WO2013107841A1

WO2013107841A1 - Method for increasing the lifespan of a fuel cell with a proton exchange membrane

Info

Publication number: WO2013107841A1
Application number: PCT/EP2013/050891
Authority: WO
Inventors: Sylvain PASSOT; Christel FAURE; Alejandro Franco; Olivier Lemaire
Original assignee: Commissariat A L'energie Atomique Et Aux Energies Alternatives
Priority date: 2012-01-19
Filing date: 2013-01-18
Publication date: 2013-07-25
Also published as: EP2805371A1; FR2986107B1; FR2986107A1

Abstract

The subject matter of the invention is a method for increasing the lifespan of a fuel cell having a proton exchange membrane comprising an electrochemical cell supplied with fuel with at least one catalytic electrode comprising carbon and one catalyst made from noble metal which can be platinum or palladium or ruthenium, and means for measuring the voltage of said cell characterised in that it comprises at least the following steps: - a first phase (Ph1) of supplying said cell with a first flow of fuel comprising hydrogen for a first period (T1) and resulting in the definition of a first slope (a) of the curve indicating the change in voltage of said cell as a function of time; - a second phase (Ph2) of supplying said cell with a second flow of fuel comprising hydrogen and a quantity of polluting species containing sulphur for a second period (T2), said second period and the quantity of polluting species being such that they result in the defining of a second slope (b) of the curve indicating the change in voltage of said cell as a function of time, the ratio of said second slope over said first slope (b/a) being strictly greater than a value of 1, said second phase generating the capture of said polluting species by said catalytic electrode; - a third phase (Ph3) for regenerating the catalytic electrode comprising the elimination of said polluting species.

Description

METHOD FOR INCREASING THE LIFETIME OF A PROTON EXCHANGE MEMBRANE FUEL CELL

The field of the invention is that of proton exchange membrane fuel cells, commonly referred to by the acronym "PEMFC" corresponding to the English term: "Proton Exchange Membrane Fuel Cell".

PEMFCs are current generators whose operating principle is based on the conversion of chemical energy into electrical energy by catalytic reaction of hydrogen and oxygen. Membrane electrode assemblies (MEAs), commonly referred to as battery cores, are the building blocks of PEMFCs. As represented in FIGS. 1a and 1b, the stack core 1 is composed of a polymer membrane 2, and catalytic layers 3, 4 present on one side and the other of the membrane 2 and respectively constituting the anode and the cathode.

The membrane 2 thus makes it possible to separate the anode and cathode compartments 6. The catalytic layers 3, 4 are generally nano-particles supported by carbon aggregates. Gaseous diffusion layers 7,8 (carbon fabric, felt, etc.) are placed on either side of the MEA 1, to ensure the electrical conduction, the homogeneous distribution of the reactive gases and the evacuation of the water produced by the reaction. A system of channels 9, 10 placed on each side of the AME brings the reactive gases and evacuates water and excess gases to the outside.

At the anode 3, the decomposition of hydrogen adsorbed on the catalyst produces protons H ⁺ and elections e ^" .The protons then pass through the polymer membrane before reacting with oxygen at cathode 4.

The reaction of protons with oxygen at the cathode leads to the formation of water and the production of heat 4.

Depending on the method of producing hydrogen, the gas may include impurities. It has been shown that carbon monoxide and sulfur compounds are particularly harmful to the operation of the battery.

In this context, maximum concentration thresholds have been proposed to standardize the quality of hydrogen used for fuel in the case of an automotive application: for example, 0.2 μιτιοΙ / mol for CO and 0.004 μιτιοΙ / mol for sulfur compounds.

Currently, improving the life span of PEMFCs is a major issue for the use and development of batteries for the consumer market. That is why the highlighting and understanding of cell aging phenomena are essential today. In this context, numerous scientific studies have shown that the aging of fuel cells is, among other things, associated with the change of the nano / microstructural properties of the catalytic active layer.

