WO2014207156A1

WO2014207156A1 - Iron sulfide based catalyst for electrolytic water reduction into hydrogen gas

Info

Publication number: WO2014207156A1
Application number: PCT/EP2014/063614
Authority: WO
Inventors: Cédric TARD; Marion Giraud
Original assignee: Centre National De La Recherche Scientifique (Cnrs); Universite Paris Diderot Paris 7
Priority date: 2013-06-26
Filing date: 2014-06-26
Publication date: 2014-12-31

Abstract

The present invention relates to the use of a catalyst for electrolytic water reduction into hydrogen gas, said catalyst comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst, an electrode coated with said catalyst, an electrolytic cell comprising said electrode, and a process for electrolytic water reduction into hydrogen gas using said electrode or electrolytic cell, and a process for preparing nanoparticles comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

Description

IRON SULFIDE BASED CATALYST FOR ELECTROLYTIC WATER REDUCTION

INTO HYDROGEN GAS

The present invention relates to catalysts for the production of hydrogen gas through electrolytic water reduction. In particular, the present invention relates to the use of a catalyst for electrolytic water reduction into hydrogen gas, said catalyst comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst, an electrode coated with said catalyst, an electrolytic cell comprising said electrode, and a process for electrolytic water reduction into hydrogen gas using said electrode or electrolytic cell, as well as a process for preparing nanoparticles comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

World energy consumption is predicted to increase at least two-fold by 2050, due to population and economic growth. Estimations of fossil fuel reserves (coal, oil and gas) suggest that this rise could be compensated in principle; but replacement of fossil fuels by renewable and sustainable energy sources, such as wind or sunlight, seems much preferable from an environmental, economic and security standpoint.

However, a major drawback associated with these sources relates to their intermittent nature, preventing consumers from being able to use the "renewable" energy during the entire day.

To overcome this inherent drawback, storing energy in the form of chemical bonds and releasing it during peak consumptions arose as a possible solution. In this prospect, molecular hydrogen, thanks to its clean cold combustion in fuel cells, appears as a promising candidate. However, the use of large scale amounts of molecular hydrogen as energy supply raises several issues, namely production, storage, and transportation.

At present, molecular hydrogen gas is mainly produced through steam reforming of hydrocarbons from fossil fuels, which is not satisfactory due to the economic and environmental cost of feedstock, and in particular because of the release of carbon dioxide through the reformation reaction. In this context, water splitting into oxygen and hydrogen gas by electrolysis appears as the most efficient hydrogen production process, be it in terms of atom economy and availability of starting material (water), or because of the carbon- free nature of the reaction.

Electrolysis is a method of using an electric current to drive an otherwise non- spontaneous chemical reaction. Crossing of said electric current through an ionic substance that is either molten or dissolved in suitable solvents results in chemical reactions at the electrodes. An electrolytic reaction is performed in an electrolytic cell, comprising at least:

- an electrolyte (a substance able to carry the electric current from one electrode to the other, preferably an aqueous solution containing salts, acids and/or bases,); - an electric current supply providing the energy necessary to create or discharge the ions in the electrolyte; and

- two electrodes, i.e. electrical conductors providing a physical interface between the electrical circuit and the electrolyte.

An electrical potential is applied across the two electrodes immersed in the electrolyte. Positively charged ions (cations) move towards the electron-providing (negative) cathode, whereas negatively charged ions (anions) move towards the positive anode. At the electrodes, electrons are absorbed or released by the atoms and ions. Atoms gaining or losing electrons, thus becoming ions, dissolve into the electrolyte. Conversely, ions gaining or losing electrons, thus becoming uncharged entities, separate from the electrolyte.

Even though an external electric power is needed to perform said water splitting reaction, renewable sources may be used to produce said electric power. Conversely, when energy is required, the obtained hydrogen can be recombined with oxygen within a fuel cell to regenerate water, in turn releasing electricity into the system. The reversible reduction of water includes proton reduction into dihydrogen (equation 1, taking place at the cathode), which is a seemingly simple reaction, but requires multistep catalysis to proceed, and involving high energy intermediates:

2 H⁺ (aq) + 2 e→H₂(gas) (1)

In parallel, oxidation of water occurs at the anode, producing oxygen gas (equation 2): 2 H₂0(liq)→ 0₂ + 4 H⁺(aq) + 4e^~ (2)

In the present invention, the terms "water reduction into hydrogen gas" and "proton reduction into hydrogen gas" will be identically used to refer to the reaction detailed in equation (1) above, occurring in an aqueous medium. The use of catalysts to perform the proton reduction into hydrogen gas enables to decrease the required driving force (or overpotential).

According to the present invention, "overpotential" is understood as a potential difference between the thermodynamic reduction potential of the H⁺/H₂ couple and the potential at which the reduction is experimentally observed. The thermodynamic reduction potential of the H⁺/H₂ couple is pH dependent, following equation (3):

E(pH) = E°(H+/H₂) - (RTln(10)/F)*pH (3) wherein E represents a potential (in Volts (V)), pH represents the (experimentally determined) pH of the electrolyte solution,

E°(H⁺/H₂) represents the thermodynamic reduction potential of the H⁺/H₂ couple (V),

R is the universal gas constant (R = 8.314 J K ¹ mol ¹)

T is the temperature in Kelvin

F is the Faraday constant, (the number of coulombs per mole of electrons: F = 9.6485x l0⁴ C mof¹). At room temperature, i.e. at around 298 K, (RTln(10)/F) equals approximately 0.059.

Homogenous as well as heterogeneous catalysts have been described, however, heterogenous catalysts are preferred for industrial applications. Currently, the most efficient heterogeneous catalysts used for water splitting are made of platinum or certain of its alloys, but platinum based catalysts are costly, thus not suitable for industrial applications.

According to the present invention, a "heterogeneous catalyst" is a catalyst which is contained in a phase which differs from the phase of the reactants. In contrast, a homogenous catalyst is dissolved in the same phase as the reactants. Therefore, in the present invention, a "heterogeneous catalyst" is not soluble in water. It is preferably a solid, which may be used in a composition to form films to coat the electrodes. Moreover, said "heterogeneous catalyst" is the active material which catalyzes the proton reduction into hydrogen gas. In particular, the surface of said heterogeneous catalyst interacts with the protons to produce intermediates in said reduction. Of note, non-precious metal catalysts, in particular molybdenum sulfide electrocatalysts, in the form of nano-crystals, amorphous electro-polymerized films, molecular complexes, or amorphous particles, operating at low overpotentials, high current densities under mild condition (ca. pH 7, 1 atm, room temperature) have been reported for their high activity towards molecular hydrogen evolution in acidic or neutral water at relatively low overpotentials.

