UK ISSN 0032-1400

P L A T I N U M METALS R E V I E W

A quarterly survey of research on the platinum metals and of developments in their application in industry

VOL. 36 APRIL 1992 NO. 2

Contents

Platinum Metals as Components of Catalyst-Membrane Systems

Platinum-Iridium Carbon Monoxide Sensor

Solvated Atoms of Platinum, Palladium and Gold

Platinum Metals Catalyst Studies

Advances and Developments in Emissions Control

Platinum Group Metals in 1991

Platinum Improves Protective Coatings

The Large Scale Production of Hydrogen from Gas Mixtures

Platinum in High-Temperature Superconductors

Ruthenium Oxide Contacts

Temperature-Programmed Reduction of Platinum Group Metals Catalysts

Iridium Protects Rocket Thrusters

Abstracts

New Patent

70

79

80

85

86

89

89

90

97

97

98

103

104

116

Communications should be addressed to The Editor, Platinum Metals Review

Johnson Matthey Public Limited Company, Hatton Garden, London E C l N 8EE

Platinum Metals as Components of Catalyst-Membrane Systems By ProfessorV. M. Gryaznov A. V. Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, Moscow

Platinum group metals are extensively used as catalysts both in dispersed form and as solids. During recent years we have been wit- nessing the rapid and successful development of a new branch of catalysis, namely the cre- ation of catalyst-membrane systems. The system combines a catalyst and a membrane which has selective permeability for one of the reagents. Platinum group metal catalysts en- sure h igher target product yields a n d durability than other catalysts.

In general, catalyst-membrane systems en- hance both reaction rate and selectivity, due to directed transfer of energy and reagents (1). Three functional variations of catalyst-mem- brane systems have been investigated:

(a) One of the initial reactants, for example hydrogen, reaches the catalyst through the membrane, which is permeable for this sub- stance only. The second reactant comes from a gaseous or liquid phase.

(b) One of the reaction products is selec- tively removed through the membrane.

(c) The substance penetrating through the membrane is being formed on the catalyst adjacent to one surface of the membrane, this then diffuses through the membrane and reacts at its other surface on the second cata- lyst with the substance being introduced from the gaseous or liquid phase.

In the first case the catalyst-membrane sys- tem provides independent control of the surface concentrations of the two reagents in order to suppress their competing adsorption which is harmful, but inevitable on conven- tional catalysts. For example during phenol hydrogenation on palladium-ruthenium alloy foil, the cyclohexanone yield decreases very rapidly with time if hydrogen is fed in as a mixture with phenol vapour, see Figure l(a). However, if hydrogen penetrates through the foil to the surface which is in contact with

k

2 6 0 W

U

U

Pd-RU

l b l

TIME

Fig. 1 The time dependence of the eyclohexanone yield during phenol hydrogenation on a palladium-ruthenium foil, supplied with: (a) hy- drogen in a mixture with phenol vapour and (h) with hydrogen passing through the membrane catalyst

Platinum Metals Rev., 1992,36, (2), 70-79 70

Fig. 2 Cyclohcxane conver- sion is shown as a function of the flow rate of: (a) argon and (b) a mixture of argon with 1,3-pentadiene vapour along the other surface of the palladium-ruthenium foil. The mole fractions of cyclohexane and 1,3-penta- diene in the initial mixtures with argon are equal to 0.17 and 0.12, respectively

Palladium-ruthenium -Membrane- Palladium-ruthenium

c Ar+ H

z (a) 0

10

2 8 05Fl = o s 10 20 30

+ cs H10

+ Ar

5 K) 20 30 P v,, RW RATE OF ARGON, (rmolesls V,,FLOW RATE OF ARGON. FENTADIENE

Urnoles/s

phenol vapour, the product yield is much higher, see Figure l(b). In this case the foil is acting as both the catalyst and the membrane. Under optimal conditions 92 per cent of phe- nol is hydrogenated on the catalyst in one step into cyclohexanone, which is used for the pro- duction of nylon-6.

In the second case the reaction rate in- creases due to the removal from the reaction zone of the obtained product as it passes through the membrane. Thus the reaction rate of cyclohexane dehydrogenation increases with the removal through the palladium-ru- thenium foil of the hydrogen formed. The catalyst in this reaction contains 0.4 per cent platinum and 0.4 per cent rhenium supported on alumina. The catalyst pellets are repre- sented by circles in Figure 2. On increasing the velocity of the inert gas flow, V,, which washes the other surface of the membrane, cyclohexane conversion rises dramatically, Figure 2(a) lower curve, and attains the equili- brium degree of conversion, indicated by the blue line, which is obtained under the same conditions but without hydrogen withdrawal. I t also works for much lower V, values when the cyclohexane feed rate is four times smaller, and its conversion reaches 0.93, as indicated by the upper curve of Figure 2(a). If, however, the palladium-ruthenium foil is replaced by a

steel plate (non-permeable to hydrogen) which is covered by a palladium-ruthenium alloy layer, then cyclohexane conversion is re- duced to the values shown by the points on the ordinate axis of Figure 2(a).

The two catalysts shown in Figure 2 permit the use of hydrogen, which is removed through the palladium-ruthenium foil, for the hydrogenation of 1,3-pentadiene (see right hand scheme). Thus reaction coupling takes place on the catalyst-membrane system. This illustrates the third type of catalyst-membrane system.

As compared with hydrogen removal by the inert gas, the degree of cyclohexane dehy- drogenation increases when reactions are so coupled; which can be seen in Figure 2(b), where the curves lie higher than the respective curves of Figure 2(a). The dotted curve shows pentadiene conversion into the products of selective hydrogenation, which are pentenes, the selectivity of which attains a maximum of 98 per cent. Under these conditions the con- versions for both cyclohexane and pentadiene reach 0.99. Experimental evidence of such re- action coupling was found independently in the U.S.S.R. (2) and in the U.S.A. (3), some 25 years ago. Later, coupled reactions were stu- died on hydrogen permeable palladium alloys in the form of foils and thin-walled

Platinum Metals Rev., 1992,36, (2) 71

Fig. 3 A laboratory-scale spiral membrane catalyst supplied by the A. V. Topchiev Institute of Petrochemical Synthesis, and a pilot plant reactor containing two hundred similar spirals, are shown

tubes (4-7). A system analogous to that shown in Figure 2 was used for coupling butane de- hydrogenation and hydrogen oxidation (6). The system consisted of a commercial alumi- n o c h r o m i u m o x i d e c a t a l y s t a n d a palladium-ruthenium alloy foil. The foil served both as the hydrogen permeable mem- brane and as the catalyst for the oxidation of hydrogen by air, which was fed along the other surface of the foil. A similar system, compris- ing an alumino-platinum catalyst and a palladium tube, has been used for cyclohexane dehydrogenation (8,9).

Types of Catalyst-Membrane Systems Monolithic Membrane Catalysts

Some materials are both catalytically active and selectively permeable. Palladium and its alloys have been extensively studied for hy- drogen evolution or consumption (10-12), and silver has been used for partial oxidation. A non-porous membrane catalyst made of these materials may be called monolithic. Foils and tubes made of palladium alloys are monolithic catalysts permeable only for hy- drogen. Such monolithic membrane catalysts with smooth surfaces are easily produced by foil rolling and seamless tube drawing and are commercially available in Russia. Palladium foil alloys can vary from 100 to 10 pm in thickness; tube has a wall thickness of 50 pn

and an outer diameter of 0.6 mm. The foils are active both for hydrogenation and dehydroge- nation reactions and also possess higher hydrogen permeability than pure palladium and some other industrially used alloys. A thin walled palladium alloy tube twisted into the form of a spiral is shown in Figure 3. This tube can be used as a liquid phase hydrogena- tion catalyst. Hydrogen is fed into it when the tube is submerged into the liquid to be hy- drogenated and heated to the required temperature. The effect of the structures of the substituted ally1 alcohol and propargyl alcohol on the hydrogenation rate (13) and on the kinetics of the hydrogenation of phenylace- tylene to styrene have been studied on such catalyst systems (14). The latter reaction is of interest because phenylacetylene admixture in styrene complicates the polymerisation. With hydrogen diffusing through the membrane catalyst the phenylacetylene hydrogenation rate is ten times higher than in experiments where hydrogen is bubbled into the hydroge- nated substance, and selectivity to styrene reaches 0.92.

Methods for the hermetic connection of foil sheets or tube bundles to other parts of the reactor system have been described (1 1).

The reactors, one of which is also shown in Figure 3, have successfully withstood pilot- plant tests at temperatures up to 973 K and pressure drops up to 10 MPa. These reactors

Platinum Metals Rev., 1992,36, (2) 72

can also be used for hydrogen extraction from reforming gases, the product gases of methane steam conversion and also the purge gases of ammonia synthesis.

The merits of monolithic membrane cata- lysts are: durability and reliability, stability at high temperatures and during thermocycling in hydrogen; resistance to corrosion and mechanical damage, so practically excluding precious metal losses; and the reaction pro- ducts are readily separated, this being especially important for creating flexible and ecologically pure technologies.

A monolithic membrane catalyst, permeable only for hydrogen, is shown in Figure 4(a), which represents a cross-section through pal- ladium alloy foil sheet or tube wall.

The drawback of monolithic catalysts is the low ratio of their surface area to noble metals volume, which can be increased by roughen- ing one or both surfaces of the membrane catalyst, for example by thermal diffusion of a chemically active metal into the palladium alloy sheet and the subsequent removal of this metal by an acid treatment. The porous layers,

Figure 4(b), formed in this way are strongly bound to the palladium alloy sheet and are not dispersed during the reaction, unlike Raney catalysts. A certain part of the chemically ac- tive metal introduced into the palladium alloy remains after the acid treatment. Thus the membrane catalyst can be modified in a differ- ent way on both surfaces if a reaction coupling is to be performed.

A monolithic membrane catalyst can also be modified by introducing ultra-dispersed par- ticles of catalytically active metal or oxide into these porous layers, as is depicted schemati- cally in Figure 4(c).