For example, a sharp decrease in the thickness of the active layer at the cathode has been observed after a few hours of operation. The damage of the carbon support to the cathode induces the loss of catalytic surface and an increase of the contact resistance between the cathode and the gas diffusion layer. This contributes to reducing the sustainability of PEMFCs. This degradation is due to the corrosion of the carbon catalyst support by water according to the following reaction:

C + 2H ₂ 0 → C0 ₂ + 4H ⁺ + 4th ^~ (1)

The redox potential of this reaction is about 0.2 V / ENH. Since the cathode potential of a cell is generally greater than 0.2 V, this reaction always takes place. In addition, the permanent presence in large quantities of water, produced by recombination of the protons with oxygen at the cathode, favors the reaction (1).

The presence of oxygen at the anode, resulting from the cathode by permeation through the membrane, can promote the proton pump effect as described in the article by Franco, M. Gerard, J. Electrochem. Soc. 155 (4) (2008) B367. : the reaction of anodic oxygen on the catalytic sites to form water (2), consumes protons at the anode.

4H ⁺ + 4th ^~ → 2H ₂ 0 (2) This proton pump effect increases carbon corrosion at the cathode by shifting the balance to proton production. The reaction (1) is then strongly shifted to the right. The reaction (1) is also shifted to the right by the electron pump effect at the cathode: the reduction of oxygen (2) at the cathode sucks the electrons produced by the corrosion of the carbon (1). Platinum better catalyzes carbon corrosion if an oxygen reduction reaction occurs on its surface.

In parallel, the reduction of oxygen at the anode on the catalytic sites can generate the production of free radicals and / or hydrogen peroxide by reactions (3) and (4).

0 ₂ + H ⁺ → HOO '(3)

HOO '+ H ^{+ -} H ₂ 0 ₂ (4)

It is these compounds that are responsible for the degradation of the membrane. The degradation of the membrane induces an increase in the proton transfer resistance of the anode to the cathode due to a change in the structure of the proton conductive polymer.

The mechanisms of membrane degradation and carbon corrosion (via the proton pump effect) therefore have in common the reduction of oxygen at the anode. This reaction occurs during a local depletion of hydrogen on the catalytic sites. This depletion can be caused by current transients occurring during the stop / start phases or during power cycles, but also by the formation of water plugs, or by stopping the supply of hydrogen.

One of the major problems remains the limitation of carbon corrosion and degradation of the membrane by acting on the oxygen reduction reaction at the anode.

Several solutions have been proposed to reduce or eliminate carbon corrosion at the cathode, among which may be mentioned in particular:

- the modification of the nature of the support carbon:

Some methods involve acting directly on the carbon support, such as the use of carbon carriers that are more resistant to corrosion, such as carbon nanotubes or fullerenes. A heat treatment of the carbon support also improves its resistance to corrosion. C + 2H ₂ 0 → C0 ₂ + 4H ⁺ + 4th ^~ (5)

Nevertheless, the use of carbon nanotubes or fullerenes increases the cost of the active layer. In addition, the electrical conductivity of fullerenes is significantly lower than conventional carbon. The electrical conductivity of carbon nanotubes is better than that of conventional carbon only in the nanotube axis. Finally, this solution does not prevent the reduction of oxygen at the anode which can produce free radicals.

- the addition of CO? in the oxidizing gas (cathode)

A proposed solution to reduce carbon corrosion at the cathode is to introduce carbon dioxide (CO ₂ ) into the air at the cathode and control its amount. This solution was recently patented by Toyota. By introducing C0 ₂ , the reaction (1) is displaced to the left and carbon consumption / corrosion is thus slowed down:

C + 2H ₂ 0 ^ _ C0 ₂ + 4H ⁺ + 4th ^~ ₍₆ )

This solution nevertheless implies the use of a supply of C0 ₂ additional to that of atmospheric air. In addition, this solution does not prevent the reduction of oxygen at the anode which can produce free radicals.

the addition of CO in the fuel gas (anode)

It has also been proposed to introduce carbon monoxide (CO) at the anode to limit corrosion of cathodic carbon (7). Indeed, by reacting with the 0 ₂ present at the anode, the CO limits the proton pump effect which moves the balance of the reaction (7) to the right.

