WO2009157033A2

WO2009157033A2 - Core-shell mono/plurimetallic carbon nitride based electrocatalysts for low-temperature fuel cells (pemfcs, dmfcs, afcs and electrolysers

Info

Publication number: WO2009157033A2
Application number: PCT/IT2009/000278
Authority: WO
Inventors: Noto Vito Di; Enrico Negro
Original assignee: Universita Degli Studi Di Padova
Priority date: 2008-06-26
Filing date: 2009-06-22
Publication date: 2009-12-30
Also published as: ITPD20080188A1; WO2009157033A3

Abstract

This invention describes the preparation of electrocatalysts to be mounted at either the anode or the cathode of fuel cells or of H2 electrolysers. The materials described in this invention are particularly suitable for application in polymer electrolyte membrane fuel cells (PEMFCs), direct methanol fuel cells (DMFCs), alkaline fuel cells (AFCs) and phosphoric acid fuel cells (PAFCs). The target of the invention is the preparation of core-shell electrocatalysts where active metal sites supported on carbon nitride nanoclusters (shell) are supported on suitable electron-conducting materials (core) such as active carbons or metal nanopowders. The prepared materials feature a well-controlled metal composition, and also include the desired amount of nitrogen. The preparation protocol consists of three steps. In the first the precursor is obtained through reactions of the type: a) sol-gel; b) gel-plastic; c) coagulation-flocculation-precipitation; d) metal-ligand complexation processes in organic solvent using as a ligand a molecule, a macromolecule or a macromolecular system. The second step consists in a pyrolysis process of the precursor in an inert atmosphere leading to the production of the supported carbon nitride materials. In the last step, the chemical and electrochemical activation of the electrocatalyst is performed. The precursors are obtained through reactions leading to: a) the complexation of a transition metal with another molecule or coordination complex acting as a ligand to obtain clusters; b) the 3-D networking of the resulting clusters through suitable organic molecules and/or macromolecules and/or macromolecular systems on the support.

Description

Description of the industrial invention titled:

"CORE-SHELL" MONO/PLURIMETALLIC CARBON NITRIDE BASED ELECTROCATALYSTS FOR LOW-TEMPERATURE FUEL CELLS (PEMFCs, DMFCs, AFCs AND PAFCs) AND ELECTROLYSERS on behalf of: University of Padova

Designated Inventors: Vito Di Noto, Enrico Negro

DESCRIPTION Field of the invention The invention concerns the anodic and cathodic electrocatalysts to be used in fuel cells operating at low temperatures such as PEMFCs, DMFCs, AFCs and PAFCs, and in H₂ electrolysers together with the methods used to prepare the materials constituting those electrocatalysts. State of the art The operation of a fuel cell at a low temperature (T < 500⁰C) is possible only if the reagents are quickly and efficiently converted into the products by suitable electrocatalysts. At the anode the fuel (i.e., hydrogen, methanol, ethanol, glycerol, etc..) is oxidized, yielding H⁺ ions and other products such as CO₂; at the cathode the oxidant (air or oxygen) is reduced to OH^" or O ^". In fuel cells operating with an acid electrolyte (e.g., PEMFCs, DMFCs, PAFCs), the protons produced at the anode migrate through the electrolyte to the cathode, where they are involved in the production of water and electric current. In AFCs, the OH^" ions produced at the cathode migrate through the electrolyte to the anode, where they are consumed in the production of water and electric current. Nowadays, the best electrocatalysts used in fuel cells operating at low temperatures are based upon noble metals such as platinum, palladium and ruthenium, either as very thin powders of the metals or of the alloys or as nanocrystals supported on active carbons characterized by a very large surface area. Different applications require systems characterized by a different chemical composition. PEMFCs operating with pure H₂ as the fuel provide the best performance when they mount electrocatalysts having a content of platinum lower than about 20 wt% at both the anode and the cathode. If a PEMFC is fuelled with H₂ deriving from the reforming of hydrocarbons, a supported Pt-Ru alloy electrocatalyst must be mounted at the anode of the cell so as to prevent the poisoning of the material by CO traces found in the fuel stream, while at the cathode the best performance is provided by an electrocatalyst characterized by supported Pt. DMFCs provide an optimal performance with electrocatalysts having a metal content higher than about 40%. At the anode it is necessary to use a Pt-Ru alloy so as to efficiently oxidize the methanol fuel, while at the cathode the electrocatalyst is supported Pt. The scientific literature reports that improved cathodic electrocatalysts can be obtained by alloying the supported platinum with suitable first-row transition metals such as Fe, Co, Ni, Cr, Ti and Mn. Palladium can also be used to prepare supported electrocatalysts, either by itself or alloyed with other metals such as Fe and Co. In both cases, these electrocatalysts are usually prepared by impregnating active carbons with salts of the desired metals followed by a subsequent reduction through a variety of methods (e.g., using chemical reagents such as NaBH₄ or formaldehyde, or under a H₂ current at a high temperature). Catalysts for Io w- temperature fuel cells capable to operate without platinum-group metals (PGM) have been proposed, but they are still characterized by quite a poor performance and a very low durability. Thus, they are unsuitable for most applications. As a consequence, the only catalysts which can be used in practical low-temperature fuel cells are those based on PGM and for this reason they are very expensive; this is a major limiting factor to a widespread commercial application of this technology. An important practical target is to obtain fuel cells capable to yield a high electrical power mounting as little weight of PGM as possible. As for PEMFCs fed with pure hydrogen, the current commercial devices are capable to yield about 1 kW per gram of PGM; several studies point to the necessity to reduce this figure by a factor between 3 and 10 to allow a widespread application of fuel cells with a particular emphasis on the automotive sector. Detailed description

