Catalytic electrode for fuel cell or electrolysis cell and method for manufacturing same

Abstract

The invention relates to a method for preparing an array (13) of vertically aligned carbon nanotubes (14) for use in a catalytic electrode of a fuel cell or an electrolysis cell, comprising the following steps: providing an array (13) of vertically aligned carbon nanotubes (14) obtained by a gas phase 5 growth process, in which a precursor of a carbon nanotube growth catalyst is continuously added to a feed gas, -depositing a plurality of platinum nanodots (15) onto the outer surface of the vertically aligned carbon nanotubes by using a vapor deposition process, such as ALD. The ALD process advantageously uses Pt (PF3) 4 as the platinum source gas. The 10 nanometer dots can be protected through nanocaging.

Description

Technical Field

[0001] The present invention relates to the field of electrochemistry, more precisely to the subfield of electrodes for electrochemical processes in aqueous systems. The present invention relates to novel catalytic electrodes containing platinum nanoparticles, which can be used in fuel cells operating in aqueous systems, in particular for fuel cells using molecular oxygen and hydrogen. They can also be used in electrolysis cells operating with aqueous systems, in particular for electrolysis cells for decomposing water into molecular oxygen and hydrogen. The present invention also relates to a novel method for manufacturing such catalytic electrodes using vertically aligned carbon nanotubes (VACNT) and a deposition process for forming platinum nanoparticles on the VACNT. The catalytic electrodes can be integrated into polymer membranes. Background Art

[0002] Catalytic electrodes for fuel cells that combine gaseous oxygen and gaseous hydrogen into water have long been known, at least since the 1950s, when early spacecraft programs required an autonomous and reliable source of electrical energy. Such electrodes for fuel cells, comprising platinum nanoparticles dispersed on a porous carbon material, were the subject of several early patents (see, for example, US 3,231,428 disclosing Pt-Ir nanoparticles), and were included in the seminal textbook Modern Electrochemistry published by J. O'M Bockris and AKN Reddy in 1973 (see Volume 2, pages 1382-1395). Today, these electrodes work quite well, but are expensive because they contain a lot of platinum: their electrochemical mass activity (expressed in Watts / gram of platinum) is quite low. In other words: only a small portion of the platinum deposited on the electrode is actually active. Obviously, it would be ideal to put in less platinum to get the same catalytic effect.

[0003] Many different approaches have been taken to improve the efficiency of platinum utilization in catalytic electrodes. Vertically aligned carbon nanotubes (VACNTs) are candidate materials for making membranes for proton exchange membrane fuel cells (PEMFCs) because their specific structure enhances gas diffusivity, drainage and efficient utilization of platinum. However, they are difficult to process because VANCTs cannot be grown on polymer substrates (such as membranes for PEMFCs) and therefore cannot be used directly on their growth substrates. In arrays formed by vertically aligned carbon nanotubes (VACNTs), most of the nanotubes are parallel, and more precisely, the main direction of the nanotubes (i.e., their long axis) is approximately perpendicular to the substrate on which they are grown; a certain degree of deviation from perpendicularity may occur and is included in the term "vertically" aligned carbon nanotubes.

[0004] The paper "Vertically aligned carbon nanotube electrodes for high current density operating proton exchange membrane fuel cells" by S. Murata et al., published in Journal of Power Sources 253 (2014), pp. 104-113, proposes a multi-step method for producing such a catalytic membrane. In the method, platinum particles are produced on VACNTs (grown on a stainless steel substrate coated with an iron catalyst and an inactive oxide) by infiltration with a platinum salt solution followed by reduction, and then the VACNTs are transferred from the steel substrate to the membrane by hot pressing after being fixed to the membrane using an ionomer impregnation process; the native steel substrate is then peeled off. This process allows the VACNTs to be coated with small platinum particles (about 2nm to 2.5nm in size) that are well distributed on the surface and do not form agglomerates. However, this is a rather tedious and complicated process. Murata's paper emphasizes that the vertical alignment of the carbon nanotubes is important for manufacturing, but has nothing to do with the performance of the resulting platinum-loaded catalytic electrode, as the hot pressing step will destroy the VACNT alignment.

