q. Tong in “Proc. 4th Int. Conf. Hydrogen Effects on Material Behaviour”, eds. N. R. Moody and A. W. Thompson, Jackson Lodge, WY, 12-16 Sept., 1989,

p‘ 249;J‘ Gm* K. Kandasamy, F. A. Lewis and x. Q. Tong, Bufl5m-KOM B ~ ~ J ~ w J . Pbs. Chem., Julich,

2 A. S. D a r k , Phtinum Metalr Rev., 1958,2, (l), 16 3 J. B. Hunter, Phtinum Me& Rev., 1960,4, ( 4), 130 4 F. A. Lewis, Phtinum Metalr Rey., 1960, 4, (l), 32;

ibid., 1960, 4, (4), 132; ibid., 1961, 5, (l), 21; ibid., 1982,26, (l), 20; ibid., 1982,26, (2), 70; 1982,26, (3), 121; F. A. Lewis, “The Palladium Hydrogen

Lewis, Z. P4s. Cbem. Neue Fake, 1985,146,171; F. A. kwis, p u n 4 p l cbem., 1990,62,2091;y, s k o t o and F. A. Lewis. in “The ExDerimental 1990, No‘7

System”, Academic press, London, 1967; F, A, Medurdcal Aim, 1990,

Determination of Solubilities”, eds. G. T. ‘Hefter and R P. T. Tomkins, John Wiley & Sons, Chichester, 2002, pp. 215230

5 T. B. Flanagan, Engelhard Ind. Tech. Bull, Commemorative Issue, Thomas Graham, 1966, VII, (1/2), 9; T. B. Flanagan, R. Balasubramaniam and R Kirchheim, Phtinum Mefalr Reu., 2001, 45, (3), 114; ibid., 45, (4), 166

6 J. Volkl and G. Alefeld, in ‘Hydrogen in Metals 1’, Top. AppL P ~ s . , 1978, 28, Ch. 12,321

7 G. L. Powell and J. P. Kirkpatrick, P4s. Rev. B., 1991,43,6968

8 A. Kiissner, Z. Nafufoorsh., 1966,21a, 515 9 R G. Hickman, J. Less-Common Met., 1969,19,360

10 E. Wicke, H. Brodowsky and H. Ziichner, in ‘Hydrogen in Metals 11’, Tq. AppL Pbs., 1978, 29, Ch. 3,73; H. Ziichner, H. Barlag and L. Opara, POL J. Cbem., 1997, 71, 1863

11 W. S. Gorsky, Pbs. Z. Sow., 1935,8,457 12 J. H. Petropoulos and P. P. Roussis, J. Chem. Pbs.,

1967,47,1491; ibid., 1496; ibid., 1968,48,4619; ibid., 1969,50,3951

13 J. CermY R o q Chem., 1976, 50,1741 14 A. K. Das, J. AppL P ~ s . , 1991,70, (3), 1355 15 F. A. Lewis, J. P. Magennis, S. G. McKee and P. J.

M. Ssebuwufu, Nature (London)), 1983,306,673 16 F. A. Lewis, B. Baranowski and K. Kandasamy, J.

Less-Common Mef., 1987, 134, L27; F. A. Lewis, K. Kandasamy and B. Baranowski, Phtinum Metah Rev., 1988,32, (l), 22

17 F. A. Lewis, K. Kandasamy and S. G. McKee, Z. Pbs. Cbem. Neue Fake, 1989,164,1019

21 X. Q. Tong and F. A. Lewis, J. Less-Common Met., 1991, 169, 157; F. A. Lewis and X. Q. Tong, ibid, 1992,179, L13

22 F. A. Lewis, X. Q. Tong and R V. Bucur, Pkafinum Metalr Rev., 1991, 35, (3), 138

23 F. A. Lewis, X Q. Tong, K. Kandasamy, R. V. Bucur and Y. Sakamoto, Tbmocbim. Acfa, 1993,218, 57; F. A. Lewis, Y. Sakamoto, K. Kandasamy and X. Q. Tong, D@cf Dflu~. Forum, 1994,115-116,39

24 Y. Sakamoto, H. Tanaka, F. Sakamoto, F. A. Lewis and X. Q. Tong, Inf. J. Hydmgen Energy, 1995,20,35; Y . Sakamoto, H. Tanaka, F. A. Lewis, X. Q. Tong and K. Kandasamy, ibid, 1996, 21, 1025

25 F. A. Lewis, X. Q. Tong, R. V. Bucur and K. Kandasamy, Defect D@s. Forum, 1997,148-149,161

26 X. Q. Tong, R. A. McNicholl, K. Kandasamy and F. A. Lewis, ap. tit, (Ref. 24), 1992,17,777; X. Q. Tong, R V. Bucur, K. Kandasamy and F. A. Lewis, Z. Pbs. Cbem., 1993, 181, 771; X. Q. Tong, Ph.D. Thesis, Queen’s University, Belfast, 1991

27 F. A. Lewis, R V. Bucur, X. Q. Tong, Y. Sakamoto and K. Kandasamy, in ‘Hydrogen Power, Theoretical and Engineering Solutions: Hypothesis 11’, Grimstad, 1997, ed. T. 0. Saetre, Kluwer, Dordrecht, 1998, p. 615

28 D. Dudek and B. Baranowski, Polish J. Cbem., 1995, 69,196; Z. Pip. Cbem., 1998,206,21

29 F. A. Lewis, in ‘Progress in Hydrogen Treatment of Materials’, ed. V. A. Goltsov, Donetsk-Coral Gables, Kassiopeya, Donetsk, 2001, p. 147; Int. Sci. J. Ahnatite E n e w E d , 2002,2,4

30 F. A. Lewis, K. Kandasamy and X. Q. Tong, Solid State Phenom., 2000, 73-75, 207; ap. cit., (Ref. 26), 2002,27,687

Phfinum Metah Rev., 2003,47, (1) 35

31 V. A. Goltsov and T. N. Vezkoglu, op. tif., (Ref. 26), 2002,27,719

32 M. V. Goltsova, Yu. A. Artemenko, G. I. Zhirov and V. I. Zaitsev, op. tif., (Ref. 26), 2002,27,757

33 X. Q. Tong, Y . Sakamoto, F. A. Lewis, R. V. Bucur and K. Kandasarny, op. tit., (Ref. 26), 1997,22,141

34 V. A. Goltsov, op. tit., (Ref. 26), 2002, 27, 853

The Authors Jan Cermak, while being retired from the Institute of Physics in Prague, is currently a Research Associate with Professor Marian Ceriiansky. His interests are hydrogen in palladium and diffusion coefficients for nickel and platinum.

Xiu Qiang Tong was a Ph.D. student in the Department of Chemistry at Queen’s University, Belfast. At present he is Director of Support and Service, at Molecular Imaging Co., in Arizona, U.S.A. His interests include hydrogen in metals, scanning probe microscopy and its applications in various disciplines, such as materials surfaces (metals, polymers and biomaterials, etc.).

