WO2012028313A1

WO2012028313A1 - Process for producing platinum-transition metal catalyst particles

Info

Publication number: WO2012028313A1
Application number: PCT/EP2011/004388
Authority: WO
Inventors: Viktorija Juhart; Markus Perchthaler
Original assignee: Elcomax Gmbh
Priority date: 2010-09-03
Filing date: 2011-08-31
Publication date: 2012-03-08
Also published as: DE102010044288A1

Abstract

The invention relates to a process for producing platinum-transition metal catalyst particles, characterized by the following steps: A) provision of M_xN_y catalyst particles, where M = Pt (platinum), N is an element of the group of transition metals with the exception of platinum or any mixed crystal or any intermetallic compound of two or more elements of the group of transition metals, x and y are the proportions by weight of the abovementioned catalyst constituents M, N, where 0 < x < 1 and 0 < y < 1 and x + y ≤ 1, B) treatment of the catalyst particles provided as per step A in or with an acid, C) deposition of Pt on the catalyst particles treated as per step B.

Description

Process for producing platinum transition metal -

catalyst particles

The present invention relates to a process for the preparation of platinum-transition metal catalyst particles, which are preferably intended for use in membrane electrode assemblies for fuel cells.

The commercial success of fuel cells is still hampered by the high cost of manufacturing the membrane-electrode assemblies acting as the core component. A major cost factor is the use of platinum as catalyst both on the anode side and on the cathode side. The minimization of the platinum requirement therefore represents a central development approach for future fuel cells. For this purpose, various approaches are pursued, such as e.g. a reduction of the platinum to be deposited by optimized deposition processes or improved electrode structures.

A study of the influence of catalyst loading on the performance of PEM fuel cells has shown that the anode-side platinum loading of the

Membrane electrode assembly can be reduced to as low as about 0.05 mg / cm ² using pure H ₂ as the fuel due to the rapid hydrogen oxidation reaction (H ₂ O) reaction, without significant voltage loss in the To accept (test) fuel cell. On the cathode side, where the oxygen reduction reaction (ORR) takes place, however, it is still necessary to increase the platinum loading by approx. 4 to 10 times to achieve the necessary power density of a sufficiently efficient fuel cell (see HA Gastei- ng). ger, JE Panels, SG Yan: "Dependence of PEM fuel cell performance on catalyst loading", Journal of Power Sources, 127 (2004), 162-171).

Taking into account the platinum resources available worldwide and the current cost of platinum, the platinum content of membrane-electrode units should preferably not exceed 0.15 mg platinum per cm ^{2 of} electrode area for commercially promising applications. A reduction of the cathode-side platinum loading to 0.1 mg / cm ² is therefore desirable, if possible without having to simultaneously accept a significant loss of power density.

A reduction of the platinum loading required for a given level of performance can, if appropriate, also be achieved with catalysts other than pure platinum. Crucial for the amount of required catalyst material is - in addition to its arrangement and

Distribution within the electrode structure of a membrane-electrode assembly - in particular its activity with respect to the cathode-side running oxygen reduction reaction, either in the form of so-called mass activity (i _m ( ₀ , 9v) in the unit A / mg _Pt ) or the so-called specific activity (i _S (o, 9v> in the unit ^ A / cm ² _Pt ) is given.

In the prior art, for the phosphoric acid fuel cell (PAFC) or polymer electrolyte membrane fuel cells (PEMFCs) has already been proposed the use of supported on carbon particles platinum-transition metal catalysts, where they 2- to 4-fold mass activities (or until to 5-fold specific activities) compared to supported (pure) platinum catalysts.