The procedure to prepare materials having a large surface area to be used as anodic and cathodic electrocatalysts consists of three steps. In the first the precursor is obtained through chemical reactions based on the following processes: a) sol-gel; b) gel-plastic; c) coagulation and/or flocculation and/or precipitation; d) metal-ligand complexation processes in organic solvent using as a ligand a molecule, a macromolecule or a macromolecular system. The second step leads to the production of supported mono-plurimetallic carbon nitrides (S-MPM-CN) through suitable thermal treatments of the precursor in an inert atmosphere. The last step consists in the procedures to perform the chemical and electrochemical activations of the carbon nitride materials obtained in the second step. Precursors are obtained as described below in a detailed way through: a) complexation reactions of transition metals with complexes such as cyanometallates or molecules such as acetonitrile acting as ligands to produce clusters; b) 3-D networking of the obtained clusters with suitable organic molecules and/or macromolecules and/or macromolecular systems over the chosen support. Suitable supports include all the electron-conducting materials. Owing to the extremely large number of materials satisfying this criterion, the following list is reported only for illustrative and not for limitative purposes. Typical examples of electron-conducting materials suitable as supports for this invention are graphite powders or platelets, powders of metals such as titanium, silver, gold, platinum, zirconium, manganese, tungsten, lead, scandium, vanadium, iron, cobalt, nickel, zinc, bismuth, copper, chromium, yttrium, niobium, molybdenum, ruthenium, rhodium, palladium, cadmium, lutetium, hafnium, tantalum, tungsten, rhenium, osmium, iridium, lanthanum, cerium, praseodymium, neodymium, samarium, europium, gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium, gallium, indium, thallium, silicon, germanium, tin, etc...

There are two main procedures to obtain the precursor, according to the solubility properties of the molecules, macromolecules or macromolecular systems employed in the networking and coordination of the metal clusters. Procedure 1 is applied when all the networking and the coordinating agents are water-soluble molecules or macromolecules. Procedure 2 is applied when at least one of the networking and coordinating agents are not soluble in water. It should be highlighted that it is possible to use more than one type of molecule, macromolecule or macromolecular systems at the same time as networking and coordinating agents for the metal clusters to obtain the precursor. The molecules, macromolecules or macromolecular systems used for Procedure 1 must not contain sulphur atoms and must be very rich of oxygen atoms and hydroxyl groups. Typical examples include polyethylene glycol, polyvinyl alcohol, glycerol, sucrose, glucose, fructose and in general all the water-soluble macromolecules such as carbohydrates, etc... The molecules, macromolecules or macromolecular systems used for Procedure 2 must be soluble in organic solvents such as N-methyl-2-pyrrolidinone, dimethylformamide, dimethylacetamide, acetic acid, acetone, acetonitrile, benzene, 1-butanol, 2-butanol, 2-butanone, t-butyl alcohol, cyclohexane, diethyl ether, diethylene glycol, diglyme, dimethylether, dioxane, ethanol, ethyl acetate, ethylene glycol, heptane, hexane, methanol, methyl t-butyl ether, nitromethane, pentane, 1-propanol, 2- propanol, pyridine, tetrahydrofuran, toluene, triethyl amine, o-xylene, m-xylene, p-xylene, etc .; they must not contain sulphur atoms and must support functional groups featuring oxygen, nitrogen or phosphorus atoms such as -C≡N, -NH₂, -NHR, -NR₂, -OH, R-O-R, -C=O, -COOH, -COOR, -PH₂, -PHR, -PR₂. Owing to the extremely large number of molecules, macromolecules or macromolecular systems satisfying these criteria, the following list is reported only for illustrative and not for limitative purposes. Typical molecules which can be used as networking agents for Procedure 2 include dianilines such as p-phenylenediamine; 4,4'-methylenedianilme; 1,4- diaminobutane; dianhydrides such as 1,2,4,5-benzenetetracarboxylic anhydride; 3,3',4,4'- benzophenonetetracarboxylic dianhydride; (+)-diacetyl-L-tartaric anhydride; diphosphines such as 1 ,2-bis(diphenylphosphino)ethane; (-)-2,3-0-isopropylidene-2,3-dihydroxy-l ,4- bis(diphenylphosphino)butane; (-)-l ,2-bis[(2R,5R)-2,5-diethylphospholano]benzene; rac-2,2¹- bis(di-p-tolylphosphino)-l,r-binaphthyl; carboxylic acids such as tetrahydrofuran-2,3,4,5- tetracarboxylic acid; glutaric acid; terephthalic acid; molecules carrying more than one type of functional group such as BOC-L-glutamic acid, acetyl tributyl citrate and L(+)-glutamic acid.

Typical macromolecules which can be used as networking agents for Procedure 2 include polyimides such as Kapton, and Apical, polyamides such as Nylon 6 and Nylon 6,6, polyurethanes, polypyrrole, polyvinyl alcohol, polymethyl metacrylate, polyacrylonitrile, poly(tetramethylene ether) glycol, polyethylene glycol, etc...

Procedure 1 starts with the preparation of two solutions, A and B. The first, solution A, consists in a "soft" transition metal complex coordinated by good leaving groups and an aliquot of the selected networking agents dissolved in water. Typical examples of "soft" transition metal complexes used in the preparation of the solution A include, but are not limited to: HAuCl₄, H₂IrCl₆, H₂PtCl₆, Li₂PdCl₄, (NH₄)₂IrCl₆, (NH₄)₂OsCl₆, (NH₄)PdCl₄, (NH₄)₂PdCl₆, (NH₄)₂PtCl₄, (NH₄)₂PtCl₆, (NH₄)₃RhCl₆, (NH₄)₂RuCl₆, KAuCl₄, KPt(NH₃)Cl₃, K₂PdCl₄, K₂PtCl₄, K₂PdCl₆, K₂PtCl₆, K₂ReCl₆, K₂RhCl₆ K₂H₂IrCl₆, K₂H₂OsCl₆, K₃IrCl₆, K₃H₃RuCl₆, Na₂IrCl₆, Na₂OsCl₆, Na₂PdCl₄, Na₂PtCl₆, Na₃RhCl₆, CrCl₃, IrCl₃, FeCl₃, NiCl₂, OsCl₃, PdCl₂, PtCl₂, PtCl₄, RhCl₃, RuCl₃, ReCl₅, SnCl₄, VCl₃, VCl₄, WCl₄, WCl₆, ZrCl₄, etc...

Solution B is obtained by dissolving in water a suitable amount of a metal complex, such as a cyanometallate, capable to act as a ligand for "soft" metals, together with an aliquot of the networking agents used in the preparation of the solution A. Typical examples of cyanometallates which can be used include, but are not limited to: KAg(CN)₂, KAu(CN)₂, K₂Ni(CN)₄ K₂Pd(CN)₄, K₂Pt(CN)₄, K₃Co(CN)₆, K₃Cr(CN)₆ K₃Fe(CN)₆, K₄Fe(CN)₆-H₂O, K₃Mn(CN)₆, K₂Pt(CN)₆, K₄Ru(CN)₆.