Porous Membrane Catalysts Porous membrane catalysts are charac-

terised by higher, but less selective, gas permeability than monolithic catalysts. Sheets obtained from metal powder can serve as the matrix for such porous membrane catalysts. Some porous metals, such as activated porous nickel, are catalysts for hydrocarbon reac- tions. A porous membrane with catalyst particles distributed throughout its volume is

I MONOLITHIC POROUS COMPOSITE I

(b)

(C)

Raney type catalyst layers Dispersed catalyst in porous support

. . . . - ~ . . . ~ ~ . ~ ~ ~ . . ~ ~ ~ , ~

Metal a oxide particles

Pmtecting film Fallodium alloy film Intermediate !ayr

L&J

Intermediate layer CatdlySt 1

Catalyst (h) I- -, 11 Intermediate layer SYSTEMS OF GRANULAR AND MEMBRANE CATALYSTS

(i )

Cdtdlyst 11

Fs. 4 A hvther explanation of these ten types of catalyst-membrane systems is in the text

Plotinurn Metals Rev., 1992,36, (2) 73

presented in Figure 4(d). An example of such a system is a catalyst prepared by inserting into a porous stainless steel sheet 1.5 weight per cent of palladium in the form of ultra- d i spersed p o w d e r , p r o d u c e d b y t h e condensation of metal and toluene vapours, followed by melting of the resultant glassy solid. This catalyst gives a 25 times larger yield of linalool than the same weight of palla- dium used in the form of a palladium-6 per cent ruthenium foil, 50 pm thick, through which hydrogen diffuses during dehydrolina- loo1 hydrogenation in to linalool, at a temperature of 443 K and at atmospheric pressure. The monolithic membrane catalyst has higher selectivity (0.99) than the porous catalyst (0.90), although a selectivity of 0.95 can be achieved on the porous membrane cata- lyst by a decrease in space velocity.

The porous metallic membrane catalyst dif- fers from the conventional supported metal catalyst by having a higher mechanical sta- bility and better heat conductivity than most supports used for industrial catalysts. The porous metallic membrane also provides cer- tain selectivity for mass transfer.

To save catalytically active metal its par- ticles need only be introduced into the sub-surface layer of the porous membrane which may be metallic, oxide, ceramic or polymeric form, see Figure 4(e).

The advantages of porous and monolithic membrane catalysts can be happily combined in composite membrane catalysts.

Composite Membrane Catalysts Composites comprising a mechanically

stable porous support and a thin but con- t i nuous pa l lad ium alloy layer enable expenditure on platinum group metals to be decreased and hydrogen permeability to be increased. Such composites can be readily ob- tained, for example by diffusion welding a palladium alloy foil, 10 pm thick, to a porous metal sheet. An intermediate layer, Figure 4(f), is used to prevent the transfer of compo- nents between the support and the alloy, and also to augment adhesion.

A palladium film 20 pm thick, deposited by chemical reduction of a palladium salt onto the outer surface of the porous glass cylinder, with an average pore diameter of 300 nm (15), has been used at 673 K to remove hydrogen from the products of the water gas shift reac- tion over an iron-chromium oxide catalyst (16, 17) and for steam conversion of methane on an industrial nickel catalyst (18, 19).

A non-porous palladium layer of about 5 p m thick has been obtained by chemical plating onto a silver disc, with a pore diameter of 0.2 pm (20). This composite membrane was tested for hydrogen permeability at tempera- tures up to 680 K and appeared to be resistant to pressure drop as well as being stable to thermocycling in hydrogen.

Composite membrane catalysts can also be assembled with polymeric supports or inter- mediate layers (21-23). The use of polyarilyde has been proposed by investigators at this In- stitute, in order to widen the temperature range of their application (24). Polyarilyde is resistant in air up to 623 K, and its hydrogen from nitrogen separation factor is about 100.

Asymmetric membranes have been created at the A. V. Topchiev Institute of Petrochemi- cal Synthesis and at the Chemical Machine Building Institute. When such polyarilyde membranes are covered with a 1 p.m palladium alloy layer they possess higher permeability for hydrogen than non-metallised membranes at temperatures greater than 373 K, and are not permeable for other gases.

At even higher temperatures composite membranes of porous metals, oxides, ceramics and thin palladium or palladium alloy layers are used. At the Patrice Lumumba Peoples’ Friendship University a method of obtaining thin alloy films containing a low melting com- ponent has been developed (25). T h i s component, being liquid, coats the porous membrane, and after it has hardened other components are introduced by magnetron sputtering. The whole system is then annealed to make it homogeneous. A composite palla- dium-indium-ruthenium alloy layer was prepared by liquid indium distribution over a

Platinum Metals Rev., 1992,36, (2) 74

porous stainless steel sheet, or over a magnesia plate, with subsequent palladium-ruthenium alloy magnetron sputtering. This 2 pm thick three-component alloy layer only allows hy- drogen to permeate, at a speed of 10 m3/m2 per hour at 673 K and at a pressure drop of 0.2 MPa. Such a composite has retained all the above mentioned characteristics after 450 cycles of heating and cooling in hydrogen.

A system comprising a porous support, an intermediate layer and a film made of a selec- tively permeable and catalytically active material, may be covered by a polymer, see Figure 4(g), which allows the feedstock to per- meate but which keeps out catalyst poisons. A porous metallic sheet covered by a polysilox- ane layer was used to fix silica gel particles, on the surface of which palladium complexes were immobilised (26). Such a system proved to be active for the selective hydrogenation of cyclopentadiene into cyclopentene.

A composite catalysts for reaction coupling should comprise five layers, as shown in Fig- ure 4(h), consisting of a porous membrane with intermediate layers and films of catalysts I and I1 on its surfaces.

Granulated and Membrane Catalysts Systems

Of even more general purpose are the sys- tems containing a conventional granulated catalyst and a membrane catalyst. Figure 4 shows two varieties of such a system: a pel- let catalyst with a monolithic catalyst, 4(i), and with a composite membrane catalyst,

A system, consisting of an industrial oxide aluminochromium catalyst and a hydrogen permeable palladium alloy foil containing 20 weight per cent of silver, has been used for butane dehydrogenation (27). Part of the hy- drogen formed was removed through the membrane, with inert gas being fed along the other membrane surface at 673 K and atmos- pheric pressure, and this resulted in butene yields of 18 to 25 per cent and butadiene yields of 0.8 to 1.4 per cent.

The influence of butane and its dehydroge-

4W.

nation products on the hydrogen permeability of a series of binary palladium alloys has been investigated (6). Butane conversion increased when an air flow was used instead of an inert gas flow to remove the hydrogen which had penetrated through the membrane. A similar result was obtained for butene-1 dehydrogena- tion on an aluminochromium oxide catalyst (28), with hydrogen withdrawn through a 25 pm thick palladium foil, on the other side of which an argon mixture containing 10 per cent oxygen was fed. For example, at 658 K and a butene feed rate of 5 ml/min, oxidation of the hydrogen which had diffused through the membrane increased the butene conver- sion rate by a factor of three, compared with equilibrium conditions without hydrogen withdrawal (29).

Systems containing a pellet catalyst and a hydrogen permeable composite membrane, Figure 4(i), have been used to intensify pro- pane aromatisation (30) and steam conversion of methane (19). ZSM-5 zeolite, with gallium ions introduced by ion exchange, catalysed propane aromatisation. The zeolite was ar- ranged in the central part of a cylindrical reactor, which had a coaxial porous alumina tube, the outer surface being coated with a palladium film 8.6 pm thick. An equimolar propane-nitrogen mixture was fed into the space between the tubes, while the inner tube volume was pumped. To elucidate the func- tion of the membrane, the coated alumina tube was replaced by a non-permeable pyrex tube. The yield of aromatic hydrocarbons was 42 per cent with the pyrex tube, and rose to 76 per cent when about 90 per cent of the hydrogen formed was removed. This phe- nomenon was explained by hydrocracking and by enhancing propane dehydrogenation into propene, out of which a considerable amount of naphthalene was formed, when a gallium-containing ZSM-5 catalyst was used (31).

Steam conversion of methane was studied in a similar reactor over a nickel catalyst, with a tubular porous glass membrane (average pore diameter 300 nm) coated with a 20 pm thick

Platinum Metak Rev., 1992,36, (2) 75

palladium layer. Methane conversion reached 78 per cent at 773 K when the hydrogen dif- fusing through the membrane was removed by a flow of nitrogen, while equilibrium methane conversion without hydrogen withdrawal amounted to 44 per cent.

This classification of catalyst-membrane systems summarises and generalises the ver- sions suggested earlier (32,33).

Catalytic Membrane Reactors Membrane reactors are subdivided into ten

types according to the following criteria: cata- lyst application in reaction side and in separation side; membrane functioning only for separation or also as a catalyst; removal of the gas passing through the membrane as it is or by the introduction of a gas into the reac- tion (9).

A mathematical simulation of a catalytic membrane reactor and an inert membrane re- actor with catalyst pellets in the feed side showed a slight advantage of the former at not too small space velocities (34) which is in good agreement with experimental data for cyclo- hexane dehydrogenation into benzene. A porous glass thimble, with 0.34 weight per cent platinum introduced into its pores, served as the membrane catalyst. In this case the reaction took place in the membrane cata- lyst itself, see Figures 4(d) or 4(e). The membrane made of the same porous glass without any metallisation was catalytically inert, and had to be used with a layer of pla- tinised porous glass particles on it.

A porous membrane catalyst of the type shown in Figure 4(e) was obtained by intro- ducing 1 weight per cent platinum into the pores of a multi-layered tubular alumina membrane (MembraloxR, Alcoa). It provided a six-times greater yield of ethylene from ethane dehydrogenation than equilibrium without hydrogen withdrawal (35). Selectivity towards ethylene was more than 0.96. Ethane transfer through the porous membrane catalyst oc- curred a t a rate four-times slower than hydrogen transfer. A mathematical model of the reactor with this catalyst has been worked

out, and it represents, very satisfactorily, the experimental results when parameter adjust- ments have been made.