Besides, the article by Vrubel et al (Energy Environ. Sci. 2012, 5, 6136-6144) describes an amorphous material comprising a mixture of MoS₄ and metal sulfur MS₄, M representing Co, Ni or Fe. The ratio of the concentrations of M and Mo is of between 1 :200 and 1 :2, and preferably of 1 :3. The obtained material is then deposited on an electrode as a film, and used as a catalyst for water reduction into hydrogen gas, exhibiting an improved catalytic activity compared to a film consisting only of amorphous MoS_x. However, the solid catalysts described in this article are not used as nanoparticles, and their main component is actually molybdenum rather than iron.

The article by Yuhas et al (J. Am. Chem. Soc. 2011 , 133, 7252-7255) discloses cubane- type Fe₄S₄-clusters, which are obtained as chalcogels. Said cubane-type clusters reduce protons into dihydrogen from weak organic acids, but the poor stability of such molecules towards water and dioxygen is problematic. It is noteworthy that these molecular Fe₄S₄-clusters can be stabilized within porous chalcogenide frameworks, these systems showing activity for heterogeneous electrocatalysis and photocatalysis for dihydrogen evolution and carbon dioxide reduction. These catalysts are thus different from nanoparticles or electrodeposited materials.

There is therefore a need for new heterogeneous iron-sulfide based catalysts, which are able to efficiently catalyze water reduction into hydrogen gas at a low overpotential and a high current density. Surprisingly, applicants found that a heterogeneous catalyst comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst, is suitable as heterogeneous catalyst for electrolytic proton reduction into hydrogen gas .According to the present invention, the terms "essentially consisting of mean that the subject material comprises mostly the designated elements, but can also contain other elements in quantities low enough to not impact the properties of said subject material, in particular regarding its physical and chemical properties. For instance, a material essentially consisting of carbon may contain trace amounts of impurities besides carbon, which do not change the sought properties of carbon. In one aspect, the present invention relates to the use of a heterogeneous catalyst for electrolytic proton reduction into hydrogen gas, said catalyst comprising at least 80%> by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst.

In another aspect, the invention relates to an electrode coated with a heterogeneous catalyst, said catalyst comprising at least 80%> by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst.

In another aspect, the invention relates to an electrolytic cell comprising the electrode of the invention, preferably as the cathode.

In another aspect, the invention relates to a method comprising performing electrolytic proton reduction into hydrogen gas with an electrolyte and the electrode according to the invention as the cathode, or the electrolytic cell of the invention.

In another aspect, the invention relates to a method for preparing nanoparticles comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

The present invention thus relates to the use of a heterogeneous catalyst for electrolytic proton reduction into hydrogen gas, said catalyst comprising at least 80%> by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst.

The concentration of iron and sulfur into the catalyst of the invention is measured according to regular techniques known to the one skilled in the art, such as for example elemental analysis. The catalyst of the invention may contain other metal elements, but in very low amounts. For instance, the catalyst may have a molybdenum content (in weight percent relative to the total weight of the catalyst) which is lower than the iron content of the catalyst. This is also true of all other metal elements, such as Nickel, Palladium, Platinum, Gold, Mercury, Cobalt, Copper, Manganese, Ruthenium or Tungsten. For instance, the other metal elements represent less than 10%, preferably less than 5%, by weight, relative to the total weight of the catalyst.

In a first embodiment, said catalyst is coated on an electrode via electrodeposition, advantageously using an electroplating process. Electrodeposition is a process that uses electrical current to reduce dissolved cations, in particular dissolved metal, so as to form a coating on a substrate surface. It is analogous to a galvanic cell acting in reverse.

Electroplating is a process using electrodeposition, i.e. an electrical current to reduce dissolved metal cations, so that they form a coherent metal coating on an electrode. The part to be plated is the cathode of the circuit. In one technique, the anode is made of the metal to be plated on the part. In another technique, the electrolyte contains metal ions of the metal to be coated. In the present invention, the electrolyte typically contains iron salts with sulfur containing ions, such as FeS0₄, Fe₂(S0₄)₃ or FeS₂0₃ salts. The electrolyte may contain additional sulfur containing salts, such as Na₂S₂0₃ or Na₂S0₄. Alternatively, the electrolyte may contain a mixture of iron salts which are free of sulfur, such as FeCl₃ or FeCl₂, and additional sulfur containing salts, such as Na₂S₂0₃ or Na₂S0₄. For instance, the electrolyte or electroplating bath may contain a mixture of a 0.01 M of Fe₂(S0₄)₃ solution and a 0.3 M of Na₂S₂0₃. The electrolyte may also contain other ions that permit the flow of electricity. Both electrodes are immersed in said electrolyte. A power supply provides a direct current to the anode. At the cathode, the dissolved metal ions in the electrolyte solution are reduced at the interface between the solution and the cathode, so as to "plate out" onto the cathode. Electroplating is a technique well known to the one skilled in the art, who will set all necessary parameters so as to obtain the desired electrodeposited catalyst on the electrode, preferably the cathode. Said electrodeposited catalyst comprises at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst. Preferably, said electrodeposited catalyst comprises between 40% and 70% by weight of Iron (Fe) relative to the total weight of the catalyst. More preferably said electrodeposited catalyst comprises between 45% and 65% by weight of Iron (Fe) relative to the total weight of the catalyst. Most preferably, said electrodeposited catalyst comprises between 46 % and 63 % by weight of Iron (Fe) relative to the total weight of the catalyst.

In a preferred second embodiment, said catalyst is made of nanoparticles comprising at least 80%) by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles. Indeed better performances and stability were observed with nanoparticles. In this embodiment, said catalyst comprises or preferably consists of nanoparticles comprising at least 80%> by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

According to the present invention, "nanoparticles" are understood as particles with a particle size lower than or equal to 500 nm, preferably lower than or equal to 200 nm, more preferably lower than or equal to 100 nm. For instance, nanoparticles of the invention have a particle size of between 0.5 and 500 nm, preferably between 0.5 and 200 nm, more preferably between 0.5 and 100 nm. In the particles of the invention, iron is advantageously the main metallic constituent. Particle size of the nanoparticles of the invention can be determined by X ray diffraction using Debye-Scherrer law.

