WO2012013940A2

WO2012013940A2 - Catalysts for hydrogen generation and fuel cells

Info

Publication number: WO2012013940A2
Application number: PCT/GB2011/001156
Authority: WO
Inventors: Shik Chi Edman Tsang; Karaked Tedsree
Original assignee: Isis Innovation Limited
Priority date: 2010-07-29
Filing date: 2011-07-29
Publication date: 2012-02-02
Also published as: GB201012789D0; WO2012013940A3

Abstract

The invention provides a process for producing H₂ from a compound of formula (I), (II), (III) or (IV): R¹COOH (I); R²OH (II); R³CHO (III); R⁴NH₂ (IV); wherein R¹ is H or unsubstituted or substituted C_1-10 alkyl; R² is unsubstituted or substituted C_1-10 alkyl; R³ is H or unsubstituted or substituted C_1-10 alkyl; R⁴is H, unsubstituted or substituted C_1-10 alkyl, or C(0)NR⁵R⁶; and R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted C_1-10 alkyl; which process comprises contacting a liquid phase which comprises said compound of formula (I), (II), (III) or (IV) with a catalyst, which catalyst comprises: (a) polymetallic nanoparticles, each of which comprises a first metal and a second metal, which first metal is selected from a Group (9), Group (10), Group (11) or Group (12) d-block metal, and which second metal is other than said first metal, wherein the polymetallic nanoparticles comprise a core which comprises said second metal and a shell surrounding said core, which shell comprises said first metal; or (b) nanoparticles which comprise a metal selected from palladium, rhodium, ruthenium, iridium, copper and silver, and which have a mean particle size of less than or equal to 50 nm;_provided that when said compound is a compound of formula (II), the catalyst comprises (a) said polymetallic nanoparticles. Further provided is a catalyst for the production of hydrogen from a compound of formula (I)-(IV) or for the electro-oxidation of a compound of formula (I)-(IV), which catalyst comprises polymetallic nanoparticles, which polymetallic nanoparticles comprise a core and a shell surrounding the core; wherein the shell comprises a first metal which is palladium, and the core comprises a second metal, wherein the second metal is other than palladium, platinum, gold, iron, cobalt, nickel, titanium, tungsten, tantalum, vanadium and niobium. Further provided is a process for producing the catalysts of the invention, and the following uses of the catalysts of the invention: use for the electro-oxidation of a compound of formula (I)-(IV); use for the electro-oxidation of formic acid; use in a fuel cell; use for the electro-oxidation of a compound of formula (I)-IV) in a fuel cell; and use for the electro-oxidation of formic acid in a direct formic acid fuel cell. Further provided is an electrode suitable for use in a fuel cell, comprising a catalyst of the invention, and a fuel cell comprising a catalyst or an electrode of the invention. A process for the electro-oxidation of a compound of formula (I)-(IV) with a catalyst of the invention is also provided.

Description

CATALYSTS FOR HYDROGEN GENERATION AND FUEL CELLS

FIELD OF THE INVENTION

The invention relates to a process for producing hydrogen, catalysts for producing hydrogen or for use in fuel cells, processes for producing the catalysts, and uses of the catalysts in fuel cell applications.

BACKGROUND TO THE INVENTION

Hydrogen has attracted increasing attention as an important alternative secondary energy resource particularly when combined with fuel-cell technology, which may play a very significant role in power generation in the future. On the other hand, the storage and transfer of hydrogen are also problematic because of its low volumetric energy density. At present most of the tested hydrogen storage materials, such as metal hydrides, carbon materials, porous metal-organic frameworks, and ammonia can only store low amounts and high temperatures are required to release the stored hydrogen. Currently most H₂ is produced industrially by reforming hydrocarbons or alcohols and by the water-gas-shift (WGS) reaction at high temperatures. However, ultra-pure hydrogen gas is required by the fuel cells. In particular, the gas stream usually has to be free from CO gas (<10ppm) otherwise the catalytic performance of the fuel cell is severely hampered. On the other hand, the cumbersome multistage water gas shift (WGS) and CO cleanup processes as well as the slow response at start-up for hydrogen production preclude the technology from many uses. Formic acid, which is nontoxic and a liquid at room temperature, with a density of 1.22 g. mL^'3, has been widely used as a hydrogen source for transfer hydrogenation. The decomposition of formic acid (HCOOH = CO2+H2), which is the reverse reaction of C0₂ hydrogenation (C0₂+¾ = HCOOH), is considered to be a promising hydrogen generation process. Recently, Beller and co-workers (Angew. Chem. 2008, 120, 4026 -4029; Angew. Chem. Int. Ed. 2008, 47, 3962 -3965; ChemSusChem 2008, 1, 751- 758) investigated the catalytic decomposition of formic acid/amine mixtures over homogeneous Ru catalysts at ambient temperature, and excellent catalytic activities of some optimised catalytic species for this reaction were claimed. In order to reduce the volatility, Laurenczy and co-workers (Angew. Chem. 2008, 120, 4030-4032; Angew. Chem. Int. Ed. 2008, 47, 3966 -3968; Chem. Eur. J. 2009, 15, 3752 -3760) developed the decomposition of HCOOH/HCOONa (9:1) under aqueous conditions, but the catalysts were only found to be effective at 80°C. Similarly, Deng and co-workers (ChemSusChem 2010, 3, 71-74) developed new Ru homogeneous catalysts in ionic liquid in order to reduce the volatile matter, but yet again elevated temperatures (60°C) were required for decent catalytic rates. Ojeda and Iglesia (Angew. Chem. Int. Ed. 2009, 48, 4800-4803) reported hydrogen production using supported gold nanoparticles for this reaction at 70-110°C. A key challenge is the development of solid catalysts for the liquid phase decomposition of formic acid to hydrogen gas at room temperature. If successful, the new technology could facilitate wide usage in ambient conditions.

Furthermore, global issues such as high energy demand, fossil fuel depletion and environmental pollution draw the attention of researchers to focus on alternative energy sources and energy conversion technologies. In particular, fuel cells as energy converting devices are rapidly gaining momentum due to their high energy conversion efficiency, low emission of pollutants and capability of operating at a low temperature. One important interest in recent years has been to develop direct formic acid fuel cells (DFAFC). The formic acid fuel crossover through proton exchange membrane is much lower than methanol and thus, highly concentrated formic acid solutions and thinner membranes can be employed in DFAFC. DFAFC also have a higher electromotive force (EMF) than either hydrogen or direct methanol fuel cells (DMFC). DFAFC technology has shown electro- oxidation activity far superior to DMFC and in some cases performances approach those of H₂-PEM fuel cells. Thus, a direct formic acid fuel cell is a preferred power source for portable devices such as cellular phones, personal digital assistants (PDAs), laptop computers, etc. Formic acid was first reported to be an excellent fuel for polymer electrolyte membrane fuel cells in 2002, and since then formic acid fuel cells have become an active area of research. The electrochemical reactions in a direct formic acid fuel cell (DFAFC) are shown below:

Anode HCOOH→ C0₂ + 2H⁺ + 2e^~ Eqn. 1.1

Cathode ½ 0₂ + 2H⁺ + 2e^~→ H₂0 Eqn 1.2

Overall HCOOH + ½ 0₂→ C0₂ + H₂0 Eqn. 1.3

It is noted that the anode reaction is very similar to the catalytic decomposition of formic acid to H₂ and C0₂, with C-H breakage being rate-limiting, but this time protons are formed instead of hydrogen gas when anodic current is used. A further important challenge is therefore the development of an efficient catalyst for formic acid electro- oxidation in formic acid fuel cells. SUMMARY OF THE INVENTION

The present inventors have provided heterogeneous nanocatalysts which are active for the decomposition of simple hydrogen-containing molecules in the liquid phase, including formic acid. The catalysts can be used to generate hydrogen from such molecules at room temperature. They can also catalyse the electro-oxidation of such compounds in fuel cells, forming protons instead of H₂.

Accordingly, in a first aspect, the present invention provides a process for producing ¾ from a compound of formula (I), (II), (III) or (IV):

R'COOH (I)

R²OH (Π)

R³CHO (HI)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted Ci.io alkyl;

R² is unsubstituted or substituted Ci.₁₀ alkyl;

R³ is H or unsubstituted or substituted CHO alkyl;

R⁴ is H, unsubstituted or substituted C_M0 alkyl, or C(0)NR⁵R⁶;

R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted C]_-10 alkyl;

which process comprises contacting a liquid phase which comprises said compound of formula (I), (II), (III) or (IV) with a catalyst, which catalyst comprises:

- (a) polymetallic nanoparticles, each of which comprises a first metal and a second metal, which first metal is selected from a Group 9, Group 10, Group 11 or Group 12 d-block metal, and which second metal is other than said first metal, wherein the polymetallic nanoparticles comprise a core which comprises said second metal and a shell surrounding said core, which shell comprises said first metal; or

- (b) nanoparticles which comprise a metal selected from palladium, rhodium, ruthenium, iridium, copper and silver, and which have a mean particle size of less than or equal to 50 nm;

provided that when said compound is a compound of formula (II), the catalyst comprises (a) said polymetallic nanoparticles. In another aspect, the invention provides a catalyst for the production of hydrogen from a compound of formula (I), (II), (III) or (IV) as defined above, or for the electro- oxidation of a said compound of formula (I), (II), (III) or (IV),

which catalyst comprises polymetallic nanoparticles, which polymetallic nanoparticles comprise a core and a shell surrounding the core; wherein the shell comprises a first metal which is palladium, and the core comprises a second metal, wherein the second metal is other than palladium, platinum, gold, iron, cobalt, nickel, titanium, tungsten, tantalum, vanadium and niobium.

In another aspect, the invention provides a process for producing a catalyst for the production of hydrogen from a compound of formula (I), (II), (III) or (IV) as defined above, or for the electro-oxidation of a said compound of formula (I), (II), (III) or (IV), which catalyst comprises polymetallic nanoparticles, which polymetallic nanoparticles comprise a core and a shell surrounding the core, wherein the shell comprises a first metal which is palladium, and wherein the core comprises a second metal, wherein the second metal is other than palladium, platinum and gold;

which process comprises:

(a) reducing a salt of said second metal in the presence of a first solvent, to produce a suspension or solution of core nanoparticles in said first solvent, which core nanoparticles comprise said second metal; and

(b) reducing a palladium salt in the presence of said core nanoparticles and a second solvent, which is the same or different from the first solvent, to produce a shell on the surfaces of said core nanoparticles, which shell comprises palladium.

In another aspect, the invention provides a catalyst which is obtainable by the process of the invention for producing a catalyst for the production of hydrogen from a compound of formula (I), (II), (III) or (IV) or for the electro-oxidation of a said compound of formula (I), (II), (III) or (IV).

In another aspect, the invention provides the use of a catalyst of the invention as defined above for the electro-oxidation of a compound of formula (I), (II), (III) or (IV):

R'COOH (I)

R²OH (Π)

R^JCHO (III)

R⁴NH₂ (IV)

wherein R¹ is H or unsubstituted or substituted CI.JO alkyl;

R² is unsubstituted or substituted Cj.io alkyl;

R³ is H or unsubstituted or substituted CMO alkyl;

R⁴ is H, unsubstituted or substituted d.i₀ alkyl, or C(0)NR⁵R⁶;

R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted CMO alkyl.

In another aspect, the invention provides the use of a catalyst of the invention as defined above for the electro-oxidation of formic acid.

In another aspect, the invention provides the use of a catalyst of the invention as defined above in a fuel cell. Typically, the fuel cell comprises a compound of formula (I), (II), (III) or (IV) as defined above.

In another aspect, the invention provides the use of a catalyst of the invention as defined above for the electro-oxidation of a compound of formula (I), (II), (HI) or (IV) in a fuel cell:

R'COOH (I)

R²OH (II)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted Ci-io alkyl;

R² is unsubstituted or substituted Ci.₁₀ alkyl;

R³ is H or unsubstituted or substituted Ci-io alkyl;

R⁴ is H, unsubstituted or substituted C_M0 alkyl, or C(0)NR⁵R⁶;

R^s and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted Ci-io alkyl.

In another aspect, the invention provides the use of a catalyst of the invention as defined above for the electro-oxidation of formic acid in a direct formic acid fuel cell.

In another aspect, the invention provides an electrode suitable for use in a fuel cell, which electrode comprises a conducting substrate and a catalyst of the invention as defined above. Typically, the fuel cell comprises a compound of formula (I), (II), (III) or (IV) as defined above. More typically, the fuel cell is a direct formic acid fuel cell.

In another aspect, the invention provides a fuel cell which comprises: a catalyst of the invention as defined above or an electrode of the invention as defined above. Typically, the fuel cell further comprises a compound of formula (I), (II), (III) or (IV) as defined above. More typically, the fuel cell is a direct formic acid fuel cell. Thus, the fuel cell typically further comprises a compound of formula (I) which is formic acid.

In another aspect, the invention provides a process for the electro-oxidation of a compound of formula (I), (II), (III) or (IV) as defined above; which process comprises contacting a liquid phase which comprises said compound of formula (I), (II), (III) or (TV) with a catalyst of the invention as defined above in the presence of an electrode.

In another aspect, the invention provides a process for the electro-oxidation of a compound of formula (I), (II), (HI) or (IV) as defined above; which process comprises contacting a liquid phase which comprises said compound of formula (I), (II), (III) or (IV) with an electrode, which electrode comprises a catalyst of the invention as defined above. Typically, the electrode comprises a conducting substrate and a catalyst of the invention as defined above.

BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 shows TEM images of polymer stabilised Pd nanoparticles a) PVP (particle size 4.5 ±0.6 nm) b) 80% hydrolysed PVA (particle size 7.0 ± 0.8 nm) c) HB-PEI (particle size 9.0 ± 1.0 nm).

Fig. 2 shows TEM images of PVP-stabilised Ru nanoparticles having particle sizes a) 1.8 nm, and b) 2.3nm.

Fig. 3 shows TEM images of PVP-stabilised Ru nanoparticles having particle sizes c) 2.7 nm d) 3.2 nm.

Fig. 4 shows TEM images of PVP-Pd nanoparticles synthesised by stepwise growth having particle sizes a) 2.3 nm b) 3.5 nm c) 5.2 nm

Fig. 5 shows TEM images of PVP-Pt nanoparticles having particle sizes a) 3.2 nm b) 3.8 nm c) 5.2 nm d) 6.0 nm.

Fig. 6 shows TEM images of PVP-Rh nanoparticles having particle sizes a) 2.0 nm b) 5.1 nm c) 10.0 nm

Fig. 7 shows TEM images of PVP-Ag nanoparticles having particle sizes a) 20.3 nm b) 30.2 nm c) 50.3 nm.

Fig. 8 shows TEM images of Au nanoparticles having particle sizes a) 10.0 nm b)

30.5 nm c) 50.0 nm.

Fig. 9 shows TEM images of bimetallic nanoparticles prepared from co-reduction of mixed metal salts: a) Pt-Pd (particle size 10.0±1.8 nm) b) Ag-Pd (particle size 20.5 ±3.5 nm) and c) Au-Pd (particle size 10.0 ± 0.8 nm). Fig. 10 shows TEM images of bimetallic nanoparticles prepared by successive reduction a) Pt-Pd (particle size 10.0 nm±1.8) b) Ag-Pd (particle size 18.5±2.1 nm) c) Ag- Pd (particle size 30.5±7.2 nm).

Fig. 11 shows TEM images of bimetallic nanoparticles prepared by successive reduction with hydrogen sacrificial protective strategy a) 1 : 1 mole ratio of Ru-Pd (particle size 3.2+0.4 nm) b) 1 : 1 mole ratio of Rh-Pd (particle size 2.8± 0.5nm).

Fig. 12 shows UV-Vis spectra of Ag, Pd and Ag@Pd nanoparticle at different molar ratios

Fig. 13 shows UV-Vis spectra of Au and Au@Pd nanoparticles (particle size 10.0 ± 0.8nm) prepared from co-reduction of mixed metal ions

Fig. 14 shows TEM images of 20% Pd/C a) 2.3 nm b) 4.5 nm

Fig. 15 shows TEM images of nanoparticles after heating at 300°C under N₂ for 0.5 hr a) Pd/C (particle size 2.7+ 0.3 nm ) b) Pt@Pd/C (particle size 42.1+4.0 nm) c)

Ag@Pd/C (particle size 35.2+5.0 nm)

Fig. 16 shows the correlation between chemical shift and mole fraction of Pd shell on Ag@Pd nanoparticles.

Fig. 17 shows cyclic voltammograms (anodic scan) of formic acid electro-oxidation on Ag@Pd catalyst.

Fig. 18 shows mass activity (at 0.2 V) of Ag@Pd catalysts containing different mole ratios of core and shell metals for formic acid electro-oxidation.

Fig. 19 shows specific activity (at 0.2V) of Ag@Pd catalysts containing different mole ratios of core and shell metals for formic acid electro-oxidation.

Fig. 20 shows electrocatalytic surface area of Ag@Pd with different mole fractions of Pd overlay er.

Fig. 21 shows the correlation between % surface expansion and nanocatalysts containing various mole ratios of Pd-shell on Ag-core.

Fig. 22(a) shows the relationship between the ¹³C chemical shifts of adsorbed bridging formates on Ag@Pd, Rh@Pd, Au@Pd, Ru@Pd and Pt@Pd bimetallics and the work functions of Ag, Rh, Au, Ru, Pd and Pt; the work functions of the fee (111) plane are used for Ag, Rh, Au, Pd and Pt; for Ru, the hep (001) plane was used, having the same surface features.

Fig. 22(b) shows the correlation between chemical shift and specific activity (at 0.2V) of both carbon supported monometallic and core-shell bimetallic catalysts for formic acid electro-oxidation. Fig. 23(a) (left) is a plot of the rates of formic acid decomposition over different metal colloids (4 x 10^'5 mole in lOmL) in water vs. d-band center.

Fig. 23(b) (right) shows the correlation between the formic acid decomposition rate with the initial rate of C0₂ formation over different sizes of gold nanoparticles.

Fig. 24(a) (left) shows the relationship between rate of formic acid decomposition over Ag@Pd, Rh@Pd, Au@Pd, Ru@Pd and Pt@Pd bimetallics and the work functions of Ag, Rh, Au, Ru, Pd and Pt. The work functions of Ag, Rh, Au, Ru, Pd and Pt from fee (111) fee planes were used, while the work function of hep (001) hep plane was used for Ru (having the same surface feature as the (111) fee plane in the other metals).

Fig. 24(b) (right) shows the correlation between the ¹³C chemical shift of adsorbed formate and rate of formic acid decomposition over monometallic and core-shell bimetallic nanoparticles.

Fig. 25(a) (left) is a plot of chemical shift values of adsorbed bridging formate over Ag@Pd at different molar ratios.

Fig. 25(b) (right) is a plot of rates of formic acid decomposition over Ag@Pd at different molar ratios.

Fig. 26 shows the volume (mL) of gas liberation over time (mins) from unstirred reactor containing lOmL of aqueous formic acid (containing 0.5M, 1M, 2M and 4M of aqueous formic acid respectively) when in contact with 4 x 10^"5 mole of 1.1 Ag@Pd catalyst.