[0005] Shen et al. (“Pt coated Vertically Aligned Carbon Nanotubes as Electrodes for Proton Exchange Membrane Fuel Cells”, Procedia Engineering 93 (2014), pp. 34-42) reported a method of depositing dense platinum films onto VACNTs by sputtering. However, the thickness of the film can reach several hundred nanometers, which will block access to very small pores, resulting in a loss of specific surface area. This technology does not seem to solve the problem of efficient utilization of platinum in catalytic membranes.

[0006] WO 2022 / 047351, assigned to Air Liquide, discloses a method for forming platinum nanodots on a substrate by using a continuous gas phase reaction process (called atomic layer deposition (ALD) or a pulsed chemical vapor deposition process (pulsed-CVD)) using a specific platinum precursor molecule Pt(PF3)4. Various carbon materials can be used as a substrate. These nanodots allow platinum utilization efficiencies of up to 75%.

[0007] The present invention aims to provide a catalytic electrode for a fuel cell having an improved mass activity and possibly also an improved efficiency, thereby reducing the platinum loading of the catalytic electrode and ultimately reducing its cost.

[0008] The present invention also aims to provide a simple method to obtain a proton exchange membrane for a fuel cell using VACNT as a porous material carrying finely distributed platinum particles, exhibiting improved mass activity and enhanced catalyst utilization, thereby reducing the overall catalyst loading. Summary of the invention

[0009] The inventors have recognized that in prior art catalytic electrodes for fuel cells, a large portion of the platinum particles deposited on the substrate are not catalytically active. According to the present invention, this problem is solved by a novel catalytic electrode comprising platinum dots deposited on vertically aligned carbon nanotubes (VACNTs).

[0010] VACNT arrays have large external surfaces (typically more than 100m 2 / g), which are easily accessible to chemicals in the gas phase (such as gaseous precursors for ALD or pulsed CVD), while limiting their migration between nanotubes forming the VACNT array. The inventors surprisingly found that certain structural characteristics of the VACNT array and the VACNT have a significant impact on the uniformity of platinum nanodots deposited by ALD or pulsed CVD, especially the uniformity of those nanodots that can be obtained by the above-mentioned vapor deposition process (ALD or pulsed CVD) using Pt(PF3)4 in the presence of an oxidant or a reductant.

[0011] According to one feature of the invention, the method according to the invention uses a VACNT array that has been grown from the gas phase in a reaction chamber from which a carbon source gas and a catalyst precursor enter simultaneously. This is a one-step growth process in which a precursor of the growth catalyst (usually droplets of ferrocene dissolved in toluene) is continuously added to the gas phase of the growth atmosphere (usually containing acetylene as a carbon source) used as a carbon nanotube growth method.

[0012] This continuous addition of the growth catalyst to the feed gas can be achieved by periodically injecting an aerosol containing said catalyst into the VACNT growth reactor, a method which itself is known from WO 2004 / 000727 (assigned to Commissariat à l'Aquila). Atomique and aux This periodic injection is usually done by supplying pulses at very short time intervals using an injection pump similar to that used in diesel engines; since this pulsed injection has no measurable consumption of the catalyst in the gas phase between two subsequent pulses, it is also called "quasi-continuous" injection.

[0013] Compared to the VACNT array obtained in a two-step process in which a catalyst is deposited onto a growth substrate and then carbon nanotubes are grown from the vapor phase, the VACNT array obtained by the above-mentioned one-step process exhibits at least two different features, namely, the presence of catalyst particles in the center of the carbon nanotubes and the presence of many defects on the outer surface of the carbon nanotubes, which can promote the anchoring and growth of platinum nanodots during subsequent processing steps.

[0014] The inventors have found that using the VACNT arrays obtained by this one-step process for the deposition process of platinum nanodots produces a catalytic film with improved properties and higher electrochemical mass activity of platinum. On the other hand, the two-step VACNT process growth process (wherein in the first step the growth catalyst is deposited onto a substrate, followed by the carbon nanotube growth process itself) does not result in a VACNT array constructed of carbon nanotubes having graphene planes (layers) that are tilted (i.e., form an angle) relative to the main direction of the nanotubes.