Mathematical models of a catalytic mem- brane reactor, an inert membrane reactor with catalyst pellets, a plug flow reactor packed with catalyst pellets, and also a mixed flow reactor have been created (36). It was found that for liquid phase reactions without any volume changes, the first two types of reactors are preferable at high space velocities. It is also true for an inert membrane reactor with catalyst pellets for reactions with a volume decrease, but it does not work as well as a catalytic membrane reactor at a low pressure drop on the membrane.

Investigations using a catalytic membrane reactor with a palladium membrane showed that when the dehydrogenated substance and the hydrogen carrier gas, coming through the membrane, flow in opposite directions the de- gree of conversion is the highest at a minimal reactor length, if compared with the regimes of co-current, ideal mixing in the reaction chamber, hydrogen collection chamber or in both of them (37). While these mathematical models were being created it was assumed that hydrogen transfer through a palladium mem- brane is described by the Sieverts law. It was proved experimentally, however, that in the process of butane dehydrogenation on palla- dium alloy foils the hydrogen permeability decreases, not because of the butane but as a result of the adsorption of butenes (6, 38). This effect is even more pronounced for palladium-antimony alloy than for ruthe- nium-palladium alloy. When the membrane ceased to be hydrogen permeable, due to strong adsorption of butenes, these butenes are then not subject to isomerisation or to hydrogenation in the hydrogen flow from the gaseous phase. If, however, hydrogen transfer through the membrane is only decreased by butene adsorption then only isomerisation of butenes occurs, and not hydrogenation. These results led to the assumption that in butene isomerisation and hydrogenation reactions hydrogen atoms entering the active centres of

Platinum Metals Rev., 1992,36, (2) 76

the membrane catalyst from its sub-surface layer are of great importance.

A study of the influence of electron ac- ceptors (S, CO) and electron donor (K) adsorption on the hydrogen permeability of palladium-silver alloy concluded that the species passing through the alloy is not H' (39). Hydrogen permeability decreased after contact with the ethylene, but in the process of ethylene hydrogenation by hydrogen, pas- sing through the same foil, a 10-fold increase in permeability was observed.

Another study has been performed to inves- tigate the effect of the directions of flow of the hydrogenated substance and of the hydrogen along adverse surfaces of the membrane cata- lyst on the hydrogen permeability and also on the depth of cyclopentadiene hydrogenation occurring on a palladium alloy containing 4 weight per cent indium (40). The hydrogen transfer rate to the hydrogenation chamber was higher with a counter flow of hydrogen- nitrogen mixture and cyclopentadiene vapour-nitrogen mixture than with co-current flow. Figure 5 shows the dependence of the degree of cyclopentadiene (CPD) conversion on VH*:VCpD feed velocity ratio, for counter- current (Curve 1) and co-current (Curve 2) flows. The amount of hydrogen transferred through the membrane catalyst during differ- ent flow directions was kept constant by changes in the composition of the hydrogen- nitrogen mixture.

During the coupling reactions of terpene alcohol: borneol dehydrogenation into cam- phor, on a copper catalyst, with cyclopentadiene

; 6 1.0

5

In w

0.5 w 3

s w

:T hydrogenation on palladium-ruthenium alloy foil, counter-current flow again proved to be more effective than co-current flow (41). As Figure 6 depicts, the cyclopentene concentra- t ion (Curve 1) in the cyclopentadiene hydrogenation products is much higher for counter-current flow and decreases more slowly during the course of the experiment than for co-current flow, see Figure 6(b). The camphor yield from borneol was 93 per cent. In the counter-current coupling of cyclohex- ane dehydrogenation and 1,3-pentadiene hydrogenation, discussed earlier and shown schematically in Figure 2, a 20 per cent in- crease in selectivity for pentenes was obtained, compared with using co-current conditions (42).

Structurally, a catalytic membrane reactor can resemble a block made up of corrugated

Fig. 6 Time dependence of the concentrations of cy- clopentene (Curve 1) and cyelopentane (Curve 2) are shown as products of cy- clopentadiene hydrogenation by hydrogen coming through the palladium-ruthenium during borneol dehydroge- nation on the other surface of the foil: (a) counter-cur- rent and (b) co-current flows

cyclopentadiene

(a )

cyclopentadoene introduction

20 2 -

20 4 0 60 TIME, minutes "I 20 40

Platinum Metak Rev., 1992,36, ( 2 ) 77

Fig. 5 The conversion of cyclopentadiene is dependent on the ratio of hydrogen: cyclopentadiene in the feed velocities for both counter-current (Curve 1) and co- current (Curve 2) flows

and plane sheets obtained by roasting mix- tures of inorganic oxide powders, binders and plasticisers. Having parallel orientation, each pair of corrugated sheets is separated by a plane sheet and thus parallel channels are formed. Every second corrugated sheet is turned through 90” with respect to the first, thus two similar channel systems will be per- pendicular to each other (43). Such blocks can be manufactured out of porous metal sheets, as mentioned earlier. Blocks with perpendicular channel systems are convenient for separating the flows along both sides of the membrane catalyst.

The catalyst material may also be deposited on alumina multi-channel membrane ele- men ts to produce catalytic membrane reactors (44). Methods of depositing platinum group metals on solid supports of different composi- tion, as well as on porous ones, have been investigated at the Patrice Lumumba Peoples’ Friendship University and at the A. V. Top- chiev Institute of Petrochemical Synthesis of the Russian Academy of Sciences.

The palladium based monolithic membrane catalysts proposed by the above named in- stitutions, in collaboration with A. A. Baikov Ins t i tu te of Metallurgy of the Russian Academy of Sciences, could be successfully combined with conventional catalysts for the dehydrogenation or steam conversion of hy- drocarbons. The apparatus for extracting pure hydrogen by diffusion (45), and hydrogen generators incorporating catalytic cracking

and silver-palladium diffusion units (46), could also be equipped with the above men- tioned palladium alloy membranes thus giving high hydrogen permeability for the products of hydrocarbon transformations.

Five years ago membrane reactor catalysts, along with two other types of catalyst, were said to have high potential to make a major impact on new catalyst technology for the fu- t u re (47). Clearly the new data about catalyst-membrane systems fully justifies those expectations.

Conclusions Combining catalysts and membranes has

become a most promising method for improv- ing existing technologies. It seems likely that the suggested classification of catalyst-mem- brane systems will be helpful in further investigations.

Several types of catalytic membrane reactor have now been studied both theoretically a n d experimentally. T h e results have proved to be very encouraging, especially for thermodynamically complicated reac- tions. Palladium alloys which maintain their hydrogen permeability when in con- tact with dehydrogenation products have been found. Such membranes are good for use in combination with hydrocarbons dehy- drogenation catalysts and can be used as catalysts for the oxidation of the removed hy- d r o g e n , t h u s c o m p e n s a t i n g f o r t h e endothermicity of dehydrogenation.

References 1 V. M. Gryaznov, Dokl. Akad. Nauk SSSR, 1969,

2 V. M. Gryaznov, U.S.S.R. Authors Certif. No.

3 W. C. Pfefferle, U.S. Patent 3,290,406; 1966 4 V. M. Gryaznuv, V. S. Smirnov, L. K. Ivanova and

A. I! Mischenko, Dokl. Akad. Nauk SSSR, 1971, 190,144

5 V. S. Smirnov, V. M. Gryaznov, N. V. Orekhova, M. M. Ermilova and A. I! Mischenko, Dokl. Akad. Nauk SSSR, 1975,224,391

6 N. V. Orekhova, and N. A. Makhota, in “Mem- brane Catalysts Permeable for Hydrogen or Oxygen” (in Russian), Moscow, Nauka, 1985,49

7 N. N. Mikhalenko, E. V. Khrapova and V. M. Gry- aznov, Kinet. Katal., 1986,27, 138

189,794

27,4092; 1969

8 N. Itoh, K. Miura, T. Shindo, K. Haraya, K. Obata and K. Wakabayashi, Sekiyu Gakkaishi, 1989,32, (I), 47

9 N. Itoh, Sekiyu Gakkaishi, 1990,31, (3), 136 10 “Membrane Catalysts Permeable for Hydrogen or

11 V. M. Gryaznov, Pkuinum Metah Rev., 1986,30,

12 J. N. Armor, Appl. Catal., 1989,49, (l), 1 13 A. N. Karavanov, in Abstracts of the Papers of 25th

Scientific Conference of the Department of Physics-Mathematical and Natural Sciences (in Russian), Moscow, Peoples’ Friendship University Publishing, 1989,75

14 V. I. Lebedeva, ibid., 80

Oxygen” (in Russian), Moscow, Nauka, 1985

(2)s 68

Plaiinum Metals Rev., 1992,36, (2) 78

15 S. Uemiya, Y. Kude, K. Sugino, N. Sato, T. Mat- suda and E. Kikuchi, c h . Len., 1988,1687

16 E. Kikuchi, S. Uemiya, N. Sato, H. Inoue, H. Ando andT. Matsuda, Chem. Ltr., 1989,489

17 S. Uemiya, N. Sato, H. Ando and E. Kikuchi, Znd. Eng. Chem. Res., 1991,30,585

18 S . Uemiya, N. Sato, H. Ando, T. Matsuda and E. Kikuchi, Sekiyu Gakkahhi, 1990,33,418

19 S . Uemiya, N. Sato, H. Ando, T. Matsuda and E. Kikuchi,Appl. Catal., 1991,67,223

20 K. Govind and D. Atnoor, Znd. Eng. Chem. Res., 1991,30,391

21 V. M. Gryaznov, V. S. Smirnw, V. M. Vdovin, M. M. Ermilova, L. D. Gogua, N. A. Pritula and I. A. Litvinov, BritishPutent 1,528,710; 1978