Fred Lewis is retired from Queen’s University, Belfast, after many years of research into hydrogen diffusion in palladium and palladium alloys. These are still his main interests.

Steven Bell is a Lecturer in Physical Chemistry at Queen’s University, Belfast, with interests in excited state porphyrins, redox enzymes, Raman spectroscopy and spectroscopic analysis.

Platinum Group Metals Technology in Ekaterinburg Research and development undertaken at the

Ekaterinburg Non-ferrous Metals Processing Plant in Russia into metallurgical aspects of the platinum group metals (pgms) is described in a recent issue of the Russian journal Trvefnye Metdh (l), translated into English.

With over 80 years’ experience, the plant is a leader in platinum (Pt) jewellery production which forms, together with jewellery alloys, a large part of the Pt output at the plant. Casting and moldings are done in an inert atmosphere or under vacuum and jewellery soldering alloys for Pt alloys are being developed. A volatile oxide-forming component is added to a jewellery alloy to counter gas absorption.

Dispersion hardened alloys for glass production,

Other technologies with improved results are: knitted gauze catalysts and catchment gauzes for ammonia oxidation, Pd alloy powders for capaci- tors, and new materials for dental alloys. Au-Cu-Pd alloy phase diagrams have also been investigated.

The plant at Ekaterinburg recycles pgms waste. Their rhodium refining scheme is very flexible. There are new processes for osmium (0s) recovery and separation, and a method for the production of high purity 0 s powder uses gaseous phase extraction at up to 2000°C. A sample from a placer deposit from the Inagli field, Yakutia, contained 0.89% Os, 71.55% Pt, etc. Iron, nickel and copper were the major impurities. The recovery of up to 95% 0 s in a full processing cycle of placer Pt, and

particularly glass fibre production, have been developed. The protection of pgm equipment and reduction of metal loss during fabrication has been achieved by plasma technology, in a TM-plasmoce- ramic’ system. This has been adopted by several industries. Sputtered coatings on the outer surface of equipment result in less metal use and metal loss.

The brittle fracture of Pt alloys caused by various molten metals and elements, such as iron, manganese, calcium and silicon, has been investigated. Iridium (Ir) research is a speciality at Ekaterinburg, so a short item from V. A. Dmitriev, N. I. Timofeyev and A. V. Ermakov on brittle frac-

100% recovery from the gaseous phase is claimed.

References 1 Non-Fmur Mefuh, 2002, (3), published by ‘Ore and

Metals’ Publishing House, PO Box 625, Moscow 119049, Russia; Fax: +7 (095) 230 4423 P. Panfilov and A. Yermakov, Platinum Metah Rm., 2 2001, 45, (4), 179

Noble and Rare Metals Conference The 4th international conference on noble and

rare metals (Nm-2003) will be held in Donetsk, Ukraine, on 22nd to 26th September, 2003. Geology, extraction, recovery and secondary refining, alloys and

ture in Ir and Ir alloys and the effects of molten additions is of note. Ir only fails by brittle fracture under tensile stress (2). Ir crucibles used in repro- cessing lead-zinc ‘cakes’ containing high concentrations of gold and silver at > 1300°C have operated for a few hundred hours without failure.

alloy properties, and industrial uses will be covered. Information can be obtained from Professor

V. A. Goltsov of Donetsk National Technical University, Ukraine; E-mail: [email protected]. donetsk.ua; and from the internet at: http:// dgtu.donetsk.ua/NRMworld/files/eng/.

Pkdnum Metah h., 2003,41, (1) 36

ABSTRACTS of current literature on the platinum metals and their alloys

PROPERTIES Control of the Magnetic Anisotropy of a Copt Nanomultilayer with Embedded Particles LJ. JEON, D.-W. KANG, D.-E. KIM, D.-H. KIM, S.-B. CHOE and S . - C . S m , A d v . Mutm, 2002,14, (16), 1116-1120

Co/Pt nanomultilayers with embedded particles were shown to have different magnetic characteristics compared with particle-free samples (made by a mag- netron sputtering system) of almost the same structure. This is particularly with respect to having (a) smaller coercivity fields, (b) biaxial magnetic anisotropies and (c) the existence of a critical field at which the easy direction changes from the parrallel direction to a perpendicular direction and vice versa.

Martensitic Transformation in Pd-Rich Fe-Pd-Pt Alloy T. WAD& T. TAGAWA and M. TAYA, Scr Mdm, 2003,48, (2), 207-21 1

The replacement of Pd with Pt in an Fe-30 at% Pd alloy decreased the f.c.c./f.c.t. martensite transformation temperature. Strengthening due to solution hardening with Pt was observed. The shape memory effects of Fe-(xPd_rPt), = 30 alloys with 5 4 at.% Pt were investigated below the austenite finishing temperatures.

Improvement of Shape Memory Characteristics in Fe-Pd Melt-Spun Shape Memory Ribbons D. VOKOLJN and C. T. HU, J. &J CaqDd, 2002,346, (1-2), 147-1 53

An Fe-Pd alloy with - 30 at.% Pd and with or without very small amounts of the b.c.t. phase exhibits shape memory properties with transformation temperatures - room temperature. The use of the free jet melt-spinning method in manufacturing the Fe-Pd ribbons was effective in suppressing the presence of b.c.t. martensite. This was due to rapid solidification.

Two-Directional Nz Desorption in Thermal Dissociation of N20 on Rh(llO), Ir(l10), and Pd(l l0) at Low Temperatures H. HORINO. I. RZE~NICKA, A. KOKALJ, I. KOBAL, Y. OHNO, A. HIRATSUKA and T. MATSUSHIMA, J. Vac. Sci. Technal. A, 2002,20, (5), 1592-1596

Two-directional N2 desorption was found in NzO dissociation on Rh(llO), Ir(l10) and Pd(l10) at < 160 K by angle-resolved thermal desorption. NzO(u) is mostly dissociated during heating, emitting Nz@ and leaving O(u). Nz showed four desorption peaks at 110-200 K. One of them commonly shows a cosine distribution due to desorption fiom N&). N2 desorbing in the other peaks sharply collimates off the surface normal in the plane along the [OOl] direction.

CHEMICAL COMPOUNDS Metallabenzenes and Valence Isomers. Synthesis and Characterization of a Platinabenzene V. JACOB, T. J. R WEAKLEY and M. M. HALEY, Angew. them. Int. Ed, 2002,41, (18), 347C3473

The synthesis of the hist example of a platinabenzene (1) has been achieved. ~t(coa)Clz] was treated with a nucleophile 3-vinylcyclopropene to give (l), which was fully characterised by NMR spectroscopy and X-ray crystallography. (1) contains two different rearrangement products of the vinylcyclopropene precursor.