[0015] Therefore, a first object of the present invention is a method for preparing an array of vertically aligned carbon nanotubes for use in a catalytic electrode of a fuel cell or an electrolytic cell, comprising the steps of:

[0016] - providing an array of vertically aligned carbon nanotubes obtained by a gas phase growth process, wherein a precursor of a carbon nanotube growth catalyst is continuously added to a feed gas,

[0017] - depositing a plurality of platinum nanodots onto the outer surface of the VACNT by using a first vapor deposition process.

[0018] In an advantageous embodiment, the platinum nanodots are preferably deposited by chemical vapor deposition (CVD) techniques, preferably by atomic layer deposition (ALD) or pulsed CVD, in the presence of an oxidant or a reductant, using a specific gaseous precursor (i.e., Pt(PF3)4). The oxidant may be oxygen radicals or H2O, O2, O3, and mixtures thereof. The reductant may be molecular hydrogen. The terms ALD and pulsed CVD will be used interchangeably in the present invention, both referring to a two-step continuous vapor deposition process, in which two separate precursor reactive species are introduced in each step, and each step is separated by a gas purge.

[0019] According to yet another aspect of the present invention, the deposition process of platinum nanodots is carried out at a temperature below 300° C., preferably below 275° C., more preferably between 25° C. and 275° C., and even more preferably between 50° C. and 250° C. Heating the surface of VACNTs to a temperature exceeding about 300° C. in the presence of platinum can result in thermal degradation of carbon nanotubes.

[0020] The first vapor deposition process advantageously comprises a sequence of alternating cycles, each cycle comprising an exposure time and a purge time, and wherein during the exposure time, the array is exposed to a Pt(PF3)4 gas and a reactive gas, preferably selected from the group formed by: H2, H2O, O2, O3, NO2, oxygen radicals and mixtures thereof, NH3, SiH4, Si2H6, Si3H8, SiH2Me2, SiH2Et2, N(SiH3)3, SiH2(NEt2)2, other Si-H containing reactants, hydrogen radicals, hydrazine, methylhydrazine, amines, NO, N2O, borane, B2H6, CH4, C2H6, CH3I and mixtures thereof. The duration of the exposure time and / or such purge time of each of the alternating cycles is advantageously between 0.1 seconds and 60 minutes, preferably between 1 second and 1000 seconds, and more preferably between 10 seconds and 100 seconds. The number of said sequences is advantageously between 2 and 100, preferably between 5 and 35, more preferably between 8 and 30, and most preferably between 10 and 25.

[0021] According to an advantageous feature of the invention, the average diameter of the platinum nanodots is between 0.7 nm and 5 nm, preferably between 1 nm and 5 nm, and more preferably between 1 nm and 4 nm. Since only the surface of the platinum nanodots is catalytically active, smaller particles will allow a larger surface area to volume ratio: the smaller the particles, the higher the specific activity (activity per gram of platinum). However, smaller particles are more susceptible to degradation over time, in particular due to Ostwald ripening. This effect is particularly evident below 2 nm. Above 5 nm, the specific activity is significantly reduced. Particles smaller than 2 nm can be used, but then they are preferably protected by nanocaging, as will be explained below.

[0022] In an advantageous embodiment of the invention, said first vapor deposition process is performed such that the total platinum loading of said array of arrays of vertically aligned carbon nanotubes is higher than 10 wt.-%, preferably higher than 20 wt.-%, more preferably higher than 30 wt.-%, and most preferably higher than 40 wt.-%.

[0023] According to another feature of the present invention, the volume mass of the VACNT array is higher than 0.10 g / cm 3 , preferably higher than 0.15 g / cm 3 , more preferably higher than 0.20 g / cm 3 , and still more preferably at least 0.25 g / cm 3 (regardless of the substrate on which the VACNT array has been deposited). Preferably no more than 0.70 g / cm 3 and more preferably not more than 0.5 g / cm 3When the bulk mass of the VACNT array is too high, the nanotubes may form bundles with low access to most of the tube surface; the deposition of platinum nanodots will be uneven. When the bulk mass is too low, the efficiency of precursor usage in the ALD or pulsed CVD process appears to decrease.

[0024] According to yet another aspect of the invention, the stability of platinum nanodots can be enhanced by encapsulating them in nanocages made of inorganic oxides deposited by a regioselective vapor deposition process. According to this advantageous embodiment, the method of the invention comprises a further step, wherein the VACNT array comprising platinum nanodots is first treated with a regioselective blocking agent (e.g., a surfactant), which is applied to the platinum nanodots but not to the nanotube surface and the substrate of the VACNT, and wherein the inorganic porous oxide is then deposited by using a second vapor deposition process. The vapor deposition process is typically an ALD process, wherein the inorganic oxide is deposited by ALD onto the carbon nanotubes but not onto the platinum nanodots.