DETAILED DESCRIPTION OF THE INVENTION

The following substituent definitions apply with respect to the compounds defined herein:

A Ci-₁₀ alkyl group is an unsubstituted or substituted, straight or branched chain saturated hydrocarbon radical. It may for example be methyl, ethyl, propyl, butyl, pentyl, hexyl, heptyl, octyl, nonyl or decyl. Typically, it is C_1-6 alkyl, for example methyl, ethyl, propyl, butyl, pentyl or hexyl, or C alky], for example methyl, ethyl, i-propyl, n-propyl, t-butyl, s-butyl, i-butyl or n-butyl. When an alkyl group is substituted it typically bears one or more substituents selected from unsubstituted Ci.6 alkyl, substituted or

unsubstituted aryl (as defined herein), cyano (CN), amino (NH₂),

alkylamino, di(Ci. io)alkylamino, arylamino, diarylamino, arylalkylamino, amido, acylamido, hydroxyl (OH), oxo (=0), halo, carboxy, ester, acyl, acyloxy, CMO alkoxy, aryloxy, haloalkyl, sulfhydryl (i.e. thiol, -SH), Ci-jo lkylthio, arylthio and sulfonyl. Examples of substituted alkyl groups include haloalkyl, hydroxyalkyl, aminoalkyl, alkoxyalkyl and alkaryl groups. The term alkaryl, as used herein, pertains to a CM O alkyl group in which at least one hydrogen atom has been replaced with an aryl group. Examples of such groups include, but are not limited to, benzyl (phenylmethyl, PhCH₂-), benzhydryl (Ph₂CH-), trityl (triphenylmethyl, PI1₃C-), phenethyl (phenylethyl, Ph-CH₂CH₂-), styryl (Ph-CH=CH-), cinnamyl

(Ph-CH=CH-CH₂-).

Typically a substituted C_1-2o alkyl group carries 1, 2, 3 or 4 substituents, for instance, 1 , 2 or 3 substituents, or more typically 1 or 2 substituents.

An aryl group is a substituted or unsubstituted, monocyclic or bicyclic aromatic group which typically contains from 6 to 14 carbon atoms, preferably from 6 to 10 carbon atoms in the ring portion. Examples include phenyl, naphthyl, indenyl and indanyl groups. An aryl group is unsubstituted or substituted. When an aryl group as defined above is substituted it typically bears one or more substituents selected from unsubstituted d-6 alkyl, unsubstituted aryl, cyano (CN), amino (NH₂), CMO alkylamino, di(C]-io)alkylamino₃ arylamino, diarylamino, arylalkylamino, amido, acylamido, hydroxyl (OH), oxo (=0), halo, carboxy, ester, acyl, acyloxy, CMO alkoxy, aryloxy, haloalkyl, sulfhydryl (i.e. thiol, - SH), C_Mo alkylthio, arylthio and sulfonyl. Typically it carries 0, 1 , 2 or 3 substituents.

As used herein the term acyl represents a group of formula: -C(=0)R, wherein R is an acyl substituent, for example, a substituted or unsubstituted C O alkyl group or a substituted or unsubstituted aryl group. Examples of acyl groups include, but are not limited to, -C(=0)CH₃ (acetyl), -C(=0)CH₂CH₃ (propionyl), -C(=0)C(CH₃)₃ (t-butyryl), and -C(=0)Ph (benzoyl, phenone).

As used herein the term acyloxy (or reverse ester) represents a group of formula: -OC(=0)R, wherein R is an acyloxy substituent, for example, substituted or unsubstituted CMO alkyl group or a substituted or unsubstituted aryl group, typically a Ci_-6 alkyl group. Examples of acyloxy groups include, but are not limited to, -OC(=0)CH₃ (acetoxy), -OC(=0)CH₂CH₃, -OC(=0)C(CH₃)₃, -0C(O)Ph, and -OC(=0)CH₂Ph.

As used herein the term ester (or carboxylate, carboxylic acid ester or oxycarbonyl) represents a group of formula: -C(=0)OR, wherein R is an ester substituent, for example, a substituted or unsubstituted C O alkyl group, or a substituted or unsubstituted aryl group (typically a phenyl group). Examples of ester groups include, but are not limited to, -C(=0)OCH₃, -C(=0)OC¾CH₃, -C(=0)OC(CH₃)₃, and -C(=0)OPh.

As used herein the term amino represents a group of formula -NH₂. The term C O alkylamino represents a group of formula -NHR^' wherein R^' is a CMO alkyl group, preferably a Ci-6 alkyl group, as defined previously. The term di(Ci-io)alkylamino represents a group of formula -NR^'R^" wherein R' and R" are the same or different and represent C O alkyl groups, preferably Ci.₆ alkyl groups, as defined previously. The term arylamino represents a group of formula -NHR' wherein R^' is an aryl group, preferably a phenyl group, as defined previously. The term diary lamino represents a group of formula -NR'R" wherein R' and R" are the same or different and represent aryl groups, preferably phenyl groups, as defined previously. The term arylalkylamino represents a group of formula -NR'R" wherein R' is a C O alkyl group, preferably a C_1-6 alkyl group, and R" is an aryl group, preferably a phenyl group.

A halo group is chlorine, fluorine, bromine or iodine (a chloro group, a fluoro group, a bromo group or an iodo group). It is typically chlorine, fluorine or bromine.

As used herein the term amido represents a group of formula: -C(=0)NR R , wherein R and R are independently amino substituents, as defined for di(C_1-1o)alkylamino groups. Examples of amido groups include, but are not limited to, -C(=0)NH₂,

-C(=0)NHCH₃, -C(=0)N(CH₃)₂, -C(=0)NHCH₂CH₃, and -C(=0)N(CH₂CH₃)₂, as well as amido groups in which R and R , together with the nitrogen atom to which they are attached, form a heterocyclic structure as in, for example, piperidinocarbonyl,

morpholinocarbonyl, thiomorpholinocarbonyl, and piperazinocarbonyl.

As used herein the term acylamido represents a group of formula: -NR' C(=0)R², wherein R¹ is an amide substituent, for example, hydrogen, a C_1-2oalkyl group, or an aryl group, preferably hydrogen or a C_1-2o alkyl group, and R² is an acyl substituent, for example, a Cj_-2o alkyl group or an aryl group, preferably a Ci_-2o alkyl group. Examples of acylamide groups include, but are not limited to, -NHC(=0)CH₃ , -NHC(=0)CH₂CH₃, -NHC(=0)Ph, -NHC(=0)C_I5H₃, and -NHC(=0)C₉H₁₉. Thus, a substituted C_M0 alkyl group may comprise an acylamido substituent defined by the formula -NHC(=O)-Ci.₁₀ alkyl, such as -NHC(=0)C₉H₁₉.

A Ci-io alkylthio group is a said Ci_-10 alkyl group, preferably a C_1-6 alkyl group, attached to a thio group. An arylthio group is an aryl group, preferably a phenyl group, attached to a thio group.

A CMO alkoxy group is a said substituted or unsubstituted CMO alkyl group attached to an oxygen atom. A Ci_-6 alkoxy group is a said substituted or unsubstituted Ci-6 alkyl group attached to an oxygen atom. A CM alkoxy group is a substituted or unsubstituted C alkyl group attached to an oxygen atom. Said Ci-₂o_> Ci_-6 and C alkyl groups are optionally interrupted as defined herein. Examples of CM alkoxy groups include, -OMe (methoxy), -OEt (ethoxy), -O(nPr) (n-propoxy), -O(iPr) (isopropoxy), -O(nBu) (n-butoxy), -O(sBu) (sec-butoxy), -O(iBu) (isobutoxy), and -O(tBu) (tert-butoxy). Further examples of Ci_-20 alkoxy groups are -O(Adamantyl), -0-CH₂-Adamantyl and -O- CH₂-CH₂-Adamantyl. An aryloxy group is a substituted or unsubstituted aryl group, as defined herein, attached to an oxygen atom. An example of an aryloxy group is -OPh (phenoxy).

Unless otherwise specified, included in the above are the well known ionic, salt, solvate, and protected forms of these substituents. For example, a reference to carboxylic acid or carboxyl group (-COOH) also includes the anionic (carboxylate) form (-COO^"), a salt or solvate thereof, as well as conventional protected forms. Similarly, a reference to an amino group includes the protonated form

a salt or solvate of the amino group, for example, a hydrochloride salt, as well as conventional protected forms of an amino group. Similarly, a reference to a hydroxyl group also includes the anionic form (- 0^'), a salt or solvate thereof, as well as conventional protected forms.

Unless otherwise specified, a reference to a particular compound also includes ionic, salt, solvated and protected forms.

As used herein the term "nanoparticle" means a microscopic particle whose size is measured in nanometres (nm). Typically, a nanoparticle has a particle size of from 0.5 to 1000 nm, from 1 nm to 1000 nm, or for instance from 0.5 nm to 800 nm or from 0.5 nm to 600 nm. Typically, a nanoparticle has a particle size of from 0.5 nm to 400 nm, or, for instance, from 0.5 nm to 200 nm, or from 1 nm to 200 nm. A nanoparticle may be crystalline or amorphous. The nonpassivated silicon nanoparticles referred to herein are typically crystalline. A nanoparticle may be spherical or non-spherical. Non-spherical nanoparticles may for instance be plate-shaped, needle-shaped or tubular. The term

"particle size" as used herein means the diameter of the particle if the particle is spherical or, if the particle is non-spherical, the volume-based particle size. The volume-based particle size is the diameter of the sphere that has the same volume as the non-spherical particle in question.

As used herein the term "d-block metal" means a metal of the d-block of the periodic table. As the skilled person will understand, this includes metals from Groups 3 to 12 of the periodic table, including metals from Group 3, namely, Sc, Y, Lu and Lr; metals from Group 4, namely Ti, Zr, Hf and Rf; metals from Groups 5 to 7; metals from Group 8, namely Fe, Ru, Os and Hs; metals from Group 9, namely Co, Rh, Ir and Mt; metals from Group 10, including Ni, Pd and Pt; metals from Group 1 1, including Cu, Ag and Au; and metals from Group 12, including Zn, Cd and Hg.

As used herein the term "a Group 8 d-block metal" means a metal from Group 8 of the periodic table, in the d-block. The Group 8 d-block metal is usually Fe, Ru or Os.

As used herein the term "a Group 9 d-block metal" means a metal from Group 9 of the periodic table, in the d-block. The Group 9 d-block metal is usually Co, Rh or Ir.

As used herein the term "a Group 10 d-block metal" means a metal from Group 10 of the periodic table, in the d-block. The Group 10 d-block metal is usually Ni, Pd or Pt.

As used herein the term "a Group 11 d-block metal" means a metal from Group 11 of the periodic table, in the d-block. The Group 11 d-block metal is usually Cu, Ag or Au.

As used herein the term "a Group 12 d-block metal" means a metal from Group 12 of the periodic table, in the d-block. The Group 12 d-block metal is usually Zn, Cd or Hg.

As used herein the term "a Group 1 metal" means a metal from Group 1 of the periodic table, i.e. an alkali metal. Typically, it is Li, Na, K, Rb or Cs.

As used herein the term "a Group 2 metal" means a metal from Group 2 of the periodic table, i.e. an alkaline earth metal. Typically, it is Be, Mg, Ca, Sr or Ba.

The invention provides a process for producing ¾ from a compound of formula (I), (II), (III) or (IV): R'COOH (I)

R²OH (II)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted CMO alkyl;

R² is unsubstituted or substituted CMO alkyl;

R³ is H or unsubstituted or substituted CMO alkyl;

R⁴ is H, unsubstituted or substituted CMO alkyl, or C(0)NR⁵R⁶;

R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted C j . _\ o alkyl ;

- (a) polymetallic nanoparticles, each of which comprises a first metal and a second metal, which first metal is selected from a Group 9, Group 10, Group 1 1 or Group 12 d-block metal, and which second metal is other than said first metal, wherein the polymetallic nanoparticles comprise a core which comprises said second metal and a shell surrounding said core, which shell comprises said first metal; or

provided that when said compound is a compound of formula (II), the catalyst comprises (a) said polymetallic nanoparticles.

Usually, when said compound is a compound of formula (II) or (III), the catalyst comprises (a) said polymetallic nanoparticles.

Typically, when said compound is a compound of formula (II), (III) or (IV), the catalyst comprises (a) said polymetallic nanoparticles.

In the process of the invention for producing ¾, H₂ is usually produced in the gaseous state. Thus, the process of the invention for producing H₂ is typically a process for producing hydrogen gas. During the process of the invention for producing H₂, the compound of formula (I), (II), (III) or (IV) is adsorbed on the catalyst surface. Hydrogen atoms are then cleaved from the adsorbed molecules, C-H bond cleavage generally being the rate-determining step. The adsorbed hydrogen atoms then recombine to form gaseous hydrogen. (When the reaction is performed in the presence of an electrode, e.g. a fuel cell electrode, the adsorbed hydrogen is instead converted into protons at the anode.)

Thus, H₂ is produced from a compound of formula (I), (II), (III) or (IV) as defined above.

Compounds of formula (I), i.e. compounds of formula R'COOH where R¹ is H or unsubstituted or substituted C_\.\ alkyl, undergo catalytic dehydrogenation in the processes of the invention, to produce ¾. Typically, C0₂ is produced as a by-product. Thus, when R¹ is H and the compound of formula (I) is formic acid (HCOOH), the catalytic dehydrogenation proceeds in accordance with the following reaction:

HCOOH -» C0₂ + H₂

Another possible decomposition reaction for formic acid is the dehydration of formic acid, which proceeds as follows:

HCOOH ^ CO + H₂0

However, the catalysts used in the process of the present invention catalyse the dehydrogenation reaction rather than the dehydration reaction, and as a result the product gases produced in the process of the present invention typically contain no more than 10 ppm CO. Thus, the product gas mixture usually comprises H₂, C0₂, and no more than 10 ppm CO. Typically the product gas mixture comprises less than 10 ppm CO.

R¹ in the compound of formula (I) is H or unsubstituted or substituted CMO alkyl, more typically H or unsubstituted or substituted Ci-₆ alkyl. More typically, R¹ is H or unsubstituted or substituted C alkyl. In one embodiment, when R¹ is a CMO alkyl group it is unsubstituted. Thus, R¹ in the compound of formula (I) may be unsubstituted C O alkyl, or for instance unsubstituted d-6 alkyl, or unsubstituted C alkyl. The compound of formula (I) may for instance be propanoic acid or acetic acid. Usually, R¹ in the compound of formula (I) is H, and the compound of formula (I) is formic acid.

Compounds of formula (II), i.e. compounds of formula R²OH where R² is unsubstituted or substituted C O alkyl, also undergo catalytic dehydrogenation in the processes of the invention, to produce ¾.

Usually, R in the compound of formula (II) is CMO alkyl, which C O alkyl is either unsubstituted or substituted with one, two or three hydroxyl groups. Thus, R in the compound of formula (II) may be unsubstituted CMO alkyl, or for instance unsubstituted C_1-6 alkyl, or unsubstituted CM alkyl. The compound of formula (II) may for instance be methanol, ethanol or propanol. Alternatively, R in the compound of formula (II) may be CMO ^kyl substituted with one, two or three hydroxyl groups, or for instance Ci_-6 alkyl substituted with one, two or three hydroxyl groups. In one embodiment, R in the compound of formula (II) is C_1-6 alkyl substituted with one or two hydroxyl groups, or for instance C alkyl substituted with one or two hydroxyl groups. Thus, the compound of formula (II) may be ethylene glycol or glycerol.

In one embodiment, the compound of formula (II) is methanol, ethanol or ethylene glycol.

Compounds of formula (III), i.e. compounds of formula R³CHO where R³ is H or unsubstituted or substituted C O alkyl, also undergo catalytic dehydrogenation the processes of the invention, to produce ¾.

Usually, R³ in the compound of formula (III) is H or unsubstituted or substituted C_1-6 alkyl; or for instance H or unsubstituted or substituted CM alkyl. Typically, it is H or unsubstituted C_1-6 alkyl. More typically it is H or unsubstituted C alkyl. The compound of formula (III) may for instance be formaldehyde, acetaldehyde or propanal

(propionaldehyde). Compounds of formula (IV), i.e. compounds of formula R⁴N¾ wherein R⁴ is H, unsubstituted or substituted CMO alkyl, or C(0)NR⁵R⁶, and wherein R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted Cj.jo alkyl, also undergo catalytic dehydrogenation in the processes of the invention, to produce H₂.

For instance, when R⁴ is H and the compound of formula (IV) is ammonia, the. catalytic dehydrogenation proceeds in accordance with the following reaction:

2NH₃ · N₂ + 3H₂

Usually, R⁴ in the compound of formula (IV) is H, unsubstituted or substituted C_\.e alkyl, or C(0)NR⁵R⁶; or for instance H, unsubstituted or substituted C alkyl, or

C(0)NR⁵R⁶. More typically, it is H, unsubstituted C_!-6 alkyl or C(0)NR⁵R⁶. Even more typically, it is H, unsubstituted CM alkyl or C(0)NR⁵R⁶. Usually, R⁴ is H or C(0)NR⁵R⁶.

R^s and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted

alkyl. Typically, R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted C_1-6 alkyl. More typically, R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted C alkyl. Usually, the alkyl group (Q.io alkyl, C\. 6 alkyl or C alkyl) is unsubstituted. More typically, R⁵ and R⁶ are both H.

Usually, R⁴ is H or C(0)NH₂ and the compound of formula (IV) is ammonia or urea. It may be ammonia, for instance.

Typically, the compound is a compound of formula (I) as defined above. In another embodiment, it is a compound of formula (II) as defined above. In another embodiment, it is a compound of formula (III) as defined above. In another embodiment, it is a compound of formula (IV) as defined above.

In one embodiment:

R¹ is H or C alkyl;

R² is C alkyl which is unsubstituted or substituted with one or two hydroxyl groups;

R³ is H or unsubstituted CM alkyl;

R⁴ is H, unsubstituted C_M alkyl or C(0)NR⁵R⁶; and

R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted CM alkyl.

Usually, the compound of formula (I) is selected from formic acid and acetic acid; the compound of formula (II) is selected from methanol, ethanol and ethylene glycol; the compound of formula (III) is selected from formaldehyde and acetaldehyde; and the compound of formula (IV) is selected from ammonia and urea.

Typically, the compound is a compound of formula (I) which is formic acid. Thus, typically in the process of the invention for producing ¾, said liquid phase comprises formic acid. In this embodiment, the catalyst is usually termed a formic acid

dehydrogenation catalyst.

In one embodiment, the catalyst used in the process of the invention for producing H₂ comprises polymetallic nanoparticles. The term "polymetallic nanoparticle" as used herein means a nanoparticle which comprises more than one metal, i.e. it comprises two or more metals. Typically, a polymetallic nanoparticle comprises only two metals, in which case the polymetallic nanoparticle is a bimetallic nanoparticle. However, in other cases, one or more further metals may be present too. A polymetallic nanoparticle may comprise an alloy of two or more metals. Alternatively, a polymetallic nanoparticle may have a "core-shell" structure. Thus, a polymetallic nanoparticle may have a central core, and a shell surrounding the core, wherein the shell comprises a first metal and the core comprises a second metal, which second metal is different from the first metal. The word

"surrounding" in this context means either completely surrounding or partially

surrounding. The word is therefore intended to cover polymetallic nanoparticles in which the shell completely surrounds the core, such that the metal atoms on the surface of the nanoparticle are atoms of the first metal. However, it is also intended to cover polymetallic nanoparticles in which the shell is incomplete, such that the majority of metal atoms on the surface of the nanoparticle are atoms of the first metal but some atoms of the second metal are also exposed. Preferably, the shell completely surrounds the core. As is discussed further below, the molar ratio of the first (shell) metal to the second (core) metal can be selected such that the shell completely surrounds the core and no core metal atoms are exposed.

The process of the invention for producing H₂ comprises contacting a liquid phase comprising said compound of formula (I), (II), (III) or (IV) with the catalyst. The term "liquid phase" in this context means that the compound of formula (I), (II), (III) or (IV) may be present in solution or as a neat liquid. Thus, in one embodiment, the liquid phase comprises said compound of formula (I), (II), (III) or (IV) as a neat liquid. Other components, e.g. additives may be present in said neat compound of formula (I), (II), (III) or (IV). Alternatively, the liquid phase may consist of said compound of formula (I), (II), (III) or (IV) as a neat liquid. In another embodiment, the liquid phase comprises said compound of formula (I), (II), (III) or (IV) and a solvent. Any suitable solvent may be used. Typically, however, the solvent comprises a polar protic solvent. For instance, the solvent may comprise water or an alcohol.