Size distribution of the nanoparticles of the invention is polydisperse.

Advantageously, nanoparticles of the invention are maintained dispersed but they can also be used when aggregated.

Preferably, the nanoparticles used in the present invention contain at least 85% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

The nanoparticles according to the present invention preferably have a particle size lower than or equal to 200 nm, even more preferably lower than or equal to 120 nm.

Advantageously, the nanoparticles used in the present invention comprise at least 80% by weight of Fe_xS, with x between 0.5 and 1 included. In a particular embodiment, x is between 0.5 included and 1 excluded. In a preferred embodiment, x is between 0.5 and 0.75 or is between 0.85 and 0.95. Therefore, in a preferred embodiment, the nanoparticles comprise at least 80% by weight of Fe₃S₄ or FeS₂ or Fe(o.9±o.os₎S, in particular Fe₃S₄ or FeS₂ or Feo.9S.

In a particular embodiment, the nanoparticles used in the present invention are greigite Fe₃S₄-nanoparticles advantageously prepared according to Zhang et al (J. Alloys Compd 2009, 488, 339-345). Gram-scale batches of nanoparticles can be prepared from cheap and abundant precursors, namely FeCl₃ and thiourea. Said greigite Fe₃S₄- nanoparticles are typically prepared using a polyol as the solvent of the reaction, and the obtained particles are grafted with organic chains resulting from an interaction between the solvent and the iron sulfide material. Said particles contain at most 15% of organic material by weight relative to the total weight of the particles. A pure greigite phase is observed by X-ray diffraction (XRD), with a crystallite size of ca. 36 nm. Specific surface area has been estimated from BET isotherm, with a measured value of 8.2 ± 0.1 m² g^"1. These nanoparticles are typically platelets and preferably have a particle size lower than or equal to 200 nm, in particular lower than or equal to 150 nm.

The catalysts used in the invention are typically stable over 18 months in air, without any significant structure degradation, unlike molecular Fe/S clusters (which require strictly anoxic conditions for storage).

Advantageously, the nanoparticles used in the present invention are pyrite FeS₂ - nanoparticles. Indeed better performances in electrolytic water reduction, in particular at high temperature, were observed with pyrite nanoparticles. Moreover, the ubiquity of iron sulfide minerals in nature, such as pyrite FeS₂ - which is the most abundant mineral on the Earth's surface - renders said catalysts very attractive: a raw comparison between the different metal prices shows that iron ore (101 $/t) is much cheaper than nickel (21 ,800 $/t), molybdenum (34,900 $/t) or cobalt (39,700 $/t), the platinum-group metal being much more expensive (18,500,000 $/t).

The present invention also provides methods for preparing nanoparticles comprising at least 80%) by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

In a particular embodiment, the method comprises the following successive steps: a) mixing an iron salt, preferably Fe(III)Cl₃, Fe(II)Cl₂, Fe(III)(acetate)3, or Fe(III)(acetylacetonate)3, Fe(II)(acetate)2, or Fe(II)(acetylacetonate)2, and a sulfur source, preferably Na₂S, elemental sulfur, thiourea or cysteine, and a solvent, preferably a polyol or an alkylamine, b) heating the obtained reaction mixture through thermal heating or microwave irradiation, and c) recovering the thus obtained nanoparticles.

In this embodiment, the sulfur source is preferably thiourea or elemental sulfur.

The iron source is preferably Fe(III)Cl3. Also, step b) may be carried out either under pressure or at atmospheric pressure.

According to the present invention, a "polyol" is understood as a C₂-Ci₂ polyol containing at least two hydroxyl functional groups, wherein, in the alkyl chain, 1 to 3 carbon atoms may be optionally replaced by an oxygen atom, provided that the chain does not contain two contiguous oxygen atoms, i.e. the polyol is not a peroxide. Preferably, the polyol is a glycol ether. An example of polyol is ethyleneglycol, diethyleneglycol, triethyleneglycol, tetraethyleneglycol or 1 ,2-propanediol, 1 ,2- butanediol or glycerol, more preferably, it is ethyleneglycol, diethyleneglycol, or 1 ,2- propanediol, 1 ,2-butanediol.

According to the present invention, an « alkylamine » encompasses a mono- or disubstituted Cs-Cis alkylamine, wherein said alkyle chain is linear or ramified, optionally comprising 1 or 2 double bonds. The alkyl amine of the invention is preferably monosubstituted, even more preferably selected from the group consisting of n-octylamine, n-dodecylamine, n-hexadecylamine, and oleylamine, preferably octylamine n-hexadecylamine, and oleylamine. Preferably, when step b) is carried out at atmospheric pressure, step b) is carried out at a temperature of between 120°C and 370°C, more preferably between 150°C and 230°C. In a particular embodiment, step b) is carried out at the reflux temperature of the solvent. When step b) is carried out under pressure, the pressure is typically of between 1 and 50 bars, preferably between 1 bar and 10 bars. Moreover, the heating temperature is preferably of between 120°C and 370°C, more preferably of between 150°C and 230°C.

In an alternative embodiment, the method comprises the following successive steps: a) mixing a precursor acting both as the iron and sulfur source, preferably Fe₄S₄(SPh)₄ or Fe₂S₂(CO)6, more preferably Fe₂S₂(CO)6, and a solvent, advantageously a polyol or an alkylamine, b) heating the obtained reaction mixture through thermal heating under pressure or at atmospheric pressure, and c) recovering the thus obtained nanoparticles.

In this embodiment, the solvent is preferably an alkylamine.

Advantageously, when step b) is carried out at atmospheric pressure, step b) is carried out at a temperature of between 120°C and 370°C, more preferably between 150°C and 230°C. In particular, step b) is carried out at the reflux temperature of the solvent.

When step b) is carried out under pressure, the pressure is typically of between 1 and 50 bars, preferably between 1 bar and 10 bars. Moreover, the heating temperature is preferably of between 120°C and 370°C, more preferably of between 150°C and 230°C.