Solutions A and B must be mixed together and stirred until they are completely transparent. Once the solution is well-mixed the resulting product must be let rest at room temperature. The chemical reaction is considered complete when one of the above-mentioned transitions (sol-gel and/or gel/plastic; coagulation and/or flocculation and/or precipitation; metal-ligand complexation process) is observed. It may be necessary to wait for a few days for these reactions to occur. The selected support may be added in each of the steps outlined above.

Should the product have the features of a suspension, it is necessary to remove the excess solvent. This result can be achieved with two different procedures: a) filtration; or b) by drying the reaction mixture. This latter treatment may be performed in a rotovapor at 60°C until a compact and homogeneous solid remains, or over a hot plate. The drying process may last a few days.

Procedure 2 starts with the dissolution of the metal complexes in an organic solvent. This is usually accomplished by a two-step procedure. At first, the minimum amount of water is used to dissolve the desired amount of each water-soluble metal complex. Typical examples include, but are not limited to: HAuCl₄, H₂IrCl₆, H₂PtCl₆, Li₂PdCl₄, (NH₄)₂IrCl₆, (NELO₂OsCl₆, (NH₄)PdCl₄, (NH₄)₂PdCl₆, (NH₄)₂PtCl₄, (NH₄)₂PtCl₆, (NH₄)₃RhCl₆, (NH₄)₂RuCl₆, KAuCl₄, KPt(NH₃)Cl₃, K₂PdCl₄, K₂PtCl₄, K₂PdCl₆, K₂PtCl₆, K₂ReCl₆, K₂RhCl₆ K₂H₂IrCl₆, K₂H₂OsCl₆, K₃IrCl₆, K₃H₃RuCl₆, Na₂IrCl₆, Na₂OsCl₆, Na₂PdCl₄, Na₂PtCl₆, Na₃RhCl₆, CrCl₃, IrCl₃, FeCl₃, NiCl₂, OsCl₃, PdCl₂, PtCl₂, PtCl₄, RhCl₃, RuCl₃, ReCl₅, SnCl₄, VCl₃, VCl₄, WCl₄, WCl₆, ZrCl₄, (NH₃)₄Pt(NO₃)₂, Pd(NO₃)₂-xH₂O, Co(NO₃)₂-xH₂O, Ni(NO₃)₂-xH₂O, Fe(NO₃)₃-xH₂O, Cr(NO₃)₃-xH₂O, Mn(NO₃)₂-xH₂O, AgNO₃, etc... Afterward, a large excess of organic solvent is used to dilute the water solution. Typical examples include, but are not limited to: N-methylpyrrolidone, dimethylformamide, dimethylacetamide, acetonitrile, etc... A further coordinating agent may be dissolved into the organic solvent. The molecules of the organic solvent may act as the coordinating agent themselves. New metal coordination compounds soluble in the desired organic solvent and characterized by labile ligands are thus formed. The desired networking agents are dissolved in a suitable organic solvent; typical examples include, but are not limited to: N-methylpyrrolidone, dimethylformamide, dimethylacetamide, acetonitrile, etc... At this point, the solutions containing the metal complexes obtained previously are added, leading to the development of the 3D networking of the precursor. The chemical reaction is considered complete when one of the above- mentioned transitions (sol-gel and/or gel/plastic; coagulation and/or flocculation and/or precipitation; metal-ligand complexation process) is observed. It may be necessary to wait for a few days for these reactions to occur. The support may be added in each of the steps outlined above. The organic solvent is removed from the product, for instance by evaporation, and the precursor can undergo the other steps of the preparation procedure. The obtained precursor undergoes a thermal treatment as described below. At first the material is kept under dynamic vacuum at 10^"3 bar for 16-24 hours at 150-200°C, then a first step is performed stabilizing the material at a higher temperature (300-1200°C) for no longer than two hours. Lastly, a further thermal treatment under dynamic vacuum (10^"3 bar) is performed at high temperature (400-1200°C), lasting between one and six hours. The resulting material is finely ground and subsequently washed repeatedly with water so as to remove soluble reaction byproducts such as halides.

The washed material is activated with hydrogen peroxide, 10% vol., and is eventually dried. The chemical reactions involved in the preparation of the precursor result in a compact and homogeneous material where the desired quantity of metallic centres is uniformly distributed. The thermal treatment performed under vacuum removes most of the oxygen and hydrogen atoms of the organic binder from the material; furthermore, it provides the energy necessary to activate the nucleation and growth of the carbon nitride-based clusters of the desired metals. The clusters are of nanometric size and have the programmed chemical composition. The electron-conducting support provides a matrix featuring an extremely large surface area where the catalytic material based on mono-plurimetallic carbon nitrides is supported; furthermore, it guarantees to the material the electrical conductivity necessary for its operation.

The main aim of the prolonged washing with bidistilled water is the removal from the electro- active material of possible halide-based soluble derivatives: should they bind permanently on the active cluster sites, they would compromise the catalytic activity of the material. The treatment with hydrogen peroxide is performed so as to "clean" the external surface of the clusters from impurities which could both poison the active catalytic sites and make the active metallic area of the material smaller.

The main features of the invention are highlighted in the following descriptions which should be considered, together with the attached graphs, specific information concerning particular examples reported only for illustrative and not for limitative purposes.