Polymer Complexes of Rhodium(l1) Trifluoroacetamidate with Pyrazine, 4,4'-Bipyridine, and 1,4-Diazabicyclo[2.2.21octane M. HANDA, Y. MURAKI, M. MIKURNA, H. AZUMA and K. KASUGA, BULL Chem. Sac. Jpn., 2002,75, (8), 1755-1756

Polymer complexes, [{Rh~(HNOCCF$.,(L)},] (L = pyrazine, 4,4'-bipyridine (4,4'-bpy), 1,Cdiazabicy- clo[2.2.2]octane), of Rh(II) trifluoroacetamide dimers bridged by bidentate hgands were prepared and characterised. For [(Rh~(HN0CCF,),(4,4'-bpy)}.], the crystallographic inversion centre is located at the centre of the dimer core. The chain structure is built up by the alternating arrangement of the Rhz dimer and 4,4'-bpy.

Kinetic Analysis and Solvent Effects in the Carbonylation of RuCI3.3Hz0 R TA"ENBAIJM,J P&. ck. 4 2002,106, (41), 950Z9505

The carbonylation of RuC1,.3HzO in refluxing alcohol represents the entry point to Ru organometallic chemistry. The overall carbonylation reaction is composed of three consecutive reactions, each step resulting in a Ru carbonyl complex with different Ru:CO ratios. A kinetic analysis of the overall reaction revealed the parameters involved in the rates of formation of each compound, and as a result, provides a method to control the product composition.

Triple C-H Activation of 1 ,[i-Bis(di-tert- butylphosphino)-2-( S)-dimethylaminopentane on Ruthenium Gives a Chiral Carbene Complex v. F. KUZNETSOV, A. J. LOUGH and D. G. GUSEV, Chem. Cammnn., 2002, (20), 2432-2433

A novel burn-chelating diphosphine, 1,5-bis(di-tert- butylphosphino)-2-(5)-dimethylaminopentane (l), 'pincer' ligand, was prepared in five steps. (1) under- went triple C-H activation in reaction with puCl~(pcymene)]z to give a chiral square-pyramidal 16-electron carbene complex of Ru (2). (2) was isolat- ed in 79% yield as air sensitive thin red needles soluble in common organic solvents.

Platinnm Metuh Rev., 2003,47, (l), 3 7 4 37

PHOTOCONVERSION Light-Emitting Cyclopalladated Complexes of 6-Phenyl-2,2’-bipyridines with Hydrogen-Bonding Functionality F. NEVE, A. CRISPINI. c. DI PIETRO and s. CAMPAGNA, Orgunometukcs, 2002,21, (17,351 1-3518

Cyclometallated Pd(II) complexes [Pd(Ln)Cl] (n = 1-3) were prepared using HL1 = 4-carboxy- (l), HL2 = 4-carboxyphenyl- (2) and HL3 = Chydroxyphenyl- 6-phenyl-2,2’-bipyridine (3). All the complexes (1-3) exhibited intense luminescence at 77 K, with lifetimes of 10-200 ps. (1) gave room temperature solid-state luminescence (emission at 650 nm and lifetime of 1 p), due to oligomeric species.

Oivalent Osmium Complexes: Synthesis, Characterization, Strong Red Phosphorescence, and Electrophosphorescence B. CXUSON, G. D. PHFLAN, W. KAMG‘lSKY, L DALTON, X JIANG, S. LIU and A. K.-Y. JEN, 1. A m . Chem. Joc., 2002,124, (47), 14162-14172

Red-emitting Os(Il)(N-N)zLL (1) (N-N = bipy or phen derivative; L L = phosphine or arsine) were synthesised. (1) feature strong MLCT absorption bands in the visible region and smong red phospho- rescent emission at 61 1 4 5 1 nm, with quantum yields 545%. Red OLEDs were fabricated by doping (1) into a blend of poly(N-vinylcarbazole) and 2-terb butylphenyl-5-biphenyl-l,3,4-oxadiazole.

High-Brightness and Low-Voltage Light-Emitting Devices Based on Trischelated Ruthenium(l1) and Tris(2,2‘-bipyridine)osmium(ll) Emitter Layers and l o w Melting Point Alloy Cathode Contacts F. G. GAO and A. J . BARD, Chem. Muter., 2002, 14, (8), 3465-3470

LEDs were fabricated based on an amorphous film of Ru@py)3(C104)z (- 100 nm thick) on I T 0 with printed low melting point alloys, such as Ga:In, Ga:Sn and Bi:In:Pb:Sn, as cathodic contacts. A device of structure: IT0 (I 10 n/square)/Ru@py)i(ClO,)z/ Ga:Sn gave red emission (3500 cd m-’ at 4.0 V) at 660 nm. C1~Ru(bpy)3(C104)Z, Ru@henanthroline),(C104)2 and Os(bpy)J(l‘F& were also used for LEDS.

Efficient Panchromatic Sensitization of Nanocrystalline TiO, Films by p-Oiketonato Ruthenium Polypyridyl Complexes A. lSLM1, H. SUGIHARA, M. YANAGIDA, K. HARA. G. FUJIHASHI, Y. TACHIBANA, R. KATOH, S. MURATA and H. ARAKAWA, New J Chem., 2002,26, (8), 966-968 Ru(4,4‘,4”-tricarboxy-2,2‘:6‘,2”-terpyridine) (1,1,1~

trifluoropentane-2,4-dionato)(NCS) and Ru(4,4‘,4”- mcarboxy-22’:6‘~- terp~~e) (1,1,1 -trifluoroeicosa.ne- 2,4-dionato)(NCS) when anchored to nanocrystalline TiOz h s achieve very efficient panchromatic sensitisation. This is over the whole visible range extending into the near IR region, yielding 70% in& dent photon to current conversion efficiencies.

ELECTRODEPOSITION AND SURFACE C OAT1 N G S Electrodeposition of Rhodium Metal on Titanium Substrates A. M. BARAKA, H. H. SHAARAWY and H . A. HAMED, Anti- Conusion Method Muter., 2002,49, (4), 277-282

Ti subsmates were pre-anodised in oxalic acid solution (100 g l-’) at high current density of 60-95 mA cm ’ and at ambient temperature. A thin, porous and conductive Ti oxide film was obtained which could then be electroplated with Rh. Rh metal was,elec- trodeposited over the anodised Ti substrates from a bath consisting of Rhz(SO& (5.2 g 1.’) and HzSO~ (100 g 1 ’). Rh electrodeposits with high adhesion and brightness can be achieved.

Plasma Based Ion Implantation Technology for High-Temperature Oxidation-Resistant Surface Coatings s. ISOGAWA, H. TOJO, A. CHAYAHARA and Y. HORINO, 514 Cout. TechnoI., 2002,158-159,186192

Ir-Re alloy surface coatings (1) were produced by plasma-based ion implantation (PBII) technology, using a coaxial type arc-vacuum metal-plasma source. (1) were used to prevent surface oxidation of WC alloy engineering tools at high temperature. The stick- ing strength of the PBII-films is similar to that of films prepared by a conventional sputtering method (SP-films). Microscopic exfoliation was observed at the surface of the SP-films, but no exfoliation was observed at the PBII-films.