The present invention also relates to an electrode coated with a heterogeneous catalyst, said catalyst comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst.

In a first embodiment, the catalyst is coated on the surface of the electrode via electrodeposition advantageously using an electroplating process.

Said electrodeposition and electroplating are carried out as described above. In a preferred case, the electrode according to the invention essentially consists of carbon, preferably vitreous carbon, and is coated with said catalyst.

In a particular case, the electrode essentially consists of Indium Tin Oxide (ITO), or Fluorine Tin Oxide (FTO), and is coated with said catalyst. In another particular embodiment, the electrode essentially consists of gold (Au) or mercury (Hg), and is coated with said catalyst.

In a second embodiment, the electrode is coated with a composition comprising a binder and a catalyst, said catalyst comprising nanoparticles comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

Said nanoparticles used in the electrode of the invention are as described above.

Said binder is advantageously selected from the group consisting of conductive polymers, ionomers or fluoropolymers. Conductive polymers are organic polymers that conduct electricity. In particular, polyacetylene, polypyrrole, polyaniline, poly(p-phenylene vinylene) (PPV), poly(3- alkylthiophenes) and their copolymers are the main classes of conductive polymers. Examples of conductive polymers are polyfluorenes, polyphenylenes, polypyrenes, polyazulenes, polynaphthalenes, poly(pyrrole)s (PPY), polycarbazoles, polyindoles, polyazepines, polyanilines (PANI), poly(thiophene)s (PT), poly(3,4- ethylenedioxythiophene) (PEDOT), and poly(p-phenylene sulfide) (PPS). Preferred conductive polymers are polypyrrole, polyazepines, polyanilines or poly(3,4- ethy lenedio xythiophene) .

An ionomer is a polymer that comprises monomer units of both electrically neutral monomer units and a fraction of ionized monomer units (usually no more than 15 mole percent) covalently bonded to the polymer backbone as lateral moieties. Most ionomers are copolymers of neutral segments and ionized units, said ionized units usually consisting of carboxylic acid groups or sulfonic acid groups. Preferred examples of ionomers are polystyrene sulfonates, and in particular Nafion^®, a sulfonated tetrafluoroethylene based fluoropolymer-copolymer. Advantageously, when Nafion^® is used, it is used at a concentration (in the composition comprising both the binder and the catalyst) higher than 20 % wt, relative to the total weight of the composition (comprising both the binder and the catalyst), in order to obtain better performances with nanoparticles of the invention. A fluoropolymer is a fluorocarbon based polymer with multiple carbon-fluorine bonds. In particular, a fluoropolymer results from a polymerization reaction using at least one type of fluorinated monomer. It is characterized by a high resistance to solvents, acids, and bases. Examples of fluoropolymers are polyvinylfluoride and polyethylenetetrafluoroethylene.

Said binder ensures: 1) a good electronic conductivity, as well as 2) a good permeation of protons through the composition by means of porosity of the polymer and/or the presence acid-base functional groups within the polymer structure.

In a preferred embodiment, said binder is a fluorinated polymer with sulfonic end groups such as a perfluorosulfonic acid.

The composition comprising a binder and a catalyst may further contain additives, in particular to improve the conductivity properties of the composition, such as carbon black, carbon nanotubes or graphene.

The coating is preferably in the form of a film of said composition. In this embodiment, the use of a composition comprising a binder and said nanoparticles for coating the electrode overcomes any scaling up problems that could arise when using an electrodeposition process. Moreover, nanoparticles allow a better control of the structure and composition of said catalyst.

In a preferred embodiment, the electrode according to the invention essentially consists of carbon, preferably vitreous carbon, and is coated with a composition comprising said catalyst.

In a particular embodiment, the electrode essentially consists of Indium Tin Oxide (ITO), or Fluorine Tin Oxide (FTO), and is coated with a composition comprising said catalyst.

In another particular embodiment, the electrode essentially consists of gold (Au) or mercury (Hg), and is coated with a composition comprising said catalyst.

It is observed that the coated electrodes of the invention are stable to air, water and oxygen for a period of at least 6 months. The present invention also provides an electrolytic cell comprising the electrode according to the invention, preferably as the cathode.

The electrolytic cell further comprises an electrolyte, as explained above. An electrolyte is an aqueous solution containing soluble salts or acids or bases. Preferably, the electrolyte is a buffered aqueous solution, with a pH of between 5 and 8, more preferably between 6.5 and 7.5. Advantageously, the electrolyte comprises potassium phosphate or sodium chloride.

The faradic yield of an electrolytic cell aimed at producing hydrogen gas through proton reduction is the ratio of the amount of electrons (in Coulomb) used to produce hydrogen gas relative to the amount of electrons (in Coulomb) furnished to the electrolytic system by the external electric source. Advantageously, the faradic yield of the electrolytic cell of the invention is greater than 75%, preferably greater than 80%. In a particular embodiment, the faradic yield is greater than 90% or 95%.

The present invention also relates to a method comprising performing proton reduction using the electrode according to the invention as the cathode and an electrolyte, or the electrolytic cell of the invention, thereby producing hydrogen gas.

Preferably, the electrolyte is a buffered aqueous solution, preferably so that the pH of said electrolyte is of between 5 and 10, more preferably between 6.5 and 7.5.

Preferably, the overpotential applied to the electrodes is below 0.5 V, preferably below 0.4 V, even more preferably below 0.3 V. For instance, the overpotential applied to the electrodes is of between 0 and 0.5 V, preferably between 0.05 V and 0.4 V, even more preferably between 0.1 V and 0.3 V. In a preferred embodiment, the potential applied to the electrode is below (0.5 + 0.059 *pH) V versus normal hydrogen electrode, preferably between (0.05 + 0.059 *pH) and (0.4 + 0.059*pH) V versus normal hydrogen electrode, even more preferably between (0.1 + 0.059*pH) and (0.3 + 0.059*pH) V versus normal hydrogen electrode, pH representing the pH of the electrolyte solution. Advantageously, the proton reduction is carried out at a temperature of between 15°C and 40°C, more preferably between 15°C and 30°C, even more preferably between 18°C and 27°C.