[0025] Finally, in a further step, the blocking agent is removed. The selective blocking agent is selected so that it can selectively adhere to the platinum dots and deposit only in the surface of the platinum nanodots; it is also selected so that it does not contaminate the ALD reactor during the deposition of the inorganic oxide. Surfactants can be used as blocking agents, such as amines, thiols or acids. In particular, the selective blocking agent is advantageously selected from alkylamines, allylamines (such as oleylamine), alkylthiols or carboxylic acids (such as oleic acid).

[0026] The second vapor deposition process is advantageously performed such that the average diameter of the platinum nanodots is between 0.7 nm and 5 nm, preferably between 1 nm and 4 nm, and more preferably between 1 nm and 3 nm.

[0027] The second vapor deposition process advantageously comprises a sequence of alternating cycles, each cycle comprising an exposure time and a purge time, and wherein during the exposure time, the array is exposed to an organometallic precursor of a metal element and a reactive gas to form an inorganic oxide of the metal element, the metal element preferably being selected from the group formed by: zirconium, niobium, tantalum, vanadium, tungsten, molybdenum, titanium, hafnium, cobalt, nickel, yttrium, cerium, lanthanum, rare earths or other elements of the lanthanide series.

[0028] Another object of the present invention is an array of vertically aligned carbon nanotubes for use in a catalytic electrode of a fuel cell or electrolytic cell, comprising a plurality of platinum nanodots on the outer surface of said nanotubes, characterized in that said carbon nanotubes have graphite planes oriented perpendicularly with respect to the main direction of the nanotubes. The volume mass of said array of vertically aligned carbon nanotubes is advantageously between 0.10 g / cm 3 Up to 0.45g / cm3 and preferably between 0.15 g / cm 3 Up to 0.30g / cm 3 The average diameter of the platinum nanodots is preferably between 0.7 nm and 5 nm, preferably between 1 nm and 5 nm, more preferably between 1 nm and 4 nm, and most preferably between 1 nm and 3 nm. The plurality of platinum nanodots advantageously comprises face-centered cubic platinum crystals.

[0029] Another subject of the invention is the use of an array according to the invention for the manufacture of a catalytic electrode for a fuel cell or an electrolysis cell.

[0030] Still another object of the present invention is a polymer film comprising an array of vertically aligned carbon nanotubes comprising a plurality of platinum nanodots on the outer surface of said nanotubes, characterized in that said carbon nanotubes have graphitic planes oriented perpendicularly with respect to the main direction of said nanotubes. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Figures 20 to 21 illustrate various aspects of the invention; some of these figures relate to the prior art for comparison.

[0032] Figure 1 A cross section of a catalytic membrane according to the prior art is schematically shown, wherein catalytic particles are dispersed on a granular porous carbon support.

[0033] Figure 2 (including Figure 2a and Figure 2b ) schematically shows a cross-section of a catalytic membrane according to the present invention, wherein catalytic particles are dispersed on an array of vertically aligned carbon nanotubes. Figure 2a and Figure 2b Different embodiments of such catalytic membranes are shown.

[0034] Figure 3 (including Figure 3a and Figure 3b ) shows that before depositing platinum nanodots ( Figure 3a ) and after ( Figure 3b ) Electron microscope image of carbon nanotubes grown in a one-step process. The arrows identify the iron-rich phase, which is the remnant of catalyst particles trapped in the hollow carbon nanotubes.

[0035] Figure 4 (including Figure 4a and Figure 4b ) shows a transmission electron microscope image of carbon nanotubes grown in a method using ferrocene as a catalyst precursor and acetylene as a carbon source. Figure 4a A two-step growth process is represented and shows perfect nanotubes with aligned graphene planes. Figure 4bRepresents a one-step growth process showing misaligned graphene planes.