C + 2H ₂ 0 ^ ^→ C0 ₂ + 4H ⁺ + 4th ^~ (7) The presence of a small amount of CO in the hydrogen thus has a beneficial effect for the corrosion of the cathodic carbon as described in the article by AA Franco, M. Guinard, B. Barthe, O. Lemaire, Electrochim. Acta, 54 (22) (2009) 5267.

This technical solution nevertheless has the disadvantage of using a mixture of hydrogen and carbon monoxide without any other impurity. It is recognized that CO is more easily removed from reform gases than sulfur compounds. The tolerance of a less pure mixture to have the right amount of CO therefore implies the presence in greater quantity of other impurities. This solution also requires continuous injection and the right amount of CO to limit the reduction of oxygen, which implies having a bottle of CO or H ₂ / CO mixture in optimal quantity; - the use of a protective layer

The use of a non-conductive protective layer based on silica (SiO ₂ ) on a set [carbon nanotube / platinum] has been proposed by the authors: S.Takenaka, H.Matsumori, H.Matsune, E. Tanabe, and M. Kishida, J. Electrochem. Soc., 155 (9) (2008) B929. This layer aims to limit the platinum migration and thus the formation of aggregates of catalytic particles responsible for a decrease in the performance of the battery. This solution therefore acts on the stability of the catalyst but not on the carbon which is already stable. The authors make it clear that carbon nanotubes have good stability to carbon corrosion. Furthermore, the SiO ₂ layer is produced over the entire [carbon nanotubes / platinum] system by hydrolysis of compounds (3-aminopropyltriethoxysilane and tetraethoxysilane) previously mixed with the entire system [carbon / platinum nanotubes]. ].

However, this technical solution is complex and expensive. In addition, the addition of a protective layer may cause a decrease in the active area and therefore performance. Finally, this solution does not prevent the reduction of oxygen at the anode which can produce free radicals.

In this context, the present invention relates to a method for increasing the lifetime of a proton exchange membrane fuel cell comprising an electrochemical cell fueled with fuel. with at least one catalytic electrode comprising carbon and a noble metal catalyst which may be platinum or palladium or ruthenium, and means for measuring the voltage of said cell characterized in that it comprises at least the steps following:

a first supply phase of said cell by a first fuel stream comprising hydrogen for a first duration and leading to the definition of a first slope a of the curve relating to the evolution of the voltage of said cell according to time ;

a second supply phase of said cell by a second fuel stream comprising hydrogen and a quantity of polluting species based on sulfur for a second duration, said second duration and the quantity of polluting species being such that they lead to the definition of a second slope b of the curve relating to the evolution of the voltage of said cell as a function of time, the ratio of said second slope on said first slope being strictly greater than a value of 1, said second phase generating the capture of said polluting species by said catalytic electrode;

a third regeneration phase of the catalytic electrode comprising supplying a third stream comprising hydrogen with a constant current density, said third stream being less than a nominal fuel flow, for an initial duration leading to a value a stoichiometric coefficient of said fuel less than 1, so as to increase the oxidation potential of said catalytic electrode and to reach the oxidation potential of said polluting species, to eliminate it.

The ratio b / a can be, for example, 3, thus corresponding to a slope b of 300 μν / h and a slope a of 100 μν / h, but also to a slope b of 30 μν / h and a slope of 10 μν / h. As a reminder, a slope of 300 μν / h can correspond to both a loss of 300 μν in one hour and a loss of 30 mV in 100 hours. The value of b / a equal to 1 would correspond to an equality of the slopes, which would mean the absence of impact of the second flow with respect to the operation with the first flow. Advantageously, the ratio b / a may be limited to a value less than 100.

The third phase can be operated under a third flow identical to the first flow, the regeneration of the electrode being facilitated, in the absence of the polluting species.

According to a variant of the invention, the third regeneration phase of the catalytic electrode is operated under a third fuel flow identical to the second flow.