The inventors have shown that, in the process of the invention for producing ¾, the catalyst can advantageously be used to generate hydrogen from the compound of formula (I), (II), (III) or (IV) at room temperature. However, higher temperatures may also be used. Typically, therefore, the step of contacting the liquid phase comprising said compound of formula (I), (II), (III) or (IV) with the catalyst is performed at a temperature which does not exceed 100 °C. The contacting may for instance be performed at a temperature of from 0 °C to 100 °C. Lower temperatures may however be employed. Thus, said contacting may for instance be performed at a temperature which does not exceed 60 °C. Usually, the contacting step is performed at a temperature of from 0 °C to 60 °C or for instance at a temperature of from 0 °C to 50 °C. In one embodiment, said contacting is performed at a temperature which does not exceed 40 °C. Typically, in this embodiment, the contacting step is performed at a temperature of from 0 °C to 40 °C or for instance at a temperature of from 0 °C to 30 °C.

In one embodiment, the catalyst comprises (b) said nanoparticles which comprise a metal selected from palladium, rhodium, ruthenium, iridium, copper and silver. These nanoparticles may be monometallic nanoparticles, i.e. nanoparticles in which the only metal present is the metal selected from palladium, rhodium, ruthenium, iridium, copper and silver. Alternatively, the nanoparticles may be polymetallic, i.e. they may comprise said metal selected from palladium, rhodium, ruthenium, iridium, copper and silver and a further metal. In one embodiment, the nanoparticles which comprise a metal selected from palladium, rhodium, ruthenium, iridium, copper and silver are monometallic nanoparticles.

Usually the nanoparticles (b) comprise a metal selected from palladium, rhodium, ruthenium and silver.

In another embodiment, the catalyst comprises said nanoparticles (b) which comprise a metal selected from rhodium, ruthenium, iridium and copper. These

nanoparticles may be monometallic nanoparticles, i.e. nanoparticles in which the only metal present is the metal selected from rhodium, ruthenium, iridium and copper.

Alternatively, the nanoparticles may be polymetallic, i.e. they may comprise said metal selected from rhodium, ruthenium, iridium and copper and a further metal. In one embodiment, the nanoparticles which comprise a metal selected from rhodium, ruthenium, iridium and copper are monometallic nanoparticles. Said nanoparticles may have a mean particle size of less than or equal to 400 nm. For instance said nanoparticles may have a mean particle size of from 0.5 nm to 400 nm, or for instance from 1 nm to 400 nm. Typically, the particle size distribution of the nanoparticles is such that 90 % of the particles have a particle size of less than 400 nm.

More typically, said nanoparticles have a mean particle size of less than or equal to

200 nm. For instance said nanoparticles may have a mean particle size of from 0.5 nm to 200 nm, or for instance from 1 nm to 200 nm. Typically, the particle size distribution of the nanoparticles is such that 90 % of the particles have a particle size of less than 200 nm.

Even more typically, said nanoparticles have a mean particle size of less than or equal to 100 nm. For instance said nanoparticles may have a mean particle size of from 0.5 nm to 100 nm, or for instance from 1 nm to 100 nm. Typically, the particle size

distribution of the nanoparticles is such that 90 % of the particles have a particle size of less than 100 nm.

Usually, said nanoparticles have a mean particle size of less than or equal to 60 nm. For instance said nanoparticles may have a mean particle size of from 0.5 nm to 60 nm, or for instance from 1 nm to 60 nm. Typically, the particle size distribution of the

nanoparticles is such that 90 % of the particles have a particle size of less than 60 nm.

In one embodiment, said nanoparticles have a mean particle size of less than or equal to 50 nm. For instance said nanoparticles may have a mean particle size of from 0.5 nm to 50 nm, or for instance from 1 nm to 50 nm. Typically, the particle size distribution of the nanoparticles is such that 90 % of the particles have a particle size of less than 50 nm.

In another embodiment, said nanoparticles have a mean particle size of less than or equal to 20 nm. For instance said nanoparticles may have a mean particle size of from 0.5 nm to 20 nm, or for instance from 1 nm to 20 nm. Typically, the particle size distribution of the nanoparticles is such that 90 % of the particles have a particle size of less than 20 nm.

In yet another embodiment, said nanoparticles have a mean particle size of less than or equal to 15 nm. For instance said nanoparticles may have a mean particle size of from 0.5 nm to 15 nm, or for instance from 1 nm to 15 nm. Typically, the particle size distribution of the nanoparticles is such that 90 % of the particles have a particle size of less than 15 nm.

Usually, said nanoparticles comprise a metal selected from palladium, rhodium and ruthenium. The nanoparticles may for instance comprise a metal selected from palladium and rhodium.

Alternatively, the nanoparticles (b) may comprise a metal selected from ruthenium and rhodium.

Rhodium is a preferred metal. Thus, typically, the nanoparticles (b) comprise rhodium.

Palladium is also preferred. Thus, the nanoparticles (b) may comprise palladium. Typically, the nanoparticles have a mean particle size of less than or equal to 10 nm. They may for instance have a mean particle size of less than or equal to 5 nm, or even less than or equal to 3 nm. Thus, the nanoparticles may have a mean particle size of from 0.5 nm to 15 nm, or for instance from 1 nm to 15 nm. Typically, the particle size distribution of the nanoparticles is such that 90 % of the particles have a particle size of less than 15 nm. Alternatively, the nanoparticles may have a mean particle size of from 0.5 nm to 5 nm, or for instance from 0.5 nm to 3 nm. Usually, in this embodiment, the particle size distribution of the nanoparticles is such that 90 % of the particles have a particle size of less than 5 nm or less than 3 nm, as the case may be.

Usually, in the process of the invention for producing ¾, the catalyst comprises (a) said polymetallic nanoparticles. In this aspect, the polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding said core, which shell comprises said first metal. "Surrounding" in this context means completely surrounding, so that none of the core atoms are exposed, or partially surrounding, so that some of the core atoms are exposed. Preferably, however, the shell completely surrounds the core such that none of the core atoms are exposed.

Typically, said first metal is selected from palladium, platinum, rhodium or iridium. More typically, the first metal is palladium.

Usually, the work function of the second metal is less than the work function of the first metal. This is advantageous in core-shell type polymetallic nanoparticles, because the lower work function of the second (core) metal compared to the first (shell) metal causes charge transfer from the core (lower work function) to the shell (higher work function). This in turn causes stronger adsorption of the compound of fomula (I), (II), (III), or (IV) onto the surface of the nanoparticle. This is thought to be due to increased back-bonding from the metal d-orbitals into the pi orbitals of the adsorbed molecule. Indeed, the inventors have observed increased absorption of formate onto the metal surface when core- shell nanoparticles have been used in which the work function of the second (core) metal is less than the work function of the first (shell) metal. "Work function" here means the work function of the (1 11) crystal plane for metals having a face centred cubic (fee) lattice structure, or the work function of the (001) crystal plane for metals having a hexagonal closest packed (hep) lattice structure.

When Pd is used as the first (shell) metal, the second (core) metal typically has a work function which is lower than the work function of Pd, i.e. less than 5.6 eV.

Typically, therefore, in the process of the invention for producing ¾, the work function of the second metal is less than 5.6 eV. More typically, it is less than or equal to 5.3 eV. In a preferred embodiment, it is less than or equal to 5.0 eV.

Typically, in the process of the invention for producing H₂, the second metal is selected from: a d-block metal of any one of Groups 8, 9, 10, 1 1 and 12, a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, provided that said second metal is other than said first metal. More typically, the second metal is selected from copper, silver, gold, nickel, palladium, platinum, cobalt, rhodium, iridium, ruthenium, iron, osmium, zinc, cadmium, lithium, sodium, potassium, rubidium, caesium, beryllium, magnesium, calcium, strontium and barium. The second metal may for instance be selected from copper, silver, gold, nickel, palladium, platinum, cobalt, rhodium, iridium and ruthenium. Usually, the second metal is selected from copper, silver, gold, platinum, rhodium and ruthenium. More typically, the second metal is silver or rhodium. In one embodiment, the second metal is silver.

Typically, in the process of the invention for producing H₂, the second metal is a d- block metal of any one of Groups 8, 9, 10, 11 and 12, a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, provided that the second metal is other than palladium, platinum, gold, iron, cobalt and nickel.

Usually, the second metal is other than palladium, platinum, gold, iron, cobalt, nickel, titanium, tungsten, tantalum, vanadium and niobium.

Typically, said second metal is selected from silver, rhodium, ruthenium, copper and iridium. More typically, said second metal is silver, rhodium, ruthenium or copper. For instance, said second metal may be silver, rhodium or copper. Preferably, it is silver or copper. Alternatively, said second metal may be selected from silver, rhodium and ruthenium.

Usually, the second metal is silver.

The second metal may however be selected from rhodium and copper. Thus, for instance, the second metal may be rhodium. In some embodiments, the second metal is copper.

Usually, however, the second metal is silver.

Usually, the first metal is palladium. Thus, typically, in the process of the invention for producing H₂, said catalyst comprises said polymetallic nanoparticles, which nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal which is palladium.

Typically, when polymetallic nanoparticles are employed as the catalyst in the process of the invention for producing ¾, the polymetallic nanoparticles have a mean particle size of less than or equal to 50 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 50 nm, or for instance from 1 nm to 50 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the particles have a particle size of less than 50 nm. More typically, said polymetallic nanoparticles have a mean particle size of less than or equal to 40 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 40 nm, or for instance from 1 nm to 40 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the particles have a particle size of less than 40 nm.

In another embodiment, said polymetallic nanoparticles have a mean particle size of less than or equal to 30 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 30 nm, or for instance from 1 nm to 30 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 30 nm. More typically, said polymetallic nanoparticles have a mean particle size of less than or equal to 25 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 25 nm, or for instance from 1 nm to 25 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 25 nm. Even more typically, said polymetallic nanoparticles have a mean particle size^*of less than or equal to 20 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 20 nm, or for instance from 1 nm to 20 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 20 nm.

In another embodiment, said polymetallic nanoparticles have a mean particle size of less than or equal to 15 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 15 nm, or for instance from 1 nm to 15 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 15 run. More typically, said polymetallic nanoparticles have a mean particle size of less than or equal to 5 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 5 nm, or for instance from 1 nm to 5 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 5 nm.

The molar ratio of the second metal to the first metal in said polymetallic nanoparticles may for instance be from 3:1 to 1 :3. However, molar ratios closer to 1:1 are often advantageous, resulting in the benefits associated with full coverage of the core by the shell (so that no core atoms are exposed) whilst at the same time the shell is thin enough for effective charge transfer from the core to the outer surface of the polymetallic nanoparticle (the metal in the shell typically having a higher work function than the metal in the core).

Usually, therefore, when polymetallic nanoparticles are employed as the catalyst in the process of the invention for producing ¾, the molar ratio of the second metal to the first metal in said polymetallic nanoparticles is from 2.5:1 to 1 :2.5. More typically, it is from 2:1 to 1:2. Thus, the molar ratio of the second metal to the first metal in said polymetallic nanoparticles is from 1.5:1 to 1 :1.5. More typically, the ratio is from 1.2:1 to 1 :1.2. Usually, said molar ratio of the second metal to the first metal in said polymetallic nanoparticles is about 1 : 1.

Typically, in the process of the invention for producing H₂, said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is silver.

Typically, these polymetallic palladium/silver nanoparticles have a mean particle size of less than or equal to 35 nm, preferably less than or equal to 25 nm, more preferably less than or equal to 20 nm. More typically, they have a mean particle size of from 1 nm to 35 nm, preferably from 10 run to 25 nm, more preferably from 17 to 20 nm. Usually, the polymetallic nanoparticles have a mean particle size of from 1 nm to 20 nm, preferably from 10 nm to 20 nm, or in some cases from 17.0 to 20.0 nm.

Typically, the standard deviation from said mean is less than or equal to 5.0 nm. More typically, it is less than or equal to 3.0 nm. In one embodiment, the polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is silver, and the polymetallic nanoparticles have a mean particle size of from 17.0 to 20.0 nm, wherein the standard deviation from said mean is less than or equal to 3.0 nm.

Typically, the silver cores of said polymetallic nanoparticles have a mean particle size of from 10.0 nm to 17.0 nm, preferably from 13.0 nm to 17.0 nm, more preferably about 15.0 nm. Typically, the standard deviation from said mean is less than or equal to 3.5 nm. More typically, it is less than or equal to 1.5 nm. In another embodiment, the silver cores of said polymetallic nanoparticles have a mean particle size of about 15.0 nm, wherein the standard deviation from said mean is less than or equal to 1.5 nm.

Typically, the molar ratio of the silver to the palladium in said polymetallic nanoparticles is from 1.5: 1 to 1 :1.5. Preferably it is about 1 :1.

In another embodiment of the process of the invention for producing H₂ said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is rhodium.

Typically, in this embodiment, the polymetallic nanoparticles have a mean particle size of less than or equal to 4 nm. Thus, for instance, the polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 4 nm. Typically, the standard deviation from said mean is less than or equal to 1.0 nm. More typically, it is less than or equal to 0.5 nm. Thus, for instance, the polymetallic nanoparticles have a mean particle size of from 2.3 to 3.5 nm, wherein the standard deviation from said mean is less than or equal to 0.5 nm.

Typically, the rhodium cores of said polymetallic nanoparticles have a mean particle size of about 0.5 to 2.5 nm, preferably from 1.5 to 2.5 nm, more preferably about 2.0 nm. Typically, the standard deviation from said mean is less than or equal to 0.5 nm, or for instance less than or equal to about 0.4 nm. More typically, it is less than or equal to 0.2 nm. More typically, the rhodium cores of said polymetallic nanoparticles have a mean particle size of about 2.0 nm, wherein the standard deviation from said mean is less than or equal to 0.2 nm.

Typically, the molar ratio of the rhodium to the palladium in said polymetallic nanoparticles is from 1.5: 1 to 1 : 1.5, preferably wherein said molar ratio is about 1 : 1. In another embodiment of the process of the invention for producing H₂ said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is ruthenium. Typically, these polymetallic nanoparticles have a mean particle size of less than or equal to 5 nm. More typically, the polymetallic nanoparticles have a mean particle size of from 0.5 nm to 5 nm. The polymetallic nanoparticles may for instance have a mean particle size of from 2.6 to 4.0 nm. Typically, the standard deviation from said mean is less than or equal to 2.0 nm. More typically, it is less than or equal to 1.0 nm, for instance less than or equal to 0.6 nm. Thus, in one embodiment the polymetallic nanoparticles have a mean particle size of from 2.6 to 4.0 nm, wherein the standard deviation from said mean is less than or equal to 0.6 nm.

The ruthenium cores of said polymetallic nanoparticles may have a mean particle size of about 1.0 to 2.6 nm. Typically, the standard deviation from said mean is less than or equal to 1.0 nm, more typically less than or equal to 0.5 nm, for instance less than or equal to 0.2 nm. Thus, for instance, the ruthenium cores of said polymetallic nanoparticles may have a mean particle size of about 2.3 nm, wherein the standard deviation from said mean is less than or equal to 0.2 nm.

Typically, the molar ratio of the ruthenium to the palladium in said polymetallic nanoparticles is from 1.5: 1 to 1 : 1.5, preferably wherein said molar ratio is about 1 : 1.

In another embodiment of the process of the invention for producing H₂ said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is gold. Typically, the polymetallic nanoparticles have a mean particle size of less than or equal to 15 nm. The polymetallic nanoparticles may for instance have a mean particle size of from 1.0 to 15.0 nm. Typically, the standard deviation from said mean is less than or equal to 2.5 nm. More typically,- it is less than or equal to 1.5 nm. The polymetallic nanoparticles may for instance have a mean particle size of from 8.0 to 12.0 nm, wherein the standard deviation from said mean is less than or equal to 1.5 nm.

Typically the molar ratio of the gold to the palladium in said polymetallic nanoparticles is from 1.5 : 1 to 1 : 1.5, preferably wherein said molar ratio is about 1 : 1.

In another embodiment of the process of the invention for producing ¾, said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is platinum. In one embodiment, these polymetallic nanoparticles have a mean particle size of less than or equal to 15 nm. Thus, the polymetallic nanoparticles may have a mean particle size of from 1 nm to 15 nm. Typically, the standard deviation from said mean is less than or equal to 4.5 nm. More typically, it is less than or equal to 2.5 nm. Thus, the polymetallic nanoparticles may have a mean particle size of from 8.0 to 12.0 nm, wherein the standard deviation from said mean is less than or equal to 2.5 nm.

Typically, the platinum cores of said nanoparticles have a mean particle size of from 2.0 to 8.0 nm, preferably from 3.0 to 7.0 nm, more preferably from 5.0 to 6.0 nm. Typically, the standard deviation from said mean is less than or equal to 3.5 nm. More typically, it is less than or equal to 2.5 nm, for instance less than or equal to 1.5 nm. Thus, in one embodiment, the platinum cores of said nanoparticles have a mean particle size of about 5.0 to 6.0 nm, wherein the standard deviation from said mean is less than or equal to 1.5 nm. Typically, the molar ratio of the platinum to the palladium in said polymetallic nanoparticles is from 1.5: 1 to 1 :1.5, preferably wherein said molar ratio is about 1 : 1.

In another embodiment of the process of the invention for producing ¾, said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise an alloy comprising said first metal and said second metal, wherein said first metal is silver or platinum, and said second metal is palladium. Typically, these polymetallic nanoparticles have a mean particle size of less than or equal to 25 nm, preferably less than or equal to 15 nm, more preferably less than or equal to 10 nm.

Typically, the polymetallic nanoparticles have a mean particle size of from 1 nm to 25 nm, preferably from 1 nm to 1 nm, more preferably from 1 nm to 10 nm.

The catalyst used in the process of the invention for producing H₂ typically further comprises a polymer. The polymer is typically referred to as a stabilising polymer. The polymer may for instance be polyvinyl pyrrolidone) (PVP), polyvinyl alcohol) (PVA) or poly(ethylenimine) (PEI)-It has been found that the presence of certain polymers can improve the activity of decomposition of a compound of formula (I), (II), (III) or (IV) (typically formic acid), to produce hydrogen. In particular, the presence of certain polymers can improve the rate and total volume of hydrogen production at lower temperatures such as room temperature. One category of polymer that has been particularly useful in this regard is a polymer bearing pendant amine groups. Poly(ethylenimine) (PEI) is an example of such a polymer. In one embodiment, the catalyst used in the process of the invention for producing H₂ further comprises a polymer. Typically, said polymer is a polymer which comprises a plurality of amino, Ci.io alkylamino and/or di(C_1-jo)alkylamino groups. Typically the polymer is poly(ethylenimine) (PEI).

The catalyst used in the process of the invention for producing H₂ may or may not further comprise a solid support material, in addition to said nanoparticles or said polymetallic nanoparticles, wherein the nanoparticles or polymetallic nanoparticles are supported on said support material. Any suitable support material may be used, for instance the solid support material may comprise carbon, alumina or titania. Typically, however, it comprises carbon.

In one embodiment, the support comprises an oxide, a nitride, carbon or nanotubes.

The nanotubes are usually carbon nanotubes. The oxide is typically a metal oxide. The nitride is typically a metal nitride.

The nanoparticles or polymetallic nanoparticles are typically present in an amount of from 10 to 30 weight %, based on the total weight of the catalyst including the support, preferably in an amount of from 15 to 25 weight %, based on the total weight of the catalyst including the support.

In other embodiments, the catalyst does not comprise a solid support material.

Thus, the nanoparticles or polymetallic nanoparticles may be unsupported. In such embodiments, the nanoparticles or polymetallic nanoparticles may be present as a suspension, for instance a colloidal suspension, in the liquid phase employed in the process of the invention. The nanoparticles or polymetallic nanoparticles may for instance initially be present in the solid state, and subsequently contacted with the liquid phase, to form a colloidal suspension of the nanoparticles or polymetallic nanoparticles in the liquid phase.

Or, for instance, a suspension, typically a colloidal suspension, of the nanoparticles or polymetallic nanoparticles in a solvent may be contacted with said liquid phase, resulting in a colloidal suspension of said nanoparticles or said polymetallic nanoparticles in said liquid phase.