The method or the electrolytic cell of the invention is particularly robust. Indeed, it has been observed that the proton reduction into hydrogen gas occurs without decrease in efficiency for at least 5 days continuously. Moreover, unlike the methods or electrical cells of the prior art, the electrolyte may be a neutral aqueous solution, i.e. with a pH of between 6.5 and 7.5, and the electrochemical reaction occurs at an overpotential which competes with the one necessary when using other not iron based non-precious metal catalysts.

DESCRIPTION OF THE FIGURES

Figure 1 represents the transmission electron microscopy images of the Fe₃S₄- nanoparticles of example 3 of the invention. Figure 2 represents X-ray diffraction patterns of the greigite type Fe₃S4-nanoparticles of example 3 prepared by the 'polyol process' method after 18 months stored in the air (the diffractogramm was recorded in the 15-100° 2Θ range only), after BET pretreatment and after controlled potential electrolysis. The vertical bars represent the theoretical greigite pattern (JCPDS 00-016-0713). Figure 3 shows the crystal structure image of the unit cell of greigite Fe₃S₄- nanoparticles of example 3 as described in the JCPDS n° 00-016-0713 pattern (balls and sticks style, dark grey for iron atoms and light grey for sulfur atoms) (left) and projection of the structure along the a axis (right).

Figure 4 shows cyclic voltammetry of the Fe₃S₄-nanoparticles of example 3 dispersed in Nafion^®. Voltammogramms A and B were recorded on a rotating disk electrode, in 0.05 M potassium phosphate at pH 7.0 and at 25 °C. Scan rate = 0.01 V s^"1; rotating rate = 4000 rpm.

Figure 5 shows controlled potential electrolysis of the Fe₃S₄-nanoparticles of example 3. Experiments performed in 1.0 M potassium phosphate buffer (pH 7.0). A) Overpotential = | applied potential - 0.059 x pH| V vs. ΝΗΕ (Normal Hydrogen Electrode). H₂ evolution rate calculation assumes that every electron is used for molecular hydrogen evolution. B) Electrolysis at an overpotential of 300 mV (-0.713 V vs. NHE), showing charge build-up versus time

for the cell with (■) and without (A) the Fe₃S₄-nanoparticles of example 3. Figure 6 shows the gas chromatogram of the head space of the cell after 5 days of electrolysis with Feo.₉S pyrrhotite nanoparticles.

Figure 7 presents a comparison of electrodeposited iron sulfide with different iron sulfide nanoparticles catalysts deposited onto a carbon electrode (intensity I in μΑ vs E in V (V vs NHE). Dash-dot line (— ·— ) electrodeposition. Nafion coated electrode, 0.1 mol/L phosphate buffer, pH 7.0, scan rate 1 mV/s. Solid line (— ): pyrrhotite Feo.₉S; dashed line ( ): greigite Fe₃S₄; dotted line (· · ·) pyrite FeS₂.

Figure 8 represents polarization curves of electrodeposited catalysts (from literature data) and of the Fe₃S₄ nanoparticles according to the invention (example 3) in water at pH 7. Ni-MoS₃ (□), Fe-MoS₃ (o) and MoS₃ (A) curves were recorded in a phosphate buffer (unknown concentration) at 1 mV s^"1. Co/P/O (■) curve was recorded in a 0.5 M phosphate buffer at 0.05 mV s^_1.Fe₃S₄ (□) curve was recorded in a 0.5 M phosphate buffer at 1 mV s^"1. The curve represents the log (J) (in A.cm^"2) vs the overpotential (in mV).

Figure 9 represents IV curves (overpotential in mV versus J in mA.cm^"2) obtained with the iron sulfide nanoparticles of the invention compared to Pt (at 80°C). The (■) represent the results obtained with pyrite nanoparticles; the (·) represent the results obtained with pyrrhotite nanoparticles; the (A) represent the results obtained with greigite nanoparticles; the (T) represent the results obtained with Pt (platinum) nanoparticles (comparative test). Figure 10 represents a comparison at different temperatures of catalytic performances of the iron sulfide nanoparticles of the invention at J = 300 mA cm^"2. The depicted curves represent the overpotential (in mV) versus temperature (in °C) at J = 300 mA cm^"2. The (■) represent the results obtained with pyrite nanoparticles; the (·) represent the results obtained with pyrrhotite nanoparticles; the (A) represent the results obtained with greigite nanoparticles; the (T) represent the results obtained with Pt (platinum) nanoparticles (comparative test). The following examples are meant to illustrate the present invention, but do not intend to limit its scope in any way.

EXAMPLES Chemicals and reagents

All chemicals were of analytical grade and were used without further purification. Iron chloride, FeCl3.6H₂0, and ethylene glycol (HO(CH₂)₂0(CH₂)₂OH) were purchased from Acros; Nafion® (perfluorosulfonic acid-PTFE copolymer, 5% w/w solution) and thiourea (SC(NH₂)₂) were purchased from Alfa-Aesar; Tetraethylammonium acetate and potassium phosphate were purchased from Sigma- Aldrich.

Physical methods

X-ray diffraction patterns were determined on a Panalytical X'pert Pro diffractometer equipped with a multichannel X'celerator detector, using Co Ka radiation (λ = 1.7889 A) in the 2Θ range 15-130 °. The data were collected at room temperature, with a step of 0.033 degrees and a time by step equal to 100 s. The size of the coherent diffraction domains (crystallite size) was determined with MAUD software which is based on the Rietveld method combined with Fourier analysis, well-adapted for nano-objects.

The crystallographic structures are classified according to the ICDD classification, referred to by their JCPDS numbers, which are available on the following website: http://www.icdd.com/.

Nitrogen adsorption-desorption isotherms were obtained at 77 K using a Belsorp max surface Area Analyzer from Bel Japan, Inc.. Samples were first degassed using a Belprep apparatus at 423 K and 10 Pa for 12 h before measurement. Specific surface areas were calculated using the well known Brunauer-Emmett-Teller (BET) equation with relative pressures in the range 0.05 - 0.20.