EXAMPLE 1 {Material PtNi-CNi 600/G}

This example reports the detailed description of the synthesis of a material of the type S-MPM- CN for the cathodic reduction of oxygen based on platinum and nickel. 1067 mg of sucrose were dissolved in the minimum amount of milli-Q water (~2 ml), yielding a viscous, transparent solution. 400 mg Of K₂PtCl₄ and 332 mg Of K₂Ni(CN)₄ with a hydration degree of 30% were each dissolved in the minimum amount of milli-Q water (~2 ml), yielding a deep red solution (A) and a clear yellow solution (B), respectively. The transparent sucrose solution was equally divided among A and B. 533 mg of XC-72R carbon black was added to each of A and B; the resulting black suspensions were diluted with about 5 ml of milli-Q water each to ensure a sufficiently low viscosity. A was added dropwise into B; the final product was stirred for 2 hours and then allowed to rest overnight. The resulting black suspension was then transferred into a ventilated oven and the water was removed at 100°C over 8 hours, yielding the precursor which was then transferred into a quartz tube. The quartz tube was connected to a vacuum line and brought to 10^"3 bar. The sample placed under vacuum underwent a first thermal treatment at 150°C for 16 hours. The resulting black solid was then thermally treated at 300⁰C for two hours under dynamic vacuum yielding a black, very rough powder. After finely grinding it into a mortar, the product was re-introduced into a quartz tube, where it underwent a further thermal treatment at 600°C under dynamic vacuum for two hours. Afterwards, the sample was removed from the tube, finely ground in a mortar and eventually washed four times with milli-Q water to remove the soluble byproducts and the chloride ions deriving from the chemical reactions involved in the preparation of the material. At the end of each washing step the mother waters were separated from the sample by centrifugation at 4500 rpm for 15 minutes.

The resulting wet slurry was then transferred to a Petri capsule and treated with about 20 ml of hydrogen peroxide, 10% vol. A significant evolution of bubbles was observed. The Petri dish was placed under an IR lamp to remove the water yielding the final material.

EXAMPLE 2 {Material PtNi-CNi 900/G}

This example includes the synthesis and characterization procedures of materials of the type S- MPM-CN for the cathodic reduction of oxygen based on platinum and nickel. The preparation of the material is exactly the same as the one described in Example 1, with the only difference that the final thermal treatment, lasting two hours, was performed at 900°. The product, once it was recovered from the quartz tube, was washed and activated as described in Example 1.

EXAMPLE 3 {Material PtFe-CN₁ 600/G}

This example reports the detailed description of the synthesis of a material of the type S-MPM- CN for the cathodic reduction of oxygen based on platinum and iron.

1067 mg of sucrose were dissolved in the minimum amount of milli-Q water (~2 ml), yielding a viscous, transparent solution. 400 mg of K₂PtCl₄ and 814 mg of K₄Fe(CN)₆-SH₂O were each dissolved in the minimum amount of milli-Q water (~2 ml), yielding a deep red solution (A) and a clear yellow solution (B), respectively. The transparent sucrose solution was equally divided among

A and B. 533 mg of XC-72R carbon black was added to each of A and B; the resulting black suspensions were diluted with about 5 ml of milli-Q water each to ensure a sufficiently low viscosity. A was added dropwise into B; the final product was stirred for 2 hours and then allowed to rest overnight. The resulting black suspension was dried, treated and activated as described in

Example 1 yielding the final PtFe-CNi 600/G material. EXAMPLE 4

{Material PtNi-CNi 900/G}

This example includes the synthesis and characterization procedures of materials of the type S-

MPM-CN for the cathodic reduction of oxygen based on platinum and iron. The preparation of the material is exactly the same as the one described in Example 3, with the only difference that the final thermal treatment, lasting two hours, was performed at 900°. The product, once it was recovered from the quartz tube, was washed and activated as described in Examples 1 and 3.

EXAMPLE 5

{Material PdCoNi-CNi 600/G} This example reports the detailed description of the synthesis of a material of the type S-MPM-

CN for the cathodic reduction of oxygen based on palladium, nickel and cobalt.

1280 mg of sucrose were dissolved in the minimum amount of milli-Q water (~2 ml), yielding a viscous, transparent solution. 555 mg OfK₂PdCl₄ and 585 mg OfK₂Ni(CN)₄ with a hydration degree of 30% were each dissolved in the minimum amount of milli-Q water (~2 ml), yielding a deep red solution (A) and a clear yellow solution (B), respectively. 595 mg of K₃Co(CN)₆ were then dissolved in B, yielding a clear yellow solution. The transparent sucrose solution was equally divided among A and B. 640 mg of XC-72R carbon black was added to each of A and B; the resulting black suspensions were diluted with about 5 ml of milli-Q water each to ensure a sufficiently low viscosity. A was added dropwise into B; the final product was stirred for a few minutes and then allowed to rest overnight. The resulting compact black gel was then transferred into a ventilated oven and the water was removed at 100⁰C over 8 hours, yielding the precursor which was then transferred into a quartz tube. The subsequent thermal treatment, washing and activation procedures to obtain the final PdCoNi-CN₁ 600/G material are the same as those described in Example 1. EXAMPLES 6-8

{Materials PdCoNi-CNi 500/G, PdCoNi-CNi 700/G, PdCoNi-CNi 900/G} T2009/000278

This example includes the synthesis and characterization procedures of materials of the type S- MPM-CN for the cathodic reduction of oxygen based on palladium, cobalt and nickel. The preparation of the material is exactly the same as the one described in Example 5, with the only difference that the final thermal treatment, lasting two hours, was performed at 500°, 700° and 900° for the PdCoNi-CNi 500/G, PdCoNi-CNi 700/G, PdCoNi-CNi 900/G materials, respectively. The product, once it was recovered from the quartz tube, was washed and activated as described in Example 1.

EXAMPLE 9

{Materials PdCoNi-CNh 600/G} This example includes the synthesis and characterization procedures of materials of the type S-

MPM-CN for the cathodic reduction of oxygen based on palladium, cobalt and nickel.

714 mg of Pd(NO₃)₂-2H₂O were dissolved into the minimum amount of milli-Q water (~1 ml), yielding a deep red solution A. 50 ml of acetonitrile were added to A, and the resulting deep red solution was brought to a small volume (~5 ml) on a hot plate. The solution was then brought to a volume of 50 ml with acetonitrile and later reduced to a small volume (~5 ml) on a hot plate. This procedure was repeated three times yielding a final deep red solution A having a volume of 50 ml.