[0036] Figure 5 (including Figure 5a and Figure 5b ) is a schematic diagram illustrating what can be seen in FIG4 : a nanotube with aligned graphene walls ( Figure 5a ) and defective nanotubes ( Figure 5b ), which shows the graphene planes oriented almost perpendicularly with respect to the major axis of the nanotube, and more typically forming an angle of between about 30° and about 90° with said major axis.

[0037] Figure 6 The Pt(PF3)4 used in the example of the first vapor deposition process is shown. 19 F NMR spectrum.

[0038] Figure 7 (including Figure 7a and Figure 7b ) relates to Example 1 and shows the platinum loading on the VACNT powder ( Figure 7a ) and the utilization efficiency of Pt(PF3)4 precursor ( Figure 7b ).

[0039] Figure 8 (including Figure 8a and Figure 8b ) also relates to Example 1 and shows the utilization efficiency of the Pt(PF3)4 precursor on a conventional porous carbon powder (No. KB) ( Figure 8a ) and platinum loading ( Figure 8b ).

[0040] Fig. 9 Also related to Example 1 and showing TGA analysis of 18 cycle Pt ALD sample (three replicates, curve (b)) compared to pristine VACNT (curve (a)). Solid line refers to residual mass (in wt%), dashed line refers to DTA.

[0041] Fig.10 It also relates to Example 1 and shows XPS analysis of Pt ALD samples at different numbers of cycles.

[0042] Fig.11 Also relates to Example 1 and shows BF-TEM images and nanodot size distribution as a function of the number of Pt(PF3)4 / H2 cycles during platinum nanodot ALD deposition. Data for 2, 4, 8, and 18 cycles are shown.

[0043] Figure 12 (including Fig.12a , Figure 12b and Fig.12c) relates to Example 4 and a VACNT array grown in a one-step growth process. It shows the platinum nanodots deposited after four ALD cycles using the ALD process as described in Example 2 at two different magnifications ( Fig.12a )and( Figure 12b ) under transmission electron microscope. The size distribution of platinum nanodots is given in the histogram ( Fig.12c ).

[0044] Figure 13 (including Fig.13a , Fig.13b and Fig.13c ) also relates to Example 4 and gives the same type of information as FIG. 12 after eighteen ALD cycles.

[0045] Figure 14 (including Fig.14a , Fig.14b and Fig.14c ) relates to Example 4 and a VACNT array grown in a two-step growth process. The same type of information as in FIG. 12 is shown after four ALD cycles.

[0046] Figure 15 (including Fig.15a , Fig.15b and Fig.15c ) also relates to Example 4 and gives the same type of information as FIG. 14 after eighteen ALD cycles.

[0047] Figure 16 (including Fig.16a and Fig.16b ) relates to Example 5 and shows the surface activity of several catalyst samples ( Fig.16a ) and mass activity ( Fig.16b ), which is the current measured at a certain potential (0.90 V and 0.95 V) divided by the mass or surface area of ​​platinum.

[0048] Fig.17 It relates to Example 6 and shows the mass activities of different catalyst samples measured at up to five different potentials (0.6V, 0.7V, 0.8V, 0.9V, 0.95V).

[0049] Fig.18 Relates to Example 7 and shows TGA measurements obtained on commercial multi-walled carbon nanotube powder after four ALD cycles (curve (b)), eight ALD cycles (curve (c)) and eighteen ALD cycles (d) using Pt(PF3)4 / H2. Curve (a) relates to the original powder.

[0050] Figure 19 (including Fig.19a and Fig.19b ) relates to Example 7 and shows the utilization efficiency of the Pt(PF3)4 precursor on conventional carbon nanotube powder ( Fig.19a) and platinum loading ( Fig.19b ).

[0051] Figure 20 (including Fig.20a and Fig.20b ) relates to Example 7 and shows a TEM-BF image of the same area of ​​the sample after four ALD cycles using Pt(PF3)4 / H2 ( Fig.20a ) and STEM-HAADF images ( Fig.20b ). DETAILED DESCRIPTION

[0052] Unless otherwise stated, all percentage values ​​are by weight.