Gas analyses for dihydrogen detection were performed on a Hewlett Packard 6890 Series GC System gas chromatograph, with a thermal conductivity detector fitted with a 2 m long Agilent Technology 1/8" Carbosieve S3 60-80 mesh column, calibrated with pure H₂ gas. Example 1

FeCl₃.6H₂0 (1.255 g, 4.64 mmol, 1 eq.) and thiourea (0.707 g, 9.3 mmol, 2 eq.) were dissolved in 100 mL of polyol in a 3-neck round flask at room temperature. The dark red solution was heated up to 180 °C at a rate of 6 °C min ¹ under mechanical stirring (500 rpm). Upon heating, a black precipitate occurred at ca. 160 °C, and the reaction mixture was kept under stirring at 180 °C overnight. The precipitate was centrifuged at 8,500 rpm and washed three times with absolute ethanol, each washing being followed by centrifugation at 8,500 rpm. The collected black powder was air-dried overnight at room temperature and then characterized as it was by X-ray diffraction and nitrogen adsorption-desorption.

Polyol Obtained structure (JCPDS number as reference)

Ethyleneglycol Greigite Fe₃S₄ (00-016-0713)

1,2-Propandiol Pyrite FeS₂ (01-071-0053)

1,2-Butandiol Pyrite FeS₂ (01-071-0053)

Diethyleneglycol Greigite Fe₃S₄ (defined as Fe₂₄S₃2) (96-016-0124)

Example 2

Fe₂S₂(CO)₆ (69 mg, 0.2 mmol, 1 eq.) is dissolved in 5 mL of oleylamine. 30 mL of alkylamine solvent are heated upon magnetic stirring in a 3-neck round-bottom flask (equipped with a condenser and a thermocouple) to a temperature of T °C under argon at 10 °C/min. Once the desired temperature is reached, the solution of iron-sulfur complex is quickly added through a syringe, a black precipitate forms immediately and the reaction is kept at temperature T °C overnight (ca. 16 h). The precipitate is centrifuged at 22,500 rpm for 15 min and washed three times with absolute methanol, each washing being followed by centrifugation at 22,500 rpm and sonication. The collected black powder is air-dried overnight at room temperature and then characterized as it is by X-ray diffraction. Alkylamine solvent T (°C) Obtained structure (JCPDS number as reference)

Oleylamine 230 Pyrrhotite Fe₀.₉iS (00-029-0726)

Pyrrhotite Fe₀.₉S (00-029-0724) and other matching

Hexadecylamine 215

phase structures

230 Fei__xS (00-024-0080) and others with x ranging from

Octylamine

(solvothermal) 0 to 0.05

Example 3

Nanoparticles synthesis and characterization

1) Synthesis FeCl3.6H₂0 (1.255 g; 4.64 mmol, 1 eq.) and thiourea (0.707 g; 9.3 mmol; 2 eq.) were dissolved in 100 mL of ethylene glycol in a 3 -neck round flask at room temperature. The dark red solution was heated up to 180 °C at a rate of 6 °C min^"1 under mechanical stirring (500 rpm). Upon heating, a black precipitate occurred at ca. 160 °C, and the reaction mixture was kept under stirring at 180 °C overnight. The precipitate was centrifuged at 8500 rpm and washed three times with absolute ethanol, each washing being followed by centrifugation at 8,500 rpm. The collected black powder was air- dried overnight at room temperature and then characterized as it was by X-ray diffraction and nitrogen adsorption-desorption.

2) Characterization Particle morphology was determined for the synthesized powders using a Transmission Electron Microscope (Jeol-JEM-lOOCXII operating at 100 kV). The powder was dispersed in ethanol and a drop placed on a 200-mesh carbon-coated copper grid.

Particle size were determined for the synthesized powders using X-Ray diffraction pattern recorded on the powder recovered after centrifugation (Figure 2A) using and Debye-Scherrer law. The crystalline phase was identified by comparison with ICSD reference files. The diffraction peaks of all samples matched perfectly the (111), (220), (311), (222) and following crystalline planes of the cubic Fe₃S₄ phase (Fd-3m space group, JCPDS n° 00-016-0713) with slightly broadened peaks suggesting the formation of fine crystals whose size was in the order of magnitude of a few tenths of nm. No characteristic peak of any foreign phase was detected.

XRD was also used as a tool to check the stability of the nanoparticles with time, temperature (BET pretreatment performed at 423K) and electrochemical measurements (Figure 2B). XRD patterns were recorded after different experiments that might damage the crystallographic structure of the material (Figure 3). There was no significant degradation of the material, the structure remaining exactly the same. This testified the high robustness of the nanoparticles.

Electrochemistry

I) Cyclic Voltammetry (Figures 4 A and 4B)

The potentiostat used for cyclic voltammetry was an Autolab PGSTAT 12. The working electrode was a 3-mm diameter glassy carbon rotating disk electrode (Tokai), carefully polished and ultrasonically rinsed in absolute ethanol before use. The counter-electrode was a platinum wire and the reference electrode an aqueous saturated calomel electrode (SCE). All experiments were carried out under argon at 20 °C.

Nafion^® films loaded with Fe3S₄-nanoparticles were prepared to evaluate the catalytic activity towards molecular hydrogen evolution. Vitreous carbon rotating disk electrodes (RDE) were coated with Fe3S₄-nanoparticles dispersed in Nafion^® and aged for 12 h at 100 °C in an oven. Specifically, from a mixture of greigite Fe3S₄-nanoparticles synthesized above (20 mg), Nafion® (66 μί) and isopropanol (200 μί), a drop was deposited on the electrode surface, which was then dried in air and left for 12 h at 100 °C in an oven. A 0.05 M phosphate buffer (pH 7.0) was prepared and used as supporting electrolyte and degassed under argon. The coated electrode was found to be stable for more than 6 months, with no particular storage care.

Under argon atmosphere, cyclic voltammetry in aqueous 0.05 M potassium phosphate (pH 7.0) exhibited a reduction peak at -0.71 V vs. normal hydrogen electrode (ΝΗΕ) (Fig. 4A). Two oxidation peaks at -0.05 and +0.19 V vs. ΝΗΕ allowed the system to recover its resting state, and the cycle was repeated several times without any current density loss. A sharp rise in current from ca. -0.9 V vs. NHE was observed, indicating a catalytic process for water activation (Fig. 4B). This current enhancement is characteristic for catalysis of molecular hydrogen evolution.