Thus in one embodiment, said step of contacting of said liquid phase with said catalyst results in a colloidal suspension of said nanoparticles or said polymetallic nanoparticles in said liquid phase.

In some embodiments, the liquid phase does not comprise a solvent, but comprises a neat liquid which is the compound of formula (I), (II), (III) or (IV). Thus, the liquid phase may comprise neat formic acid. In some embodiments, the liquid phase consists of, or consists essentially of, the compound of formula (I), (II), (III) or (IV). Thus, the liquid phase may consist of, or consist essentially of, formic acid.

In other embodiments, the liquid phase comprises said compound of formula (I), (II), (III) or (IV) and a solvent. Thus, the liquid phase may comprise formic acid and a solvent. Any suitable solvent may be used. Typically the solvent comprises a polar protic solvent, for instance an alcohol or water. Typically the solvent comprises water.

The concentration of said compound of formula (I), (II), (III) or (IV) in said liquid phase is typically from 0.01 M to 26.52 M, more typically from 0.1 M to 10.0 M. Thus, the concentration of said compound of formula (I), (II), (III) or (IV) may be from 0.01 M to 20.0 M, or for instance from 0.5 M to 10.0 M. In another embodiment the concentration of the compound in the liquid phase is from 0.2 M to 5.0 M, from 0.3 M to 5.0 M. More typically, it is from 0.5 M to 5.0 M, or for instance from 0.5 to 4.0 M.

In one embodiment, the concentration of the compound of formula (I), (II), (III) or (IV) in said liquid phase is from 0.3 M to 6.0 M, or for instance from 0.3 M to 5.0 M. In other embodiments, however, it is from 0.4 M to 3.0 M, of for instance from 0.5 M to 2.5 M.

In some embodiments, the concentration of the compound of formula (I), (II), (III) or (IV) in said liquid phase is from 0.5 M to 1.6 M. More typically, in these embodiments, it is from 0.6 M to 1.4 M, or for instance from 0.7 M to 1.3 M. Usually, it is from 0.8 M to 1.2 M, or for instance from 0.9 M to 1.1 M, i.e. about 1 M. Typically, in these

embodiments, the compound of formula (I), (II), (III) or (IV) is formic acid. The catalyst may be any of the catalysts defined above. More typically, however, it is a catalyst which comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises silver and a shell surrounding the core, which shell comprises palladium. The hydrogen generation activity of such systems is shown in Fig. 26 filed herewith. ₅

Typically in the process of the invention for producing H₂, the concentration of said catalyst in the liquid phase is less than or equal to 0.01 M, preferably less than or equal to 0.005 M, more preferably less than or equal to 0.0005 M.

The process of the invention for producing ¾ may further comprise recovering said H₂. Typically, the process of the invention produces a mixture of gases comprising hydrogen gas, for instance a mixture of ¾ and C0 . The step of recovering said H₂ typically therefore comprises collecting the product gas mixture and separating the ¾ from said mixture. The separation may be effected by any suitable method known in the art, for instance by using a filter material which selectively retains contaminants and lets the hydrogen pass through. The separated ¾ gas may also for instance be compressed and/or stored for later use.

Usually, the compound of formula (I), (II), (III) or (IV) is formic acid. In some embodiments of the process of the invention for producing ¾, said compound of formula

(I), (II), (III) or (IV) is formic acid and said contacting results in the production of a gaseous product mixture comprising H₂ and C0₂. Usually, the gaseous product mixture comprises a 1 :1 molar ratio of H₂ and C0₂. Typically, the gaseous product mixture comprises no more than lOppm by volume carbon monoxide.

Typically in the process of the invention for producing H₂, said compound of formula (I), (II), (III) or (IV) is formic acid and the production of ¾ from formic acid occurs in a single step, in accordance with the following reaction:

HCOOH -» C0₂ + H₂

and substantially without any dehydration of formic acid in accordance with the following reaction:

HCOOH - CO + H₂0

In another aspect, the present invention provides a process for producing H₂ from a compound of formula (I), (II), (III) or (IV): R'COOH (I)

R²OH (II)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted CMO alkyl;

R² is unsubstituted or substituted Q.io alkyl;

R³ is H or unsubstituted or substituted CMO alkyl;

R⁴ is H, unsubstituted or substituted Ci_-!0 alkyl, or C(0)NR⁵R⁶;

R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted CMO alkyl;

- nanoparticles which comprise a metal selected from palladium, rhodium,

ruthenium, iridium, copper and silver; or - polymetallic nanoparticles, each of which comprises a first metal and a second metal, which first metal is selected from a Group 9, Group 10, Group 1 1 or Group 12 d-block metal, and which second metal is other than said first metal.

In one embodiment of this process, the polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding said core, which shell comprises said first metal. As discussed above, "surrounding" in this context means completely surrounding, so that none of the core atoms are exposed, or partially surrounding, so that some of the core atoms are exposed. Preferably, however, the shell completely surrounds the core such that none of the core atoms are exposed. Alternatively, the polymetallic nanoparticles may comprise an alloy comprising said first metal and said second metal.

This process may be as further defined hereinbefore on pages 13 to 27. The catalyst used in this process may also be as further defined anywhere herein for the processes or the catalysts of the invention.

Some of the catalysts used in the process of the invention for producing H₂ are novel. Thus, the invention provides a catalyst for the production of hydrogen from a compound of formula (I), (II), (III) or (IV), or for the electro-oxidation of a compound of formula (I), (II), (III) or (IV):

R'COOH (I)

R²OH (II)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted d-io alkyl;

R² is unsubstituted or substituted C_1-10 alkyl;

R³ is H or unsubstituted or substituted C O alkyl;

R⁴ is H, unsubstituted or substituted C_M0 alkyl, or C(0)NR⁵R⁶;

which catalyst comprises polymetallic nanoparticles, which polymetallic nanoparticles comprise a core and a shell surrounding the core; wherein the shell comprises a first metal which is palladium, and the core comprises a second metal, wherein the second metal is other than palladium, platinum, gold, iron, cobalt, nickel, titanium, tungsten, tantalum, vanadium and niobium. In one embodiment, the catalyst of the invention is for the production of formic acid or for the electro-oxidation of formic acid.

Typically, the second metal in the catalyst of the invention is a d-block metal of any one of Groups 8, 9, 10, 11 and 12, a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, provided that the second metal is other than palladium, platinum, gold, iron, cobalt and nickel. More typically, the second metal is a said d-block metal of any one of Groups 8, 9, 10, 11 and 12.

Usually, the work function of the second metal is less than the work function of the first metal, palladium. This is advantageous in core-shell type polymetallic nanoparticles, because the lower work function of the second (core) metal compared to the palladium shell causes charge transfer from the core to the shell. This in turn causes stronger adsorption of the compound of fomula (I), (II), (III), or (IV) onto the surface of the nanoparticle. This is thought to be due to increased back-bonding from the metal d-orbitals into the pi orbitals of the adsorbed molecule. Indeed, the inventors have observed increased absorption of formate onto the metal surface when core-shell nanoparticles have been used instead of pure palladium nanoparticles, in which the work function of the second (core) metal is less than the work function of palladium. "Work function" here means the work function of the (11 1) crystal plane for metals having a face centred cubic (fee) lattice structure, or the work function of the (001) crystal plane for metals having a hexagonal closest packed (hep) lattice structure.

Thus the second (core) metal typically has a work function which is lower than the work function of Pd, i.e. less than 5.6 eV. Typically, therefore, in catalysts of the invention, the work function of the second metal is less than 5.6 eV. More typically, it is less than or equal to 5.3 eV. In a preferred embodiment, it is less than or equal to 5.0 eV.

Thus, in the catalysts of the invention, the second metal is an electron-rich

(electropositive) metal, such as a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, or a metal on the right hand side of the d-block: a d-block metal of any one of Groups 8, 9, 10; 1 1 and 12 provided that the second metal is other than palladium, platinum, gold, iron, cobalt and nickel.

Typically, said second metal is selected from silver, rhodium, ruthenium, copper, iridium, osmium, zinc, cadmium, lithium, sodium, potassium, rubidium, caesium, beryllium, magnesium, calcium, strontium and barium. More typically, said second metal is selected from silver, rhodium, ruthenium, copper and iridium. In one embodiment, the second metal is selected from copper, silver, rhodium and ruthenium. Thus, the second metal may for instance be silver or rhodium. In a preferred embodiment, the second metal is silver.

Said second metal may for instance be silver, rhodium or copper. Preferably, it is silver or copper. Alternatively, said second metal may be selected from silver, rhodium and ruthenium.

Usually, the second metal is silver.

The second metal may however be selected from rhodium and copper. Thus, for instance, the second metal may be rhodium.

In some embodiments, the second metal is copper.

Typically, the polymetallic nanoparticles of the catalyst of the invention have a mean particle size of less than or equal to 50 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 50 nm, or for instance from 1 nm to 50 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the particles have a particle size of less than 50 nm. More typically, said polymetallic nanoparticles have a mean particle size of less than or equal to 40 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 40 nm, or for instance from 1 nm to 40 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the particles have a particle size of less than 40 nm.

In another embodiment, said polymetallic nanoparticles of the catalyst of the invention have a mean particle size of less than or equal to 30 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 30 nm, or for instance from 1 nm to 30 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 30 nm. More typically, said polymetallic nanoparticles of the catalyst of the invention have a mean particle size of less than or equal to 25 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 25 nm, or for instance from " ^" ' 1 nm to 25 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 25 nm. Even more typically, said polymetallic nanoparticles have a mean particle size of less than or equal to 20 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 20 nm, or for instance from 1 nm to 20 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 20 nm. In another embodiment, said polymetallic nanoparticles have a mean particle size of less than or equal to 15 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 15 nm, or for instance from 1 nm to 15 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 15 nm. More typically, said polymetallic nanoparticles have a mean particle size of less than or equal to 5 nm. For instance said polymetallic nanoparticles may have a mean particle size of from 0.5 nm to 5 nm, or for instance from 1 nm to 5 nm. Typically, the particle size distribution of the polymetallic nanoparticles is such that 90 % of the polymetallic nanoparticles have a particle size of less than 5 nm.

The molar ratio of the second metal to the first metal in said polymetallic nanoparticles of the catalyst of the invention may for instance be from 3:1 to 1 :3. However, molar ratios closer to 1 : 1 are often advantageous, resulting in the benefits associated with full coverage of the core by the shell (so that no core atoms are exposed) whilst at the same time the shell is thin enough for effective charge transfer from the core to the outer surface of the polymetallic nanoparticle (the metal in the shell typically having a higher work function than the metal in the core). Usually, therefore, the molar ratio of the second metal to the first metal in said polymetallic nanoparticles is from 2.5:1 to 1 :2.5. More typically, it is from 2:1 to 1 :2. Thus, the molar ratio of the second metal to the first metal in said polymetallic nanoparticles is from 1.5: 1 to 1 : 1.5. More typically, the ratio is from 1.2: 1 to 1:1.2. Usually, said molar ratio of the second metal to the first metal in said polymetallic nanoparticles is about 1 :1.

Typically, in the catalyst of the invention, the second metal is silver. Typically, these polymetallic palladium/silver nanoparticles have a mean particle size of less than or equal to 35 nm, preferably less than or equal to 25 nm, more preferably less than or equal to 20 nm. More typically, they have a mean particle size of from 1 nm to 35 nm, preferably from 10 nm to 25 nm, more preferably from 17 to 20 nm. Usually, the polymetallic nanoparticles have a mean particle size of from 1 nm to 20 nm, preferably from 10 nm to 20 nm, or in some cases from 17.0 to 20.0 nm.

Typically, the standard deviation from said mean is less than or equal to 5.0 nm.

More typically, it is less than or equal to 3.0 nm. In one embodiment, the polymetallic nanoparticles have a mean particle size of from 17.0 to 20.0 nm, wherein the standard deviation from said mean is less than or equal to 3.0 nm. Typically, the silver cores of said polymetallic nanoparticles have a mean particle size of from 10.0 nm to 17.0 nm, preferably from 13.0 nm to 17.0 nm, more preferably about 15.0 nm. Typically, the standard deviation from said mean is less than or equal to 3.5 nm. More typically, it is less than or equal to 1.5 nm. In another embodiment, the silver cores of said polymetallic nanoparticles have a mean particle size of about 15.0 nm, wherein the standard deviation from said mean is less than or equal to 1.5 nm.

Typically, the molar ratio of the silver to the palladium in said polymetallic nanoparticles is from 1.5:1 to 1:1.5. Preferably it is about 1 :1.

In another embodiment of the catalyst of the invention, the second metal is rhodium. Typically, in this embodiment, the polymetallic nanoparticles have a mean particle size of less than or equal to 4 nm. Thus, for instance, the polymetallic

nanoparticles may have a mean particle size of from 0.5 nm to 4 nm. Typically, the standard deviation from said mean is less than or equal to 1.0 nm. More typically, it is less than or equal to 0.5 nm. Thus, for instance, the polymetallic nanoparticles may have a mean particle size of from 2.3 to 3.5 nm, wherein the standard deviation from said mean is less than or equal to 0.5 nm.

Typically, the molar ratio of the rhodium to the palladium in said polymetallic nanoparticles is from 1.5: 1 to 1 : 1.5, preferably wherein said molar ratio is about 1 : 1.

In another embodiment of the catalyst of the invention, the second metal is ruthenium. Typically, in this embodiment, the polymetallic nanoparticles have a mean particle size of less than or equal to 5 rim. More typically, the polymetallic nanoparticles have a mean particle size of from 0.5 nm to 5 nm. The polymetallic nanoparticles may for instance have a mean particle size of from 2.6 to 4.0 nm. Typically, the standard deviation from said mean is less than or equal to 2.0 nm. More typically, it is less than or equal to 1.0 nm, for instance less than or equal to 0.6 nm. Thus, in one embodiment the polymetallic nanoparticles have a mean particle size of from 2.6 to 4.0 nm, wherein the standard deviation from said mean is less than or equal to 0.6 nm. The ruthenium cores of said polymetallic nanoparticles may have a mean particle size of about 1.0 to 2.6 nm. Typically, the standard deviation from said mean is less than or equal to 1.0 nm, more typically less than or equal to 0.5 nm, for instance less than or equal to 0.2 nm. Thus, for instance, the ruthenium cores of said polymetallic nanoparticles may have a mean particle size of about 2.3 nm, wherein the standard deviation from said mean is less than or equal to 0.2 nm.

Typically, the molar ratio of the ruthenium to the palladium in said polymetallic nanoparticles is from 1.5: 1 to 1 : 1.5, preferably wherein said molar ratio is about 1 :1.

In some embodiments, the catalyst of the invention further comprises a polymer. The polymer is typically refened to as a stabilising polymer. The polymer may for instance be polyvinyl pyrrolidone) (PVP), poly( vinyl alcohol) (PVA) or poly(ethylenimine) (PEI).

It has been found that the presence of certain polymers can improve the activity of decomposition of a compound of formula (I), (II), (III) or (IV) (typically formic acid), to produce hydrogen. In particular, the presence of certain polymers can improve the rate and total volume of hydrogen production at lower temperatures such as room temperature. One category of polymer that has been particularly useful in this regard is a polymer bearing pendant amine groups. Poly(ethylenimine) (PEI) is an example of such a polymer. In one embodiment, the catalyst of the invention further comprises a polymer. Typically, said polymer is a polymer which comprises a plurality of amino, CMO alkylamino and/or di(Ci- io)alkylamino groups. Typically the polymer is poly(ethylenimine) (PEI).

The catalyst of the invention may or may not further comprise a solid support material, in addition to said polymetallic nanoparticles, wherein the polymetallic nanoparticles are supported on said solid support material. Any suitable support material may be used, for instance the solid support material may comprise carbon, alumina or titania. Typically, however, it comprises carbon. A carbon is preferred when the catalyst is used in a fuel cell.

In one embodiment, the support comprises an oxide, a nitride, carbon or nanotubes. The nanotubes are usually carbon nanotubes. The oxide is typically a metal oxide. The nitride is typically a metal nitride.

The polymetallic nanoparticles are typically present in an amount of from 10 to 30 weight %, based on the total weight of the catalyst including the support, preferably in an amount of from 15 to 25 weight %, based on the total weight of the catalyst including the support. Typically, when a solid support is present, the second metal in the catalyst of the invention is silver and the polymetallic nanoparticles have a mean particle size of less than or equal to 40 nm. More typically, in this embodiment, the polymetallic nanoparticles have a mean particle size of from 1 nm to 40 nm, from 10 nm to 40 nm, or from 20 nm to 40 nm. Thus, the polymetallic nanoparticles may have a mean particle size of from 30.0 to 40.0 nm. Typically, the standard deviation from said mean is less than or equal to 5.0 nm. Usually, in this embodiment, the molar ratio of the silver to the palladium in said polymetallic nanoparticles is from 1.5:1 to 1 : 1.5. More typically, wherein said molar ratio is about 1: 1.

In other embodiments, the catalyst does not comprise a solid support material. Thus, the polymetallic nanoparticles may be unsupported. In such embodiments, the nanoparticles or polymetallic nanoparticles may be present as a suspension in a solvent (for instance a colloidal suspension in the solvent), or simply in powder form. Any suitable solvent may be used, but the solvent typically comprises a polar protic solvent, for instance an alcohol or water.

In another aspect, the invention provides a catalyst for the production of hydrogen from a compound of formula (I), (II), (III) or (IV), or for the electro-oxidation of a compound of formula (I), (II), (III) or (IV):

R'COOH (I)

R²OH (Π)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted C O alkyl;

R is unsubstituted or substituted C_1-10 alkyl;

R³ is H or unsubstituted or substituted C O alkyl;

R⁴ is H, unsubstituted or substituted C_1-10 alkyl, or C(0)NR⁵R⁶;

R^s and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted C]_-10 alkyl;

which catalyst comprises polymetallic nanoparticles, which polymetallic nanoparticles comprise a core and a shell surrounding the core; wherein the shell comprises a first metal which is palladium, and the core comprises a second metal, wherein the second metal is other than palladium, platinum and gold. In one embodiment, the catalyst is for the production of formic acid or for the electro-oxidation of formic acid.

Typically, the second metal in the catalyst is a d-block metal of any one of Groups 8, 9, 10, 11 and 12, a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, provided that the second metal is other than palladium, platinum and gold. More typically, the second metal is a d-block metal of any one of Groups 8, 9, 10, 1 1 and 12.

Usually, the work function of the second metal is less than the work function of the first metal, palladium. This is advantageous in core-shell type polymetallic nanoparticles, because the lower work function of the second (core) metal compared to the palladium shell causes charge transfer from the core to the shell. This in turn causes stronger adsorption of the compound of fomula (I), (II), (III), or (IV) onto the surface of the nanoparticle. This is thought to be due to increased back-bonding from the metal d-orbitals into the pi orbitals of the adsorbed molecule. Indeed, the inventors have observed increased absorption of formate onto the metal surface when core-shell nanoparticles have been used instead of pure palladium nanoparticles, in which the work function of the second (core) metal is less than the work function of palladium. "Work function" here means the work function of the (1 1 1) crystal plane for metals having a face centred cubic (fee) lattice structure, or the work function of the (001) crystal plane for metals having a hexagonal closest packed {hep) lattice structure.

Thus, in the catalyst defined above, the second metal is an electron-rich

(electropositive) metal, such as a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, or a metal on the right hand side of the d-block: a d-block metal of any one of Groups 8, 9, 10, 11 and 12.

Typically, said second metal is selected from silver, rhodium, ruthenium, copper, iridium, nickel, cobalt, iron, osmium, zinc, cadmium, lithium, sodium, potassium, rubidium, caesium, beryllium, magnesium, calcium, strontium and barium. More typically, said second metal is selected from silver, rhodium, ruthenium, copper, iridium, nickel and cobalt. In one embodiment, the second metal is selected from copper, silver, rhodium and ruthenium. Thus, the second metal may for instance be silver or rhodium. In a preferred embodiment, the second metal is silver. The catalyst may be as further defined hereinbefore for the catalyst of the invention (see pages 29 to 35).