Some examples of relevant prior publications are listed below:

[1] S. Chen, H.A. Gasteiger et. al. : "Lytinum-Alloy Cathode Catalyst Degradation in Proton Exchange Membrane Fuel Cells: Nanometer Scale Compositional and

Morphological Changes, J. Electrochem Soc., 157, A82-A97 (2010)

[2] H.A. Gasteiger, S.S. Kocha, B. Sompalli, and F.T. Wagner, Applied Catalysis B: Environmental 56: 9 (2005)

[3] S. Mukerjee, S. Srinivasan, M.P. Soriaga, and J. McBreen, Journal of the Electrochemical Society

142: 1409 (1995)

[4] T. Toda, H. Igarashi, H. Uchida, and M. Watanabe, Journal of the Electrochemical Society 146: 3750 (1999) [5] UA Paul, A. Wokaun, GG Sherer, TJ

Schmidt, V. Stamenkovic, V. Radmilovic, N. Markovic, and P.N. Ross, J. Phys. Chem B 106: 4181 (2002)

[6] E. Antolini, R.R. Passos, and E.A. Ticianelli, Electro-chimica Acta 48: 263 (2002)

[7] S. Chen, P.J. Ferreira, W. Sheng, N. Yabuuchi, L.F. Allard, and Y. Shao-Horn, J. Am. Chem. Soc.

130: 13818 (2008)

[8] S. Koch, J. Leisch, M.F. Toney, and P. Strasser, J. Phys. Chem C 111: 3744 (2007)

The following explanations are given as reasons for the increased activity of platinum transition metal catalysts over platinum catalysts:

Platinum has a cubic face centered (fcc) crystal lattice with a lattice constant of 3.93 Å. If platinum is alloyed with a metal of lesser lattice constant, a contraction of the crystal lattice occurs. The increase in the electrocatalytic activity of platinum alloys is correlated with the reduction of Pt-Pt interatomic distances.

A second hypothesis is that, due to the alloy, there is an inhibition of the formation of OH _ad (adsorbed OH) and thus more reaction sites are available for the reaction at a constant potential. XA ES data (X-ray adsorption near edge structure) show that the adsorption of OH, which in the case of Pt / C (supported on carbon platinum catalysts) at 0.8 V starts, with corresponding alloying catalysts shifted to higher potentials.

The leaching of the alloy catalysts leads to a roughening of the catalyst surface and thus to an increase in the electrochemically active surface as well as the electrochemical activity as a whole.

Furthermore, electronic effects are mentioned: The alloying of platinum with transition metals shifts the position of the center of the d-band to higher energies. The literature also discussed the formation of so-called "Pt-skin" and "Pt-skeleton" structures, the formation of which leads to increased activities with respect to the ORR through weakening of the Pt-OH _ad bond in the electronically modified structure.

However, the transition metals used in this case (Fe, Mn, Cr, Co, Cu, Ir,...) Did not show sufficient stability in the acidic environment of the aforementioned fuel cells, since the oxidative, acidic conditions segregate the transition metal (eg cobalt) used to the Surface induce and the segregated to the surface transition metal then triggers in an acidic medium. In this context, it should also be noted that released in fuel cell operation

Transition metal ions produce an effective proton concentration gradient in the MEA, which significantly reduces the performance of the MEA. For this reason, it has already been proposed to add supported platinum-transition metal catalysts prior to their use in a fuel cell in an acid

treat to dissolve easily soluble (in particular in the uppermost atomic layers of the surface of the catalyst particle) transition metals from said catalyst particles, which reduces the amount of dissolved in the later fuel cell operation transition metal and thereby taking place efficiency reduction of the fuel cell. Of the

However, the segregation effect described above is still present in such acid-treated platinum-transition metal catalysts and leads to a drastic loss of efficiency of a membrane-electrode unit loaded with such a catalyst, so that the abovementioned platinum-transition metal catalysts are still not suitable for long-term operation have sufficient stability of a fuel cell.

Cyclic voltammetric measurements with rotating disk electrodes show that the mass activity or the specific activity of previously known supported platinum-transition metal catalysts drastically breaks down after only a few cycles, which has the advantage of high specific activity at the beginning (compared to supported platinum catalysts). completely destroyed for practice.

In light of the above-described prior art, the object of the present invention is to to provide a method for producing a platinum transition metal catalyst, which - in addition to an increased activity over a pure platinum catalyst - also characterized by a high long-term stability, especially in the operation of a fuel cell. Furthermore, the invention is directed to the provision of a correspondingly prepared and thus long-term stable platinum transition metal catalyst.

The above object is achieved by a method according to claim 1 and a platinum transition metal catalyst prepared accordingly.