According to a variant of the invention, the second fuel stream comprises a type of gas H ₂ S or CS ₂ or S0 ₂ or COS.

According to a variant of the invention, the first stream consists of reformed hydrogen.

According to a variant of the invention, the first fuel stream comprises pure hydrogen.

According to a variant of the invention, the first fuel stream comprises a first quantity of H ₂ S, the second fuel stream comprising a second quantity of H ₂ S, the said second quantity being greater than the said first quantity.

According to a variant of the invention, the third phase comprises the application of a power pulse on said cell in potentiostatic operation by an external load.

According to a variant of the invention, the third phase comprises a voltammetry cycle on said catalytic electrode in order to reach the oxidation potential of said polluting species.

According to a variant of the invention, the third phase further comprises the following succession of steps:

decreasing the third fuel flow from a nominal value, corresponding to a first stoichiometric value for an initial duration, passing the potential of said catalytic electrode from an initial value to a first value greater than or equal to the oxidation potential the polluting species;

increasing said third fuel flow corresponding to a second stoichiometric value for a secondary duration to accelerate the fuel flow to the catalytic sites of the anode; the return of said third stream to said nominal operating value for the reaction of the fuel with the catalytic electrode.

According to a variant of the invention, the initial duration and the secondary duration are of the order of a few minutes.

According to a variant of the invention, the duration of the first phase is greater than or equal to the duration of the second phase.

According to a variant of the invention, the second stream comprises an amount of H ₂ S of between about a few ppb and a few ppm, the durations of the first and second phases being typically between 30 minutes and 50 hours depending on the concentration of H ₂ S this duration being inversely proportional to the concentration.

According to a variant of the invention, the method comprises a cycle of steps comprising at least the first, the second and the third phases.

According to a variant of the invention, the method of the invention advantageously comprises the following steps:

the first phase under feed in first fuel flow;

the acquisition of the voltage of the electrochemical cell;

the second phase under supply of a second fuel stream comprising a polluting species for a second duration and enabling the attainment of a voltage threshold value of said electrochemical cell, without prior determination of said second duration and the quantity of polluting species; in said second fuel stream;

the regeneration phase of the catalytic electrode.

According to a variant of the invention, the catalytic electrode comprises platinum nanoparticles supported by carbon aggregates.

The invention will be better understood and other advantages will become apparent on reading the description which follows given by way of non-limiting example and by virtue of the appended figures among which:

FIGS. 1a and 1b illustrate the diagrams of the operating principle of a PEMFC fuel cell;

FIG. 2 schematizes the evolution of the voltage of an electrochemical cell of a proton exchange fuel cell fed by a pure H ₂ flux according to the known art and the evolution of the voltage of an electrochemical cell of a proton exchange fuel cell fed in successive cycles by a flow in H ₂ and by a flow further comprising a polluting species H ₂ S according to the invention;

FIG. 3 illustrates the results obtained in an exemplary method of the invention using a second fuel stream comprising 45 ppb of H ₂ S, in terms of evolution of the cell voltage in comparison with those obtained with a constant flux. of H ₂ according to the known art.

In general, the present invention provides a method for improving the life of a fuel cell by reducing the irreversible degradation of the carbon present at the catalytic electrode and by resorting discontinuously to an operation. reversible poisoning of said electrode preventing said degradation.

The Applicant proposes to exploit a reversible poisoning operation which is naturally a handicap, but which becomes an asset in the process of the invention, making it particularly clever.

According to the process of the invention, in a first phase, a first fuel stream based on pure hydrogen is used in a conventional manner, or reformed, during a first period during which the carbon of the catalytic electrode begins to degrade by undergoing an oxidation reaction.

In a second phase, the catalytic electrode is poisoned by feeding the electrolytic cell with a gas flow comprising said species capable of poisoning said electrode. In this step, it is important to measure the duration and rate of polluting species by following the evolution of the voltage of the electrochemical cell to properly control the process as will be developed in more detail in the following description.