[0053] The present invention aims to provide an improved catalytic VANCT array, which can be used to manufacture catalytic electrodes, especially catalytic electrodes used in fuel cells and electrolytic cells. Such electrodes generally include a polymer membrane, which includes a porous carbon material layer, and a catalyst is formed on the porous carbon material layer. Figure 1 A catalytic membrane 1 according to the prior art is schematically shown, which comprises a porous carbon layer 3 comprising porous carbon particles 4 which have been deposited on a carbon substrate 2. Platinum particles 5, 6 are deposited on the porous carbon layer and into the open pores of the carbon layer 3, and a suitable polymer sheet 7 (which is typically an ionomer) is deposited on the carbon layer. The platinum particles 6 buried inside the porous network of the carbon particles may become easily deactivated, while the platinum particles 5 on the surface of the porous carbon particles have no risk of being buried due to structural modification of the reaction products or the catalyst support.

[0054] Figure 2 Two different embodiments of a catalytic membrane 10 using a VACNT array 13 as the porous carbon material are schematically shown. The VACNT array is deposited on a carbon substrate 12 and is formed of individual, parallel carbon nanotubes 14 having approximately the same length. Figure 1 By comparison, it can be seen that the use of VACNT array 13 gas molecules (in Figure 1 2) into the ionomer membranes 2, 12 much more easily because their diffusion paths form a straight line, whereas the diffusion paths through the porous carbon particles 4 can be quite tortuous. In a catalytic membrane for a fuel cell, the gaseous molecules may be molecular hydrogen and oxygen that recombine into water. All platinum particles 6 are formed on the outside of a tube that has no pores except for the space inside the tube. The ionomer sheet 17 covers the platinum nanodots 15. It should be noted that Figure 2a and Figure 2b The difference between the illustrated embodiments resides in the presence of a protective layer 18 which protects the platinum article from oxidation; this protective layer 18 may be formed by niobium oxide particles.

[0055] The inventors have surprisingly found that there are significant differences between catalytic electrodes based on VACNT arrays grown in a one-step process and catalytic electrodes based on VACNT arrays grown in a two-step process when the same amount of platinum is deposited using the same platinum deposition process according to the present invention. In particular, the inventors have found that VACNT arrays obtained in a two-step process (wherein in a first step, a growth catalyst is deposited onto a substrate followed by the VACNT growth process itself as a second step) do not improve the efficiency of platinum in the catalytic membrane or catalytic electrode.

[0056] Here, we will briefly review the principle of the one-step VACNT growth method for preparing carbon nanotubes as originally described in WO 2004 / 000727, as already mentioned above. The method is carried out by pyrolyzing a carbon source gas (i.e., acetylene) in the presence of a catalyst precursor (e.g., ferrocene) in a reaction chamber at high temperature. A liquid containing a liquid hydrocarbon precursor of at least one carbon and a metal compound precursor of at least one catalyst metal is formed into finely divided liquid particles (e.g., droplets) under pressure by a periodic injection system, and the finely divided particles (e.g., droplets) formed in this way are transported by a carrier gas stream containing a gaseous carbon source; and introduced into the reaction chamber, where deposition and growth of carbon nanotubes occur. Ferrocene, toluene, and acetylene can be used as solvents for catalyst precursors, liquid hydrocarbon precursors, and catalyst precursors, as well as gaseous carbon sources, respectively. The method produces an array of vertically aligned carbon nanotubes, which are multi-walled nanotubes.

[0057] The inventors observed two differences between VACNT arrays grown in a one-step process and VACNT arrays grown in a two-step process: the presence or absence of catalyst particles within the carbon nanotubes, and the appearance of defects on the surface of the carbon nanotubes. Although catalyst particles are a relatively simple means of distinguishing the two types of carbon nanotubes, their presence is not directly related to the differences between the two types of carbon nanotubes observed in terms of platinum nanodot deposition. On the other hand, the presence of surface defects in carbon nanotubes has been shown to be related to TiO2 nanoparticle deposition by atomic layer deposition (see L.Acauane et al., "Influence of Different Defects in Vertically Aligned Carbon Nanotubes on TiO2 Nanoparticle Formation trough Atomic Layer Deposition", ACS Applied Materials & Interfaces, vol. 8 (25), p. 16444-50; 2016). However, these surface defects are very difficult to detect and identify.