The same procedure was applied for the preparation of the two solutions B and C; B was obtained after dissolving 785 mg of Ni(NOs)₂-OH₂O, while C was obtained after dissolving 786 mg of

Co(NO₃)₂-6H₂O. The colour shown by B was deep green, while C was deep purple. 1000 mg of polyacrylonitrile was dissolved in 200 ml of dimethylformamide at a temperature of about 130°C; the solutions A, B and C were later added in this order and the resulting dark solution was stirred for 1 hour at 150°C. 1000 mg of XC-72R carbon black were added to this product; the resulting black suspension was stirred at 150°C for 2 hours and then at 130°C for 4 hours. The precursor was obtained after removing all the remaining solvent by placing the product in a ventilated oven at 100°C for 16 hours. The precursor was then transferred into a quartz tube, thermally treated, washed and activated as described in the Example 1, yielding the final material

PdCoNi-CN_h 600/G.

EXAMPLES 10-12

{Materials PdCoNi-CN_h 500/G, PdCoNi-CN_h 700/G, PdCoNi-CN_h 900/G} This example includes the synthesis and characterization procedures of materials of the type S-

MPM-CN for the cathodic reduction of oxygen based on palladium, cobalt and nickel. The preparation of the material is exactly the same as the one described in Example 9, with the only difference that the final thermal treatment, lasting two hours, was performed at 500°, 700° and 900° for the PdCoNi-CN_h 500/G, PdCoNi-CN_h 700/G, PdCoNi-CN_h 900/G materials, respectively. The product, once it was recovered from the quartz tube, was washed and activated as described in Example 1.

COMPARATIVE EXAMPLE

The catalytic activity of the materials described in Examples 1-12 was tested and compared with that of commercial catalysts according to the procedure described below. The chemical composition of the materials was accurately determined with ICP-AES and microanalysis, yielding the results reported in Table 1 and 2. Table 1. Chemical composition of the Pt-X-CNi/G materials.

Atomic weight%

Material Formula aK ^aPt ^aNi ^aFe ^bC ^bN

0.712 10.4 3.06 2U

PtNi-CN, 600/G - 71.47 0.52 ±0.009 ±0.2 ±0.04 Ko.34[PtNio.9₈C_mNo.7θ]

0.92 10.79 3.22

PtNi-CN, 900/G - 75.99 0.35 ±0.01 K0.43ptNi0.99cll4NO.45] ±0.08 ±0.03

2.11 9.82 5.45

PtFe-CN, 600/G - 66.29 1.95 ±0.01 K1.07tPtFe1.94C 110N2.77] ±0.03 ±0.02

1.12 10.09 5.83 25

PtFe-CN, 900/G - 74.72 0.71 ±0.01 ±0.03 ±0.01 Kθ.55[PtFe2.O2Cl2θNo.98]

a Determined by ICP-AES spectroscopy. b By elemental analysis. Table 2. Chemical composition of the PdCoNi-CN/G materials.

Atomic weight% „ ,

Material — ^ _apd a_C() ^ b_c b^^~ Formula

PdCoNi-CN₁ SOOZG j^ ^₂ ^₃ JJj⁹J₃ 67.92 2.57 Kc₇₉[PdC_0L38Ni_L40C₁₁₉N₃.₉]

PdCoNi-CN₁600/G 70.90 2.90 Ka₈₇[PdC₀L₂₆NiL₂₇C₁₁₁N₃₅]

PdCoNi-CN₁ 700/G 66.63 2.14 Kc₈₇[PdCo_L30Ni_L3OC₁O₇N₂.,]

PdCoNi-CN₁, 700/G -° ^₂ ^₅ ^₄ 64.74 6.63 PdC_0L11Ni_L15C₆₂N₅.,

PdCoNi-CN₁, 900/G -° ^₀ ^g ^⁷J₃ 57.58 1.94 PdC_0L19Ni_L20C₅₈N_L7

a Determined by ICP-AES spectroscopy. b By elemental analysis. c Not revealed

Each material was mixed in a 1:1 weight ratio with XC-72R carbon black and extensively

ground in an mortar. The resulting homogeneous black mixture was used to prepare the cathode catalytic layer of a membrane-electrode assembly (MEA) according to the decal procedure. The Nafion/carbon ratio was set equal to 0.3; the overall loading on the cathode ranged between 3 and 4 mg/cm²; the overall loading of precious metals on the various cells was about 0.12 mgp_{t Or} p_d/cm². The anode catalytic layer of each MEA was prepared using a commercial EC-20 electrocatalyst provided by ElectroChem Inc. using a Pt loading of about 0.4 mgp_t/cm². The gas diffusion layers applied to the anodic and cathodic sides of each MEA were P50 and P50T carbon paper provided by Ballard Power Systems. The active layer of each MEA was characterized by an area of about 4

cm . Nafion 117 membranes provided by Ion Power were used for all the MEAs as polymer electrolytes. A commercial MEA provided by ElectroChem Inc. was used as the reference; it was assembled on a Nafion 117 membrane with the EC-20 electrocatalyst on both the anode and the cathode side. Each electrode featured an area of 1 cm² and was characterized by a platinum loading of 1 mgp_t/cm². The polarization curves of all the MEAs were collected in the same conditions: temperature of the anode/cell/cathode: 85/85/85°C; hydrogen flow rate: 800 seem; air flow rate: 1.7 slpm; oxygen flow rate: 1 slpm; relative humidity of the reactants: 100%; back pressure of the reactants: 65 psig. Each MEA was allowed to operate for a few hours before collecting the final polarization curves reported in this patent. The current and power values were normalized on the mass of noble metals effectively located on each cathode. This operation was performed since in fuel cells fed with pure hydrogen the loading of precious metals at the anode can be as low as 1/8 with respect to the cathode without any significant degradation in the performance of the overall

device. This is due to the much better kinetics of the hydrogen oxidation reaction with respect to the oxygen reduction reaction, which is the main bottleneck in the operation of the system. The polarization curves were further normalized on the effective electrode area. In order to benchmark the performance of the various materials, it was decided to determine the minimum amount of noble metal mounted at the cathode in order to obtain 1 kW of electric power. This figure is easily determined as the inverse of the maximum of each power curve reported in Fig. 1-6 using as

oxidant either air or pure oxygen. These values are reported in Table 3, together with the requirements set by the Department of Energy of the U.S. government for applications in the

automotive sector.

Table 3. Minimum amount of PGM required to produce 1 kW of electrical power.