APPARATUS AND TECHNIQUE Fabrication of Sharp Tetrahedral Probes with Platinum Coating M. KLTALAWA and A. TODA,Jpn.J. @s., Part I, 2002, 41, VB), 492W931

Metal-deposited cantilever probes (1) with a 13 nm Up radius for conducting scanning probe microscopy are obtained by sputter coating a Si cantilever with Pt on a single side. (1) are free from cantilever curling and have 350 probe resistance on average. Al was deposited as a reflex coating on the back of (1).

Analytical Performance of a Glucose Biosensor Prepared by Immobilization of Glucose Oxidase and Different Metals into a Carbon Paste Electrode S. A. MISCORLA, G. D. BARRERA and G. A. WAS, Ek&unu&, 2002,14, (14), 981-987

Enzymatic metallised C paste electrodes (CPEs) prepared by incorporation of metal mixtures into the paste offer a substantial decrease in the overvoltage for HzOz oxidation and reduction. Combining Ir into CPE containing glucose oxidase with Pd, Cu or Ru gave enhancements in sensitivity and selectivity. A linear relationship between current and glucose concentration was obtained at i 1.5 x 10 ’ M glucose for a Pd(2.65°/~)-Ir(8.00%-glucose oxidase(7.00°/o)-CPE.

Phtinwn Metah Rev., 2003, 47, (1) 38

Heterogeneous Pd-Catalyzed Biphenyl Synthesis under Moderate Conditions in a Solid-Liquid Two-Phase System S. MUKHOPADHYAY, S. RATNER, A. SPERNAT, N. QAFISHEH and Y. SASSON, 0%. Pmcess Res. Dw., 2002,6, (3), 297-300

The coupling of substituted halobenzenes to give biphenyls was achieved at 65°C using a reducing agent, such as a formate salt, and NaOH, in the presence of tetrabutylammonium bromide phase-transfer catalyst and 5% Pd/C catalyst. The occurrence of a competing hydrodehalogenation reaction was mir- imised by altering the process variables. The highest selectivity to biphenyl was 71%. The solid Pd/C catalyst was recycled without losing its catalytic activity by filtration and washing with MeOH.

Novel Biomass Gasification Method with High Efficiency: Catalytic Gasification at Low Temperature M. ASADULLAH, T. MIYAZAWA, S. ITO, K KUNIMORI, M. YAMADA and K. TOMISHIGE, Gmen Cbem., 2002, 4, (4), 385-389

Rh/CeOZ/SiOz (1) was shown to be an efficient catalyst for the gasification of biomass at 823-923 K in a fluidised bed continuous feeding reactor. This temperature range is much lower than conventional gasification methods (973-1073 K for catalytic and 107S1223 K for non-catalytic). With (l), - 98% C conversion can be achieved. The tar was completely gasified and only a very small amount of char formed.

Low-Temperature Catalytic Decomposition of N20 on Platinum and Bismuth-Modified Platinum: Identification of Active Sites R. BURCH, G. A. AlTARD, S. T. DANIELLS, D. J . JENKINS, J . P. BREEN and P. HU, Cbem. Commm.., 2002, (22), 2738-2739

Surface modification of 5% Pt/C with Bi was carried out by irreversible adsorption of aqueous Bi ions from a Bi nitrate solution. The Bi/5?h Pt/C catalysts were used to identify the active centres on Pt for NzO decomposition. Terrace sites were not active for this reaction, whereas edge or defect sites appeared to be the active sites. Thus, in contrast to NO dissociation on Rh, where large metal particles are favoured, for Pt, the highest activity for N20 dissociation occurs on very small metal particles.

Catalyst-Support Interaction in Fluorinated Carbon-Supported Pt Catalysts for Reaction of NO with NH, W. AN, K T. CHUANG and A. R SANGER, J. cat&, 2002,211, (2), 308-315

Pt/fluorinated C (0, 10,28 and 65% F) catalysts (1) were characterised and their activities were compared for the reaction of NO with NH3. Optimum activiq and selectivity were found for (1) when the support has 28% F. Above this F content, catalytic activity is inhibited due to bloc- of the Pt sites. The increase in selectivity with (1) is due to the electronic interaction between the fluorinated C and Pt which gives rise to enhanced dissociative chemisorption of NO on active Pt sites.

Dehydrogenation of Neohexane to Neohexene on Platinum Polymetallic Catalysts S. B. KOGAN and u HERSKOWITZ, Id En& C h . h., 2002, 41, (24), 5949-5951

Selective dehydrogenation of neohexane to neohexene was carried out on Pt polymetallic catalysts and Cr supported catalysts (1) in steam at 52CL55O"C. Pt- Sn-K/Al2O3 and Pt-Sn-K-Fe/Alz03 (2) gave similar initial performances of high selectivity of 83 mol%, significantly higher than those of (1). (2) with Fe has much better long-term stability, in 26 cycles, and was regenerated by steam treatment at 550°C.

Membrane Reactor Microstructured by Filamentous Catalyst L. KIWI-MINSKER, 0. womm and A. RENKEN, Cbem Eng. Sci, 2002,57, (22-23), 49474953

A microstructured membrane reactor (1) with micro-channels formed between closely packed catalytic filaments, 7 pn in diameter, of Pt/Sn on AlZO3 has been designed. (1) with a Pd/Ag membrane was used in non-oxidative propane dehydrogenation. The catalytic flaments were active/selective and with- stood periodic regeneration. At propane conversions exceeding equilibrium, selectivities towards propene I 97% were obtainable.

H 0 M 0 G E N EO U S CATALYSIS First Application of Secondary Phosphines as Supporting Ligands for the Palladium-Catalyzed Heck Reaction: Efficient Activation of Aryl Chlorides A. SCHNYDER, T. -R, A F. INDOLESE, u. P-OW and M. STUDER, Ady. Syntb. CataL, 2002, 344, (5), 495498

Secondary diakylphosphines (1) were employed as supporting hgands for the Pd-catalysed Heck reactions of electron-rich and electron-poor aryl chlorides with olefins. PdClz solution in concentrated HC1 was used as the source of Pd. The yields with HP(f-butyl)z and HP(adamanty1)z were comparable or better than those with P(t-butyl), and P(cyclohexyl),. (1) are readily available at low cost on a technical scale.