2) Controlled potential electrolysis (Figure 5 A) The potentiostat used for controlled potential electrolysis was an Auto lab PGSTAT 12. The working electrode was a 3-mm diameter glassy carbon rotating disk electrode (Tokai), carefully polished and ultrasonically rinsed in absolute ethanol before use. The counter-electrode was a platinum wire and the reference electrode an aqueous SCE electrode. All experiments were carried out under argon at 20 °C. From a mixture of greigite Fe3S₄-nanoparticles (10 mg), Nafion® (33 μί) and isopropanol (100 μί), 2μΙ_^ were deposited on the electrode surface, which was then dried in air and left for 12 h at 100 °C in an oven. A 0.5 M phosphate buffer (pH 7.0) was prepared and used as supporting electrolyte and degassed under argon. Experiments were then conducted at different controlled potential on a rotating disk electrode at a rotating rate of 4,000 rpm for 600 seconds. Straight slopes were then subtracted from the blank solution and from the film reduction.

Figure 5 A shows molecular hydrogen rates measured at different overpotentials (overpotential = | applied potential - 0.059 x pH| V vs. ΝΗΕ) after subtracting the contribution from the blank solution and the film catalyst reduction (see below). Slow catalysis was observed at low overpotentials, and a sharp increase arises from 500 mV, reaching ca. 57 nmol s^"1 cm^"2 at an overpotential of 600 mV.

3) Bulk Electrolysis (Figure 5B)

The potentiostat used for cyclic voltammetry was an Amel 552. The working electrode was a glassy carbon crucible (Tokai), carefully polished and ultrasonically rinsed in absolute ethanol before use. The counter-electrode was a platinum wire soaking in a 1.0 M tetraethylammonium acetate solution, separated from the solution through a porous glass frit. The reference electrode was an aqueous SCE electrode, separated from the solution through a porous glass frit. All experiments were carried out under argon at 20 °C. A mixture of greigite Fe3S₄-nanoparticles (20 mg), Nafion® (66 μί) and isopropanol (200 μί) was deposited into the crucible, which was then dried in air and left for 12 h at 100 °C in an oven. A 1.0 M phosphate buffer (pH 7.0) was prepared, degassed under argon for 1 hour and used as supporting electrolyte.

Figure 5B presents bulk electrolysis in 1.0 M potassium phosphate (pH 7.0), carried out over five days at an overpotential of 300 mV (-0.713 V vs. ΝΗΕ). The slope is perfectly straight for the first two days of the experiment, and then subsequently curves due to a slight increase of the pH, the phosphate buffer being modified during the electrolysis due to proton consumption (pH 7.3 after 5 days). Taking apart the first hour of experiment, quantitative (> 0.95) Faradic yield for molecular hydrogen evolution was confirmed by gas chromatography analysis as well as by volumetric measurements. Long-duration controlled-potential electrolysis (CPE) was performed to assess the durability and robustness of the pyrrhotite Feo.9S nanoparticle-coated electrode. Quantitative (>0.99) Faradaic yield for molecular hydrogen evolution was confirmed by gas chromatography analysis (figure 6) as well as by volumetric measurements. Only molecular hydrogen is observed, the very small peak of dioxygen corresponds to a probable small leak of the cell that allows air to enter in the cell. No other gas is detected.

It is worthy to note that after five days no significant structural change for the electrolyzed catalyst was observed by XRD (Fig. 2). The stability of the material was also tested by stirring pyrite FeS₂ powder in 1 M aqueous solution of sulfuric acid. No degradation was observed after 3 days of immersion: no Fe³⁺ ions are detectable in the supernatant because this latter remains colorless.

Example 4

From polarisation curves in neutral water, comparison of the activity of the nanoparticles of example 3 with prior art electrodeposited non-precious catalysts is presented in Figure 8 and Table 1.

At a given current density, J = 0.1 mA cm^"2, the overpotential of the nanoparticles of example 3 is shifted positively by 250 mV compared to N1-M0S₃, the best in the molybdenum sulfide series reported by Hu and co-workers (Merki et al Chem. Sci. 2012, 3, 2515-2525), and by 120 mV compared to the cobalt phosphate material reported by Artero and co-workers (Cobo et al Nat. Mat. 2012, 11, 802-807). Table 1. Exchange current densities (Jo) of different electrocatalysts, in water at pH 7 (data extracted from polarization curves, Fig. 8).

Jo Slope Range of η

Material References

(mA cm^"2) (mV dec^-1) (mV)

Ni-MoS₃ 1.0 x 10-² 96 100-150 Merki et al

Fe-MoS₃ 5.0 IO-³ 95 125-175 Merki et al

Co/P/O 1.9 x IO-³ 134 200-300 Cobo et al

M0S3 1.2 x IO-³ 86 150-200 Merki et al

Present

Fe₃S₄ 3.6 x IO-⁴ 147 290-350

invention

This comparison demonstrates that the catalyst of the invention for molecular hydrogen evolution from neutral water is robust and efficient, operates at a mild overpotential. High turnover frequencies were observed at a moderate overpotential, together with catalytic stability exceeding five days. Despite the fact that the particles of example 3 (Fe₃S₄-nanoparticles) are less active than cobalt phosphate or molybdenum sulfide materials at a given overpotential, the stability and the ease of preparation are clear advantages, in particular in view of industrial large scale applications. Taking into account the abundance and cheapness of the different materials, the nanoparticles of the present invention are clearly superior to catalysts of the prior art, in particular cobalt phosphate or molybdenum sulfide.

Example 5

I) Electrodeposition of iron sulfide material

Electrodeposition of iron sulfide material was conducted by immersing a vitreous carbon electrode into an aqueous solution of 0.35 M of Na₂S₂C"3 and 0.035 M of FeS0₄ and cycling the potential hundred times between + 0.1 V and - 1.0 V vs. SCE at 0.1 V/s. 2) Comparison of catalysis performances between electrodeposited iron sulfide and iron sulfide nanoparticles (Figure 7)

To compare the different catalysts, all the electrocatalysis experiments are done at pH 7.0 in a phosphate buffer at 1 mV/s.

For the greigite Fe₃S₄ and the pyrite FeS₂ nanocatalysts, a pre-wave is observed prior to the catalytic wave. This pre-wave corresponds to the reduction of the film. The catalytic current for the electrodeposited material is much weaker than the one of the iron sulfide nanocatalysts. No significant current enhancement was observed between 0 and -0.85V which show lesser performances of electrodeposited iron sulfide compared to nanoparticles.