The above-defined catalyst may be produced by the process of the invention for producing said catalyst, which process comprises:

(a) reducing a salt of the second metal in the presence of a first solvent, to produce a suspension or solution of core nanoparticles in said first solvent, which core nanoparticles comprise said second metal; and

Step (b) is typically performed after step (a), in a sequential reduction process. However, for some second metals, step (b) be performed at the same time as step (a), in a co-reduction process.

Any suitable palladium salt may be used: such Pd salts are well known in the art. Typically, however, the palladium salt is palladium nitrate.

Similarly, any suitable solvents may be employed as the first and second solvents. Typically, however, the first solvent comprises a polar protic solvent. Thus, an alcohol or water, or a glycol, for instance ethylene glycol, may be employed as the first solvent. The second solvent also usually comprises a polar protic solvent. Thus, an alcohol or water, or a glycol, for instance ethylene glycol, may be employed as the second solvent.

The process of the invention for producing the catalyst of the invention may advantageously make use of stabilising polymers. Thus, the process of the invention for producing the catalyst of the invention typically comprises:

(a) reducing a salt of said second metal in the presence of a first solvent and a first stabilising polymer, to produce a suspension or solution of core nanoparticles in said first solvent, which core nanoparticles comprise said second metal; and

(b) reducing a palladium salt in the presence of said core nanoparticles, a second solvent, and a second stabilising polymer, to produce a shell which comprises palladium on the surfaces of said core nanoparticles, wherein the second solvent is the same or different from the first solvent and the second stabilising polymer is the same or different from the first stabilising polymer.

Step (b) is typically performed after step (a), but may, for some second metals, be performed at the same time as step (a). 001156

38

The sequential reduction process may comprise a hydrogen sacrificial protective step. Without such a step, metals which have a reduction potential which is lower than that of Pd, such as Rh and Ru, tend to segregate to the surface of particles produced during reduction, instead of remaining at the core. Thus, a hydrogen sacrificial protective strategy can advantageously be used in order to obtain a Pd shell around metals having a lower reduction potential than Pd ("reversed" core-shell nanoparticles).

Thus, in some embodiments, the process of the invention for producing the catalysts of the invention comprises:

(a) reducing a salt of said second metal in the presence of a first solvent and a first stabilising polymer, to produce a suspension or solution of core nanoparticles in said first solvent, which core nanoparticles comprise said second metal;

(bl) contacting a suspension or solution of said core nanoparticles in a second solvent with hydrogen gas, wherein the second solvent is the same or different from the first solvent; and

(b2) reducing a palladium salt in the presence of said core nanoparticles, said second solvent, and a second stabilising polymer, to produce a shell which comprises palladium on the surfaces of said core nanoparticles, wherein the second stabilising polymer is the same or different from the first stabilising polymer.

Typically, in the processes of the invention for producing the catalysts of the invention, the first and second stabilising polymers are independently selected from polyvinyl pyrrolidone) (PVP), polyvinyl alcohol) (PVA) and poly(ethylenimine) (PEI).

Usually, said suspension or solution of core nanoparticles produced in step (a) is a colloidal suspension or solution.

Typically, in step (a) the step of reducing said salt of said second metal is performed by heating said salt in the presence of said first solvent and, when present, said stabilising polymer, under an inert atmosphere. Usually, the salt of said second metal is heated to a temperature of at least 80 °C, preferably to a temperature of from 80 °C to 170 °C, preferably for a duration of at least 2 hours.

Usually, in step (b), or step (b2), the step of reducing said palladium salt is performed by heating said salt in the presence of said second solvent and, when present, said second stabilising polymer, under an inert atmosphere. Typically, the palladium salt is heated to a temperature of at least 80 °C, preferably to a temperature of from 80 °C to 100 °C, preferably for a duration of at least 2 hours. Step (a) of the process of the invention may or may not further comprise isolating said core nanoparticles from said solvent.

Similarly, step (b), or step (b2), may or may not further comprise isolating said polymetallic nanoparticles from said solvent.

Supported catalysts of the invention may be produced by employing a further step,

(c), in the process of the invention. Thus, in some embodiments, the process further comprises (c) depositing said polymetallic nanoparticles on a support material, thereby producing a catalyst which comprises said polymetallic nanoparticles and a support material, wherein the polymetallic nanoparticles are supported on said support material.

The process may further comprise: (d) heating said catalyst to remove any residual stabilising polymer.

The catalysts of the invention may be used as catalysts for the electro-oxidation of a compound of formula (I), (II), (III) or (IV), as defined hereinbefore, to produce protons (H*). As used herein the term "electro-oxidation" means oxidation at the surface of an electrode. It is thought that when a compound of formula (I), (II), (III) or (IV) is contacted with an electrode comprising the catalyst of the present invention, the compound is adsorbed onto the surface of the catalyst, and the catalyst catalyses cleavage of hydrogen atoms from the adsorbed molecules. The adsorbed hydrogen atoms are then converted into protons at the electrode. C-H bond cleavage is thought to be the rate-determining step.

Accordingly, the catalysts of the present invention can catalyse the electro- oxidation of a compound of formula (I), (II), (III) or (IV), as defined hereinbefore, to produce protons (H⁺) at an electrode.

Thus, catalysts of the invention can be used as the catalyst in a fuel cell electrode, in fuel cells in which the fuel is a compound of formula (I), (II), (III) or (IV) as defined hereinbefore. Such fuel cells are powered by the electro-oxidation of a said compound of formula (I); (II), (III) or (IV).

The electrochemical reactions in a direct formic acid fuel cell (DFAFC) are shown below:

Anode HCOOH→ C0₂ + 2H* + 2e Eqn. 1.1

Cathode ½ 0₂ + 2H* + 2e^~ -→ H₂0 Eqn 1.2

Overall HCOOH + ½ 0₂→ C0₂ + H₂0 Eqn. 1.3

It can be seen that the electro-oxidation reaction at the anode is very similar to the catalytic decomposition of formic acid to H₂ and C0₂, and it is this reaction at the anode which is catalysed by the catalysts of the present invention. Again, C-H breakage is the rate-limiting, step, but this time protons are formed instead of hydrogen gas.

Accordingly, the present invention provides a process for the electro-oxidation of a compound of formula (I), (II), (III) or (IV), which compound is as defined hereinbefore, which process comprises contacting a liquid phase which comprises said compound of formula (I), (II), (III) or (IV) with a catalyst of the invention in the presence of an electrode. Typically, the process is for the electro-oxidation of a compound of formula (I), (II), (III) or (IV) to produce protons at said electrode. The catalyst of the invention may be as defined hereinbefore. The liquid phase with which the catalyst is contacted may also be as further defined hereinbefore for the process of the invention for producing ¾.

In another aspect, the invention provides a process for the electro-oxidation of a compound of formula (I), (II), (III) or (IV) as defined above; which process comprises contacting a liquid phase which comprises said compound of formula (I), (II), (III) or (IV) with an electrode, which electrode comprises a catalyst of the invention as defined hereinbefore. Typically, the process is for the electro-oxidation of a compound of formula (I), (II), (III) or (IV) to produce protons at said electrode. Typically, the electrode comprises a conducting substrate and a catalyst of the invention as defined hereinbefore. The liquid phase may be as further defined hereinbefore for the process of the invention for producing ¾.

The invention further provides the use of a catalyst of the invention as defined hereinbefore for the electro-oxidation of a compound of formula (I), (II), (III) or (IV), which compound of formula (I), (II), (III) or (IV) is as defined hereinbefore.

The invention further provides the use of a catalyst of the invention as defined hereinbefore for the electro-oxidation of a compound of formula (I), (II), (III) or (IV) as defined hereinbefore to produce protons at an electrode. Typically, the electrode comprises the catalyst of the invention. More typically, the electrode comprises a conducting substrate and a catalyst of the invention as defined hereinbefore.

The compound is typically a compound of formula (I) which is formic acid. Thus, the invention further provides the use of a catalyst of the invention as defined hereinbefore for the electro-oxidation of formic acid. Typically, said electro-oxidation proceeds in accordance with the following reaction:

HCOOH→ C0₂ + 2H⁺ + 2e^~

Further provided is the use of a catalyst of the invention as defined hereinbefore for the electro-oxidation of formic acid to produce protons at an electrode. Typically, the electrode comprises the catalyst of the invention. More typically, the electrode comprises a conducting substrate and a catalyst of the invention as defined hereinbefore. Typically, said electro-oxidation proceeds in accordance with the following reaction:

HCOOH→ C0₂ + 2H⁺ + 2e^~

The invention further provides the use of a catalyst of the invention as defined hereinbefore in a fuel cell. Typically, the fuel cell comprises a compound of formula (I), (II), (III) or (IV) as defined hereinbefore.

The invention further provides the use of a catalyst of the invention as defined hereinbefore for the electro-oxidation of a compound of formula (I), (II), (III) or (IV) as defined hereinbefore in a fuel cell.

Further provided is the use of a catalyst of the invention as defined hereinbefore, for the electro-oxidation of formic acid in a direct formic acid fuel cell.

The invention further provides an electrode suitable for use in a fuel cell, which electrode comprises a catalyst of the invention as defined hereinbefore. Typically, the electrode comprises a conducting substrate and a catalyst of the invention as defined hereinbefore.

The invention further provides a fuel cell which comprises: a catalyst of the invention as defined hereinbefore, or an electrode of the invention as defined above.

In one embodiment, the fuel cell is a direct formic acid fuel cell.

The present invention is further illustrated in the Examples which follow:

EXAMPLES

EXAMPLE 1; Synthesis of nanocatalysts

Materials

Ruthenium (III) chloride (RuCl₃ XH₂0, 40.14%), hydrogen hexachloroplatinate (IV) (H₂PtCl₆, 39.51%), palladium nitrate (Pd(N0₃)₂, 40.70%), rhodium(III) nitrate (Rh(N0₃)₃, 99.9%) and were obtained from Johnson Matthey, pic. Hydrogen tetrachloroaurate (III) (HAuCl₄, 52 %), silver nitrate (99%), ethylene glycol 99%, diethylene glycol 99%, polyvinyl pyrrolidone) (PVP, MW 40,000), 98% polyvinyl alcohol) (PVA, MW 13,000- 23,000), 80% hydrolysed polyvinyl alcohol) (PVA, MW 9000-10000), H¹³COOH (95% weight in H₂0, 99 atom % ¹³C), sulphuric acid (97.5%, analytical grade), perchloric acid (70%) and absolute ethanol were purchased from Aldrich. Poly(ethylenimine) (branched PEI, MW 10,000), acetone (laboratory reagent grade), nitric acid (70%, analytical grade), sodium hydrogen carbonate were obtained from Fischer. Triethylene glycol (99%), formic acid puriss (-98%) were procured from Fluka. .Perfluorosulfuric acid PTFE copolymer 5% w/w solution was purchased from Alfa Aesar. Vulcan^® carbon XC 72R was obtained from Cabot Company Limited. 1% CO in helium gas was purchased from BOC gases.

Deuterium oxide (99.99% atom of D) was purchased from Cambridge Isotope

Laboratories. Deionized water was purified by a Milli-Q water purification system.

Methods r

Synthesis of polymer stabilised metal nanoparticles was carried out by the polyol process using PVP as a main stabiliser and polyol such as ethylene glycol, diethylene glycol, triethylene glycol as both solvent and reducing agent. The size of nanoparticles was controlled by using various reaction parameters such as reaction temperature, reaction time, polyol type, concentration of metal salt and mole ratio of stabiliser to metal salt.

1.1 Synthesis of Polymer Stabilised Pd Nanoparticles

PVP-Pd nanoparticles: In a 100 mL round bottom flask, 0.0560g of Pd(N0₃)₂

(2.1417x1ο^"4 moles) and 0.24 g (2.1628xl0^"3 moles) of PVP were dissolved in 30 mL of ethylene glycol. The solution was then heated to 160 °C for 2 hrs under N₂ atmosphere. The obtained colloidal PVP-Pd nanoparticles were then precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere.

PVA-Pd and PEI-Pd nanoparticles: The method for the synthesis of PVA-Pd and PEI- Pd nanoparticles were the same as those described for PVP-Pd nanoparticles except that the polymer (PVA 0.1 g or PEI 0.4g) and ethylene glycol were heated to 120 °C for 30 min before the addition of Pd(N0₃)₂.

Polymer stabilized Pd nanoparticles in colloidal form were prepared by using a water soluble polymer, Polyvinyl pyrrolidone) (PVP), 98% Poly(vinyl alcohol) (98% PVA), 80% hydrolyzed poly( vinyl alcohol) (80% PVA), or hyper-branched Poly(ethylenimine) (HB-PEI).

In order to obtain stable nanoparticles, the factors which can affect to stability of different polymer-stabilised nanoparticles such as the solubility of polymers in ethylene glycol and _ type of the obtained nanoparticles mixture after reduction reaction were studied. The results are shown in Table 1. The TEM images of obtained Pd nanoparticles with different polymer stabiliser are shown in Fig 1. Table 1

The reaction parameters are controlled; ratio of polymer-stabiliser to Pd(II) ion is 10 times, reaction time is 2 hr, heating rate is 7°C/min. Reaction temperature is 160 °C. Concentration of Pd ion is 7.1 mM. The solubility of polymer in polyol media affects the stability of the polymer stabilised nanoparticles. PVP with amide group can be well dissolved in ethylene glycol in homogeneous solution at room temperature while HB-PEI and 80% PVA are partially soluble. PVA with high degree of hydrolysis (98%) is only dissolved in ethylene glycol at temperature above 80°C.

Polarity of polymer: After the reduction reaction, Pd nanoparticles stabilised by PVP and 80%PVA are in suspension form while Pd stabilised by HB-PEI and 98% PVA are precipitated from the reaction. The stabilisation of metal particles by polymers strongly depends on the extent of solvation of the polymer in the surrounding medium (ethylene glycol). Therefore, in order to stabilise nanoparticles, polymers should typically not only dissolve in a solvent but also have more polarity than solvent. In the polyol method ethylene glycol acts as solvent and reducing agent. Ethylene glycol is very weak acidic (pKa = » 15) which can be oxidised to aldehydes, carboxylic acid (glycolic acid and oxalic acid). This oxidized form is believed to act as a stabiliser for the metal nanoparticle colloid. Therefore, there seems to be a competition between the functional group of the polyol solvent and polymer stabilisation to coordinate and stabilise nanoparticles.

The amide group of PVP and acetyl group of 80% hydrolysed PVA have a higher polarity than the hydroxyl group and the oxidized forms of ethylene glycol whereas the amine group in HB-PEI and the hydroxyl group in 98% PVA have a lower polarity. A higher polarity increases the interactions between the functional groups of the polymer and the metal causing of the precipitation of metal nanoparticles. The solvent used in the reaction medium is therefore an important factor. The selected solvent should be a good solvent for the metal salt and the polymer. The optimum solvent system should have lower polarity than the most polar constituent of the polymer, and the polymer must coordinate to the nanoparticle's surface to be an effective stabilizer. Isolation of polymer stabilised-metal nanoparticles; PVP, 80% hydrolysed PVA, 98% PVA and HB-PEI stabilized Pd nanoparticles in ethylene glycol media can easily be separated from the reaction by acetone. However, the solubility of the polymer-stabilised nanoparticles in ethylene glycol affects the degree to which precipitation occurs. All types of polymer stabilised-Pd nanoparticles (PVP, 80% hydrolysed and 98% PVA, HB-PEI) can re-dispersed well in polar solvents such as alcohol and water. It can be confirmed that PVP is the optimum stabiliser for the preparation polyer-stabilised nanoparticles in ethylene glycol media. 1.2 Size Controlled Synthesis of PVP Stabilised Ru Nanoparticles (PVP-Ru

Nanoparticles)

PVP-Ru nanoparticles of various sizes were synthesized according to the following procedure. In a 100 mL round bottom flask, 0.0520g (2.0652 1ο-⁴ moles) of RuCl₃xH₂0 and 0.24 g (2.1628^χ10^'3 moles) of PVP were dissolved in 30 mL of diethylene glycol under stirring. The solution was then heated (7 °C/min) to 160°C under N₂ atmosphere. After 2 hrs at 160 °C, a dark brown solution of colloidal PVP-Ru nanoparticles was obtained. The PVP-Ru nanoparticles were precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere. PVP-Ru nanoparticles obtained by this method were 2.3 nm. Different sizes of PVP-Ru nanoparticles were synthesized by varying the mole ratio of PVP to Ru³⁺ ion, reaction temperature, heating rate and polyols. The list of experiments with various reaction parameters along with particle size achieved are presented below. In order to prepare stable metal nanoparticles with well defined particle size and narrow distribution, suitable and effective methods of controlling the metal particle size were studied. This research examined a number of variables and parameters affecting the characteristics of the Ru nanoparticles. These include reaction conditions such as reaction temperature, time, heating rate, the polymer to metal ratios and various reducing agents. It was found that these parameters mainly affect the particle size of the colloidal Ru nanoparticles and its distribution within the polymer, as shown in Table 2.

Table 2: Reaction conditions and particle dimensions of PVP-Ru nanoparticles

Reaction time is controlled to 2 hr. Concentration of Ru ion is 6.9 mM

Size, shape and distribution of nanoparticles can be characterized by transmission electron microscope. TEM images of PVP-Ru nanoparticles are shown in Fig 2. Standard deviations were calculated from the distribution of particle size about 100 particles. The standard deviation is less than 0.3 for all the reactions carried out under various parameters. It is indicated that mono-dispersed PVP-Ru nanoparticles were obtained by this preparation method

1.3 Size Controlled Synthesis of PVP Stabilised Pd Nanoparticles fPVP-Pd

Nanoparticles): Step Growth Formation Method

In a 100 mL round bottom flask, 0.0560g (2.1417x10^ moles) of Pd(N0₃)₂ and 0.24 g (2.1628χ10^'3 moles) of PVP were dissolved in 30 mL of diethylene polyol under stirring. The solution was then heated (7 °C/min) to 90°C under N₂ atmosphere. After 2 hrs at 90 °C/min, a transparent dark brown colloidal solution of PVP-Pd nanoparticles was obtained. The PVP-Pd nanoparticles were precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere. The particle size of PVP-Pd nanoparticles obtained with the above conditions was 2.7 nm.

Different sizes of Pd nanoparticles were synthesized by step growth using pre-formed PVP-Pd nanoparticles as seed. The first step growth of PVP-Pd nanoparticles was carried out by mixing 10 mL of colloidal PVP-Pd nanoparticles with a solution containing 0.056g (2.1417x 1 ο^"4 moles) of Pd(N0₃)₂, 0.24g of PVP and 30 mL of diethylene glycol, followed by heating the mixture at 90°C for 2 hrs. The second step growth was conducted in the same way.

PVP-Pd nanoparticles' sizes in range 2.3-5.2 nm were prepared by polyol process using Pd(NO)₃ as CI^" free precursor and PVP as stabilizer. PVP-Pd nanoparticles size of 2.7 nm were used as the seed for stepwise growth reduction in order to obtain the larger particles with a small change of size distribution as shown in Table 3. The TEM images of PVP-Pd nanoparticles are shown in Fig 4. Table 3: Sizes and distribution of PVP-Pd nanoparticles prepared by step growth formation

Solvent for reaction No. 1 is EG and reaction NO. 2-6 is DEG. Reaction temperature is 90°C and reaction time is 2hr. The ratio of PVP to Pd ion is 10 times. Concentration of Pd ion is 7.1 raM. 1.4 Size Control Synthesis of PVP Stabilised Ft Nanoparticles (PVP-Pt Nanoparticles)

In a 100 mL round bottom flask, 0.098g (1.9848 10^"4 moles) of H₂PtCl₆.6H₂0 and 0.2400 g (2.1628 l0^"3 moles) of PVP were dissolved in 30 mL of ethylene glycol under stirring. The solution was then heated (7 °C/min) to 160°C for 2 hrs under N₂ atmosphere. The resulting dark brown solution of colloidal PVP-Pt nanoparticles was precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere. The particle size of PVP-Pt nanoparticles was 3.8 nm. The particle size of PVP-Pt nanoparticles were controlled by reaction temperature and heating rate. The various reaction parameters were maintained during the synthesis of PVP-Pt nanoparticles and the obtained particle sizes are shown in Table 4 below.