The process according to the invention for the preparation of platinum-transition-metal catalyst particles is characterized by the following steps:

A) providing M _x N _y catalyst particles, wherein

M = Pt (platinum),

N is an element of the group of transition metals other than platinum or any solid solution or any intermetallic compound of two or more elements of the group of transition metals,

x and y are the proportions by weight of the above-mentioned Katalystorbestandeile with

0 <x <1 and 0 <y <1 and x + y ^ 1 B) Treatment of the catalyst particles provided in step A in or with an acid

C) Deposition of Pt on the catalyst particles treated according to step B

In other words, a platinum-N-catalyst of platinum and at least one further transition metal is thus initially provided in step A. If appropriate, the catalyst composition may also contain further catalytically active components or other additives.

The group of transition metals which may be used in the context of the present invention is preferably the group comprising cobalt (Co), nickel (Ni), iron (Fe), chromium (Cr), copper (Cu), titanium (Ti) , Manganese (Mn) and tungsten (W).

Binary (eg Pt-Co) or ternary (eg Pt-Co-Cu) catalyst compositions in which the weight fractions x (of platinum) and y (N) add up to exactly 1 are particularly preferred in the context of the present invention. since these are either already commercially available in the form of supported on carbon catalyst particles (eg Pt-Co / C) or - for example by deposition of suitable catalyst precursors (eg suitable platinum and N precursors) on a suitable support material (eg carbon or a be - already constructed electrode layer for a membrane electrode assembly) - can be produced.

In step B, the catalyst particles provided according to step A are then exposed to an acid, whereby soluble N components are dissolved out of the platinum N catalyst, which is a first stabilization of the catalyst with regard to a long-term use in a fuel cell

Increasing the catalytic activity or the electrochemically active surface result.

According to step C platinum is then deposited on the acid-pretreated platinum-N-catalyst particles, whereby a drastic stabilization effect is achieved for the catalyst particles prepared according to the invention, based on a - almost complete - suppression of that described above

Segregation effect also in fuel cell operation, i. under acidic conditions and / or varying potentials. Incidentally, the particles produced according to the invention show a higher catalytic activity compared with platinum. the oxygen reduction reaction (ORR). Preferably, in step C, a layer of at least two or three atomic layers of platinum is deposited on the catalyst particles.

In the context of the present invention, it has been found that the production process according to the invention The final deposition of platinum on the previously acid-treated platinum-N catalyst particles is not associated with a significant reduction in the activity (= specific activity and / or mass activity) of the catalyst with respect to the ORR, which is particularly not obvious since pure platinum catalysts are one opposite Platinum-transition metal catalysts have significantly decreased activity.

As a result, in the context of the present invention, platinum-N catalyst particles are thus produced which have an activity with respect to the ORR similar to that of commercially available platinum-transition metal catalysts, but at the same time a drastically increased long-term stability under fuel cell conditions.

In a preferred embodiment of the present invention, the catalyst particles provided according to step A are selected in their composition such that the platinum content of the catalyst particles to be further processed according to the invention is less than or equal to 80% by weight (x.fwdarw.0.8), more preferably less than or equal to 60% by weight. (x ^ 0.6). In this way, the platinum loading of a membrane electrode electrode unit equipped with catalyst particles according to the invention can be kept within an acceptable range. Furthermore, in a likewise preferred development of the invention, it can be provided that the catalyst particles provided in accordance with step A are selected in their composition such that x + y = 1, that is to say that the catalyst, if appropriate on carbon particles or the electrode structure of a Membrane electrode unit is deposited, in addition to platinum and the one or more transition metals for use has no further components. Furthermore, cobalt (Co) and / or nickel (Ni) is used in a particularly advantageous manner as the transition metal.

The acid used in step B is preferably HC10 ₄ , H ₃ PO ₄ , HNO ₃ , H ₂ SO ₄ , HCl, HCOOH or CH ₃ COOH. The catalyst particles must preferably be exposed to the acid for only a few minutes for effective discharge of the readily soluble and near-surface N-type component of the catalyst.

If the method of electrochemical deposition is selected in step C, this preferably comprises the following steps:

Cl) introducing the catalyst particles treated in step B into a platinum-containing electrolyte

C2) applying a platinum precursor specific potential so that platinum is reduced on the catalyst particles. The platinum precursor-specific potential, as is customary for an electrochemical deposition process, is typically preferably a sawtooth-like potential profile which causes the reduction of the platinum from the precursor and thereby the deposition of platinum on the catalyst particles.