22 R. V. Bucur, and V. Mercea, Surf: Coating Technol., 1986,28,387

23 E! Mercea, L. Murian, V. Mercea, and D. SilipasJ Membrane Sci., 1988,35,19

24 M. M. Ermilova and S. I. Zavodchenko, in “Mem- brane Catalysts Permeable for Hydrogen or Oxygen” (in Russian), MOSCOW, Nauka, 1985,33

25 V. A. Nesmeyanm, S. I. Zavodchenko, 0. S. Sere- bryannikova, Yu. M. Serov and V. M. Gryaznw, in Abstracts of the Papers of 25th Scientific Con- ference of the w e n t of Physics-Mathematical and Natural Sciences, (in Russian), Moscow, Peo- ples’ Friendship University Publishing, 1989,77

26 V. M. Gryaznov, V. S. Smirnw, V M. Vdovin, M. M. Ermilova, L. D. Gogua, N. A. Pritula and G. K. Fedorova, US. Patent 4,394,294; 1983

27 N. V. Orekhova and N. A. Makhota, in “Metals and Alloys as Membrane Catalysts” (in Russian), Mos- cow, Nauka, 1981,168

28 R. Zhao, R. Govind and N. Itoh, Sep. Sci. Technol., 1990,25, (13-15), 1473

29 R. Zhao, N. Itoh and R.Govind, in ”Novel Ma- terials in Heterogeneous Catalysis”, Am. Chem. SOC. SR., 1990,Vol. 437, p. 216

30 S. Uemiya, T. Matsuda and E. Kikuchi, Chem. Len., 1990,1335

31 M. Shibata, H. Kitagawa, Y. Sendoda and Y. Ono, Stud. Surf: Sci. Catal., 1986,28,717

32 V. M. Gryaznw, in “Metals and Alloys as Mem- brane Catalysts” (in Russian), Moscow, Nauka, 1981,4

33 V. M. GryaznovJ D. Z. M e d b e v All-Union Chem. SOC. (in Russian), 1989, (6), 604

34 Y. M. Sun and S. J. Khang, Znd. Eng. Chem. Res., 1988,27,1136

35 A. M. Champagnie, T. T. Tsotsis, R. G. Minet and I. A. Webster, Chem. Eng. Sci., 1990,45, (8), 2423

36 Y. M. Sun and S. J. Khang, Znd. Eng. Chem. Res., 1990,29,232

37 N. Itoh, Y. Shindo and K. Haraya,J. Chem. Eng. Jpn., 1990,23, (4), 420

38 V. M. Gryaznw, M. M. Ermilova, N. V. Orekhova and N. A. Makhota, Proc. 5th Int. Symp. on Hete- rogeneous Catalysis, Varna, Bulgaria, 1983, Vol. 1, 225

39 S. B. Ziemecki, in Abstracts 1990 Spring Meeting of Material Research SOC., San Francisco, 207

40 N. N. Mikhalenko, E. V. Khrapova and V M. Gry- aznov, Russ. J. Phys. C h . , 1986, (2), 5 1 1

41 V. M. Gryaznov, M. M. Ermilova, L. S. Morozova, N. V. Orekhova, V. f! Polyakova, N. R. Roshan, E. M. Savitsky and N. I. Parfenova, J. Lss-Common Met., 1983,89,529

42 N. V. Orekhova, M. M. Ermilova and V. M. Gryaz- nov, Dokl. A M . Nauk SSSR, 1991,321, (l), 141

43 R. De Voss, V. Hatziantoniou and N. H. Schoon, Chem. Eng. Sci., 1982,37, (ll), 1719

44 H. E! Hsieh, R. R. Bhave and H. L. Fleming,J. Membrane Sci., 1988,39,221

45 J. E. Philpott,PhtinumMetaLFRev., 1985,29, (l), 12 46 J. E. Philpott,PhtinumMetaIs Rev., 1989,33, (2), 58 47 J. E Roth, in “Catalysis 1987”, Proc. of the 10th

North American Meeting of the Catalysis Soc., San Diego, U.S.A. May 17-22, 1987, ed. J. W. Ward, Elsevier, Amsterdam, 1988, p. 925

Platinum-Iridium Carbon Monoxide Sensor Various heat of oxidation and doped metal

oxides types of catalytic sensors have been used in gas detectors, but in general they suffer from interference caused by water vapour. These changes in humidity can produce spurious sig- nals which have in the past been overcome by the use of high power heaters.

In order to solve the humidity-effect problem that occurs with catalytic carbon monoxide sen- sors and to eliminate the requirement for heaters, researchers at the Chalk River Labora- tories of AECL Research, in co-operation with Asahi Electronics Inc., Ontario, Canada, have developed and tested several new bimetallic platinum group metal catalysts, (K. Marcin- kowska, M. €! McGauley and E. A. Symons,

Sens. Acrivafurs B, 1991,5, (1-4), 91-96). The optimised catalyst contained a total of 10

weight per cent of platinum and iridium which was supported on porous, inertly hydrophobic polystyrene-divinylbenzene granules contained in nylon mesh thimbles.

This new carbon monoxide sensor was found to be independent of humidity and even after testing for 10 months, no affect on the carbon monoxide oxidation activity of the catalyst was detected despite exposure to carbon monoxide concentrations of up to 250 ppm.

The sensor requires no heater as the catalyst is active at ambient temperatures down to around -10°C. This has facilitated production of a portable, battery-powered detector.

Platinum Metals Rev., 1992,36, (2) 79

Solvated Atoms of Platinum, Palladium and Gold PRECURSORS TO COLLOIDS, FILMS AND CATALYSTS

By Professor Kenneth J. Klabunde Department of Chemistry, Kansas State University, U.S.A.

Atoms of the noble and other metals can be trapped in cold solvents, and solvated metal atom solutions can be prepared and manipu- lated at low temperatures. Such solvated atoms have been useful in: (a) the preparation of non-aqueous colloidal metal solutions which, in turn, can be used to prepare metallic films of platinum, palladium and gold; (b) preparing ultra-fine bimetallic powders of gold-tin; (c) trapping platinum-tin fine par- ticles on a lumina in order to prepare bimetallic PtoSno heterogeneous catalysts. Ultra-fine particles of metals are usually pre- pared by high temperature metal salt reduction methods (1-3). Under such conditions, the approach to the most thermo-dynamically stable state has often moved further than is desirable. Very small, nano-scale particles and kinetically stable phases are often not attain- able by such reduction techniques. This is particularly true of bimetallic combinations; meta-stable bimetallic particles either revert to the most thermodynamically stable state or may phase-separate under high temperature reducing conditions. In order to prepare meta-stable states or possibly new phases of nano-scale metal particles, low temperature, kinetic growth methods should be employed (4). Zero-valent atoms should be used rather than salts or oxides since reduction steps can thus be avoided. Actually, in recent years we have witnessed the development of several methods for the low temperature, kinetically controlled growth of metal clusters from free atoms. These include the gas phase “cluster beam” approach (metal clusters, Buckyballs, etc.) and the clustering of metal atoms in low temperature matrices. In order to carry out

cluster syntheses on a large scale with rela- tively low expense, we have studied such cluster growth in low temperature organic sol- vents (5). I n fact this method has been practiced for about 20 years and serves as a forerunner of other clustering methods (6). Kinetic control of cluster growth can be re- alised, and the unique structure/reactivity of such materials has been demonstrated many times (4). Magnetic properties have also been studied (7).

In this method, metal atoms are first trapped in frozen solvents by codepositing the evaporated atoms with excess solvents at a temperature of 77 K. Upon warming, a liquid of “solvated metal atoms” is formed, often stable in the 180 to 250 K range. Further warming leads to atom agglomeration to form particles 2 to 9 nm in size, but further growth is precluded by particle solvation. Under the right conditions, stable colloidal metal solutions are formed; such as palladium particles in acetone (8).

In the presence of a catalyst support, metal atom nucleation and cluster growth occurs on the surface of the support; an example being very small platinum clusters on alumina. In this way “solvated metal atom dispersed” (SMAD) catalysts have been prepared (5 ) .

We describe recent findings of interest to the users of platinum, palladium and gold.

Non-Aqueous Palladium and Gold Colloids

The codeposition of palladium or gold atoms with acetone followed by slow warming leads to non-aqueous metal colloidal solutions. In the case of palladium, black solutions result whereas with gold, dark red-purple liquids are

Platinum Metals Rev., 1992,36, (2), 80-84 80

formed (8-1 1). The individual colloidal par- ticles are 4 to 9 nm in size and are indefinitely stable under the correct conditions of concen- tration and solvent polarity.

Polar, organic solvents can be employed, such as acetone, ethanol, isopropanol, di- methylsulphoxide and dimethylformamide

Colloid stability is apparently feasible due to the solvation/ligation effects of the solvent (steric effects), combined with electronic ef- fects possible because of electron scavenging to form negatively charged colloidal particles (8, l l) , as shown by electrophoresis studies.

Perhaps the most remarkable feature of these colloidal particles is their “living na- ture”. These particles will grow if their solvating medium is perturbed. Solvent remo- val leads to metallic films. Thus, these solutions can be “spray painted” onto sur- faces. The solvent evaporates leaving films that look like palladium or gold films. The films do retain some of the organic solvent, however, which poorly affects their conducti- vities. Heating tends to drive out the organic impurities, and the films become smoother and better conductors (12, 13).

The best results have been obtained when surfaces that interact with the metal particles are coated. For example palladium on stain- less steel, gold on copper, gold on silver, or better still, gold on polyphenylenesulphide polymer. In this case, the sulphur containing surface “ligated” well with the depositing gold particles, and upon heating a perfect, strongly adhering gold film was formed (13).

(8-10).