Monodispersed PVP-Pt nanoparticles were prepared by polyol process. ¾PtCl₆ as precursor for Pd(IV) was reduced at different temperature and heating rate for 2 hr. In ethylene glycol, the Pt(IV) ion can be reduced at a wider range of temperature (110-180°C) than in alcohol. At these reduction temperatures, near spherical shape of Pt particles were obtained with narrowed-size distribution as shown in Table 4 and Fig 5. It can be seen that the obtained particle sizes of Pt particles strongly depend on the heating rate of reaction mixture as the nucleation and growth step are mainly depending on reduction rate during the particle formation period.

Table 4: Reaction conditions for the preparation of various sizes of PVP-Pt nanoparticles

The reaction time is 2 hr. Mole ratio of PVP to Pt ion is 10. Solvent is ethylene glycol.

Concentration of Pt ion is 6.2 mM. 1.5 Size Controlled Synthesis of PVP Stabilised Rh Nanoparticles (PVP-Rh

Nanoparticles) In a 100 mL round bottom flask, 0.025g (2.4270x 10^"4 moles) of Rh(N0₃)₃ was dissolved in 300μ1 of water and then 0.24 g (2.1628 l0^"3 moles) of PVP was added. The mixture was dissolved in 30 mL of ethylene glycol under stirring and heated (7°C/min) to 160°C under N₂ atmosphere. A dark brown of colloidal PVP-Rh nanoparticles was obtained after 2 hrs at 160 °C. The PVP-Rh nanoparticles were precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere. The particle size of obtained PVP-Rh nanoparticles was 2.0 nm. The particle size of PVP-Rh nanoparticles was controlled by varying reaction temperature, heating rate and polyol solvent as shown in Table 5 below.

PVP-Rh nanoparticles were prepared from Rh(N0₃)₃ as precursor. Since this precursor gave a low solubility in the polyol solvent, it was then dissolved in a minimum amount of water before mixing with ethylene glycol. PVP-Rh nanoparticles sizes in range 2-10 nm were prepared by varying the reduction rate which strongly influence on controlling the particle size. The sizes of the obtained PVP-Rh nanoparticles are shown in Table 5. The size distribution was found to increase with an increase of particle size and the particle shape seems to change from cubo-octaheral to polyhedral when the particle size is increased. The TEM images of the obtained particles are shown in Fig 6.

Table 5: Reaction parameters for the preparation of various sizes of PVP-Rh nanoparticles

Mole ratio of PVP to R (III) ion is 10. The reaction time is 2 hr. Concentration of Rh ion is 8.1mM. 1.6 Size Controlled Synthesis of PVP Stabilised Ag Nanoparticles (PVP-Ag

Nanoparticles)

In a 100 mL round bottom flask, 0.025g (2.2945 10^'4 moles) of AgN0₃ and 0.48 g (4.3256x 10^'3 moles) of PVP were dissolved in 30 mL of ethylene glycol under stirring. The solution was then heated to 120°C under N₂ atmosphere. A clear yellow solution of colloidal PVP-Ag nanoparticles was obtained after an hour at 120°C. The PVP-Ag nanoparticles were precipitated in acetone, washed with acetone twice and dried under N₂ atmosdphere. The particle size of obtained PVP-Rh nanoparticles was 15.0 nm. Different sizes of PVP-Ag nanoparticles were synthesized by varying the concentration of PVP and AgN0₃, reaction temperature, and polyol solvents, as shown in Table 6 below.

PVP-Ag nanoparticles particle sizes in range 15-50 nm were prepared by the polyol process with different reaction conditions as shown in Table 6. TEM images of some sample are shown in Fig. 7.

Table 6: Reaction conditions for the preparation of various size PVP-Ag

nanoparticles

Reaction time is 1 hr. heating rate is 7°C/min 1.7 Size Controlled Synthesis of PVP Stabilized Au Nanoparticles (PVP-Au

Nanoparticles)

In a 100 mL round bottom flask, 0.028g (7.3921xl0^'5 moles) of AuHCl₄, 0.48 g

(4.3256xl0^"3 moles) of PVP and 0.1 g of NaHC0₃ were dissolved in 30 mL of ethylene glycol under stirring. The solution was then heated to 90°C under N2 atmosphere. A red solution of colloidal PVP-Au nanoparticles was obtained after 2 hrs at 90°C. The PVP-Au nanoparticles were precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere. The particle size of obtained PVP-Au nanoparticles was 10.0 nm. Different sizes of PVP-Ag nanoparticles were synthesized by varying concentration of the metal salt, the molar ratio of PVP to AuHCl₄ and reaction temperature as presented in Table 7 below.

Colloidal Au nanoparticles size in range 10-50 nm were prepared by polyol process using ethylene glycol as reducing agent. The optimum condition for the preparation of nanosized Au particles with high dispersity was examined from change of various parameters. The results are shown in Table 7. The TEM images of colloidal PVP-Au nanoparticle are shown in Fig 8.

Table 7: Reaction conditions for the synthesis of collidal PVP-Au nanoparticles

The heating rate is controlled to 7°C/min. Synthesis of Bimetallic Nanoparticles

Bimetallic nanoparticles were prepared by polyol process using PVP as stabilizer. Salts of two different metals were mixed together and simultaneously reduced to obtain bimetallic nanoparticles with alloy structure whereas core-shell structure was obtained by reduction of two different metal salts in two stages. The first stage involved the reduction of metal salt to form the metal core while the second stage involved the reduction of another metal salt on the surface of the metal core to form the shell.

The alloy bimetallic of Pt/Pd and Ag/Pd were prepared by continuous reduction. In addition, core-shell structured nanoparticles having a Pd shell and a core of Pt, Ru, Rh or Ag were prepared by successive reduction. The core-shell structured Au@Pd was prepared by co-reduction. 1.8 Synthesis of Pt/Pd Alloy 1:1 Mole Ratio by Co-Reduction

In a 100 mL round bottom flask, 0.028g of Pd(N0₃)₂ (1.0709x 1ο^"4 moles), 0.049g H₂PtCl₆.6H₂0 (0.9924x10 moles) and 0.24 g (2.l 628xl0^"3 moles) of PVP were dissolved in 30 mL of ethylene glycol under stirring. The mixture was then heated (20°C/min) to 160°C under N₂ atmosphere. After heating for 2 hrs, a dark brown solution colloidal Pt/Pd nanoparticles was obtained. The Pt/Pd alloy nanoparticles were precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere.

1.9 Synthesis of Metal-Core Pd-Shell Nanoparticles by Successive Reduction

Nanoparticles of Ru, Pt, Rh and Ag metal cores were prepared first by following the procedure outlined in sections 1.2, 1.4, 1.5 and 1.6, respectively. Pd salts were then reduced in the presence of these metal cores to obtain bimetallic nanoparticles of various metal-cores with Pd-shell. A typical procedure for the synthesis of a polymer protected bimetallic nanoparticles with 1 : 1 mole ratio of metal-core Pd-shell is described here. 15 mL of colloidal solution of metal core was mixed with a solution containing 0.028 g (1.0709 1ο-⁴ moles) of Pd(N0₃)₂, 0.024g ( 2.1628x l0-³ moles) of PVP and 30 mL of ethylene glycol. The resulting solution was heated (7°C/min) to 90°C for 2 hrs under N₂ atmosphere. In the case of Rh@Pd and Ru@Pd nanoparticles, Pd has higher reduction potential than Ru and Rh, the colloidal solutions of metal core (PVP-Rh and PVP-Ru) were purged with hydrogen gas for 30 min before mixing with the solution of Pd(N0₃)₂. The M@Pd nanoparticles were precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere.

Ag-core Pd-shell nanoparticles with different mole ratios of Ag : Pd were prepared. In order to maintain the same concentration of Pd salt in all the reactions, the volume of ethylene glycol was adjusted, as shown in Table 8. Table_8: Composition of various ingredients employed during synthesis of Ag-core Pd-shell nanoparticle

1.10 Synthesis of Au(¾Pd Core-Shell Nanoparticles by Co-Reduction In a 100 mL round bottom flask, 0.028g (1.0709x 10^"4 moles) of Pd(N0₃)₂, 0.020g

(1.0159X10^"4 moles) HAuCl₄, 0.24 g (2.1628xl0^"3 moles) of PVP were dissolved in 30 mL of ethylene glycol under stirring. The solution was then heated (7 °C /min) to 160°C under N₂ atmosphere. After heating for 2 hrs, a dark brown solution of Au@Pd nanoparticles was obtained. The nanoparticles were precipitated in acetone, washed with acetone twice and dried under N₂ atmosphere.

TEM Observations

TEM images were recorded on JOEL 2000FX. TEM samples were prepared by placing a drop of colloidal dispersion of metal nanoparticles in methanol onto a carbon-coated copper grid followed by natural evaporation of the solvent. At least 100 nanoparticles were considered for reporting the mean diameter with standard deviation. Pt-Pd bimetallic 1 : 1 molar ratio was prepared as detailed above by co-reduction of Pd( 0₃)₂ and H₂PtCl by polyol process in the presence of ethylene glycol as reducing agent and PVP as stabilizer. The mixture of the metal salts was reduced at 160°C with high heating rate (20°C/min). The TEM image of the obtained nanoparticles is shown in Fig 9 a). Ag-Pd and Au-Pd bimetallic nanoparticles were also prepared by the same method as the Pt-Pd bimetallic nanoparticles, AgN0₃ and HAuCI₄ being used as the metal precursor respectively. The TEM images of obtained nanoparticles are shown in Fig 9 b) and c). From the images in Fig 9, the obtained Pt-Pd, Ag-Pd and Au-Pd bimetallic nanoaprticles are stable (no aggregation). The results suggest these bimetallic nanoparticles are not mixtures of monometallics of both metals but consist of single bimetallic particles.

Pt-Pd and Ag-Pd bimetallic nanoparticles were also prepared by successive reduction as detailed above. Pt and Ag nanoparticles were prepared as a core metal first follow by the reduction of Pd ion deposited on these metal cores. By this method, the core-shell structures Pt@Pd and Ag@Pd are expected. The reduction temperature for the reduction of Pd shell is low at 90°C in order to avoid re-oxidation of core atom and the creation of new nuclei of Pd shell replacing deposit core surface. The obtained bimetallic nanoparticles are uniform and have a narrow size distribution as shown in Fig 10. The Pt-Pd bimetallic nanoparticles were prepared from 6.0 nm Pt cores and the Ag-Pd bimetallic nanoparticles were prepared from 15.0 nm Ag cores with the ratio of metal core to shell being 1 : 1. It can be seen from the TEM images in Fig. 10 a and b that the obtained bimetallic nanoparticles are larger than the metal core. The implication is that the Pt and Ag cores are covered by Pd atoms. Also, the difference in contrast between the core and shell metals is clearly observed in the larger bimetallic nanoparticles shown in Fig 10 c.

Rh-Pd and Ru-Pd bimetallic nanoparticles were also prepared by successive reduction. As the reduction potential of Rh and Ru are lower than Pd, Rh and Ru atom tend to segregate to the surface of particles during reduction. In order to obtain a Pd shell (reversed core- shell) a hydrogen sacrificial protective strategy was used to modify these nanoparticles. Noble metals like Pd, Pt, and many others have the ability to adsorb hydrogen and split it to form metal-H bonds on the metal surface. Hydrogen atoms adsorbed on noble metals have a very strong reducing ability, implying a very low redox potential. Therefore, the hydrogen gas was passed through Rh and Ru monometallic colloidal nanoparticles at room temperature for 1 hr before mixing and reduction of the Pd presursor. The TEM images of the obtained bimetallic nanoparticles are shown in Fig. 11. The images show very good dispersity of the particles.

UV-Visible Spectroscopy

UV-visible absorption spectra of Ag, Au and their bimetallic nanoparticles in aqueous solution were recorded at room temperature on Cintra 10 UV-Vis spectrometer using a quartz cuvette with an optical path length of 1 cm. Important information about the structure of nanoparticles such as alloy or core-shell structure is obtained by analysing the change of the position of plasmon absorption peak. UV-visible spectra of Ag compared with Ag@Pd nanoparticle prepared at different molar ratios are shown in Fig. 12. Ag nanoparticle has a characteristic surface plasmon resonance at 420 nm while Pd nanoparticle hasn't strong absorption and only shows a broad absorption tail (not shown). From Fig 12, it can be seen that no plasmon peak is observed for the obtained nanoparticle prepared with 1 : 1 and higher metal ratio of Ag core to Pd shell. This is indicative of full shell cover of Pd atom. The same result was observed for obtained 1 :1 mole ratio of Au-Pd bimetallic nanoparticles prepared from co-reduction (Fig 13). The Au@Pd core-shell structure was confirmed by CO adsorption analysis, for the Au-Pd bimetallic nanoparticles prepared from co-reduction (see under heading 1.10 above).

EXAMPLE 2: Pd-Based Bimetallic Nanoparticles as Catalysts for Formic Acid Electro-Oxidation

A series of metal-core Pd-shell (M@Pd, where M = Ru, Rh, Pt, Ag and Au) bimetallics were designed, synthesised and characterised as detailed in Example 1.

2.1 Preparation of Carbon Supported Catalysts (M/C Catalysts)

Pretreatment of carbon: lg of Vulcan^® XC-72R was pretreated with 60% nitric acid under reflux for 3 hrs. It was then filtered, washed with water until water washings neutral pH and dried overnight at 120°C.

Deposition of colloidal metal nanoparticles on pretreated carbon: 0.08 g of pretreated carbon was dispersed in 50 ml of ethanol under sonication for 0.5 hr. The reaction mixture containing PVP-Pd nanoparticles was added dropwise onto the dispersed carbon in ethanol under sonication. The resulting mixture was stirred overnight, and filtered. The filtered nanoparticles were washed with water and ethanol to remove residual chloride and surfactant, which would otherwise negatively affect the catalytic activity. The prepared M/C nanocatalysts had a metal loading of 20% wt.

Removal of Stabilizer: The stabilizer PVP in PVP-M/C nanocatalyst was removed by heating at 300°C for 0.5 hr under N₂. Preparation of Catalyst Ink: Catalyst ink was prepared by ultrasonically dispersing 5.0 mg of M/C catalyst in a mixture of 200 μΐ of ethanol and 50μ1 of 5% w/w solution of perfluorosulfuric acid PTFE copolymer for 15 min.

Discussion: Vulcan carbon has been thought to be one of the best candidate supports for electrocatalysts for PEMFCs due to its proper surface area and pore structure. Therefore, it was used as a catalyst supported for this study. The as-prepared mono and bimetallic nanoparticles were loaded on Vulcan carbon with 20% metal loading by physical deposition. The obtained metal/C supported nanoparticles were filtered and washed many times with water and ethanol to remove excess PVP stabilizer, polyol solvent and inorganic ion impurity. TEM images of Pd/C are shown in Fig 14. The images clearly show that the obtained Pd/C nanoparticles are well dispersed and do not aggregate after deposition on the carbon support.

Using a polymer stabilizer such as PVP in a colloidal method has been found to inhibit fuel cell catalysis. Heat treatment may be employed to get rid of the PVP. In this work, the minimum temperature studied (maximum activity of catalyst) was found to be 300°C under N₂ for 0.5 fir based on our thermogravimetry results (not shown). In order to verify whether there is any surface reconstruction and sintering of our mono and bimetallic supported catalysts after heat treatment, TEM characterization of the catalysts were employed. Typical TEM images of our samples after the heat treatment are presented in Fig 15.

Fig. 15a shows that the particle distribution still maintains at narrow range (SD= ±0.4 nm) with some degree of aggregation. From the figure, the particle sizes of Pd/C nanoparticles appear to be larger (1.2 times) than those before the heat treatment. In case of Pt@Pd and Ag@Pd nanoparticles (Fig. 15b and c), it is clearly shown that the core-shell nanoparticles maintain their uniform dispersity after the removal of PVP although, the heating treatment apparently affects the particle size and the surface composition. It can be seen that after the heating, the particle size becomes larger and the contrast between metal core and shell is clearly observable. The particle sizes of Pt@Pd and Ag@Pd after heating are larger by around 4 times and 2 times, respectively compared to the corresponding particles without the heat treatment. 2.2 Preparation of Working Electrode

Cleaning of Electrode: Glassy carbon working electrode (surface area 0.0706 cm²) was polished with 0.05 μπι alumina suspension (water) followed by ultrasonic cleaning in deionized water. This cleaned electrode was used as substrate for the carbon support catalysts.

Preparation of Electrode: 1.0 of an ultrasonically redispersed catalyst ink was pipetted onto the glassy carbon substrate and dried at room temperature. The deposited catalyst was 56 μg metal cm after solvent evaporation.

2.3 Electrochemical Measurements

Electrochemical measurements were carried out on an electrochemical cell with conventional three electrode configuration. The glassy carbon thin film electrode coated with the metal/C catalyst under investigation was used as working electrode. The saturated calomel electrode (SCE) was used as the reference electrode. A Pt disk electrode with Surface area 0.0314 cm² was used as the counter electrode. All potentials were recorded with respect to SCE. The cell was filled with 0.5 M H2SO4 as an electrolyte and purged with N₂ to remove dissolved oxygen. All formic acid oxidation experiments were performed in 0.5M H2SO4 + 2.0 HCOOH solution. MilliQ water was used to prepare the solutions. Electrochemical experiments were performed on compactstat, Ivium electrochemical analysis instrument. 2.4 Electrochemical Pretreatment

Before electrochemical measurements, a set of potential cycles (between -0.2 V and 1.2 V vs. SCE at 50 mVs^-1) was applied to the electrode for surface cleaning. The cyclic voltammetry measurement was carried out in 0.5M H S0 until a steady voltammogram was obtained (around 10 cycles).

2.5 Cyclic Voltammetry Measurements Cyclic voltammetry measurements for formic acid electro-oxidation were carried out at room temperature in 0.5 M H₂S0₄ and 2 M aqueous solution of formic acid at potential range between -0.2 and 1.2 V vs SCE and at a potential sweep rate of 50 mV/s.

2.6 Ag(¾Pd Core-shell Catalysts for Formic Acid Electro-Oxidation

Probing an Adsorption Strength of Adsorbed Formate by Solution ¹³ C NMR

Spectroscopy

Core-shell Ag@Pd catalysts having different molar ratios of Ag to Pd were prepared and characterized as detailed in Example 1. The chemical shifts of adsorbed formate species on different compositions of Ag to Pd are presented in Table 9. The correlation between chemical shift of bridging formate and mole fraction of Pd overlayer in Ag@Pd core-shell is shown in Fig. 16.