As far as the catalyst particles are concerned - e.g. is carried on carbon-supported catalyst particles, which are usually introduced in a conventional manner at a later date in the electrode structure for a fuel cell or a paste forming the electrode layer, so these catalyst particles in step Cl simply said

Electrolytes admixed and later, i. after completion of the actual deposition process according to

Step C2, again filtered out of the electrolyte. If, however, the catalyst particles are already deposited on an electrode structure, then the electrode structure can either be immersed completely or only on one side in the platinum-containing electrolyte and then be suitably cleaned of electrolyte residues.

The platinum precursors are, for example, H ₂ PtCl ₆ , Pt (NO ₂ ) 3,

(NH ₄ ) ₂ PtCl ₆ , Na ₂ PtCl ₆ , K ₂ PtCl ₆ , H ₂ Pt (OH) ₆ , PtO ₂ , PtCl ₄ , H ₂ Pt (S0 ₄ ) ₂ or [Pt (NH ₃ ) ₃ N0 ₂ ] N0 ₂ into consideration. The method of electrochemical deposition offers the advantage that the platinum deposited on the catalyst particles from the platinum precursor is ultimately deposited at those points at which the catalyst particles are accessible to the process media during later operation of a fuel cell.

Alternatively, however, the method of chemical deposition is also suitable, which-especially for supported catalyst particles-preferably makes use of the following steps:

Cl) suspending the catalyst particles treated according to step B

C2) mixing this suspension with a platinum precursor in dissolved or solid form

C3) deposition of platinum on the catalyst particles by addition of reducing agent

C4) filtering, washing and drying the catalyst particles obtained.

It is understood that the method of chemical deposition of platinum is also applicable to existing in or on an electrode layer platinum-transition metal catalyst particles, in which case loaded with platinum-transition metal catalyst particles surface of the electrode structure with a suspension having a platinum precursor contained therein is charged and

Subsequently, the deposition of the platinum can be carried out by adding reducing agent. Suitable reducing agents are, for example, solutions containing KBH, NaBH ₄ , LiALH ₄ and / or N ₂ H ₄ . Furthermore, in principle, gaseous reducing agents such as hydrogen, S0 ₂ , CO, CH ₄ and / or NH ₃ can be used.

The final platinum weight fraction of the catalyst particles prepared according to steps A) to C) should preferably be less than or equal to 95% by weight, more preferably less than or equal to 90% by weight, and again preferably less than or equal to 80% by weight of the final catalyst composition.

As already mentioned, the catalyst particles produced according to the invention are preferably so-called "supported" catalyst particles in which the catalyst composition is deposited on suitable carrier particles. Particularly suitable carrier particles are carbon (nano) particles, in particular in the form of (industrial) carbon black, which is e.g. powdered or

can be pulverized. However, so far are also suitable carrier particles of other materials, such as graphite, graphitized carbon black, Ti0 ₂ , Ti ₄ 0 ₇ , Sn0 ₂ , WC (tungsten carbide) or Tic (titanium carbide). Furthermore, these may also be present, for example, in the form of carbon nanotubes, Ti0 nanotubes, carbonized Ti0 ₂ nanotubes, etc. The catalyst particles provided according to step A advantageously have a mean diameter in the nanometer range, for example in the range of 1 nm to 900 nm, thus ensuring a particularly effective distribution of the catalyst within the electrode structure of a membrane electrode assembly for a fuel cell can.

The catalyst particles provided in step A can also already be part of an electrode layer for fuel cells, whereby in the context of the present invention, the advantages of structurally produced electrode or catalyst layers with a platinum-transition metal catalyst can advantageously be exploited at the same time.

Finally, the present invention is directed not only to the process described above but also to platinum N catalyst particles prepared by the process of the invention and thus stabilized. For this purpose, all the aspects, advantages and structural properties already explained above apply, so that in this respect reference may be made to the above statements.