Bimetallic Fine Particles of AuSn as a Model of PdSn and PtSn

In order to learn more about the atom-atom clustering process in solvated metal atom media, two metals were simultaneously iso- lated in excess cold solvents, and clustering was allowed to occur on warming. The metal pair chosen was goldtin because these two metals form several well characterised inter- metallic compounds throughout their entire compositional range. If selectivity in growth

Aubolv). Sn(s01v) - AuSn(sMv1 (I1 A d - 1 ~ ) . AuSnkolv) - Au2Snl(solv) tni AulSn~(solv)~ AuSntsolv) - Au,Sn,(solv) trnl AunSno(sorv). solv - AunSnntR&lsMv> ( IV)

5tilQh5ed toward further growth

Fig. 1 A generalisation of the rate pro- cesses for cluster growth and cluster reac- tions with the host solvent, where: solv=solvent molecule and Rzfragment of solvent that serves as a &and. Mobility de- creases with size and therefore the rate of reaction (111) decreases; mobility is low at high viscosities, therefore reaction (IV) competes more effectively with (111)

was to be found, this bimetal system seemed to lend itself best to these experiments.

Experimental parameters such as evapora- tion rate and evaporation method, solvent polarity and viscosity, and warming rate dur- i ng cluster formation were varied, and cluster/crystallite sizes were monitored. Addi- t i ona l i n fo rma t ion was g leaned f rom Mossbauer spectrometry, Differential Scan- ning Calorimetry (DSC), X-Ray Powder Diffraction (XRD) and X-Ray Photoelectron Spectroscopy (XPS) (9).

The results demonstrated that the cluster growth process was somewhat selective toward the growth of clusters of AuSn, Au,Sn and tin metal. Solvent viscosity had an effect only in cases where large viscosity changes occurred in a narrow temperature range during which cluster growth occurred (about 150-200 K). A more sensitive parameter was solvent polarity - highly polar ethanol allowed the crystallites to grow larger, to 27 nm. The most sensitive parameter, however, was the rate of matrix warming. A slow warm-up yielded smaller particles with higher surface areas, results are given in the Table. These findings can be ex- plained by considering a competition between cluster growth and the reaction of the growing cluster with the host solvent, so that ligand stabilisation (solvation) occurs and stops fur- ther growth. In Figure l this is expressed as a competition between Reactions (IV) and (111).

Overall these results are promising with re- gard to the possibility of controlling the

Platinum Metals Re%, 1992,36, (2) 81

Solvent Properties Compared with Surface Areas and Crystallite Sizes of Resultant AuSn Powders

Solvent

Pentane

Acetone

Toluene

Ethanol

Cyclo- hexane

Ether

Hexane

Melting point, “C

-1 30

-95

-95

-1 17

6.5

-1 16

-95

Dielectric constant, Ee

1.8

20.7

2.4

24.3

2.0

4.3 (20)

1.9 (20)

Viscosity”

0.289

0.399

0.772

1.733

1.02 (20)

0.284

0.401

a Viscosities are at 0°C unless indicated otherwise in parentheses b Fast warm-up: -196°C to 25°C in -0.5 hours

Slow warm-up: -196°C to 25°C in 3-4 hours c BET method using nitrogen adsorption d From XRD data using Scherrer equation

selectivity of cluster growth to certain bime- tallic compositions while still maintaining a small particle size. This is demonstrated most clearly by considering Figure 2, which shows Differential Scanning Calorimetry spectra of AuSn clusters grown in acetone by fast or slow warm-up. As shown, the endothermic (up- ward) peaks are sharp, indicating the melting points of small particles of specific composi- tions (tin -219”C, eutectic mix of Au,Sn and AuSn -275”C, and AuSn -419°C). Note that more selective growth to A u k , Au,Sn and tin is evident in the slow warm-up case (b), which

Narm-upb

fast

slow

fast

slow

fast

slow

fast

slow

fast

slow

fast

slow

fast

slow

Surface area‘, m’/g

16.5

23.8

14.9

18.3

17.6

43.7

11.6

16.5

19.4

28.0

16.6

-

23.4

19.8

:rystal sized, nm

16.3

10.2

17.8

11.2

17.6

17.1

26.9

23.6

13.1

12.7

17.8

9.3

13.1

12.5

is good evidence for selective growth. In the fast warm-up case (a), it is obvious that a “wilder” growth process took place leading to more components (10).

PtSn/Al,O, SMAD Catalysts The commercial importance of the Pt-

Sn/AI,O, catalyst systems, as well as the poor understanding of the role of tin, led us to investigate this bimetallic system in some de- tail. These solvated metal atom dispersed (SMAD) catalysts were prepared in two ways. The “half S M A D process refers to treating

Platinum Metals Rev., 1992,36, (2) 82

L

t w I

preformed, conventional Pt/Al,O, catalysts with solvated tin atoms, thereby ensuring the deposition of metallic tin on the platinum metal cluster. The "full SMAD" process refers to the evaporation/trappingolvation of plati- num and tin atoms simultaneously, followed by warming, bimetallic cluster growth and trapping on high surface area alumina (5,14).

Two conventionally prepared Pt-SnO/Al,O, catalysts were also prepared and compared with SMAD catalysts.

The unique feature of the SMAD catalysts is that Sn" is present, and the study and com- p a r i s o n of t hese Pt'Sn" sys t ems w i t h conventional PtoSnz+"+ systems should be of help in quelling the debate about whether Sno plays a role in the commercial catalysts.

Through the combined use of catalytic

,'

/' ,'

SCCOnd run \

I

probe reactions, IL9Sn Mossbauer, Extended X-Ray Absorption Fine Structure, XPS and XRD it has been demonstrated that the half- SMAD catalytic particles are made up of the expected platinum particles with a partial thin Sn" coating. Interestingly, this type of catalyst showed the highest activity for n-heptane re- forming to benzene and toluene. On the other hand, the full-SMAD catalyst

particles were shown to be alloy-like, probably rich in Pt" on the outer surface of the catalyst layer, Figure 3. This type of catalyst showed lower activity but better selectivity to benzene and toluene. Undesirable hydrogenolysis reac- tions were greatly depressed, and this is perhaps due to an ensemble effect (Sn" dilut- ing surface Pt") decreasing these unwanted, surface sensitive reactions.

Platinum Metals Rev., 1992,36, ( 2 ) 83

Fv. 2 Differential scanning calorimetry spectra of AuSn particles prepared in acetone by (a) fast warm-up and (b) slow warm-up

n Sn

. S”(sol”)n (rero valent tin atom)

I I l l I I l l

Alumina

~t atom 0 sn atom

Fig. 3 Proposed sequence fo r the formation of full- SMAD Pt-Sn bimetallic par- ticles on alumina (Sn:Pt atomic ratio of 2.5 is illus- trated)

These results also demonstrate that zero- valent tin does affect catalytic performance in beneficial ways. So, although zero-valent tin is rarely detected in conventional Pt-Sn/Al,O, catalysts, small amounts possibly formed on the platinum particles (by hydrogen reduction of Sn2+) may be at least partially responsible for beneficial changes in this important class of bimetallic catalysts.

A comparison of catalytic properties for the SMAD and conventional catalyst systems sug- gested some generalisations: (i) the presence of SnO, and SnO in the con- ventional catalysts may play a role in improving lifetimehtability by blocking plati- num particle sintering; (ii) the presence of Sno in combination with

Pto can affect catalyst activity and selectivity; (iii) the presence of Pto-Sno alloy (rich in Sn’) can depress unwanted hydrogenolysis, while activity for the desired dehydrocyclisation is only lowered slightly (5, 14).

I n summary, the use of metal vapour methods, especially solvated metal atoms, shows promise for producing ultra-fine mono- metallic and bimetallic particles for many interesting applications, including catalysts, magnetic materials, colloids and films.

Acknowledgements The author would like to acknowledge the work of

many fine students, in particular Dr. Y. X. Li, Yi Wang, Ellis Zuckerman and Greg Youngers. The sup- port of the National Science Foundation, for nearly 20 years now, is also deeply appreciated.

References 1 V Haensel and R. Bunvell, Sci. Am., 1971,225,46 2 J. R. Anderson, “Structure of Metallic Catalysts”,

Academic Press, New York, 1975 3 R. D. Srivastava, “Heterogeneous Catalytic

Science”, CRC Press, Boca Raton, Florida, 1988 4 K. J. Klabunde, G. H. Jeong and A. W. Olsen, in

“Molecular Structures and Energetics”, ed. J. A. Davies, I! L. Watson, J. E Liebman and A. Green- berg, VCH, NewYork, 1990, pp. 433463

5 K. J. Klabunde, Y.-X. Li and B.-J. Tan, Chem. Muter., 1991,3, (l), 30

6 K. J. Klabunde, “Chemistry of Free Atoms and Particles,” Academic Press, NewYork, 1980

7 C. E Kemizan, K. J. Klabunde, C. M. Sorensen and G. C. Hadjipanayis, C k Muter., 1990,2, (l), 70

8 G. Cardenas-Trivino, K. J. Klabunde and B. E. Dale, Langmuir, 1987,3, (6), 986

9 S.-T Lin, M. T Franklin and K. J. Klabunde, Lnngmuir, 1986,2, (2), 259

10 Y. Wang, Y.-X. Li and K. J. Klabunde, “Selectivity in Catalysis: Clusters, Alloys, and Poisoning”, ACS Symp. Ser., ed. S. Suib and M. Davis, in press, 1992

11 M. T. Franklin and K. J. Klabunde, “High Energy Methods in Organometallic Chemistry”, ACS Symp. Ser. 333, ed. K. Suslick, 1987, pp. 246-259

12 G. Cardenas-Trivino, K. J. Klabunde and B. E. Dale, “Thin Metallic Films from Solvated Metal Atoms”, SPIE Proc. (Opt. Eng. paper 821-29 (8 pages), 1987, pp. 206-213

13 K. J. Klabunde, G. Youngers, E. Zuckerman, B.-J. Tan, S. Antrim and I! M. A. Sherwood, invited paper for Ew.3 Solidstate Inorg. Chem., submitted

14 Y.-X. Li and K. J. KlabundeJ Caral., 1990, Us, 173

Platinum Metals Rev., 1992,36, ( 2 ) 84

Platinum Metals Catalyst Studies Catalysis Volume 9: A Specialist Periodical Report EDITED BY J. J. SPIVEY, Royal Society of Chemistty, Cambridge, 1992,279 pages, ISBN 0-85186-604-2, Ji97.50

Three of the five chapters that make up this report will help to increase the understanding of catalyst deactivation and regeneration mech- anisms in the industrially significant areas of naphtha reforming and pollution control, where platinum metals catalyst systems con- tinue to find important applications. All catalysts become deactivated during use, and thus there is a growing interest in gaining a much greater understanding of the reactions involved.