Table 9: Variation of chemical shifts of adsorbed formate species in different modes on Ag@Pd nanocatalysts with different mole ratios of Ag and Pd

adsorbed formate on Ag atom ** chemical resonance peaks are not clearly seperated The surface modification of Ag-core with different molar ratios of Pd shell brought changes in the chemical shift of the adsorbed formate as presented in Table 9. For example, for Ag-core Pd-shell catalysts, the resonances for 2: 1 , 1 : 1 , 1.2 and 1 :3 mole ratios were shifted to lower field (higher chemical shift) compared to that of adsorbed formate on monometallic Pd catalyst but the 3:1 mole ratio gave a similar value to the pure Pd. The Ag-core Pd-shell catalyst with 3 : 1 mole ratio exhibited two ¹³C resonances corresponding to adsorbed monodentate formate on Ag and Pd surfaces. This is in agreement with surface plasmon characterization results, which revealed the presence of Ag atoms on the surface (see Example 1). The ¹³C resonance of adsorbed formate on Ag@Pd appeared to give higher chemical shift compared to that of adsorbate on monometallic Pd indicating the stronger interaction of adsorbed formate. When Ag and Pd are in contact at the interface, there could be electrons transfer from Ag at higher Fermi level/lower electronegativity to Pd due to ligand effect and thereby the adsorption strength of formate on Pd shell is expected to be stronger (increase back donation to adsorbed formate). An alternative explanation for the higher chemical shift could originate from the lattice strain effect. The obtained Ag@Pd samples also showed an increase in lattice constant, which is indicative of surface expansion (~0.5-3%) compared to monometallic Pd. At lower ratio of Pd overlayers, an increase in expansion of Pd lattice was clearly observed. This could also lead to stronger adsorption of formate. Thus, the adsorption strength of adsorbed formate on the catalysts was analysed by measuring the chemical shift; this indicated that charge transfer and lattice strain effects influence adsorption strength. The chemical shift of adsorbed formate in bridging form varies with the mole fraction of Pd overlayer as shown in Fig 16. As seen from the figure, as Pd atoms cover Ag surface with increasing mole fraction of Pd, the adsorbed formate exhibits a higher chemical shift than those of pure Pd in the ¹³C NMR spectra. The resonance for ^l3C nuclei of adsorbed formate on Ag@Pd catalyst containing 0.33 mole fraction of Pd had almost a similar chemical shift value to that of monometallic Pd, while the resonances for adsorbed formate on Ag@Pd catalysts containing 0.50 and 0.66 mole ratio of Pd appeared further downfield (higher chemical shifts).

Catalytic Activity Testing by Electrochemical Techniques

Ag@Pd nanoparticles were deposited on carbon support and the polymer stabilizer was burnt off before catalytic testing. The electrochemical activity of Ag@Pd/C catalysts for formic acid oxidation was examined as a function of overlayer thickness over different mole fractions of Pd by using cyclic voltammetry. The current densities achieved for Ag- core Pd-shell nanoparticles with different mole ratio of Ag:Pd are shown in Fig 17. The catalytic activity of various Ag@Pd catalysts is presented in terms of their mass activity and specific activity (see Figs. 18 and 19, respectively).

The maximum mass activity and specific activity was achieved for Ag@Pd catalyst with 1 : 1 mole ratio, which is in a very good agreement with the results obtained from solution ¹³C NMR spectroscopy as detailed above. The mass activity of Ag@Pd was found to decrease with either increasing or decreasing mole ratio of Ag or Pd. If some of Ag atoms are exposed on the surface of the Ag@Pd catalysts with high mole ratio of Ag-core, the less active Ag atoms causes a decrease in the activity per gram of the catalyst. When the molar ratio of Pd increases above 1, the particle size of Ag@Pd catalyst become larger, thus, lowering the total surface area and lowering the mass activity of the catalyst.

The specific activity of a catalyst depends on its electrochemical surface area (ECSA). The change in ECSA of the catalysts with mole fraction of Pd overlayer in Ag@Pd catalysts is presented in Fig. 20. From Fig. 20 it is clear that the maximum ECSA was observed for Ag@Pd catalyst with 1:1 mole ratio. Both the specific activity (Fig 19) and ECS A (Fig 20) for Ag@Pd catalysts with various molar ratios of Ag:Pd exhibited the same trend.

1 ^"¾

The surface probing by solution C NMR spectroscopy for determination of adsorption strength of Ag@Pd catalysts can clearly reflect their electro-oxidation activities. Weaker adsorption of formate results in lower activity of the catalyst while stronger adsorption of formate results in higher activity. The activity of Ag@Pd catalyst also depends on electronic promotion from Ag-core to Pd-shell and the lattice strain of the Pd overlayers. The percent of lattice expansions at various mole fractions of Pd overlayer on Ag-core were calculated by comparing with the lattice constant of 5.2 nm Pd nanoparticle ( =

3.872). The results are presented in Fig. 21. It was evident that Pd overlayer on Ag surface gave higher lattice value (2-3% expansion) in particular, when Pd-shell was in thin layer i.e., when the mole fraction of Pd-shell was less than 0.3, as shown in Fig. 21. However, the degree of lattice expansion decreased sharply to 1% when the mole fraction of Pd increased to 0.5 but at mole fractions of Pd > 0.5 the percentage of expansion decreased only marginally. It is interesting to note that the results from ¹³C NMR analysis of Ag@Pd nanoparticles and the specific activity for formic acid electro-oxidation showed the maximum at 1 : 1 Ag to Pd mole ratio. It indicates that the stronger adsorption of the formate species (higher chemical shift) results in higher activity for formic acid electro- oxidation. At this ratio, analysis showed that the catalyst surface is fully covered with Pd atoms. On the other hand, lattice expansion should reach the maximum at the 3:1 mole ratio. Probing the surface by UV -visible (surface plasmon technique) suggested that this mole ratio contained some Ag atoms on the surface. The exposure of the Ag atoms on surface is possible to cause the decrease in number of ensemble sites for the specific reaction resulting in the decrease in activity. Thus, the fully covered shell with a minimum thickness of Ag@Pd nanoparticle corresponds to the catalyst with the maximum activity.

Adsorption Strength of Formate on Various Metal-Core Pd-shell Nanoparticles .

The surface structure of the bimetallic catalysts was investigated by CO adsorption using ATR-IR spectroscopy. The electronic properties of core-shell type catalysts were inferred by analysing the chemical shifts of adsorbed formate (i.e., to assess the adsorption strength for formate on various core-shell catalysts by solution ¹³C NMR spectroscopy) and their catalytic activity for formic acid electro-oxidation was examined by electrochemical techniques. The surface probing experiments conducted on 1 : 1 M@Pd by means of CO adsorption indicated that the Pd-shell layers in these cases were fully covered.

Table 10: Variation of chemical shifts of different modes of adsorbed formate species on various metal-core Pd-shell catalysts

The adsorption strengths of formate on Ag@Pd nanoparticles are discussed above. The chemical shifts of adsorbed formate on Ru@Pd and Au@Pd were similar to each other but the chemical shift of Rh@Pd shifted more downfield (higher chemical shift). Pd has higher work function (5.60 eV) than Ru (5.52 eV) and results in the charge transfer from the Ru core atoms to the Pd shell atoms (ligand effect), giving stronger interaction between Pd and adsorbed formate (shifting the chemical resonance to downfield). However, in term of strain effect (data obtained from DFT calculations), adding Pd overlayers on Ru surface is proposed as the d-band center shifts away from the Fermi level (-0.47 eV) and the interaction strength between adsorbed formate and Pd surface should be decreased. This corresponds to the compression of lattice parameter (~1 %) of Pd shell from the XRD characterization. The ligand effect in the case of Au@Pd causes charge transfer from Au- core (lower work function (5.47 eV) to Pd-shell (5.60 eV) and thus, a stronger adsorption of formate is expected due to the increased back donation. In addition, XRD results indicated approximately 1% expansion of Pd-shell and thus, a stronger adsorption was also anticipated. The ¹³C resonance of adsorbed formate on Rh@Pd core-shell nanoparticle appeared more downfield (166.44 ppm). Based on the electronic effect in a Rh@Pd core- shell catalyst, the charge transfer from Rh (4.98 eV) to Pd (5.60 eV) is expected to result in stronger interaction of adsorbed formate on the catalyst whereas lattice strain effect in the catalyst based on XRD results (0.23 % compression). With smaller Rh metal-core (2.0 nm) in Rh@Pd at 1 : 1 mole ratio and, the Pd-shell covers Rh-core as thin layer, it is possible that the charge transfer effect is dominant over the strain effect.

Fig. 22(a) shows the relationship between C chemical shifts of three modes of adsorbed formates on Ag@Pd, Rh@Pd, Au@Pd, Ru@Pd and Pt@Pd bimetallics and work functions of Ag, Rh, Au, Ru, Pd and Pt from (111) fee lattice plane are used while (001) of hep Ru with identical surface feature is used. Work function is the minimuim energy needed to remove an electron from the Fermi level. In chemical language, it is similar to the ionization potential (binding energy) that reflects the energy state of electrons on solid. It is clear from Fig. 22(a) that the charge transfer between metal-core and metal-shell is related to the work function and the work function in turn is related to the adsorption strength. In addition, work function is determined by surface orientation and surface charge redistribution. Charge transfer can determine the sign and magnitude of surface dipole change leading to a strong dependence on the orientation of the substrate. Therefore, the work function is governed by the details of the charge redistribution. Charge redistribution is a factor which contributes to the shift of d-band of Pd-shell. Therefore, the data on work-function anisotropy obtained from the (1 1 1 ) crystal plane of fee structure of Pt, Pd, Rh, Ag and Au metals were used for the plot as shown in Fig. 22(a), while (001) of Ru metal with hep structure having the same surface feature as (11 1) of fee metals was also included. The excellent correlation of the line clearly indicates the charge transfer between core metal and Pd shells play a key role in determining the adsorption energy of formate. Thus, the electronegativity of metal-core is believed to have an influence on the overlayer. If the metal-core is less negative (lower work function) than the metal-shell then an electron transfer from metal-core to metal shell occurs, resulting in partial negative charge on the surface layer and positive charge on metal-core layer. This leads to a dipole pointing inward that enhances the original surface dipole due to electron "spilling out" upon an increase in the metal work function, which in turn leads to stronger adsorption strength of the adsorbed formate and the ¹³C resonance shifts downfield. Catalytic Activity of Different Pd-Based Core-shell Catalysts for Formic Acid Electro- Oxidation

As detailed above, the electronic properties of the metal-core Pd-shell (M@Pd, where M = Ru, Rh, Pt, Ag and Au) bimetallic catalyst precursor colloids were inferred by analysing the chemical shifts of adsorbed formate. Their catalytic activity for formic acid electro- oxidation was then examined by cyclic- voltametry (CV) over the carbon supported bimetallic particles after the removal of stabiliser. Prior to this experiment CO

chemisorption confirmed the Pd-shell layers were fully covering the inner core M (no CO chemisorption peak) in the 1 :1 @Pd samples. The electron promotion effect of metal core on Pd shell to further enhance the chemical shift value was evident in term of relative work function of the metals. An excellent linear correlation was indeed observed as shown in Fig 22(a).

A plot of specific activity against the chemical shift of adsorbed formate for different monometallic and Metal-core Pd-shell bimetallic catalysts is shown in Fig. 22(b). The maximum mass activity for formic acid electro-oxidation was observed for the

nanocatalyst based on Ag-core Pd-shell structure.

It is clear from Fig. 22(a) that there is a charge transfer between metal-core and metal-shell due to difference in work function which affects the dynamic adsorption strength (change in chemical shift value) under the same catalytic operation conditions in liquid phase. The excellent linear correlation clearly indicates that the charge transfer between core metal and Pd shells of these bimetallic nanoparticles plays a key role in determining the dynamic adsorption energy of formate.

The plot of specific activity against the chemical shift of adsorbed formate for the different monometallic and metal-core Pd-shell bimetallic catalysts synthesised in Example 1 is shown in Fig. 22(b). An excellent linear relationship between chemical shift and specific activity of electrocatalysts is observed over both monometallic and core-shell catalysts. So far from this work, Ag@Pd offers the highest specific activity of all catalyst samples, the activity of this new electrocatalyst being over 30% greater than Au@Pd, the best reported electro-catalyst for formic acid oxidation in the literature (Zhou, W. J.; Lee, J. Y.

Electrochem. Commun. 2007, 9, 1725-1729). EXAMPLE 3: Catalytic hydrogen gas production from formic acid in water at room temperature

Formic acid is nontoxic and a liquid at room temperature with a density of 1.22 g. mL^"3 and it can be safely handled in aqueous solution. The present work shows for the first time that solid Ag-Pd core-shell nano-catalysts can rapidly decompose HCOOH in water to form H₂ and C0₂ (1 : 1) gas mixture with CO concentration of lower than 10 ppm at room temperature (20°C). A turnover frequency (TOF) per surface metal site of around 88 h^"1 is estimated, giving a large amount of hydrogen gas (2.8L.g^"1.h^'1) at ambient conditions with no additive(s) required.

Figure 23(a) shows the rates of formic acid decomposition in water over different metal colloids at room temperature. As seen from the figure, the rate of formic acid

decomposition increases from electron rich elements primarily from the right hand side of periodic table (weak adsorption) to electron poor elements on the left hand in periodicity (strong adsorption). Density Function Theory (DFT), summarized by the d-band center model, has been successfully used in the literature to model various chemisorption systems, and has shown that the closer the d-band center is to the Femi level, the higher the adsorption energy is. Thus, the plot in Figure 23(a) is consistent with the fact that the rate of formic acid decomposition depends on the strength of formic acid chemisorption (the measured catalytic activity of Pt was only slightly lower than expected due to surface fouling/poisoning). From our results, the order of activity in the metals studied, for formic acid decomposition activity in water when 4 nm particle size was used, is as follows: Pd >Rh > Pt ~ Ru >Au > Ag. Our order apparently does not fit to the order of calculated static adsorption energies of bridging formate on (1 10) and (11 1) metal surfaces using First Principle DFT calculations. On the other hand, the calculated hydrogen atom adsorption enthalpies on (111) metal surfaces appear to follow nicely with the experimental order. Regarding the mechanism of formic acid decomposition to CO₂ and H₂ on metal surface in water, the initial O-H breakage of formic acid is a facile reaction but further C-H breakage of adsorbed formate to CO₂ and H is shown to be rate limiting. One can envisage that the metal surface not only adsorbs the formate specie through the interaction of oxygen atoms, the interaction of M-H in transition state will also facilitate its dissociation (account for the activity order over transition metals following their H atom adsorption enthalpies). Thus, the dynamic interactions of M-O-C and M-H-C in transition state of the adsorbate and catalyst surface prior to products formation is expected to be important. This shows the direct relationship with the catalytic activity of the metal particles. It is interesting to point out that the solvent (water) effect on formic acid oxidation over metal surface has recently been stressed in a modeling approach, which reported the C-H adsorption and activation of formic acid on metal surface due to local chemical environment in aqueous phase. Fig. 23(b) shows the dramatic size effect of typical Au particles on the rate of decomposition. Clearly, the smaller Au size gave higher NMR chemical shift in formic acid adsorption, which resulted in the higher reaction rate in a linear manner. It is therefore important to investigate the size effect of these small nanocatalysts.

It is interesting to know how the change in the surface potential of transition metal nanoparticles can help to catalyse formic acid decomposition in water. Formic acid is a weak acid (Ka = 1.8 x 10^"4) in water and thus only a very small fraction is in free formate species when no catalyst is added. From viewpoint of molecular structure, we expect that the adsorption of formic acid would favor a readily dissociative adsorption, giving electron delocalized formate species that can bind stronger on metal surface (sigma and pine bonding). The formation of two strong M-0 chemisorption bonds in the most stable bridging mode through the sigma and pine interactions with the d-band electrons could further facilitate the activation of C-H in transition state, giving C0₂. For smaller electron rich nanoparticles such as Au, there is a significant quantum size effect (the downshift of Fermi level owing to change in average coordination numbers). This renders the metal surface to form stronger M-0 and M-H interactions hence withdrawing more electron density from ¹³C which facilitates decomposition of the adsorbed formate species to C0₂. In order further to enhance the activity of Pd monometallic catalyst but with retention of the Pd surface feature, a series of metal-core Pd-shell (M@Pd, where M = Ru, Rh, Pt, Ag and Au) bimetallics were designed, synthesised and characterized as described above in Example 1. The electronic properties of core-shell type catalyst colloids were inferred by analysing their chemical shifts of adsorbed formate and their catalytic activity for formic acid decomposition to C0₂/H₂ at room temperature was also examined. Prior to this experiment CO chemisorption confirmed the Pd-shell layers were fully covering the inner core M (no CO chemisorption peak) over the 1 :1 M@Pd samples. The electron promotion effect of metal core on Pd shell to further enhance its electron density was evident in terms of relative work function of the metals. An excellent linear correlation was indeed observed as shown in Fig 24(a). It is noted that work function is the minimum energy needed to remove an electron from the Fermi level. In chemical language, it is similar to the ionization potential (binding energy) that reflects the energy state of electrons on solid. It is clear from Fig. 24(a) that there is a charge transfer from metal-core to metal-shell due to difference in work function which affects the dynamic adsorption strength (change in chemical shift value) under the same catalytic operation conditions in liquid phase. The excellent linear correlation clearly indicates the charge transfer between core metal and Pd shells of these bimetallic nanoparticles play a key role in determining the dynamic adsorption energy of formate. A plot of specific activity against the chemical shift of adsorbed formate for all different monometallic and metal-core Pd-shell bimetallic catalysts is shown in Fig. 24(b). An excellent linear relationship between chemical shift and activity of formic acid decomposition is observed over both monometallic and core- shell catalysts. Thus, the silver core-Pd shell, Ag@Pd nanocatalyst clearly shows the highest charge transfer from silver to palladium (see difference in work function) as reflected by the highest chemical shift value of adsorbed formate on its surface. This catalyst also gives the highest activity for formic acid decomposition to CO and H₂ at room temperature. The surface modification of Ag-core with different molar ratios of Pd overlayer has brought changes in the chemical shift of adsorbed formate, as shown in Fig.25(a). For example, for Ag-core Pd-shell catalysts, the resonances for different molar ratios (3:1, 2: 1 and 1 : 1 Ag to Pd) were shifted to higher chemical shifts compared to that of adsorbed formate on monometallic Pd catalyst but the 1 :3 molar ratio gave a very similar value as the pure Pd. The ¹³C resonance of adsorbed formate on Ag@Pd appeared to give higher chemical shift compared to that of adsorbate on monometallic Pd indicating stronger interaction of adsorbed formate. When Ag and Pd are^'in contact at the interface, there is electron transfer from Ag of higher Fermi level/lower electronegativity to Pd due to ligand effect and thereby the adsorption strength of formate on Pd shell is expected to be stronger (increase a higher degree of back donation to adsorbed formate).

The catalytic activity for formic acid decomposition of various Ag@Pd catalysts at different molar ratios is presented in Fig. 25(b). The highest activity was achieved for Ag@Pd catalyst with 1 : 1 molar ratio, which is in very good agreement with the highest chemical shift values observed from solution C NMR spectroscopy. The activity of Ag@Pd was found to decrease with either increasing or decreasing mole ratio of Ag or Pd. If some of Ag atoms are exposed on the surface of the Ag@Pd catalyst when a high molar ratio of the Ag-core is used, the relatively inactive Ag atoms can cause a decrease in catalyst activity. On the other hand, too thick a Pd shell can also decrease catalytic activity, which is presumably influenced by the lower degree of electronic perturbation from the Ag inner core.

Table 11. Initial results on catalytic decomposition of formic acid in water over 1:1 Ag@Pd nanocatalyst

Reaction conditions: 4M (or 8M) formic acid in lOmL total volume, Temperature as specified in Table 11 ; Reaction Time= 3.5 hours; total gas measured by gas burette (GC analysis confirmed the gas contained H₂:C0₂=1 :1 with CO <10 ppm); formic acid conversion was estimated by volumetric titration; mass balance= > 95% when compared to the quantity of C0₂ liberated. Turnover number (TON) is defined as (number of moles of formic acid converted to H₂/C0₂)/(number of mole of catalyst) for 3.5 h reaction.