Among iron sulfide nanoparticles, greigite and pyrite seem to have similar performances which are better than those observed with nanoparticles of pyrrhotite.

The results of these comparative experiments are presented in figure 7. Example 6

Membrane pre-treatment

Nafion 115 membrane was boiled for 1 h 3 wt. % H₂0₂, rinsed with deionized water, soaked for 1 h in 50 vol. % FiN0₃, rinsed with DI water, boiled in 1 M H₂SO₄ and rinsed with deionized water. The membrane was dried at 80 °C for 4 h.

Ink recipe

Anode: 25 mg of Ir0₂ and 312.5 of Nafion (5 wt. %, Aldrich) were mixed with 1 mL of deionized water and 3 mL of isopropanol (IPA). The solution was sonicated for 1 h. The ink was then sprayed using an aerograph onto a 6.25 cm² Teflon sheet until a 2 mg/cm² catalyst loading was reached. The electrode was transferred onto pre-treated a Nafion 115 membrane using the following procedure: the MEA (Membrane Electrode Assembly) was heated in a hot-press at 80°C, the pressure was then increased up to 0.5 Tonn; the temperature was then increased up to 135°C and when stabilized, the pressure was increased up to 1 Tonn; the MEA was left at 135°C, 1 Tonn for 90 s.

Cathode: 50 mg of catalyst, 12.5 mg of carbon Vulcan XC72R (Cabot) and 625 of Nafion (5 wt. %, Aldrich) were mixed with 2 mL of deionized water and 6 mL of isopropanol. The solution was sonicated for 1 h. The ink was then sprayed using an aerograph onto a 6.25 cm² Sigracet 10 BC gas diffusion layer until a 5 mg/cm² catalyst + carbon loading was reached. The resulting gas diffusion electrode (GDE) was hot pressed onto pre-treated a Nafion 115 membrane using the same procedure as for decal transfer. The specifications of the catalyst layers prepared are listed in Table 2.

Table 2. Specifications of the catalyst layers prepared

In-situ testing

Compared to compositions with 18 % of Nafion®, compositions with 30 % of Nafion® afforded the best results for FeS nanoparticles.

10 MEA is tested on a home-made test station using a 6.25 cm² single cell. Deionized H₂0 is supplied at the anode with 200 g/h flow rate, cathode is not hydrated. IV curves obtained at 80 °C with iron sulfide nanoparticles MEA of the invention and with Pt- based MEA are shown on the figure 9.

15 On the figure 9, two different regimes can be identified. The straight line observed above 500 mA/cm² for iron sulfide nanocatalysts or 200 mA/cm² for Pt-based MEA corresponds to the cell assembly resistance. Below, these values are characteristic of the performances and efficiency of the catalyst, and the iron sulfide nanocatalysts are presenting an overpotential vs. Pt-based MEA.

20 Among the three materials studied, pyrite FeS₂ nanoparticle is the best one.

Effect of the temperature

Comparison at J = 300 mA cm^"2. Pyrite FeS₂ nanoparticle is the most efficient catalyst with an overpotential of 0.36 V at 80 °C, 0.38 V at 90 °C and 0.35 V at 100 °C (figure

Claims

1. Use of a heterogeneous catalyst for electrolytic proton reduction into hydrogen gas, said catalyst comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst.

2. Use of claim 1, wherein said catalyst is made of nanoparticles comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

3. Electrode coated with a heterogeneous catalyst, said catalyst comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the catalyst.

4. Electrode of claim 3, wherein said catalyst is coated on the electrode via electrodeposition advantageously using an electroplating process.

5. Electrode of claim 3, wherein said electrode is coated with a composition comprising a binder and said catalyst, said catalyst being made of nanoparticles comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles.

6. Electrode according to claim 5, wherein said nanoparticles have a particle size lower than or equal to 200 nm.

7. Electrode according to claim 5 or 6, wherein said nanoparticles comprise at least 80% by weight of Fe_xS relative to the total weight of the nanoparticles, with x between 0.5 and 1 included, preferably x is between 0.5 and 0.75 or is between 0.85 and 0.95.

8. Electrode according to any of claims 5 to 7, wherein said binder is selected from the group consisting of conductive polymers, ionomers or fluoropolymers, preferably a fluorinated polymer with sulfonic end groups such as a perfluorosulfonic acid.

9. Electrode according to any of claims 3 to 8, wherein it essentially consists of carbon, coated with said catalyst.

10. Electrode according to any of claims 3 to 8, wherein the electrode essentially consists of Indium Tin Oxide (ITO), or Fluorine Tin Oxide (FTO), gold or mercury, and is coated with said catalyst.

11. Electrolytic cell comprising the electrode according to any of claims 3 to 10, preferably as the cathode.

12. Method comprising performing proton reduction using the electrode according to any of claims 3 to 10 as the cathode or the electrolytic cell of claim 11, and an electrolyte, thereby producing hydrogen gas.

13. Method according to claim 12, wherein the electrolyte is a buffered aqueous solution, preferably so that the pH of said electrolyte is of between 5 and 8, more preferably between 6.5 and 7.5, and wherein the proton reduction is advantageously carried out at a temperature of between 15°C and 30°C.

14. Method according to claim 12 or 13, wherein the overpotential applied to the electrodes is below 0.5 V, preferably below 0.4 V, even more preferably below 0.3 V.

15. Method according to claim 12 or 13, wherein the potential applied to the electrode is below (0.5 + 0.059 *pH) V versus normal hydrogen electrode, preferably between (0.05 + 0.059*pH) and (0.4 + 0.059*pH) V versus normal hydrogen electrode, even more preferably between (0.1 + 0.059*pH) and (0.3 + 0.059*pH) V versus normal hydrogen electrode, pH representing the pH of the electrolyte solution.

16. Method for preparing nanoparticles comprising at least 80% by weight of Iron (Fe) and Sulfur (S) relative to the total weight of the nanoparticles, where it comprises the following successive steps: a) mixing a precursor of formula Fe₂S₂(CO)6 and a solvent, preferably a polyol or an alkylamine, preferably an alkylamine, b) heating the obtained reaction mixture through thermal heating under pressure or at atmospheric pressure, and c) recovering the thus obtained nanoparticles.