Synthesis of Monocarbenepalladium(0) Complexes and Their Catalytic Behavior in Cross- Coupling Reactions of Aryldiazonium Salts K. SELVAKUMAR A. ZAPF, A. SPANNENBERG and M. BELLER, Cbem. Eur. J., 2002,8, (17,3901-3906

The first monocarbenepalladium(0) complexes (1) having benzoquinone and naphthoquinone as additional ligands were prepared. (1) are air- and moisture-stable. NMR spectroscopy and X-ray analysis of (1) show a bidentate bonding mode of the quinone. When (1) were used as catalysts in Heck and Suzuki coupling reactions with aryldiazonium salts, yields > 90% were achieved.

Phtinum MetaLr Rev., 2003,47, (1) 39

H ETE R 0 G E N E 0 US CATALYSIS

Carbonylation Reactions of lodoarenes with PAMAM Dendrimer-Palladium Catalysts Immobilized on Silica S. ANTEBI, P. ARYqL E. MANZER and H. ALPER,]. 0%. Chem., 2002,67, (19), 6623-6631

Pd complexes, such as Pd(RCN)zClz (R = Me, Ph) or Pd(tetramethylethylenediamine)Mez, immobilised onto generation 6 3 PAMAM dendrimers supported on SiOz were used as catalysts for the carbonylation of iodobenzene in MeOH to form methyl benzoate. High yields were obtained and the catalyst was recycled 4-5 times without significant loss of activity.

Addition Polymerization of Norbornene-Type Monomers. High Activity Cationic Ally1 Palladium Catalysts J. LIPIAN, R. A. MIMNA, J. C. FONDRAN, D. YANDULOV, R. A. SHICK, B. L. GOODALL, L. F. RHODES andJ. C. HUFFMAN, Mammohks, 2002,35, (24), 8969-8977

High activity cationic (q3-allyl)Pd catalysts (1) for the vinyl addition polymerisation of norbornene-type monomers have been developed. [(q’-allyl)Pd(tricy- clohexylphosphine)(ether)] p(3,5-(CF&(C6H&] catalyst copolymerises 5-butylnorbomene and 5-triethoxysi- lylnorbomene (95:5 molar ratio) and is capable of producing more than a metric ton of copolymer per mole Pd per hour. (1) can be generated in situ by abstraction of leaving groups such as C1, CH’CO; and NO,- from Pd using a salt of a weakly coordi- nating anion.

Dendritic Nanoreactors Encapsulating Pd Particles for Substrate-Specific Hydrogenation of Olefins M. OOE, M. MU RAT^ T. MIZUGAKI, K EBITANI and K KANEDA, Nuno Lett.., 2002,2, (9), 999-1002

Dendrimer-encapsulated Pd(0) nanoparticles (2-3 nm diameter) inside poly@ropylene imine) dendrimers functionalised with triethoxybenzamide were obtained by extraction of Pd” and subsequent chemical reduction with KBK. The resulting Pd-dendrimer nanocomposites were used as catalysts for substrate- specific hydrogenation of polar olefins.

Synthesis of Pyrimido[4,5-blindoles and Benzo[4,5lfuro[2,3-1flpyrimidines via Palladium- Catalysed Intramolecular Arylation Y.-M. ZHANG, T. RAZLER and P. F. JACKSON, Tetruhedmn Lett., 2002,43, (46), 82358239

A Pd-catalysed intramolecular aiylation of various pyrimidine substrates gave pyrimido[4,5-b]indoles and benzo[4,5]furo[2,3-d]pyrimidines. Thus, 4-aryl- oxy- or 4-anilino-5-iodopyrimidines were treated with Pd(OAc)Z(PPh3)2 and a base, such as NaOAc, Et3N, CszC03 and NaOBu‘, in DMF to give the regioselective cyclised heterocycles. A study of the influence of different solvents suggested that DMF (85OC) or DMA (lOO°C) were the best. No reaction was observed at I 70°C.

Ir and Rh Complex-Catalyzed Intramolecular Alkyne-Alkyne Couplings with Carbon Monoxide and lsocyanides T. SHIBATA, K. YAMASHITA, E. KATAYAMA and K. TAKAGI, Tetrahedmn, 2002,58, (43), 8661-8667

Catalytic intramolecular akyne-alkyne couplings with CO and isocyanide have been developed. Carbonylative coupling was efficiently catalysed by Vaska’s complex (IrC1(CO)(PPh3)z). Various diynes were transformed into bicyclic cyclopentadienones in good to high yields. The first catalytic synthesis of iminocyclopentadienes was achieved using m C l ( c ~ d ) ] ~ in allcyne-alkyne coupling with isocyanides. Portions of isocyanides were added at appropriate time intervals in BuzO.

Water-Gas Shift Reaction Catalyzed by Mononuclear Ruthenium Complexes Containing Bipyridine and Phenanthroline Derivatives P. AGUIRRE. s. A. MOYA, R. SARIEGO, H. LE BOZEC and A. J. PARDEY, AppL Organornet. Chem., 2002,16, (lo), 597400

~d&o)zclz], p&ch], [R~~(co)(&o)](PF6)z, [R~LzC~]Z(PF&, [RULZ(CO)C~](PF~) and [RuL2(C03)].3Hz0 (L = bipy or phen derivative) with KOH, triethylamine or trimethylamine in aqueous 2- ethoxyethanol catalyse the WGSR. Mild conditions were used: Pc;o = 0.9 a m at 100°C. Any ligand that increased the electronic density on the Ru, such as 6- methyl- and 6,6’-dimethyl-4,4’-di-ter~-butyl-2,2’-bipyridine, produced a higher catalytic activity.

FUEL CELLS Microwave-Assisted Synthesis of Carbon Supported Pt Nanoparticles for Fuel Cell Applications w. x. CHRN, J. Y. LEE and z. IJU, Chem. Commun., 2002, (21), 258g2589

Spherical and uniform Pt nanoparticles (1) (3.54 nm diameter) were prepared by microwave irradiation. Pt/C nanocomposites containing 10,15 and 20 wt.% of Pt were also successfully prepared. (1) exhibited very high electrocatalytic activity in the room temperature oxidation of MeOH and are therefore useful in DMFC applications.

Electro-Oxidation of Ethanol on Pt, Rh, and PtRh Electrodes. A Study Using DEMS and in3i tu FTlR Techniques J. P. 1. DE SOUZA, S. 1.. QUEIROZ, K. BERGAMASKI, E. R. GONZALEZ and F. C. NART,]. P&. Chem. B, 2002,106, (38), 9825-9830

Three products, COz, acetaldehyde and acetic acid, were detected from the electrooxidation of EtOH by the title electrodes. Rh was the far less active electro- catalyst for EtOH electrochemical oxidation. Pure Pt and Pt&h,o gave similar normalised current densities, but Pt9uRh,,l gave a better COZ yield than pure Pt. Pt73Tih27 electrodes gave the best COz yield. The ratio CO2:CH3CHO increases when Rh is added to the electrode.