Material

Oxidant: Air Oxidant: Pure Oxygen

PtNi-CNi 600/G 0.43 0.30

PtNi-CNi 900/G 0.43 0.31

PtFe-CNi 600/G 0.53 0.32

PtFe-CNi 900/G 0.68 0.40

PdCoNi-CNi 500/G 0.84 0.54

PdCoNi-CNi 600/G 0.88 0.51

PdCoNi-CNi 700/G 0.59 0.39

PdCoNi-CNi 900/G 0.81 0.46

PdCoNi-CN_h 500/G 2.36 1.29

PdCoNi-CN_h 600/G 2.03 0.91

PdCoNi-CN_h 700/G 1.67 0.88

PdCoNi-CN_h 900/G 1.65 0.90

Reference MEA 1.37 1.11

Status DOE 2006 0.6

Target DOE 2010 0.3 It is observed that the supported platinum-based materials (i.e., PtNi-CNi 600/G, PtNi-CNi 900/G, PtFe-CNi 600/G, PtFe-CNi 900/G) provide very good performances, much better than those obtained with the reference MEA. In particular, the performance always falls between the DOE status of 2006 and the target set for 2010, almost reaching it in the case of PtNi-CNi 600/G, PtNi-CNi 900/G and PtFe-CNi 600/G materials mounted at the cathode of MEAs fed with pure oxygen. However, it should be highlighted that even if the MEAs are fed with air very good performances are achieved, much better than those obtained with the reference MEA and surpassing the DOE status of 2006.

As for the palladium-based materials, (i.e., PdCoNi-CNi 500/G, PdCoNi-CNi 600/G, PdCoNi-CN, 700/G, PdCoNi-CNi 900/G, PdCoNi-CN₁, 500/G, PdCoNi-CN_h 600/G, PdCoNi- CN_h 700/G, PdCoNi-CN_h 900/G), it is observed that those prepared starting from precursors obtained following Procedure 1 (i.e., PdCoNi-CNi 500/G, PdCoNi-CNi 600/G, PdCoNi-CNi 700/G, PdCoNi-CNi 900/G) yield very good performances as they always surpass the results obtained with the reference MEA, and when applied on MEAs fed with pure oxygen they are capable to overcome the DOE status of 2006.

On the other hand, the materials prepared starting from precursors obtained following Procedure 2 (e.g., PdCoNi-CN_h 500/G, PdCoNi-CN_h 600/G, PdCoNi-CN_h 700/G, PdCoNi-CN_h 900/G) provide results comparable with those of the reference MEAs when they are applied to MEAs fed with pure oxygen. However, it should be pointed out that at present (May 2008) the cost of palladium on the open market is much lower with respect to platinum (between three and four times less). While palladium-based systems provide a worse performance with respect to those based on platinum, the former are nevertheless interesting as they lead to the possibility to prepare fuel cells capable to produce electric power at a lower unit cost. Even if the descriptions reported above refer to particular cases in the field of the present invention, it is to be highlighted that it is possible to apply several modifications without T2009/000278 diverging too much from the basic concept of the invention itself. The following claims are aimed at covering these modifications applied in the whole possible range of the present invention.

The descriptions reported above should be considered as illustrative, and not exclusive. The whole possible field of the present invention is indicated in the claims and not in the previous descriptions; all the modifications and the alterations falling inside the field of the claims must therefore be included in the present invention.

Brief description of the figures

Fig. 1. Performance curves of the materials indicated in the graph mounted on the cathode of MEAs prepared as described in the text, a) Polarization curves normalized on both the electrode surface are and the PGM mass mounted on the cathode, b) Polarization curves normalized only on the PGM mass mounted on the cathode, c) Power curves normalized on the PGM mass mounted at the cathode. Test conditions: temperature of the anode/cell/cathode: 85/85/85°C; hydrogen flow rate: 800 seem; air flow rate: 1.7 slpm; relative humidity of the reactants: 100%; back pressure of the reactants: 65 psig.

Fig. 2. Performance curves of the materials indicated in the graph mounted on the cathode of

MEAs prepared as described in the text. Data were normalized as reported in the description of

Fig. 1. Tests were performed as reported in the description of Fig. 1 ; the only difference is that the flow rate of pure oxygen is 1 slpm. Fig. 3. Performance curves of the materials indicated in the graph mounted on the cathode of

Fig. 1. Tests were performed as reported in the description of Fig. 1.

Fig. 4. Performance curves of the materials indicated in the graph mounted on the cathode of

MEAs prepared as described in the text. Data were normalized as reported in the description of Fig. 1. Tests were performed as reported in the description of Fig. 2. Fig. 5. Performance curves of the materials indicated in the graph mounted on the cathode of MEAs prepared as described in the text. Data were normalized as reported in the description of Fig. 1. Tests were performed as reported in the description of Fig. 1.

Fig. 6. Performance curves of the materials indicated in the graph mounted on the cathode of MEAs prepared as described in the text. Data were normalized as reported in the description of Fig. 1. Tests were performed as reported in the description of Fig. 2.

Claims

1. Anodic (intended for fuel oxidation) and cathodic (intended for oxygen reduction) core-shell electrocatalysts made of supported mono- and plurimetallic cabon nitride clusters (shell) to be used in PEMFCs (polymer electrolyte membrane fuel cells), DMFCs (direct methanol fuel cells), AFCs (alkaline fuel cells), PAFCs (phosphoric acid fuel cells) and in H₂ electrolysers. The support consists of electron conducting nanoparticles (core) of graphite powders or platelets, powders of metals such as titanium, silver, gold, platinum, zirconium, manganese, tungsten, lead, scandium, vanadium, iron, cobalt, nickel, zinc, bismuth, copper, chromium, yttrium, niobium, molybdenum, ruthenium, rhodium, palladium, cadmium, lutetium, hafnium, tantalum, tungsten, rhenium, osmium, iridium, lanthanum, cerium, praseodymium, neodymium, samarium, europium, gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium, gallium, indium, thallium, silicon, germanium, tin, etc...

2. Method for the production of the materials described in Claim 1 consisting of the following three steps: in the first the precursor is prepared, in the second the precursor undergoes suitable thermal treatments and in the third the resulting product is activated, in a chemical and/or in an electrochemical way.

3. Method for the preparation of the precursor through sol-gel reactions by mixing two distinct solutions indicated as A and B.