Finally, in a third phase, the regeneration of said electrode is carried out by removing the polluting species at the catalytic electrode. The invention will be described in the context of the use of a pollutant species based on sulfur H ₂ S type but could also be implemented with other polluting species.

In general, hydrogen sulphide (H ₂ S) is an impurity present in hydrogen from the reforming of hydrocarbons (carbon, natural gas, petroleum). Other impurities are also present, such as carbon monoxide CO which is however more easily removed than H ₂ S, which is an advantage over the solution disclosed in the article by S.Takenaka, H.Matsumori , H.Matsune, E.Tanabe, and M.Kishida, J. Electrochem. Soc., 155 (9) (2008) B929.

According to the present invention, it is thus proposed to use hydrogen sulfide in a suitable amount in the fuel and this in a sequenced manner. Indeed, the dissociative adsorption mechanism of H ₂ S on platinum, a catalyst conventionally used for catalytic electrodes, is the following as described in the article by M.-V. Mathieu, M. Primet, Applied Catalysis, 9 (3) 361-370 (1984):

H ₂ S + Pt → Pt - S + H ₂ (8)

H ₂ S + Pt - H → Pt - S + Y ₂ H ₂ ( ₉ ) These reactions have two key points:

the catalytic oxidation of H ₂ S locally creates a Pt-S adsorbed sulfur layer on the catalysts. This local poisoning of catalytic sites prevents the catalytic reduction of oxygen at the anode.

In the case of the prior art previously mentioned and concerning carbon monoxide CO, it does not create a protective layer on the platinum, it reacts with oxygen. This requires a continuous supply of CO. According to the solution proposed in the present invention, exposure to H ₂ S is limited to a certain duration;

- the catalytic oxidation of H ₂ S produces hydrogen. This reaction supplies locally neighboring sites with hydrogen, which reduces hydrogen depletion.

These two processes help to reduce the reduction of oxygen at the anode and thus the corrosion of the carbonaceous catalytic support at the cathode, as well as the formation of H ₂ 0 ₂ at the anode and the degradation of the membrane that follows.

While the injection of H ₂ S in a continuous manner would generate a strong poisoning of the catalytic sites and the beneficial effect on the durability would be masked, it is proposed in the present invention a strategy based on the cycling of the exposure to H ₂ S .

Figure 2 illustrates for this purpose, the main steps implemented in the method of the invention.

A first phase Ph1 corresponds to operation with the main supply with a first pure hydrogen stream.

A second phase Ph2 corresponds to a controlled exposure of the AME with a second stream comprising H ₂ S to implement the processes of attenuation of the degradation phenomena of the materials as previously described. This phase is followed by a return to pure H ₂ and regeneration of the electrode to recover the performances.

FIG. 2 thus makes it possible to compare the evolution of the cell voltage under pure H _2, illustrated by the curve C _2a , with the evolution of the cell voltage with cycles of exposure to H ₂ S, illustrated by the lines broken C ₂ b. The slope represents the average speed of performance loss in pure hydrogen (in μν h ^"1) corresponding to material degradation mechanisms and in particular those described above.

According to the present invention, in the second phase Ph2, the slope b greater than the slope a represents the average speed of loss of performance due to the poisoning of the catalytic sites by H ₂ S.

During this phase, the degradation mechanisms of the materials are attenuated. These performance losses are therefore reversible provided that the catalytic sites are decontaminated by a method of regeneration of the electrode and corresponding to the Ph 3 phase.

This regeneration method requests, for example, the following depollution reaction:

Pt - S + H ₂ 0 -> SO ^~ + & H ⁺ + 6th ^~ + Pt ₍ i ₀ )

This reaction occurs for example during the application of a power pulse on the battery in potentiostatic operation by an external charge, while decreasing the hydrogen flux, as described in the FA Uribe article, TQT Rockward, Cleaning (de-poisining) PEMFC electrodes from strongly adsorbed species on the catalyst surface. US 20060249399 A1. 2006.