Phtinum Mefuh Rev., 2003,47, (1) 40

N E W PATENTS METALS AND ALLOYS Pt-Co Based Sputtering Targets HERAEXJS INC WonZAppL 02/083,974

A Co-Cr-B-Pt sputtering target alloy having multi- ple phases and with enhanced product performance may also include Ta, Nb, C, Mo, Ti, V, W, Zr, Zn, Cu, Hf, 0, Si or N. The alloy is prepared by mixing Pt powder with a Co-Cr-B master alloy, ball milling the powders and densifying the resultant alloy to form a magnetic sputtering target at pressures of - 15,000-30,000 psi, 150Cb1900"F for 1 to 6 hours.

Colloid Solution of Metal Nanoparticles POSTECH FOUNDATION World AppL 02/087,749 A Pt, Pd, Ag, Cu and/or Ni nanopartide colloid

solution (1) and metal-polymer nanocomposites (2) are prepared by dissolving a salt of the metal and a H20-soluble polymer, such as polyvinyl pyrrolidone, in HzO and an alcoholic solvent, purging with NZ or Ar gas and radiating the solution with radioactive rays. (1) and (2) have uniform particle diameter and shape, and are used as antibacterial agents, conductive adhe- sives and inks, electromagnetic wave shields, etc.

PHOTOCONVERSION Photocatalyst and Gas Deposition sow COW U.S. Patent 6,471,929 A photocatalyst with superior durability has a

fullerene polymer film (1) obtained by polymerising Cm or CT0 fullerene molecules using electron beam, electromagnetic wave or electronic polymerisation. (1) is layered on a substrate and fine Pt and Pd par& cles (0.5 nm to 100 pn in size) are applied by sputtering, evaporation or coating. An apparatus to decompose gas includes a light source so that the gas is contacted with (1) under light illumination.

Photocatalyst Material NIHON TETRA PAK KK Japane~e AppL 2002/ 186,860

A photocatalyst material (1) with excellent ability for decomposing pollutants comprises a composite film of TiOz and 3.0-80 wt.% Pd formed on the SUE face of a substrate by vacuum evaporation. The film is prepared by simultaneously evaporating Ti02 and Pd by ion plating. (1) is highly hydrophilic and can decompose many types of organic substances.

Photocatalyst Manufacture ISHIHARA SANGYO KAISHA LTD

Japanese Appl. 2002/239,395 A photocatalyst (1) able to be excited by visible

light irradiation is manufactured by adding a Pt halide compound to the surfaces of photocatalyst particles of TiOz, etc. Particles of (1) and the Pt halide compound are then heated in a liquid medium. An accelerator containing hypophosphorous acid can be further added to the liquid medium at the time of heating. (1) is stable with high photocatalytic activity.

E LECTR 0 D EPO S IT1 0 N AN D S U R FACE C OAT1 N G S Surface Coating of 'Black Platinum' OMG AG CO KG Workl AppL 02/095,088

A surface coating (l), 1 m-10 p thick, is formed from a fine dispersoid of modified Pt black particles, and comprises Pt and Si and/or a Si compound. An organic Pt(0) complex with 1,3-divinyl-l,1,3,3,-tetra- methyldisiloxane, which can decompose at < 200"C, is applied to the surface of a substrate and is then thermally decomposed. (1) protects against mechani- cal, chemical and/or thermal effects, and can be used as an antiadhesion, antireflective or catalytic layer.

CVD Ruthenium Seed layer and Ruthenium Thin Film APPLIED MATERIAIS INC U.S. Patent 6,479,100

A Ru seed layer is formed on a substrate by introducing a Ru-containing compound (1) and 0 2 into a CVD apparatus and maintaining an 02-rich environment for the initial formation of a Ru oxide seed layer. (1) is vaporised; and the Ru oxide seed layer (2) is deposited on the substrate by CVD. (2) is annealed in a gas ambient to form a Ru seed layer. A Ru thin metal film can also be deposited by MOCVD.

Electroless Platinum Plating Solution TANAKA KlIUNZOKU KOGYO KK

Japanese AppL 2002/173,780 An electroless Pt plating solution contains a hexa-

aminoplatinum complex (1). A salt of (l), which is used as the raw material, is made into an aqueous solution, and COZ is passed through to produce the carbonate of (1). This carbonate is then dissolved with acid. The plating solution enables continuous plating operation with extremely high stability, and further enables the production of a thin Pt film of high quality.

APPARATUS AND TECHNIQUE light-Emitting Devices SRI INTERNATIONAL WorldAppL 02/094,910

Conjugated polymers (1) with good solubility and semiconductivity which display high photolumines- cent and electroluminescent efficiency are claimed. (1) contain a luminescent dopant of If, Os, Pt, W, Eu and Au complexes, with a bidentate or tetradentate ligand. Electroluminescent devices and other devices containing (1) are also provided.

Enzyme Electrode SANKYO co LTD Japanse Appl. 2002/189,012 A Pt electrode, which can be used as part of an

enzyme electrode and is stable over time, comprises a Cu foil, a Ni layer, a Pd/Ni layer, and a Pt layer sequentially laminated on an insulating substrate. Superior accuracy in analysis and reproducibility are obtained. The Pt maintains very close contact and has high surface smoothness.

Phfinwn Metah Rev., 2003,47, (l), 4143 41

Sensor for Measuring Hydrocarbon Concentration NATI.. INST. ADV. IND. TECHNOL.

Jqanese AppL 2002/202,281 A gas sensor to measure the concentration of a spe-

cific gas, such as a hydrocarbon in a gaseous mixture of CO, NO, Hz, etc., is claimed. It comprises a reference electrode (1) containing baked Pt paste formed on one surface of an 0 ion conductive solid electrolyte of ZrO-stabilised YzO,, and a detection electrode (2) of 10 wt.% SrCeO3 and 90 wt.% Pt formed on the other surface. The difference in elec- tromotive force between both electrodes provides the measurement; (1) is exposed to air at 550-750°C while (2) is exposed to the gas to be measured.

H ETE R OG EN E 0 US CATALYSIS Palladium Hydrogenolysis Catalyst N E CHEMCAT COW Eumpan Appl. 1,238,700

A hydrogenolysis catalyst with high hydrogenolysis performance at low temperatures comprises: (a) a component selected from Pd oxide, Pd oxide mono- hydrate and Pd(0H)Z where Pd is in the divalent oxidation state, and (b) a component selected from Pt, Ru, Rh, Ir and Au, carried on a non-organic porous support. The Pd catalyst can be used for debenzylation, hydrodesulfurisation and dehalogena- tion reactions, and for the hydrogenolysis of esters.

Purifying Styrene Feedstock FlNA TECHNOLOGY Eumpean AppL 1,256,559

A method to remove contaminants, such as phenylacetylene, &om a crude styrene feedstock involves the catalytic reduction of phenylacetylene via injection of a reducing agent, such as Hz, to produce styrene. The reaction proceeds in the presence of a catalyst in the form of cylindrical pellets comprising < 0.3 wt.% Pd/Ca aluminate. Two reactor units, both containing the Pd/Ca aluminate catalyst, are used and output of phenylacetylene is reduced to < 10 ppm.