4. Solution A is made of an organic solvent (OS) such as N-methyl-2-pyrrolidinone, dimethylformamide, dimethylacetamide, acetic acid, acetone, acetonitrile, benzene, 1- butanol, 2-butanol, 2-butanone, t-butyl alcohol, cyclohexane, diethyl ether, diethylene glycol, diglyme, dimethylether, dioxane, ethanol, ethyl acetate, ethylene glycol, heptane, hexane, methanol, methyl t-butyl ether, nitromethane, pentane, 1-propanol, 2-propanol, pyridine, tetrahydrofuran, toluene, triethyl amine, o-xylene, m-xylene, p-xylene, water, etc...;, one or more networking agents (NA), and by one or more compounds containing a T/IT2009/000278 transition metal coordinated by halides such as chlorine, bromine and iodine to be selected, for instance, among molecules like: HAuCl₄, H₂IrCl₆, H₂PtCl₆, Li₂PdCl₄, (NH₄)IIrCl₆, (NH₄)₂OsCl₆, (NH₄)PdCl₄, (NH₄)₂PdCl₆, (NH₄)JtCl_4, (NH₄)₂PtCl₆, (NH₄)₃RhCl₆, (NH₄)₂RuCl₆, KAuCl₄, KPt(NH₃)Cl₃, K₂PdCl₄, K₂PtCl₄, K₂PdCl₆, K₂PtCl₆, K₂ReCl₆, K₂RhCl₆ K₂H₂IrCl₆, K₂H₂OsCl₆, K₃IrCl₆, K₃H₃RuCl₆, Na₂IrCl₆, Na₂OsCl₆, Na₂PdCl₄,

Na₂PtCl₆, Na₃RhCl₆, CrCl₃, IrCl₃, FeCl₃, NiCl₂, OsCl₃, PdCl₂, PtCl₂, PtCl₄, RhCl₃, RuCl₃, ReCl₅, SnCl₄, VCl₃, VCl₄, WCl₄, WCl₆, ZrCl₄ and others.

5. Networking agents must be soluble in the selected organic solvent, must not contain sulphur atoms and must support functional groups featuring oxygen, nitrogen or phosphorus atoms such as -C≡N, -NH₂, -NHR, -NR₂, -OH, R-O-R, -C=O, -COOH, -COOR, -PH₂, -PHR,

-PR₂. Examples include but are not limited to dianilines such as p-phenylenediamine; 4,4'- methylenedianiline; 1 ,4-diaminobutane; dianhydrides such as 1, 2,4,5 -benzenetetracarboxylic anhydride; 3,3',4,4'-benzophenonetetracarboxylic dianhydride; (+)-diacetyl-L-tartaric anhydride; diphosphines such as l,2-bis(diphenylphosphino)ethane; (-)-2,3-O- isopropylidene-2,3-dihydroxy-l ,4-bis(diphenylphosphino)butane; (-)-l ,2-bis[(2R,5R)-2,5- diethylphospholano]benzene; rac-2,2'-bis(di-p-tolylphosphino)-l,r-binaphthyl; carboxylic acids such as tetrahydrofuran-2,3,4,5-tetracarboxylic acid; glutaric acid; terephthalic acid; molecules carrying more than one type of functional group such as BOC-L-glutamic acid, acetyl tributyl citrate and L(+)-glutamic acid; polyimides such as Kapton, and Apical, polyamides such as Nylon 6 and Nylon 6,6, polyurethanes, polypyrrole, polyvinyl alcohol, polymethyl metacrylate, polyacrylonitrile, poly(tetramethylene ether) glycol, polyethylene glycol and others.

6. Solution B is constituted by an organic solvent (OS) such as N-methyl-2-pyrrolidinone, dimethylformarnide, dimethylacetamide, acetic acid, acetone, acetonitrile, benzene, 1- butanol, 2-butanol, 2-butanone, t-butyl alcohol, cyclohexane, diethyl ether, diethylene glycol, diglyme, dimethylether, dioxane, ethanol, ethyl acetate, ethylene glycol, heptane, hexane, methanol, methyl t-butyl ether, nitromethane, pentane, 1-propanol, 2-propanol, pyridine, tetrahydrofuran, toluene, triethyl amine, o-xylene, m-xylene, p-xylene, water, etc., by one or more networking agent (NA) and by one or more compounds containing complexes of a transition element where there are ligands like cyano, isocyano, thiocyano, amino and amido groups. Cyanometallates can be selected among compounds such as

KAg(CN)₂, KAu(CN)₂, K₂Ni(CN)₄ K₂Pd(CN)₄, K₂Pt(CN)₄, K₃Co(CN)₆, K₃Cr(CN)₆ K₃Fe(CN)₆, K₄Fe(CN)₆-SH₂O, K₃Mn(CN)₆, K₂Pt(CN)₆, K₄Ru(CN)₆ and others.

7. The support may be added either to the solutions A and/or B or to the final mixture and is an electron-conducting material. Typical examples of electron-conducting materials suitable as supports for this invention are graphite powders or platelets, powders of metals such as titanium, silver, gold, platinum, zirconium, manganese, tungsten, lead, scandium, vanadium, iron, cobalt, nickel, zinc, bismuth, copper, chromium, yttrium, niobium, molybdenum, ruthenium, rhodium, palladium, cadmium, lutetium, hafnium, tantalum, tungsten, rhenium, osmium, indium, lanthanum, cerium, praseodymium, neodymium, samarium, europium, gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium, gallium, indium, thallium, silicon, germanium, tin, etc...

8. Method to prepare a precursor as in the Claim 3, 4, 5, 6 and 7 but placing all the reagents in one solution only.

9. Method for the preparation of the precursor as in the Claims 3, 4, 5, 6 and 7 through reactions: a) sol-gel; and/or b) gel-plastic.

10. Method for the preparation of the precursor like in the Claims 3, 4, 5, 6 and 7 through coagulation and/or flocculation and/or precipitation reactions.