Another method that can also be used is to carry out a cyclic voltammetry on the anode to reach a high enough potential to oxidize the sulfur as described in the article by FH Garzon, T. Rockward, Urdampilleta IG, EL Brosha, FA Uribe, ECS Trans., 3 (1) 695-703 (2006). This method nevertheless involves feeding the anode with nitrogen and the cathode with hydrogen.

Another method that can also be used for the regeneration phase comprises the passage at zero current for several hours as explained in the references: IG Urdampilleta, FA Uribe, T. Rockward, EL Brosha, BS Pivovar, FH Garzon, ECS Trans. 1 (1) 831-842 (2007) and D. Imamura, Y. Hashimasa, ECS Trans., 11 (1) 853-862 (2007).

It is also interesting to use a method of regeneration of the usable electrode without adaptation of the galvanostatic operating mode. More precisely, as described above, the poisoning of an electrode by sulfur compounds leads to the formation of a sulfur layer on the surface of the catalyst which can only be oxidized at a high electrode potential, following a possible reactions such as reaction (10).

It is proposed in the invention to act on the flow of hydrogen via the stoichiometric coefficient St _Hi .

Anode side, according to the overall hydrogen oxidation reaction, a molecule of hydrogen thus produces 2 protons and 2 electrons. The flow of hydrogen necessary to establish a current / (at A) corresponds to the following equation:

with F the constant of Faraday (in Asmol ^"1 ).

In fuel cell systems, hydrogen at the anode and air (or oxygen) at the cathode are very often injected in excess (super-stoichiometric mode operation). The flow of hydrogen necessary for the establishment of the current is thus multiplied by the stoichiometric coefficient St _Hi :

Q "(mol s ^{~ l} ) = St" x- -

" ^{2 Hl} 2x F

The stoichiometric coefficient corresponding to the excess of gas injected.

By working at a constant current density, the decrease in the stoichiometric hydrogen coefficient below unity means that the hydrogen flow will no longer be sufficient to supply the quantity of electrons necessary to maintain the current at its setpoint. . Thus, in a first step, the current is maintained by the oxidation of the hydrogen present. Then, the hydrogen being in insufficient quantity, the anodic potential increases. Still with the aim of maintaining the constant current, other reaction mechanisms than the hydrogen oxidation reaction are solicited. These other mechanisms, for example the reaction which is the oxidation of sulfur adsorbed on the surface of platinum to form sulfuric acid, are made possible by the increase of the potential.

Thus, the flow of hydrogen should be reduced to a stoichiometry below unity (stoichiometric coefficient between 0 and 1) for a certain initial period. Typically, a necessary time can be between 1 and 10 minutes to reach an anode potential of at least greater than 0.9 V - 1, 1 V.

Indeed, this initial duration of application must be short enough not to solicit other reaction mechanisms that could degrade the electrode (for example carbon corrosion, electrolysis of water ...) and long enough to oxidize all the sulfur species adsorbed on the catalyst.

In general, the method of the present invention uses a strategy of limiting loss of performance due to the degradation of the materials by the cyclic exposure to traces of hydrogen sulphide and this can be operated both manually and dynamically. .

In the case of a manual application of the method, the prior operator the duration of exposure and the concentration in H ₂ S. The traces of Hydrogen sulfide can typically be between 4ppb and 1ppm. The duration of exposure to H ₂ S may typically be between 30 minutes and 50 hours depending on the concentration of H ₂ S, this duration being inversely proportional to the concentration.

In the case of a dynamic application of the method, the duration of exposure to H ₂ S can be adjusted automatically and dynamically so that performance losses due to poisoning do not exceed a certain percentage. This tolerance can be for example 10%. A control and monitoring system may allow not know the concentration values of H ₂ S and corresponding exposure time. The steps of such a control system can be as follows:

Step E0: operation under a flow of pure hydrogen;

Step E1: the acquisition of the cell voltage;

Step E2: the phase of poisoning by H ₂ S whose concentration is not known a priori (for example H ₂ reformed). The end of this poisoning phase occurs when the voltage decrease due to poisoning remains below a certain tolerance value on the cell voltage compared to the acquired value in E2;

Step E3: the return under a pure hydrogen flow;

Step E4: the regeneration phase of the electrode and return to step

E0. The operator determines the duration of the pure hydrogen phase separating two cycles of exposure to H ₂ S (step E0). This duration can be at least equal to the duration of exposure to H ₂ S.