The invention will be explained in more detail with reference to an embodiment of the present invention and the accompanying drawings. In one embodiment of the invention 47.6 mg first was dispersed in H ₂ 0 in an ultrasonic bath of commercially available carbon nanoparticles (Ensaco 350 ^® G carrier particles). This was followed by the addition of a 10-fold excess of NaBH ₄ in alkaline solution. First, a co-precursor aqueous Co (NO ₃ ) ₂ solution was added, followed by an H ₂ PtCl ₆ solution functioning as a Pt precursor. NaBH4 acts as a reducing agent for the chemical deposition of a Pt-Co composition on the carrier particles, wherein the reduction was carried out at room temperature overnight, ie over a period of about 14 hours. The solid was then separated using glass frit and filter paper, washed with isopropanol and water and finally dried at 80 ° C overnight in a drying oven.

As a result, in the sense of step A of the process according to the invention, supported Pt _x Co _y catalyst particles were supported on carbon carrier particles, in which the platinum constituted a weight fraction of x = 0.516 (corresponding to 51.6% by weight). The weight fraction y of the cobalt accordingly corresponded to y = 0.484 (corresponding to 48.4% by weight). The average particle size of the Pt-Co catalyst particles (supported on carbon) was determined by X-ray diffractometry (XRD) and amounts to 5.2 nm (without the carrier particles). These supported catalyst particles were then - in accordance with step B of the process according to the invention - for a period of 1 min. in 1 M HC10 ₄ acid dipped, whereby soluble, in particular in the uppermost atomic layers of the Pt-Co catalyst particles cobalt was dissolved out of the catalyst particles. The result is a Pt-Co catalyst particle with respect to the original composition higher Pt content.

Finally, according to step C of the process according to the invention, the previously acid-treated

Pt-Co / C catalyst particles platinum electrochemically deposited, which as a result supported Pt ₀ , 73Co ₀ , 27 catalyst particles were prepared in which the catalyst to about 73 wt.% Of platinum and about 27 wt.% Of cobalt consists. The metal loading of the supported Pt-Co / C particles prepared by the above-described method was 52.4 wt%, so that the carbon carrier particles accounted for only 47.6 wt% of the total weight.

The Pt-N catalyst particles (here: N = Co) produced according to the invention are particularly preferably characterized in that, because of the final platinum deposition process, at least the outer two to three atomic layers of at least a majority of the particles contain exclusively platinum and no other transition metal. Subsequently, the stability of the platinum transition metal catalyst particles prepared and stabilized as described above was determined by cycloolametry (CV) and compared with commercially available catalyst particles. Of the

The experimental setup consisted of a standard electrochemical cell with a three-electrode arrangement, a potentiostat and a triangular voltage generator. All electrochemical experiments in this work were done in a Teflon cell

which is of particular importance for the experiments in the alkaline medium (see KJJ Mayrhofer, GKH Wiberg, M. Arenz, Journal of the Electrochemical Society, 155, p 1 (2008) and KJJ Mayrhofer, AS Krampton, GKH Wilberg, M. Arenz, Journal of the Electrochemical Society, 155, p. 79 (2008)). In particular, any influence of glass corrosion on the measurement results could be ruled out. The Teflon cell is divided into two chambers. The main chamber containing the working electrode (diameter: 5 mm) and the counter electrode has a connection to the reference electrode chamber and a gas connection. The gas connection allows gases to be introduced into the electrolyte through a porous Teflon plug installed in the bottom of the main chamber. The porous Teflon is surrounded by a small shield plate to prevent gas bubbles from reaching the surface of the working electrode and blocking it. The saturation of the electrolyte is not disturbed by this. In such a Teflon cell cyclic voltammetric stability test were then carried out according to the triangular voltage method, in which a triangular or

sawtooth-shaped potential course was passed through cyclically. In the present case, the potential was varied cyclically linearly between the maximum values 0.05 and 1 V with the feed rate 1 V / s. The investigated samples were simultaneously exposed to simulate the acidic conditions in a fuel cell with 0.1 M HC10 ₄ and it was - at the beginning and at the end of the 30,000 cycles series - each of the specific activity of the respective catalyst particles with respect to the ORR and - to

For comparison purposes - the roughness factor r _f in the unit cm ² / cm t). The test results are shown in Table 1:

Table 1 The first four rows of the table relate to supported catalyst particles that are commercially available.