After the identification of the best catalyst (1 :1 Ag@Pd) an initial study on the gas liberation was carried out. As seen from Table 1 1, 1 :1 H₂ to C0₂ gas mixture was produced from the catalyst over a fixed time of 3.5 hours in water with no stirring when the catalyst was in contact with formic acid solution. A detailed analysis of this catalyst for formic acid decomposition at room temperature (20°C) was carried out. As seen from Figure 26, the rate of gas liberation depends on the concentration of formic acid used. There is an apparent decrease in decomposition rate at high concentration of formic acid. Taking the initial gas production rate as ImL of hydrogen produced (2mL total gas) over 5 minutes with 4.26mg catalyst used at room temperature, the estimated hydrogen gas production rate = 1 x 12/(4.26 x 10^"3) = 2.8 Litre-¾ h^'1 g^"1. For comparison, the estimated TON for 3.5h at room temperature = 43.8 (taking assumption of no deactivation at the prolonged reaction time). In terms of turnover frequency over the surface Pd atoms on this

2 1

new core-shell catalyst, the estimated surface area of 80m g^' of this catalyst was obtained by the electrochemical surface area technique (ECSA). Thus, a number of Pd surface atoms of 4.26 x 10^"3 x 80 x 1 x 10¹⁹ = 3.408 x 10¹⁸ (1 x 10¹⁹ as the surface packing density of Pd atoms) was obtained. Thus, the turnover frequency (TOF) for hydrogen production of 12x [l/(24xl0³)x 6.023 x 10²³]/3.408 x 10¹⁸ = 88.4 h^'1 was calculated. It is noted that the activity of this presently un-optimised solid nano-catalyst in aqueous formic acid at room temperature can reach a comparable formic acid decomposition activity to that reported using a homogeneous platinum phosphine catalyst (TOF of 100 h^"1 in an organic solvent measured after their initial 15 min at room temperature; Yoshida, Y. Ueda, S. Otsuka, J. Am. Chem. Soc. 1978, 100, 3941 - 3942).

Claims

1. A process for producing H₂ from a compound of formula (I), (II), (III) or (IV): R'COOH (I)

R²OH (II)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted CMO alkyl;

R is unsubstituted or substituted Ci.₁₀ alkyl;

R³ is H or unsubstituted or substituted CMO alkyl;

R⁴ is H, unsubstituted or substituted C_1-10 alkyl, or C(0)NR⁵R⁶;

- (b) nanoparticles which comprise a metal selected from palladium, rhodium, ruthenium, iridium, copper and silver, and which have a mean particle size of less than or equal to 50 run;

provided that when said compound is a compound of formula (II), the catalyst comprises (a) said polymetallic nanoparticles. . . 2. A process according to claim 1 provided that when said compound is a compound of formula (II) or (III), the catalyst comprises (a) said polymetallic nanoparticles.

3. A process according to claim 1 provided that when said compound is a compound of formula (II), (III) or (IV), the catalyst comprises (a) said polymetallic nanoparticles.

4. A process according to any one of claims 1 to 3 wherein the compound of formula (I) is selected from formic acid and acetic acid; the compound of formula (II) is selected from methanol, ethanol and ethylene glycol; the compound of formula (III) is selected from formaldehyde and acetaldehyde; and the compound of formula (IV) is selected from ammonia and urea.

5. A process according to any one of the preceding claims for producing H₂ from formic acid, wherein said liquid phase comprises said compound of formula (I) which is formic acid.

6. A process according to any one of the preceding claims which further comprises recovering said ¾. 7. A process according to any one of the preceding claims wherein said contacting is performed at a temperature which does not exceed 100 °C.

8. A process according to any one of the preceding claims wherein said contacting is performed at a temperature which does not exceed 40 °C.

9. A process according to any one of the preceding claims wherein said catalyst comprises (b) said nanoparticles which comprise a metal selected from palladium, rhodium, ruthenium, iridium, copper and silver. 10, A process according to any one of the preceding claims wherein said nanoparticles (b) comprise a metal selected from rhodium, ruthenium, iridium and copper.

1 1. A process according to claim 9 or claim 10 wherein said nanoparticles have a mean particle size of from 1 nm to 50 nm.

12. A process according to claim 9 or claim 1 1 wherein said nanoparticles comprise a metal selected from palladium, rhodium and ruthenium.

13. A process according to any one of claims 9 to 12 wherein said nanoparticles comprise a metal selected from rhodium and ruthenium.

14. A process according to any one of claims 9 to 13 wherein said nanoparticles comprise rhodium.

15. A process according to any one of claims 1 to 9, 1 1 and 12 wherein said nanoparticles (b) comprise palladium. 16. A process according to any one of claims 12 to 15 wherein the nanoparticles have a mean particle size of less than or equal to 10 nm.

17. A process according to any one of claims 1 to 8 wherein said catalyst comprises (a) said polymetallic nanoparticles.

18. A process according to claim 17 wherein said first metal is selected from palladium, platinum, rhodium or iridium.

19. A process according to claim 17 or claim 18 wherein said first metal is palladium.

20. A process according to any one of claims 17 to 19 wherein the work function of the second metal is less than the work function of the first metal.

21. A process according to any one of claims 17 to 20 wherein the work function of the second metal is less than 5.6 eV.

22. A process according to any one of claims 17 to 21 wherein said second metal is selected from: a d-block metal of any one, of Groups 8, 9, 10, 1 and 12, a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, provided that said second metal is other than said first metal.

23. A process according to any one of claims 17 to 22 wherein said second metal is selected from copper, silver, gold, nickel, palladium, platinum, cobalt, rhodium, iridium and ruthenium.

24. A process according to any one of claims 17 to 22 wherein the second metal is a d- block metal of any one of Groups 8, 9, 10, 11 and 12, a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, provided that the second metal is other than palladium, platinum, gold, iron, cobalt and nickel.

25. A process according to any one of claims 17 to 22 wherein the second metal is as defined in any one of claims 62 and 66 to 74. 26. A process according to any one of claims 17 to 24 wherein the second metal is silver.

27. A process according to any one of the preceding claims wherein said catalyst comprises (a) said polymetallic nanoparticles, which nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal which is palladium.

28. A process according to any one of the preceding claims wherein the polymetallic nanoparticles have a mean particle size of less than or equal to 50 nm.

29. A process according to any one of the preceding claims wherein the polymetallic nanoparticles have a mean particle size of less than or equal to 25 nm.

30. A process according to any one of the preceding claims wherein the molar ratio of the second metal to the first metal in said polymetallic nanoparticles is from 2:1 to 1 :2.

31. A process according to any one of the preceding claims wherein the molar ratio of the second metal to the first metal in said polymetallic nanoparticles is 1 : 1. 32. A process according to any one of the preceding claims wherein said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is silver.

33. A process according to claim 32 wherein said polymetallic nanoparticles are as further defined in claim 77 or claim 78. 34. A process according to any one of claims 1 to 31 wherein said catalyst comprises (a) said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is rhodium.

35. A process according to claim 34 wherein said polymetallic nanoparticles are as further defined in claim 79 or claim 80.

36. A process according to any one of claims 1 to 31 wherein said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is ruthenium.

37. A process according to claim 36 wherein said polymetallic nanoparticles are as further defined in claim 81 or claim 82.

38. A process according to any one of claims 1 to 31 wherein said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is gold.

39. A process according to claim 38 wherein the polymetallic nanoparticles have a mean particle size of from 1.0 nm to 15.0 nm. 40. A process according to claim 38 or claim 39 wherein the molar ratio of the gold to the palladium in said polymetallic nanoparticles is from 1.5:1 to 1 :1.5.

41. A process according to any one of claims 1 to 31 wherein said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise a core which comprises said second metal, and a shell surrounding the core, which shell comprises said first metal, wherein the first metal is palladium and the second metal is platinum.

42. A process according to claim 41 wherein the poly metallic nanoparticles have a mean particle size of from 1 nm to 15 nm.

43. A process according to claim 41 or claim 42 wherein the molar ratio of the platinum to the palladium in said polymetallic nanoparticles is from 1.5:1 to 1 :1.5. 44. A process according to any one of claims 1 to 31 wherein said catalyst comprises said polymetallic nanoparticles, which polymetallic nanoparticles comprise an alloy comprising said first metal and said second metal, wherein said first metal is silver or platinum, and said second metal is palladium. 45. A process according to claim 44 wherein the polymetallic nanoparticles have a mean particle size of less than or equal to 25 nm.

46. A process according to any one of the preceding claims wherein said catalyst further comprises a polymer.

47. A process according to claim 46 wherein the polymer comprises polyvinyl pyrrolidone) (PVP), polyvinyl alcohol) (PVA) or poly(ethylenimine) (PEI), preferably wherein the polymer comprises poly(ethylenimine) (PEI). 48. A process according to any one of the preceding claims wherein said catalyst comprises said nanoparticles or said polymetallic nanoparticles and a solid support material, wherein the nanoparticles are supported on said solid support material.

49. A process according to claim 48 wherein the solid support material comprises carbon, an oxide, a nitride, or nanotubes.

50. A process according to any one of claims 1 to 47 wherein said catalyst does not comprise a solid support material.

51. A process according to claim 50 wherein said contacting of said liquid phase with said catalyst results in a colloidal suspension of said nanoparticles or said polymetallic nanoparticles in said liquid phase. 52. A process according to any one of the preceding claims wherein the liquid phase consists of said compound of formula (I), (II), (III) or (IV).

53. A process according to any one of claims 1 to 51 wherein the liquid phase comprises said compound of formula (I), (II), (III) or (IV) and a solvent.

54. A process according to claim 53 wherein the liquid phase comprises formic acid and a polar, protic solvent, preferably wherein the solvent comprises water.

55. A process according to any one of the preceding claims wherein the concentration of said compound of formula (I), (II), (III) or (IV) in said liquid phase is from 0.01 M to 26.52 M, preferably from 0.1 M to 10.0 M

56. A process according to any one of the preceding claims wherein the concentration of said compound of formula (I), (II), (III) or (IV) in said liquid phase is from 0.6 M to 1.4 M, preferably from 0.7 M to 1.3 M, more preferably from 0.8 M to 1.2 M.

57. A process according to any one of the preceding claims wherein the concentration of said catalyst in said liquid phase is less than or equal to 0.01 M, preferably less than or equal to 0.0005 M.

58. A process according to any one of the preceding claims wherein said compound of formula (I), (II), (III) or (IV) is formic acid and said contacting results in the production of a gaseous product mixture comprising ¾ and C0₂. 59. A process according to claim 58 wherein the gaseous product mixture comprises a 1 : 1 molar ratio of ¾ and C0₂.

60. A process according to claim 58 or claim 59 wherein said gaseous product mixture comprises no more than lOppm by volume carbon monoxide.

61. A process according to any one of the preceding claims wherein said compound of formula (I), (II), (III) or (IV) is formic acid and the production of ¾ from formic acid occurs in a single step, in accordance with the following reaction:

HCOOH C0₂ + ¾

HCOOH CO + H₂0 62. A catalyst for the production of hydrogen from a compound of formula (I), (II), (III) or (IV), or for the electro-oxidation of a compound of formula (I), (II), (III) or (IV):

R^OOH (I)

R²OH (II)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted C O alkyl;

R² is unsubstituted or substituted CMO alkyl;

R³ is H or unsubstituted or substituted C O alkyl;

R⁴ is H, unsubstituted or substituted C_M0 alkyl, or C(0)NR⁵R⁶;

which catalyst comprises polymetallic nanoparticles, which polymeta!lic nanoparticles comprise a core and a shell surrounding the core; wherein the shell comprises a first metal which is palladium, and the core comprises a second metal, wherein the second metal is other than palladium, platinum, gold, iron, cobalt, nickel, titanium, tungsten, tantalum, vanadium and niobium.

63. A catalyst according to claim 62 wherein the second metal is a d- block metal of any one of Groups 8, 9, 10, 1 1 and 12, a Group 1 (alkali) metal, or a Group 2 (alkaline earth) metal, provided that the second metal is other than palladium, platinum, gold, iron, cobalt and nickel.

64. A catalyst according to claim 62 or claim 63 wherein the work function of the second metal is less than the work function of palladium.

65. A catalyst according to any one of claims 62 to 64 wherein the work function of the second metal is less than 5.6 eV, preferably wherein the work function of the second metal is less than or equal to 5.3 eV.

66. A catalyst according to any one of claims 62 to 65 wherein said second metal is selected from silver, rhodium, ruthenium, copper and iridium.

67. A catalyst according to any one of claims 62 to 66 wherein said second metal is silver, rhodium, ruthenium or copper.

68. A catalyst according to any one of claims 62 to 67 wherein said second metal is silver, rhodium or copper.

69. A catalyst according to any one of claims 62 to 67 wherein said second metal is silver, rhodium or ruthenium. 70. A catalyst according to any one of claims 62 to 68 wherein said second metal is silver or copper.

71. A catalyst according to any one of claims 62 to 70 wherein the second metal is silver.

72. A catalyst according to any one of claims 62 to 68 wherein said second metal is rhodium or copper.

73. A catalyst according to any one of claims 62 to 69 and 72 wherein said second metal is rhodium.

74. A catalyst according to any one of claims 62 to 68, 70 and 72 wherein said second metal is copper.

75. A catalyst according to any one of claims 62 to 74 wherein the polymetallic nanoparticles have a mean particle size of less than or equal to 50 run, preferably wherein the polymetallic nanoparticles have a mean particle size of less than or equal to 25 nm. 76. A catalyst according to any one of claims 62 to 74 wherein the molar ratio of the second metal to the first metal in said polymetallic nanoparticles is from 2: 1 to 1:2, preferably wherein the molar ratio of the second metal to the first metal in said

polymetallic nanoparticles is 1 :1. 77. A catalyst according to any one of claims 62 to 76 wherein the second metal is silver and the polymetallic nanoparticles have a mean particle size of from 1 nm to 35 nm, preferably from 10 nm to 25 nm, more preferably from 17 to 25 nm.

78. A catalyst according to claim 77 wherein the molar ratio of the silver to the palladium in said polymetallic nanoparticles is from 1.5:1 to 1 :1.5, preferably wherein said molar ratio is about 1 :1.

79. A catalyst according to any one of claims 62 to 76 wherein the second metal is rhodium and the polymetallic nanoparticles have a mean particle size of less than or equal to 4 nm.

80. A catalyst according to claim 79 wherein the molar ratio of the rhodium to the palladium in said polymetallic nanoparticles is from 1.5:1 to 1 :1.5, preferably wherein said molar ratio is about 1 :1.

81. A catalyst according to any one of claims 62 to 76 wherein the second metal is ruthenium and the polymetallic nanoparticles have a mean particle size of less than or equal to 5 nm. 82. A catalyst according to claim 81 wherein the molar ratio of the ruthenium to the . palladium in said polymetallic nanoparticles is from 1.5: 1 to 1 :1.5, preferably wherein said molar ratio is about 1 :1.

83. A catalyst according to any one of claims 64 to 82 wherein the catalyst comprises the polymetallic nanoparticles and a solid support material, wherein the polymetallic nanoparticles are supported on said solid support material, optionally wherein the support material comprises an oxide, a nitride, carbon or nanotubes.

84. A catalyst according to claim 83 wherein the support material comprises carbon.

85. A catalyst according to claim 83 or claim 84 wherein the polymetallic nanoparticles are present in an amount of from 10 to 30 weight %, based on the total weight of the catalyst including the solid support material.

86. A catalyst according to any one of claims 83 to 85, wherein the second metal is silver and the polymetallic nanoparticles have a mean particle size of less than or equal to 40 run.

87. A catalyst according to claim 86 wherein the molar ratio of the silver to the palladium in said polymetallic nanoparticles is from 1.5: 1 to 1 : 1.5, preferably wherein said molar ratio is about 1 :1. 88. A process for producing a catalyst, which catalyst comprises polymetallic nanoparticles, which polymetallic nanoparticles comprise a core and a shell surrounding the core, wherein the shell comprises a first metal which is palladium, and wherein the core comprises a second metal, wherein the second metal is other than palladium, platinum and gold; which process comprises:

(a) reducing a salt of said second metal in the presence of a first solvent, to produce a suspension or solution of core nanoparticles in said first- solvent, which core nanoparticles comprise said second metal; and

(b) reducing a palladium salt in the presence of s id core nanoparticles and a second solvent, which is the same or different from the first solvent, to produce a shell on the surfaces of said core nanoparticles, which shell comprises palladium.

A process according to claim 88 wherein the palladium salt is palladium nitrate.

90. A process according to claim 88 or claim 89 wherein the first and second solvents are independently selected from ethylene glycol, an alcohol, and water.

91. A process according to any one of claims 88 to 90 which comprises:

92. A process according to any one of claims 88 to 91 which comprises:

(bl) treating a suspension or solution of said core nanoparticles in a second solvent with hydrogen gas, wherein the second solvent is the same or different from the first solvent; and

(b2) reducing a palladium salt in the presence of said core nanoparticles, said second solvent and/or another solvent, and a second stabilising polymer, to produce a shell which comprises palladium on the surfaces of said core nanoparticles, wherein the second stabilising polymer is the same or different from the first stabilising polymer.

93. A process according to claim 91 or claim 92 wherein the first and second stabilising polymers are independently selected from polyvinyl pyrrolidone) (PVP), polyvinyl alcohol) (PVA) and poly(ethylenimine) (PEI). 94. A process according to any one of claims 88 to 93 wherein said suspension or solution of core nanoparticles produced in step (a) is a colloidal suspension or solution.

95. A process according to any one of claims 88 to 94 wherein, in step (a) the step of reducing said salt of said second metal is performed by heating said salt in the presence of said first solvent and, when present, said stabilising polymer, under an inert atmosphere. 96. A process according to any one of claims 88 to 95 wherein, in step (b), or step (b2), the step of reducing said palladium salt is performed by heating said salt in the presence of said second solvent and, when present, said second stabilising polymer, under an inert atmosphere. 97. A process according to any one of claims 88 to 96 wherein step (a) further comprises isolating said core nanoparticles from said solvent.

98. A process according to any one of claims 88 to 97 wherein step (b), or step (b2), further comprises isolating said polymetallic nanoparticles from said solvent.

99. A process according to any one of claims 88 to 98 which further comprises (c) depositing said polymetallic nanoparticles on a support material, thereby producing a catalyst which comprises said polymetallic nanoparticles and a support material, wherein the polymetallic nanoparticles are supported on said support material.

100. A process according to claim 99 which further comprises (d) heating said catalyst to remove any residual stabilising polymer.

101. A process according to any one of claims 88 to 100 wherein the catalyst is as further defined in any one of claims 62 to 87.

102. A catalyst which is obtainable by a process as defined in any one of claims 88 to 101. ^■ 103. Use of a catalyst as defined in any one of claims 62 to 87 and 102 for the electro- oxidation of a compound of formula (I), (II), (III) or (IV):

R'COOH

R²OH R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted Ci-io alkyl;

R² is unsubstituted or substituted C O alkyl;

R³ is H or unsubstituted or substituted d-io alkyl;

R⁴ is H, unsubstituted or substituted CMO alkyl, or C(0)NR⁵R⁶;

104. Use of a catalyst as defined in any one of claims 62 to 87 and 102 for the electro- oxidation of formic acid.

105. Use of a catalyst as defined in any one of claims 62 to 87 and 102 in a fuel cell.

106. Use of a catalyst as defined in any one of claims 62 to 87 and 102 for the electro- oxidation of a compound of formula (I), (II), (III) or (IV) in a fuel cell:

R'COOH (I)

R²OH (II)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted CMO alkyl;

R² is unsubstituted or substituted CMO alkyl;

^' R³ is H or unsubstituted or substituted CMO alkyl;

R⁴ is H, unsubstituted or substituted C_M0 alkyl, or C(0)NR⁵R⁶;

107. Use of a catalyst as defined in any one of claims 62 to 87 and 102 for the electro- oxidation of formic acid in a direct formic acid fuel cell.

108. An electrode for use in a fuel cell, which electrode comprises a catalyst as defined in any one of claims 62 to 87 and 102.

109. A fuel cell which comprises: a catalyst as defined in any one of claims 62 to 87 and 102 or an electrode as defined in claim 108.

1 10. A process for the electro-oxidation of a compound of formula (I), (II), (III) or (IV)

R'COOH (I)

R²OH (Π)

R³CHO (III)

R⁴NH₂ (IV)

wherein

R¹ is H or unsubstituted or substituted C O alkyl;

R² is unsubstituted or substituted Cj.io alkyl;

R³ is H or unsubstituted or substituted CMO alkyl;

R⁴ is H, unsubstituted or substituted C^o alkyl, or C(0)NR⁵R⁶;

R⁵ and R⁶, which are the same or different, are independently selected from H and unsubstituted or substituted Ci_-10 alkyl;

which process comprises contacting a liquid phase which comprises said compound of formula (I), (II), (III) or (IV) with a catalyst as defined in any one of claims 62 to 87 and 102, in the presence of an electrode.