The deactivation and regeneration of naph- tha reforming catalysts is described by J. M. Parera and N. S. Figoli of INCAPE, Santa Fey Argentina. The Universal Oil Products “plat- forming” catalyst, introduced in 1949, based on a bimetal-acid catalyst, for example, platinum- rhenium-sulphur/alumina, is more selective and more stable than platinum/alumina, and is currently the most widely used com- mercially. Catalyst deactivation is discussed in terms of coke deposition, poisoning by sulphur and nitrogen compounds, decrease of metallic and support areas and chloride concentration, heavy metal deposits, and fines formation and deposition. All these characteristics are revers- ible except the decrease in support area and heavy metals deposition. With the most com- monly used commercial, naphtha-reforming catalysts, the main reactions of the process are controlled by the acid function of the catalyst, in spite of the great initial deactivation of the metallic function. These catalysts have plati- num contents of about 0.3 per cent and the activity of the metal in the operating conditions used is enough to produce all the olefins that can be isomerised or dehydrocyclised on the acid function. The steps used for catalyst regeneration are also indicated, including coke elimination by controlled burning, oxy- chlorination to redisperse the metal function and restore the acid function, reduction

with hydrogen, and passivation by sulphiding. In the chapter by D. B. Dadyburjor of West

Virginia University, Morganstown, U.S.A., the effects of deactivation in changing catalyst se- lectivity are examined in a number of hydrocarbon reactions including those in- volving p la t inum-rhenium-sulphur on alumina. The overall conclusions from this chapter are that the various modes of deacti- vation including coking, poisoning and sintering, all change the selectivities of the different reactions.

The deactivation of stationary source air emission control catalysts is reviewed by J. R. Kittrell, J. W. Eldridge and W. C. Conner of the University of Massachusetts and KSE Inc., Amherst, Massachusetts, U.S.A. Supported platinum catalysts feature strongly among the favoured candidates both for the oxidation of volatile organic compounds to carbon dioxide and water, and for the reduction of nitrogen oxides to nitrogen and water, where ammonia may be used as the reducing agent (selective catalytic reduction). The authors review the morphological changes which take place during catalyst deactivation, deposition of poisons on the active surface, reactions between the feed and active catalyst sites, and solid-state trans- formations of the catalyst to form inactive solids. The mobility of platinum on oxide sup- ports is well known; in reducing environments at higher temperatures the platinum particles grow to raft-like structures and eventually to large three dimensional particles. In oxidising environments, the platinum is also mobile on the surface as PtO,, where x < 1, but the inter- actions wi th the suppor t can be more favourable as the PtO, wets the surface. Conse- quently a platinum on alumina catalyst which has become deactivated in a reducing environ- ment can be redispersed in an oxygen-rich environment. D.T.T.

Platinum Metals Rev., 1992,36, (2), 85 85

Advances and Developments in Emissions Control A REVIEW OF THE 1992 SAE INTERNATIONAL CONGRESS

The Society of Automotive Engineers Inter- national Congress, held in Detroit, Michigan, U.S.A. from 24th to 27th February 1992, con- tinues to be the worldwide forum for the presentation and discussion of matters relating to vehicle emissions control. Despite the re- cession the sessions were very well attended, and papers were presented by speakers from around the globe.

Broadly speaking, the contributions can be divided into two main categories, gasoline- and diesel-related. A recurrent theme throughout was the emphasis upon the fact that noble metal catalysts, particularly in gasoline applications, are very much part of a control system involv- ing engine management strategies and other engine components.

Catalyst Design for Diesel Applications

Noble metal catalysts for use on diesel en- gines were the subject of several papers, with interest in the U.S.A. focused on heavy duty engine applications.

The optimisation of noble metal formula- tions for diesel catalysts which could exhibit good control of the volatile organic fraction of the particulate matter, and also limit the forma- tion of sulphates which can otherwise increase particulate emissions was reviewed by M. G. Henk, W. B. Williamson and R. G. Silver of Allied Signal Inc. (SAE 920368). Platinum, pal- ladium and rhodium can all cause the formation of sulphate but it was demonstrated that palladium systems can be optimised to limit this undesirable effect.

The role of a noble metal flow-through oxi- dation catalyst as part of a strategy for the development of a low emissions specification heavy duty engine was considered by E Brear of Perkins Technology Ltd., and S. Fredholm and

E. Anderson of Svenska Emissionsteknik AB (SAE 920367). The problem of sulphate gener- ation over noble metal catalysts was again highlighted. The authors conclude that while it is possible to modify the platinum content and thus its interaction with other catalyst compo- nents, in order to reduce sulphate emissions, the solution may have to include a substantial reduction in the sulphur content of diesel fuel. This conclusion was reinforced in a paper by R. J. Farrauto and J. J. Mooney of Engelhard Cor- poration (SAE 920557).

Autocatalyst Systems Two papers in particular provoked excite-

ment and debate. Both considered the application of noble metal autocatalysts as part of advanced development systems designed to speed-up activation of the catalyst when cold. It is generally recognised that the major portion of both hydrocarbon and carbon monoxide emissions are produced within the first two minutes of the vehicle drive cycle, that is while the catalyst is cold. One potential control strategy involves the use of electrically heated catalysts and these have been the subject of papers at previous Congresses.

A more radical approach to the cold-start emissions problem was described by T. Ma of Ford Motor Co., N. Collings of Cambridge University and T. Hands of Combustion Ltd., (SAE 920400). Substantial reductions in cold- start emissions have been demonstrated by a strategy which causes the engine to run rich, together with air-injection and the use of electrodes - positioned in front of the conven- tional noble metal catalyst - for the first few seconds of vehicle operation.

L. S. Soucha, Jr. and D. E Thompson of Corning Inc., detailed their investigations (SAE 920093). They concluded that future

Platinum Metals Rev., 1992,36, (2), 86-89 86

emissions standards legislation could be achieved with an electrically heated platinum- rhodium-containing converter, using an extruded metal substrate in conjunction with a conventional noble metal catalyst.

A paper by W. A. Whittenberger and D. T. Sheller of Camet Co., and J. Walters of Gordon- Darby Inc., gave experiences of user vehicles equipped with electrical heated catalysts (SAE 920722). It indicated a number of areas where improved technology might make such equip- ment more viable; these included battery and power control technologies.

Catalyst Design for Gasoline Vehicles

It was reported by J. C. Dettling and Y. K. Lui of Engelhard Corporation that platinum-palla- dium catalysts could be made to perform more like rhodium-containing catalysts for the con- trol of oxides of nitrogen, than had previously been shown (SAE 920094). The platinum-palla- dium systems described are 85 per cent as effective as a platinum-rhodium system for the control of nitrogen oxides emissions on a ve- hicle, at reduced overall noble metal loadings. The authors conclude that it may be possible to meet forthcoming emission standards without increasing the usage of rhodium per vehicle.

The utilisation of palladium, this time as a substitute for platinum, was reviewed by C. N. Montreuil, S. C. Williams and A. A. Adamczyk of Ford Motor Co., as part of a programme to generate an experimental data base of catalyst conversion efficiency (SAE 920096). Platinum- rhodium and palladium-rhodium catalysts at equivalent concentrations and ratios were examined in a tubular flow reactor. Steady state conversion efficiencies for carbon monoxide, nitric oxide, propane, propylene, hydrogen and oxygen through the catalysts were determined for a variety of inlet species, concentrations and inlet gas temperatures. The results of these ex- periments show significant improvements in carbon monoxide and nitric oxide conversion eficiencies for both of these catalyst systems compared with previous generation catalyst for- mulations, when the feed-gas stoichiometric

ratio was on the rich side. The conversion efi- ciencies obtained with the platinum-rhodium formulation were similar to those obtained with the palladium-rhodium formulation over a wide range of conditions. Differences were noted, however, at low temperatures or when a high concentration of lo^" burning hydro- carbons (propane) was present.

The subject of noble metal cost optimisation was addressed by M. A. H&konen and €! Tal- vitie of Kemira Oy (SAE 920395). Various dual bed catalysts containing different platinum- rhodium, palladium-rhodium, platinum and palladium loadings, and combinations thereof, were subjected to oven and bench engine ageing techniques, and then examined for perfor- mance. It was found that catalyst performance is not necessarily proportional to the cost of the noble metals used, and that although palladium and palladium-rhodium catalysts have good thermal ageing resistance they are more sensi- tive to the presence of poisons than either platinum or platinum-rhodium. Also, the addi- tion of ceria to a three-way catalyst is more beneficial for platinum-rhodium, platinum and rhodium systems than for palladium-rhodium catalysts.

A prominent topic emerging from the papers dealing with three-way catalysts was catalyst deactivation modes, and their effects on emissions. A number of these papers dealt with thermal deactivation and the role of air to fuel ratio.

A paper by N. A. Hannington, R. J. Brisley and R. D. O’Sullivan of Johnson Matthey de- scribed the effects of catalyst ageing under different temperature and air to fuel ratio con- ditions, and compared the emissions of these catalysts to those from catalysts aged on ve- hicles driven under European conditions (SAE 920399). It was concluded that bench engine ageing which gave catalyst inlet temperatures in excess of 85OoC, with significant amounts of lean running, most closely simulated “real life” European ageing on vehicles and that these high temperature, lean ageing conditions resulted in greater deterioration of platinum- rhodium catalysts.