This method can be applied with a main supply of pure hydrogen and a supplementary supply of pure H ₂ S, or with an additional supply of mixture H ₂ / H ₂ S or a subsidiary supply of reformed hydrogen.

This method can be applied with a main feed of reformed hydrogen of superior quality and an auxiliary supply of pure H ₂ S, or with an auxiliary supply of mixture H ₂ / H ₂ S or an annexed supply of reformed hydrogen of inferior quality to the main supply.

The minimum flow rate of hydrogen that must imperatively be provided, is given by the following relation and previously explained:

with / the operating current (in A) and F the Faraday constant (in Asmol ^"1 ).

This flow rate can be multiplied by a factor called stoichiometric coefficient of between 1 and 100 for example, corresponding to the excess of hydrogen that is supplied to the battery.

In the case of the injection of H ₂ S traces, the hydrogen flow rate must be greater than or equal to the minimum flow rate. There can therefore be an increase in the total flow rate when the battery is fed with H ₂ + H ₂ S.

For example, for an operation at 0.6 A cm- ² of a cell of 25 cm ² (thus a total current of 15 A), with a hydrogen stoichiometry of 1.5, the total flow of pure hydrogen must be 1, 17 10 ^"4 mol ^{s" 1} at 157 ml min ^"1. Under the same conditions and with a supply of H ₂ S from a bottle of premix H ₂ / H ₂ S with a concentration of 1 ppm of H ₂ S in H ₂ and to inject 45 ppb of H ₂ S to at the anode, the flow rate of premix must be 7 ml_ min ^-1 and the flow rate of pure H ₂ must be 150 ml_ min ^-1 .

This method can be applied to the cathode to act on the electron pump effect. When injecting sulfur compounds into the fuel, the second process occurs at the cathode. Fewer platinum sites are available for oxygen reduction, the electron pump effect is decreased and therefore carbon corrosion is slowed down.

Example of realization

FIG. 3 illustrates the results obtained in an exemplary method of the invention using a second fuel stream comprising 45 ppb of H ₂ S, in terms of evolution of the cell voltage in comparison with those obtained with a constant flow of H ₂ of the prior art.

For that two tests were carried out: the first test begins with the injection of 45 ppb of H ₂ S in hydrogen for 45 hours (curve C _3a ). The poisoning is then stopped and pure hydrogen feeds the anode for 25 hours. The characterization phase Ph _c is then followed by a phase in pure hydrogen with a regeneration phase according to the previously described method using a third stream lower than a nominal fuel flow and denoted Fe _va , for an initial duration leading to a value of the stoichiometric coefficient of said fuel less than 1, so as to increase the oxidation potential of said catalytic electrode and to reach the oxidation potential of said polluting species, to eliminate it. Performance recovery is 98% of the initial voltage value;

for the second test, the same protocol with a variable flow supply phase is followed but the electrode is fed throughout the pure hydrogen test (curve C ₃ t>) - The method of variable flux feed Fe _of the electrode does not recover more performance. The cell voltage corresponds to 93% of initial performance.

This exemplary embodiment describes the obtaining of a gain during a single H ₂ S exposure cycle. This performance gain amounts in this case to 5% thanks to the exposure to 45 ppb of H ₂ S for 45 hours.