11. Method for the preparation of the precursor through networking processes. Each complex of the selected transition metal is dissolved in an organic solvent (OS) such as N-methyl-2- pyrrolidinone, dimethylformamide, dimethylacetamide, acetic acid, acetone, acetonitrile, benzene, 1-butanol, 2-butanol, 2-butanone, t-butyl alcohol, cyclohexane, diethyl ether, diethylene glycol, diglyme, dimethylether, dioxane, ethanol, ethyl acetate, ethylene glycol, heptane, hexane, methanol, methyl t-butyl ether, nitromethane, pentane, 1-propanol, 2- propanol, pyridine, tetrahydrofuran, toluene, triethyl amine, ø-xylene, m-xylene, ^-xylene, etc., possibly using a further coordinating agent. This process yields the solutions Aj. The starting metal complexes include, but are not limited to HAuCl₄, H₂IrCl₆, H₂PtCl₆, Li₂PdCl₄,

(NH₄)₂IrCl₆, (NH₄)₂OsCl₆, (NH₄)PdCl₄, (NH₄)₂PdCl₆, (NH₄)₂PtCl₄, (NH₄)₂PtCl₆, (NH₄)₃RhCl₆, (NH₄)₂RuCl₆, KAuCl₄, KPt(NH₃)Cl₃, K₂PdCl₄, K₂PtCl₄, K₂PdCl₆, K₂PtCl₆, K₂ReCl₆, K₂RhCl₆ K₂H₂IrCl₆, K₂H₂OsCl₆, K₃IrCl₆, K₃H₃RuCl₆, Na₂IrCl₆, Na₂OsCl₆, Na₂PdCl₄, Na₂PtCl₆, Na₃RhCl₆, CrCl₃, IrCl₃, FeCl₃, NiCl₂, OsCl₃, PdCl₂, PtCl₂, PtCl₄, RhCl₃, RuCl₃, ReCl₅, SnCl₄, VCl₃, VCl₄, WCl₄, WCl₆, ZrCl₄, (NH₃)₄Pt(NO₃)₂,

Pd(NO₃)₂-χH₂O, Co(NO₃)₂-xH₂O, Ni(NO₃)₂-xH₂O, Fe(NO₃)₃-xH₂O, Cr(NO₃)₃-xH₂O, Mn(NO₃)₂-xH₂O, AgNO₃.

12. The selected networking agents (NA) are dissolved in an organic solvent such as N-methyl- 2-pyrrolidinone, dimethylformarnide, dimethylacetamide, acetic acid, acetone, acetonitrile, benzene, 1-butanol, 2-butanol, 2-butanone, t-butyl alcohol, cyclohexane, diethyl ether, diethylene glycol, diglyme, dimethylether, dioxane, ethanol, ethyl acetate, ethylene glycol, heptane, hexane, methanol, methyl t-butyl ether, nitromethane, pentane, 1-propanol, 2- propanol, pyridine, tetrahydrofuran, toluene, triethyl amine, o-xylene, m-xylene, p-xylene, etc ., yielding the solution B. Networking agents must be soluble in the selected organic solvent, must not contain sulphur atoms and must support functional groups featuring oxygen, nitrogen or phosphorus atoms such as -C≡N, -NH₂, -NHR, -NR₂, -OH, R-O-R, - C=O, -COOH, -COOR, -PH₂, -PHR, -PR₂. Examples include but are not limited to dianilines such as p-phenylenediamine; 4,4'-methylenedianiline; 1,4-diaminobutane; dianhydrides such as 1,2,4,5-benzenetetracarboxylic anhydride; 3,3',4,4'- benzophenonetetracarboxylic dianhydride; (+)-diacetyl-L-tartaric anhydride; diphosphines such as l,2-bis(diphenylphosphino)ethane; (-)-2,3-O-isopropylidene-2,3-dihydroxy-l,4- bis(diphenylphosphino)butane; (-)-l ,2-bis[(2R,5R)-2,5-diethylphospholano]benzene; rac- 2,2'-bis(di-p-tolylphosphino)-l,r-binaphthyl; carboxylic acids such as tetrahydrofuran- 2,3,4,5-tetracarboxylic acid; glutaric acid; terephthalic acid; molecules carrying more than one type of functional group such as BOC-L-glutamic acid, acetyl tributyl citrate and L(+)- glutamic acid; polyimides such as Kapton, and Apical, polyamides such as Nylon 6 and

Nylon 6,6, polyurethanes, polypyrrole, polyvinyl alcohol, polymethyl metacrylate, polyacrylonitrile, poly(tetramethylene ether) glycol, polyethylene glycol and others.

13. The solutions Aj are added to the solution B.

14. The support may be added either to the solutions Aj and/or B or to the final mixture and is an electron-conducting material such as graphite powders or platelets, powders of metals such as titanium, silver, gold, platinum, zirconium, manganese, tungsten, lead, scandium, vanadium, iron, cobalt, nickel, zinc, bismuth, copper, chromium, yttrium, niobium, molybdenum, ruthenium, rhodium, palladium, cadmium, lutetium, hafnium, tantalum, tungsten, rhenium, osmium, iridium, lanthanum, cerium, praseodymium, neodymium, samarium, europium, gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium, gallium, indium, thallium, silicon, germanium, tin, etc...

15. Synthesis of the precursor as described in the Claims 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14.

16. Synthesis of the electrocatalysts and of their precursors as described in Claims 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 where the organic solvent is any liquid capable to dissolve the above-described compounds to yield clear solutions.

17. Synthesis of the elctrocatalysts described in the Claims 1 and 2 by applying to the precursors a thermal treatment consisting of at least two steps. The first step is executed at temperatures lower than 400⁰C, while the second step is executed at a temperature ranging from 400 and 900°C. It is possible to apply to the materials further optional thermal treatments at temperatures lower than 400°C and/or higher than 500°C.

18. The processes of Claim 17 are performed in an inert atmosphere (e.g., He₅ Ar, N₂ and others), or under vacuum.

19. Activation of the electrocatalysts described in Claims 1 and 2 through electrochemical processes simulating the cyclic voltammetry techniques in the presence of inert gases (e.g. nitrogen, argon, etc.), alternated to oxidant gases (e.g., oxygen) and/or reducing agents

(e.g., hydrogen, hydrazine and others).

20. Activation of the electrocatalysts described in Claims 1 and 2 through chemical processes alternating: a) oxidation through O₂, H₂O₂, and others; and b) reduction through reducing reagents such as H₂, formaldehyde, hydrazine, hydrazine dichloride and others.