Platinum Metals Rev., 1992,36, (2) 87

Information on the effect of oxygen concen- tration on the ageing of three-way catalysts using inlet temperatures in excess of 850°C was presented by R. M. Heck, J. K. Hochmuth and J.C. Dettling of Engelhard Corporation (SAE 920098). Higher catalyst temperatures are seen as the converter is moved closer to the engine to reduce cold start emissions. This, in combina- tion with lean conditions resulting from a fuel cut-off strategy on deceleration, may give the high temperature, lean conditions de- scribed in the paper. I t was found that catalyst deactivation increased with higher temperatures and with increasing oxygen concentration.

High catalyst temperatures leading to deacti- vation may also be experienced during ignition-induced misfire. This topic was de- s c r i b e d by C . D. Tyree o f t h e U.S. Environmental Protection Agency (SAE 920298). Vehicles were run over the Federal Test Procedure cycle with varying degrees of ignition-induced misfire. It was found that hy- drocarbon emissions were increased the most by this type of misfire, whereas much higher misfire rates were required to increase carbon monoxide tailpipe levels to the same extent. Nitrogen oxides emissions were reduced as the rate of misfire increased. Catalyst temperatures could be increased by up to 300°C while driving over the Highway Cycle with 20 per cent mis- fire, al though dur ing the Federal Test Procedure cycle, with the equivalent of one cylinder of a six cylinder engine permanently misfiring, the catalyst temperature reached an average of 600°C and a maximum of 851°C.

Catalysts may also become deactivated by poisons contained in both fuel and lubricating oil. The effect of oil-derived phosphorus and sulphated ash on catalyst deactivation was de- scribed by K. Inoue, T. Kurahashi and T. Negishi of Nippon Oil Company and K. Akiyama, K. Arimura and K. Tasaka of Toyota Motor Corporation (SAE 920654). They found that the catalyst surface phosphorus concentra- t ion increased wi th increasing carbon monoxide and nitrogen oxides emissions. The presence of sulphated ash reduced the amount

of phosphorus on the catalyst but also had a negative effect on catalytic activity. Catalyst deactivation was also much more apparent at 800°C than at 720°C. They concluded that en- gine oils low in phosphorus and sulphated ash can reduce catalyst deactivation.

The effect of fuel sulphur and air to fuel ratio on catalyst deactivation was described by J. C. Summers, J. E Skowran, W. B. Williamson and K. I. Mitchell of Allied Signal Inc., (SAE 920558). It was reported that relatively low con- centrations of fuel sulphur resulted in loss of catalyst performance, but little additional poi- soning effect is seen on increasing the sulphur content of the fuel to comparatively high levels. This deterioration in emissions is due to the diminished performance of a catalyst at its operating temperature, rather than to an in- crease in light-off temperature.

Catalyst selection for the reduction of formal- dehyde emissions from methanol-fuelled vehicles was presented by M. S. Newkirk and L. R. Smith from Southwest Research Institute, M. Ahuja, S. Albu and S. Santoro from Califor- nia Air Resources Board, and J. Leonard from South Coast Air Quality Management District (SAE 920092). The use of methanol as a fuel is seen as a viable approach to reduce air pollu- tion; concern, however, has been expressed about t he relatively high formaldehyde emissions from vehicles using this fuel. This paper reported on investigations during which the catalyst systems of several vehicles were modified in order to reduce formaldehyde emissions. In general the strategy was to reduce cold start emissions by using close coupled and electrically heated catalyst systems. These tech- niques were successful, but supplemental air injection above the catalyst was required when using an electrically heated catalyst to optimise this system.

Conclusions The Society of Automotive Engineers Con-

gress is always a barometer to the current areas of work and interest of the car companies and of the emissions control industry. The need to meet forthcoming emissions legislation clearly

Platinum Metals Rev., 1992,36, (2) 88

drives development activities. This was high- lighted by the papers which considered improved cold-start performance and in the attention given to catalyst deactivation mechanisms, by

The application of noble metal catalysts for diesel emissions control continues to be a major part of the Congress, as are papers considering the substitution of platinum or rhodium by

temperature and/or poisoning. palladium. C.J., R.D.O’S.

Platinum Group Metals in 1991 The Ayrton Metals Platinum Yearbook 1992 BY B. H. NATHAN, Woodhead Publishing, Cambridge, 1992,195 pages, ISBN 1-855734847, E45.00

Following in the footsteps of the “Platinum Yearbook 1991”, this new publication sets out to present a panorama of the many events that influenced the market for platinum group me- tals during 1991. These are covered in general terms, and in considerable detail in Chapters 1 and 2, respectively, both being supported by statistical data. A number of these events were industrial announcements or scientific reports, but most were not, and the whole makes fasci- nating reading, perhaps especially for those whose involvement with the platinum group metals does not embrace metal dealing.

While the detailed review of 1991 occupies ninety-nine pages, the prospects for the six platinum group metals in 1992 are contained in some five pages. It is suggested that of these metals, “platinum has the best prospects for a sound and widely-based recovery in 1992”.

A chapter on the Tokyo Commodity Ex- change by Kazuhiko Noma is complemented by two further chapters by Brian Nathan, in which the dealing arrangements for the London Plati- num and Palladium Market and the New York Mercantile Exchange are considered.

An interesting overview of fuel cells and their state of development for stationary and trans- portation applications is given by Jocelyn Cloete. Once again, it is concluded that proof of phosphoric acid fuel cell technology will be provided in the imminent future, from the re- sults of on-going field tests.

In the penultimate chapter Peter Gaylard out- lines the lengthy, complex and expensive processes that are necessary to extract the plati- num group metals from the limited number of major ore bodies in which they occur, in minute quantities, and to then refine them to the re- quired high purity levels. In response to the needs of the market, and to a greater awareness of environmental considerations, improved re- covery processes have been introduced in recent years.

A brief chapter on the history of platinum and the platinum group metals, based in part upon “A History of Platinum and its Allied Metals” by D. McDonald and L. B. Hunt, con- cludes this informative and interesting book which will serve as more than a record of the events of the past year. I.E.C.

Platinum Improves Protective Coatings Gas turbine engines are widely used for both

stationary and mobile applications, and the tur- bine blades, which are highly stressed during service, are required to operate at high tempera- tures in oxidising atmospheres which may be contaminated with corrosive fuel residues and ingested salts. To some extent nickel-based superalloy turbine components can be pro- tected against both oxidation and hot corrosion by nickel aluminide diffusion coatings, but in more severe environments the protective coat- ing may break down, reducing service life.

The development of platinum-containing coating systems has been reported here on sev- eral occasions over the past decade as materials

scientists have sought both to improve the pro- tection given by such coatings and to establish the precise role of the platinum in the process.

A further contribution on the subject has been published recently (H. M. Tawancy, N. M. Abbas and T. N. Rhys-Jones, Sut$ Coat. Tech-

Following an investigation of the microstruc- ture of platinum-modified aluminide coatings on selected nickel-based superalloys, the authors identify a number of ways by which the platinum improves the protective ability of the coating. Oxidation behaviour depends upon the composition of the superalloy substrate, espe- cially on its rare earth content.

not?., 1991,49, (1-3), 1-7).

Platinum Metak Rev., 1992,36, (2) 89

The Large Scale Production of Hydrogen from Gas Mixtures A USE FOR ULTRA THIN PALLADIUM ALLOY MEMBRANES

By V. Z. Mordkovich, Yu. K. Baichtock and M. H. Sosna State Institute of Nitrogen Industry, Moscow, Russia

A method of pure hydrogen production using palladium alloy membranes to separate hydrogen from hydrogen-rich gas mixtures has been employed for many years in labora- tory and industrial practices. Generally, palladium-silver membranes have been em- ployed. Over the years a number of papers describing this method have been published in this journal (1-4). There are many publica- tions by British, Japanese, Russian, etc., producers describing such units with produc- tive capacities ofup to 100 Nm’/h. These units are intended for the purification of technical grade hydrogen and operate in areas such as rare element metallurgy, the electronic indus- try and general laboratory practice. Recently it has been reported by Johnson Matthey (3, 4) that units capable of separating up to 50 Nm’/h of hydrogen from methanol-water cracking gas have been used as constituents of self-contained hydrogen generators.

One may ask the obvious question: why not use palladium alloy membranes for large-scale hydrogen production? It is easy for anyone to obtain the answer that this method does not provide a sufficient return on capital invest- ment, due to the high cost of the noble metals involved. It is less evident that both the du- rability and the hydrogen recovery level of the usually employed membranes are also some- wha t de f i c i en t . Indeed , t h e ex i s t ing membrane units are generally used only for relatively small-scale purification of technical grade hydrogen, and not for its separation from mixtures containing less than 95 per cent hydrogen.

Nonetheless, in principal, membrane tech- nology promises significant advantages in

comparison with conventional technologies. Palladium alloy membrane units combine the compactness of polymer membrane units, the high product purity of the pressure swing ad- sorption technique, the recovery level of cryogenic separation and furthermore the ability to utilise great pressure drops as the directives for rapid permeation. There are, however, a number of reasons which prevent the attainment of either the investment return or the durability and recovery levels demanded in the case of large-scale hydrogen-rich mixture separation technologies.

In this respect, five key problems may be emphasised, namely, choice of membrane ma- terials, minimisation of membrane thickness, design of the membrane holder, design of the membrane apparatus and start-up/shut-down technology. All the listed advantages can be realised and the disadvantages overcome by solving the key problems.