Preparation of l-Methyl-3-phenyl-piperazine NEULAND LABORATORIES ~ n , WO& Appl.02/090,339

1 -Methyl-3-phenyl-piperazine (1) is produced by mixing 1 -benzyl-2-phenyl-piperazine first with formic acid solution, then with formaldehyde solution, and heating to 70430°C. After treating with NaOH solution at < 25”C, filtration, washug and dry- ing l-benzyl4methyl-2-phenyl-piperazine is obtained. Acetic acid is then added in the presence of Pt/C catalyst at a HZ pressure of 3.54.0 kg ern-'. After further treatments, (1) is obtained in an air oven.

Purification of Diesel Engine Exhaust Gas KH CHEMICALS CO LTD WorkiAppl. 02/092,224

A thermally and chemically durable catalyst for purification of diesel engine exhaust gas comprises 0.01-90 wt.% Pt, Pd, Rh, Ru and/or Re supported on a S-resistant refractory oxide, such as SOz, etc. A solid acid precursor and/or H2S04 are also added. The catalyst is effective at removing particulate matter, hydrocarbons and NOx at low temperatures.

Direct Synthesis of Hydrogen Peroxide EN1 SPA Wot~2Apph. 02/092,501-502

HzOZ is produced from Hz and O2 in a reaction solvent that contains a halogenated promoter and/or an acid promoter, in the presence of a supported catalyst, such as 0.01-5 wt.% Pd and 0.01-1 wt.% Pt on activated C. The reaction solvent consists of alcohol(s), an aliphatic ether and optionally HZO, and may contain 5C-32C hydrocarbons. The process operates under high safety conditions with a high productivity and molar selectivity towards the formation of Hz02.

Fuel Reformer SUZUKI MOTOR cow Japanese AppL 2002/226202

A fuel reformer capable of a shorter warm-up time to the star t of operation of a reforming unit comprises a reforming part which generates HZ gas for steam reforming in a reforming catalyst layer while feeding heat to MeOH. A combustion part supplies the heat generated by burning a fuel in the combustion catalyst layer to the reformer. The reforming and combustion layers and catalysts are on either side of a thin metal sheet. The combustion catalyst contains 0.5-5 wt.% Pt. Excellent reforming is obtained.

H 0 M OG E N EO U S CATALYSIS Suzuki-Miyaura Cross-Coupling UNIV. NEW ORLEANS RES. WorMAppf. 02/072,511

A Pd(0Ac)Jdiazabutadiene catalytic system for cross-coupling aryl halides with arylboronic acids is claimed. A combination of Pd(OAc), and the diaza- butadiene, 1,iV,”dicyclohexyl-l,Cdiazabutadiene, is an efficient catalyst for the Suzuk-Miyaura cross- coupling of various aryl bromides and activated aryl chlorides with arylboronic acids.

Water Soluble Palladium Complexes COUNCIL SCI. IND. RES. U.S. Patent 6,469,169

A HzO-soluble Pd complex is claimed for use as a catalyst in organic transformations, such as carbonylation, oxidation, hydrogenation, etc. The complex contains a phosphhe ligand with three substituents selected from H, alkyl, arylalkyl and cycloaliphatic, at least one of which carries a sulfonic acid, and their salts. The Pd also carries a ligand of aryl or alkyl sul- fonato; aryl or alkyl carboxylato; formato; or halide, such as C1 , Br-, I-. A further anionic cheladng ligand consists of an N donor and an 0- group.

Acid Activation of Metathesis Catalysts CALIFORNIA INST. TECHNOL. U.S. Patent 6,486,279

Highly active and stable Ru metal carbene complexes used as catalysts in olefin metathesis reactions have general formula A&,$u=CHR’ where x = 0, l or 2; y = 0, 1 or 2; and = 1 or 2; R‘ is H or a substituted or unsubstituted alkyl or aryl, L is any neutral electron donor, X is any anionic ligand, and A is a ligand of covalent structure connecting a neutral electron donor and an anionic ligand. Activation with HCl improves rates and yields of olehn metathesis reactions, such as ROMP, RCM, ADMET and cross-metathesis.

Phtinum Metah Rev., 2003,47, (1) 42

Production of Optically Active Alcohols TORAY IND. INC Japanese Appl. 2002/155,096

A method to produce an optically active alcohol uses an optically active quadridentate Ru complex composed of an optically active compound contair- ing two phosphines and two amide bonds in the same molecule and Ru, and asymmetric reduction of a carbonyl compound. An optically active amide is used as a source of asymmetry. The asymmetric reduction of a wide range of carbonyl groups, from aliphatic carbonyl compounds (a p-ketoester) to aromatic ketones. is claimed.

FUEL CELLS Proton-Conducting Electrode SONY cow Eumpean Appl. 1,255,314

A proton conducting electrode (1) for a fuel battery comprises a mixture of an electron conducting catalyst (2) such as 1-50 wt.% Pt atoms, porous C powder fullerenes, C,,,, where m = 36,60,70,76, etc., (3) and a proton (€I+) dissociating group introduced into the C atoms. (1) is made by coating a mixture of (2) and (3) onto a gas transmitting current collector.

Anode Catalyst OMG AG CO KG Eumpean A#L 1,260,269

A Pt-Ru catalyst (1) for use as a fuel cell anode is prepared by suspending a support material in H20 and heating to 5 the boiling point. Solutions of H2PtCL and RuC13 are then added to the suspension, followed by addition of an alkaline solution. At pH 6.5-10, Pt and Ru are precipitated onto the support. Carboxylic acids and/or their salts are then added. (1) may optionally be calcined at 3061000°C. In a fuel cell, (1) has high tolerance to CO poisoning.

ShiA Converter with Improved Catalyst UTC FUEL CELLS LLC WorkiApl02/090,247

A shift converter in the fuel processing subsystem of a fuel cell reduces the CO content in the process gas by a water gas shift reaction using a catalyst (1) selected from Pt, Pd, Rh and Au, preferably Pt. The Pt is supported by mixed metal oxides of CeOz and ZiOz in the ranges of 30-50 moWo and 7 6 5 0 mol%, respectively. Additional metal oxides may also be present. (1) obviates the need for prior reduction and minimises the need to protect the catalyst from 0 2

during operation and/or shutdown.

Seawater/Acid/Catholyte Electrolyte us. SECRETARY OF THE NAVY U.S. Patent 6,465,124

A Mg semi-fuel cell (1) has a Mg anode, a seawater/catholyte electrolyte (preferably containing acid to solubilise solid precipitates) and an electrocat- alyst composed of Pd and Ir on C paper. The acid added to the electrolyte is preferably H2S04, HCl, phosphoric acid, acetic acid or their mixtures. (1) provides a k h energy density source for underwater vehicle applications with energy densities approach- ing 6 7 times that of AgZn.