Pt / C 9100 JM and Tanaka Pt / C are platinum catalysts supported on carbon particles, which are long-term stable but have a significantly lower specific activity of only 0.29 and 0.36 mA / cm ² compared to platinum-transition-metal catalysts. The specific activity of the pure platinum catalyst particles does not change from the initial value shown in Table 1 even after 30,000 cycles of the stability tests carried out.

The commercially available Pt Co / C

Catalyst particles according to lines 3 and 4 of Table 1 have an initial specific activity of 0.79 mA / cm ² , which, however, has dropped to a value of only 0.46 mA / cm ² after 30,000 potential cycles.

On the other hand, the Pt-Co catalyst particles prepared according to the above-described embodiment of the invention have an initial specific activity of 0.69 mA / cm ² (which is already significantly higher than that of the platinum catalysts) even after 30,000 cycles the stability test carried out in acidic conditions has remained the same, which is sufficient for use in fuel cells long-term stability of the inventively produced or stabilized platinum transition metal - catalyst particles could be detected.

Fig. 1 illustrates again schematically the process according to the invention for the preparation of platinum transition metal catalyst particles.

In this case, in the left-hand illustration of FIG. 1, a catalyst particle 1 provided as part of step A of the process according to the invention is shown, which consists of a composition of platinum 2 (dark particles) and cobalt 3 (light-colored particles) Catalyst material 2, 3 located carrier particles for the sake of clarity in Fig. 1 is not shown.

After carrying out process step B, the Pt-Co catalyst particle 4 shown in the middle in FIG. 1 results in a higher Pt content than the original composition.

After carrying out process step C, a catalyst particle shown schematically on the right in FIG. 1 finally results, in which preferably at least the outer two or three atomic layers consist entirely of platinum.

Finally, FIG. 2 shows a TEM image of the supported Pt-Co / C catalyst particles 7 produced according to the illustrated exemplary embodiment. The carbon carrier particles 6 are known together with a multiplicity of Pt-Co catalyst particles 7 carried thereon and produced according to the invention.

Claims

claims

1. A process for the preparation of platinum transition metal catalyst particles characterized by the following steps:

A) providing M _x N _y catalyst particles, wherein

M = Pt (platinum),

x and y are the proportions by weight of the above-mentioned Katalysatorbestandeile M, N are with

0 <x <1 and 0 <y <1 and x + y 1

B) Treatment of the catalyst particles provided in step A in or with an acid

C) Deposition of Pt on the catalyst particles treated according to step B

2. The method according to claim 1,

characterized,

that N comprises one element of the group

Co, Ni, Fe, Cr, Cu, Ti, Mn and W; and or any intermetallic compound of two or more elements of this group. Method according to one of the preceding claims, characterized

that it is in the deposition according to

Step C is an electrochemical or chemical deposition of platinum.

Method according to one of the preceding claims, characterized

in that the catalyst particles provided according to step A are selected in their composition such that x <0.8.

Method according to one of the preceding claims, characterized

in that the catalyst particles provided according to step A are selected in their composition such that x + y = 1.

Method according to one of the preceding claims, characterized

the acid used in step B is HC10 ₄ , H ₃ PO ₄ , HNO ₃ , H ₂ SO ₄ , HCl, HCOOH or CH ₃ COOH.

Method according to claim 3,

characterized,

that the electrochemical deposition according to

Step C) comprises the following steps: Cl) introducing the catalyst particles treated according to step B into an electrolyte containing platinum prepurate

C2) applying a platinum precursor specific potential so that platinum is reduced on the catalyst particles.

Method according to claim 3,

characterized,

the chemical deposition according to step C) comprises the following steps:

Cl) suspending the treated according to step B.

catalyst particles

C2) Mixing of this suspension with a platinum precursor in dissolved or solid form C3) Separation of platinum on the catalyst particles by addition of reducing agent C4) Filtration, washing and drying of the catalyst particles obtained

Method according to one of the preceding claims, characterized

that these are the catalyst particles supported according to step.

Method according to claim 9,

characterized,

the supported catalyst particles provided in step A are nanoparticles with a mean diameter in the nanometer range.

11. The method according to any one of the preceding claims, that the catalyst particles provided in step A are already part of an electrode layer for fuel cells.

12. platinum N catalyst particles prepared by a process of claims 1 to 11.