Key Problems of Palladium AUoy Membrane Technology

The choice of membrane material is the first highly significant problem. The palladium alloy has to possess excellent hydrogen per- meabili ty a n d be resistant to specific start-up/shut-down cyclic stresses. It is well known that hydrogen-palladium alloy interac- tions can induce the a-p phase transformation, which consequently alter atom spacings in the metal lattice, causing dimensional changes which are large enough to distort the mem- brane ( 5 , 6 ) . Thus conditions which favour the existence of p- and P-like hydride phases are dangerous in regard to the possibilities of membrane destruction. The task is therefore

Platinum Metals Rev. , 1992,36, (2), 90-97 90

Fig. 1 The variation of hy- drogen permeability, W, with pressure drop for six differ- ent palladium alloys:

ure=O.l MPa

loo

T=50°C, output press- 80

E

E 60

0 X c

'- 4 0

20

-/

Pd-lOAg-55N

Pd-4 TI

e- Pd;5.5Ni

5 10 15 2 0 PRESSURE DROP, MPa

to develop an alloy which forms hydride phases only at temperatures substantially dif- fe ren t from t h e opera t ing conditions. Palladium membranes have usually been operated in the temperature range 400 to 600°C, within which there i s good per- meability and satisfactory durability.

The start-up/shut-down procedure has to be able to avoid hydride phase formation during its operation. Additionally, it should be em- phasised tha t internal stresses and the consequent cracking of membranes are often induced by reasons other than p-phase forma- tion. Certain palladium alloys seem unsuitable for use as membrane material for large-scale technology as they have insufficient resistance to specific start-up/shut-down stresses. For example, the widely employed 25 palladium- 75 per cent silver alloy has low cyclic resistance and often cracks even if apparently suitable start-up/shut-down procedures are employed.

Different alloys also vary in hydrogen per- meability, and this is illustrated by Figure 1. It may be noted that the permeability of the best alloy is about five times as high as that of the worst. In recent years a number of alloys with

large permeability and high cyclic resistance, such as the Soviet B-X group of alloys, have been developed (7). Such alloys may contain

(O-1)Pt-(O.O1-O.5)Al-(balance)Pd. Minimisation of membrane thickness is the

second key problem. Clearly, a decrease in membrane thickness while maintaining a con- stant membrane surface area, results in a proportional reduction in the cost of equip- ment. Furthermore, permeability through thin membranes is proportionally greater than through thick membranes. When calculating the dependence of the overall investment (namely, capital cost per unit of productive capacity) on membrane thickness, the decisive significance of the membrane thickness can be demonstrated convincingly. Such dependence is shown in Figure 2 for the investment level of the entire process of hydrogen production from natural gas where the cost of the 0.1 mm thick membranes constitutes about half of the total capital cost. It can be seen that 0.05 mm thick membranes provide an investment level near that of conventional technologies, while the use of 0.01 mm thick membranes would result in the most inexpensive technology for

( ~O-~~)A~-(O.O~-~)AU-(O-~)Y-(O.O~-~)RU-

Platinum Metals Rev., 1992,36, (2) 91

005 0.10 0.15 0 2 0 MEMBRANE THICKNESS, mm

Fig. 2 Effect of membrane thickness on specific investment, L, is shown for the en- t ire large-scale process of hydrogen pro- duction from natural gae (output of about 10,000 Nm3/h of hydrogen). This process principdy includes steam-& conversion of methane followed by diffusion throqyh palladium alloy membranes. The asterisk and consequent dotted line indicate the in- vestment level of an alternate process where pressure swing adsorption is used instead of membrane diffusion

the large-scale production of hydrogen from gas mixtures.

Such thin membranes could be destroyed, however, if there were substantial increases of the operating pressure drop across the mem- brane, and at present membranes for the separation of mixtures are required to work at pressure drops of between 1 and 20 MPa. Nevertheless, an increase of pressure drop would result in a greater productive capacity and recovery level. To summarise, there could be a difficult situation in trying to balance the problem of minimisation of membrane thick- ness; on the one hand, a decrease in the thickness can critically improve the entire technology, while on the other hand very thin membranes may not be strong enough to work at the conditions demanded.

This problematic situation can be partly re- solved by the use of a suitably designed membrane holder, which constitutes the third key problem area. A simple classification of

membrane holders is presented in Figure 3, the holders being divided into “tube”, “plane- foil” and “folded-foil”. Three examples of holder design are also shown in Figure 3, namely “disc”, “capillary with external press- ure” and “capillary with internal pressure”.

Each type of membrane holder may be fur- ther characterised by the type of connection between the palladium alloy and the adjoining material. Welding or soldering has been used, but it is difficult to weld thin membranes and hence soldering often seems to be the more attractive method. On the other hand strong erosion can occur during the course of the soldering operation due to dissolution of pal- ladium alloy in the liquid solder; and even if this dissolution is not complete, the boundary alloy formed can be easily damaged by hy- drogen and by the mutual diffusion of components at high operating temperatures. The use of hard silver or gold solder can be advantageous, although the operational life of capillary holders with silver solder is usually not more than one or two months. Different kinds of welding may be used, such as arc welding, pressure welding, electron beam and laser welding. Pressure tight welds can be pro- duced by each of these methods, although mutual diffusion of palladium alloy and steel components at the operating temperatures can still constitute a significant problem which may result in the destruction of either the membrane or the weld. Residual after-weld defects and stresses also have to be taken into account, although multilayer welding tech- niques can usually avoid these difficulties.

There can be a large variety of alternative membrane holders, many of them having the same defects, such as low ranges of practical pressure drop and short operational life. It seems that the first of these limitations can be removed in the case of “plane-foil” holders with membrane support, and also by the use of “capillary” holders. In the present paper “tube” holders with tube diameters of less that 1 mm and wall thicknesses of less than 0.1 mm are designated as “capillary” holders. “Plane- foil” holders have an evident advantage of

Platinum Metals Rev., 1992,36, (2) 92

MEMBRANE HOLDERS

folded-toi I plane-foil

Fig. 3 Membrane holders can be classified as “folded-foil”, “tube” and “plane-foil” and the latter two ca- tegories can be further sub-divided

others disc %

lower cost, due to the lower cost of foil in comparison with that of tubes. On the other hand, “tube” holders also have certain par- ticular advantages especially in the case of “capillary” holders, which allow thin mem- branes to be operated under large pressure drop conditions without any membrane sup- port. Experience has shown, however, that the membrane support can provoke corrosion of the palladium alloy, and in addition the sup- port can create difficulties for optimum gas flow and can reduce the purity of the output hydrogen. The design of compact gas collec- tors is more successful in the case of “tube” holders, but this last consideration is close to another key problem, which is discussed below.

The fourth key problem is the overall design of the membrane assembly. As mentioned above, one aim is to create compact equipment

and this is easy in the case of “tube” membrane holders but “plane-foil” and “folded-foil” holders demand more complex collectors for the purified hydrogen. Designs must also pro- vide optimum rates of gas flow near the membrane surface in order to attain maxi- mum levels of productive capacity and recovery.

Necessary temperature conditions near the membranes can be achieved either by external heating of the apparatus or by preliminary heating of the gas mixture. The later seems preferable for large scale hydrogen produc- t ion, a n d i t should be noted t h a t if preliminary heaters are used without heat ex- changers then the expenditure on energy can constitute up to half the cost price of the hydrogen produced. The membrane apparatus adopted needs to be equipped with internal heaters for the gas mixture and internal heat

Platinum Metals Rev., 1992,36, (2) 93

exchangers in order to conserve the heat, while the outlet gas temperature should not be more than 200°C.

The development of optimum start-up/shut- down technology, which is the fifth key problem, has been partially discussed above. I t should again be noted that this technology requires specific control of temperature and pressure, and the use of certain alternative blow gases has also to be considered. For example, heating to the operating temperature and cooling during the course of shut-down can be usefully carried out in a nitrogen at- mosphere in order to obtain a long operational life for the membranes.

GIAP Style Development The problems reviewed above are common

key problems for any organisation which is going to attempt to improve palladium alloy membrane technology. The solutions to these problems and the relative significance of each depend, however, on which branch of industry the proposed technology is principally in- tended for.

The State Institute of Nitrogen Industry

(GIAP) began to investigate palladium alloy membrane technology for the nitrogen indus- try more than twenty years ago. T h e conditions and the needs of the nitrogen in- dustry may be characterised as:

(a) high pressures of technological and waste hydrogen-rich gas, from 3 to 32 MPa

(b) a wide range of hydrogen content in the mixtures, typically 40 to 75 volume per cent

(c) the presence of nitrogen, methane, am- monia, water, carbon dioxide, and carbon monoxide as possible components of the mix- ture and also the probable occurrence of small amounts of oil

(d) rapid growth in the demand for rela- tively low cost hydrogen. With regard to the last item it should be emphasised that the hydrogen-rich mixtures occurring in the ni- trogen industry can provide a cheap source of pure hydrogen if appropriate technology can be developed for its recovery.

Accordingly, certain directions of research were decided upon and resulted in the devel- opment of an original GIAP style in regard to metal membrane technology.

The use of the B-X series of alloys has been

Platinum Metals Rev., 1992,36, (2) 94

FK. 4 A variety of capillary membrane holders designed by GIAP are displayed

one of the most important directions of im- provement. These alloys possess good hydrogen permeability and the membranes can provide operational lives of more than two years. The B-X alloys are resistant to “poison- ing” by constituents present as impurities in the gases from the nitrogen industry. These conclusions were formed in the course of long term testing carried out by GIAF!

Different types of membrane holders were also designed and tested in GIAP, the most advanced being “capillary with internal press- ure” and “capillary with external pressure”. These are shown in Figure 4. B-X capillary tubes with wall thicknesses of from 0.05 to 0.1 mm are used and the free end of the tubes are sealed. The hydrogen-rich mixture flows along the external surface of the tubes and pure hydrogen is ejected from the internal tube space due to the pressure difference be- tween the open and the sealed ends of the tube. The “capillary with internal pressure” holders have two heads, and both ends of the tubes are open. The mixture flows into and around the internal space of the tubes and pure hydrogen is removed from their external surfaces. Due to the successful design of the holder head and the development of steel to B-X powder welding technology (8, 9), the GIAP membrane holders in both “internal’’ and “external” variants can operate at pressure drops of up to 30 MPa and are characterised by ope