Claims

1. A method of increasing the lifetime of a proton exchange membrane fuel cell comprising an electrochemical cell fueled with at least one catalytic electrode comprising carbon and a noble metal catalyst which may be platinum or palladium or ruthenium, and means for measuring the voltage of said cell characterized in that it comprises at least the following steps:

a first supply phase (Ph1) of said cell by a first fuel stream comprising hydrogen for a first duration (T1) and leading to the definition of a first slope (a) of the curve relating to the evolution the voltage of said cell as a function of time;

a second phase (Ph2) for supplying said cell with a second fuel stream comprising hydrogen and a quantity of polluting species based on sulfur for a second duration (T2), said second duration and the quantity of polluting species being such that they lead to the definition of a second slope (b) of the curve relating to the evolution of the voltage of said cell as a function of time, the ratio of said second slope on said first slope ( b / a) being strictly greater than a value of 1, said second phase generating the capture of said polluting species by said catalytic electrode;

a third phase (Ph3) for regenerating the catalytic electrode, comprising feeding a third stream comprising hydrogen with a constant current density, said third stream being less than a nominal fuel flow, for an initial duration leading to at a value of the stoichiometric coefficient of said fuel less than 1, so as to increase the oxidation potential of said catalytic electrode and to reach the oxidation potential of said polluting species, to eliminate it.

2. A method for increasing the lifetime of a proton exchange membrane fuel cell according to claim 1, characterized in that the second fuel stream comprises a gas type H ₂ S or CS ₂ or S0 ₂ or COS .

3. A method for increasing the lifetime of a proton exchange membrane fuel cell according to one of claims 1 or 2, characterized in that the first stream consists of reformed hydrogen.

4. A method for increasing the lifetime of a proton exchange membrane fuel cell according to one of claims 1 or 2, characterized in that the first fuel stream comprises pure hydrogen.

Method for increasing the lifetime of a proton exchange membrane fuel cell according to one of the preceding claims, characterized in that the first fuel flow comprises a first quantity of H ₂ S, the second flow of fuel comprising a second quantity of H ₂ S, said second quantity being greater than said first quantity.

6. Process for increasing the lifetime of a proton exchange membrane fuel cell according to one of the preceding claims, characterized in that the third phase comprises the application of a power pulse on said cell in operation. potentiostatic by an external charge.

7. A method for increasing the lifetime of a proton exchange membrane fuel cell according to one of the preceding claims, characterized in that the third phase comprises a voltammetric cycle on said catalytic electrode in order to reach the potential. oxidizing said polluting species.

8. Process for increasing the lifetime of a proton exchange membrane fuel cell according to one of the preceding claims, characterized in that the third phase further comprises the following succession of steps:

increasing said third fuel flow corresponding to a second stoichiometric value for a secondary duration to accelerate the fuel flow to the catalytic sites of the anode;

the return of said third fuel flow to said nominal operating value for the reaction of the fuel with the catalytic electrode.

9. A method for increasing the lifetime of a proton exchange membrane fuel cell according to claim 8, characterized in that the initial duration and the secondary duration are of the order of a few minutes.

Method for increasing the lifetime of a proton exchange membrane fuel cell according to one of Claims 1 to 9, characterized in that the duration of the first phase is greater than or equal to the duration of the second phase. phase.

1 1. A method for increasing the lifetime of a proton exchange membrane fuel cell according to one of the preceding claims, characterized in that the second stream comprises an amount of H ₂ S of from about a few ppb to a few ppm, the second duration of the second phase being typically between 30 minutes and 50 hours depending on the concentration of H ₂ S, this duration being inversely proportional to the concentration of polluting species.

12. A method for increasing the lifetime of a proton exchange membrane fuel cell according to one of claims 1 to 1 1, characterized in that it comprises a cycle of steps comprising at least the first, the second and third phases.

13. A method for increasing the lifetime of a proton exchange membrane fuel cell according to one of claims 1 to 12, characterized in that it comprises the following steps:

the first phase under feed in first fuel flow;

the acquisition of the voltage of the electrochemical cell;

the second phase under supply of a second fuel stream comprising a polluting species during a second period enabling a voltage threshold value of said electrochemical cell to be reached, without prior determination of said second duration and the quantity of polluting species in said second fuel stream;

the regeneration phase of the catalytic electrode.

14. Process for increasing the lifetime of a proton exchange membrane fuel cell according to one of claims 1 to 13, characterized in that the third flow is identical to the first flow.