ELECTRICAL AND ELECTRONIC ENGINEERING Contact Structure for Integrated Semiconductor STMICROELECTRONICS SRL W d A#L 02/086,965

An integrated semiconductor device contains: a first conductive region (1); a second conductive region containing Pt (2); an insulating layer (3) between the two regions; and a contact structure (4) made of a conductive Ti- and TN-layer. (4) coats the through opening in (3) and electrically connects (1) and (2). (4) is used in ferroelectric memory devices of the 'stacked' type, and is suited to the integration needs of the new CMOS technology.

Dielectric Composition with Reduced Resistance E. I. DU PONT DE NEMOURS co WorkiAppL 02/092,533

A dielectric composition (1) comprises a dielectric that is fireable in air and a conductive oxide selected from: Sb-doped Sn oxide, Sn-doped In oxide, a transition metal oxide with mixed valence states or which will form mixed valence states after hnng in Nz at 450-550°C, and conducting Pt group metal oxides, such as Ru02. (1) has reduced electrical resistance and is used in electron field emission devices to avoid charging the dielectric near the electron emitter.

Composite Barrier Structure SHARP LABORATORIES AMERICA INC U.S. Patent 6,479,304

An Ir combination film (1) used to make an electrode for a ferroelectric capacitor, also includes Ta and 0. (1) effectively prevents 0 diffusion, and is resistant to high temperature annealing in 0 2 . When used with an underlying Ta or TaN layer, the resulting barrier suppresses Ir diffusion into any underlying Si substrate, so Ir silicide is not formed. (1) remains conductive, with no pee% or hillock formation.

Giant Magnetoresistive Stack SEAGATE TECHNOLOGY L X U.S. Patent 6,490,140

A giant magnetoresistive (GMR) stack for use in a magnetic read head includes a NiFeCr seed layer, a ferromagnetic free layer (l), a ferromagnetic pinned layer (2), a nonmagnetic spacer layer between (1) and (2), and a PtMnX pinning layer (3), where Xis selected from Pd, Rh, Ru, Os, Cr, Nb, Re, Ta, Zr, Hf, Ni, Co and Fe. (1) has a rotatable magnetic moment and (2) has a fixed magnetic moment and is next to (3). A GMR read sensor has good thermal and magnetic stability and is used in magnetic data storage systems.

Perpendicular Magnetic Recording Medium SAMSUNG ~ECTRONICS KK Japanese AfpL 2002/216,341

A perpendicular reinforcement layer (1) which enhances perpendicular orientation between a substrate and a perpendicular magnetic recordmg (Pm) layer, is laminated in thicknesses of 2 15 nm. (1) contains Pt, Pd, Au, or their alloy. A base Ti layer may be placed between the substrate and (1). The effect caused by the difference in the diameter of the crystal lattice between the base layer and PMR layer is relaxed, while Perpendicular orientation is enhanced.

Platnum Metals Rm., 2003,47, (1) 43

Catalysts - Myths and Reillties A modem dictionary (1) definition of a catalyst

as ‘a substance that increases the rate of a chemical reaction without itself undergoing any permanent chemical change’ does not completely describe what occurs. Changing the rate of a chemical reaction, as opposed to just inmasing it, broadens the scope of catalysis. Indeed the dictionary acknowl- edges this extension by introducing concepts of ‘positive catalysts’ (increased reaction rates) and ‘negative catalysts’ (reduced reaction rates).

A commercial example of a negative catalyst is Lindlar’s catalyst. Here, controlled amounts of lead (Pb) are added to a palladium on calcium carbonate (Pd/CaCOS powder catalyst to inhibit alkane production from the reduction of alkynes.

Hz

Pb/Pd/CaCO, G C - -CH=CH- Jt, -CH&H?

akyne akene alkane

In the reaction A + B or C, catalysts can often be designed to favour the formation of B (with inhibition of C) or promote the production of C (with inhibition of B) - catalyst selectivity.

If a catalyst did not undergo any ‘permanent chemical change’, then presumably it could last indefinitely. In practice every catalyst has a finite lifetime. There are two main reasons why a catalyst deactivates: aghmeration of active sites and poisoning (both chemical and physical).

Aghmeration. Over time, the metal surface area of a platinum group metal (pgm) heterogeneous catalyst (and hence its activity) will fall due to metal crystallite agglomeration. A compromise has often to be made between increased reaction rate @gh temperature) and low agglomeration rate (low temperature). A similar effect can occur with mononuclear pgm homogeneous catalysts at much lower temperatures when polynuclear clusters may form. This is inhibited to some extent by excess free ligand being present.

Cbemicafpoisoning. A typical mode of chemical poisoning is when an impurity in the reagents reacts irreversibly with the active metal. It is thus

common practice to ensure that the feedstocks are as pure as possible - often achieved by installing upstream purification units.

Physicafpoisoning- external Another mode of catalyst deactivation is when the active sites become covered by ‘dirt’ from external sources (perhaps oil drops (from compressors) or condensed spray from upstream wet scrubber units). Installation of a sacrificial filter immediately upstream of the catalytic reactor can often be beneficial.

Physicaf poisoning - intemaf. Sometimes, physical masking of catalytic sites can be caused by solid deposition from an internal side reaction. In the petrochemical industry, one factor affecting catalyst lifetime in hydrocarbon reforming is the rate of elemental carbon lay-down. Such carboniferous deposits may be removed by controlled burning. The resulting exotherm must be limited to ensure minimal metal crystallite agglomeration. Such catalysts never attain 100% of their original activity, so in practice the number of reactivations is limited.

Metal h.rs can clearly be an important factor in reaction efficiency. In gas-phase heterogeneous catalysis, pgm loss can occur by volatilisation of metal or by abrasion/dusting losses from the surface of individual pellets. In liquid-phase reactions, metal loss can occur by dissolution from the support into the reaction medium. A possible loss mechanism (also can occur with homogeneous catalysts) is the soluble pgm salt plating out onto the reactor wall.

At some stage the process operator will decide that the overall reaction performance is unsatisfac- tory. Then, spent catalyst residues can be returned to the manufacturer for pgm recovery and (if required) conversion to fresh catalyst for reuse.

0. E. GROVE

Reference 1 “Oxford Dictionary of Chemistry”, 4th Edn., ed.

J. Daintith, Oxford University Press, Oxford, 2000

David E. Grove is a former Marketing Manager in Johnson Matthey Catalysts and Chemicals Division. His many years experience of the platinum group metals catalyst industry gives him a unique insight into typical user problems.

Pl;ltinum MefaLr Rey., 2003,47, (l), 44 44

PLATINUM METALS REVIEW · Platinum Group Metals and Their Oxides in Semiconductor Photo...

Documents

Transcript of PLATINUM METALS REVIEW · Platinum Group Metals and Their Oxides in Semiconductor Photo...