[0058] Figure 3 An electron microscope image of carbon nanotubes grown in a one-step process using ferrocene as catalyst precursor and acetylene as carbon source is shown. Carbon-containing iron particles were found to be located in the center of the carbon nanotubes; their chemical nature has been identified by energy dispersive spectroscopy performed in an electron microscope. Such iron-rich particles are almost never found in carbon nanotubes grown in a two-step process. A possible scientific explanation for this is that in a two-step process, in the first step, the catalyst is deposited onto a growth substrate, “the carbon source gas attaches to the catalyst surface and decomposes into carbon atoms, which then dissolve and diffuse into the catalyst. When the carbon atoms reach supersaturation in the catalyst, they separate from the top of the catalyst and form VACNTs on the top, while the catalyst particles attach to the substrate; this is also known as the “bottom growth mode” (cited from the right column of page 1595 of the publication “Recent progress in the synthesis and applications of vertically aligned carbon nanotube materials” by S. Huang et al., Nanotechnology Review, 2021; 20: 1592-1623). On the other hand, the one-step process in which the iron catalyst floats and has no interaction with the substrate follows the so-called “top growth mode,” in which “carbon atoms separate from the connection between the catalyst and the substrate to form a VACNT structure. That is, the catalyst particles are lifted and sealed by the catalyst particles during VACNT growth. Generally speaking, when the catalyst has a weak interaction with the substrate, the VACNT undergoes a “top growth mode” (cited from Huang, supra).

[0059] Figure 3a shows carbon nanotubes grown in a one-step process prior to deposition of platinum particles, while Figure 3b Similar carbon nanotubes grown in a one-step gas phase process are shown after platinum particles have been deposited using the method of the present invention. The platinum particles are visible as small black dots, while two large black dots with diameters on the order of 10 nanometers represent carbon-containing iron particles, the growth catalyst.

[0060] Figure 4 Shown is a transmission electron microscope image of carbon nanotubes grown in a one-step process using ferrocene as a catalyst precursor and acetylene as a carbon source. Figure 4a A two-step growth process is represented and shows perfect nanotubes with aligned graphene planes. Figure 4bFigure 1 shows a one-step growth process and demonstrates misaligned graphite planes: It can be seen that some of the graphite planes form angles typically between about 30° and about 90° with the major axis of the nanotube; some of the graphite planes are oriented almost perpendicularly to the major axis of the nanotube. This structural defect leads to specific surface defects. Figure 5 Schematic diagrams of these two structures are given: Figure 5a Schematic representation of perfectly aligned graphene planes, while Figure 5b Schematic representation of a defective carbon nanotube in which structural defects lead to specific surface defects. The inventors believe (but do not wish to be bound by this theory) that the presence of graphite planes that form angles typically between about 30° and about 90° with the major axis of the nanotube (and can even be oriented almost perpendicularly with respect to the major axis of the nanotube) indicates other more subtle surface defects distributed over the entire outer surface of the carbon nanotube, which explains why platinum nanodots can be deposited much more uniformly on such carbon nanotubes than on defect-free nanotubes grown by a two-step process.

[0061] We will now describe the platinum nanodot deposition process and the differences between the results obtained after depositing platinum nanodots using a VACNT array grown using a one-step growth process and the results obtained after depositing platinum nanodots using a VACNT array grown using a two-step process.

[0062] The method of the present invention includes the step of depositing a plurality of platinum nanodots onto the outer surface of the VACNT by using a first vapor deposition process. The first vapor process uses a specific gaseous platinum precursor, which is preferably Pt(PF3)4. For the VACNT array grown in the two-step process, the first vapor deposition process results in a distribution of platinum nanodots (represented by the coverage of the carrier surface) showing poor uniformity, and in particular, many agglomerates and uncovered areas are found. For the VACNT array grown in the one-step process, good uniformity of the distribution of platinum nanodots is found; in particular, no agglomerates and large uncovered areas are found. When Pt(PF3)4 is used for the first vapor deposition process, the plurality of platinum nanodots include face-centered cubic platinum crystals.

[0063] In this first vapor deposition process, the efficiency of platinum utilization is similar on VACNT arrays grown in a two-step process and on VACNT arrays grown in a one-step process: in a specific embodiment of the platinum deposition process according to the present invention, approximately 18 deposition cycles are required to achieve a 40 wt.-% platinum loading.

[0064] However, it was found that at high platinum loadings (40 wt.-% platinum) using the vapor deposition process of the present invention, the distribution of platinum nanodot diameters on the VACNT arrays grown in the two-step process was very non-uniform; in particular, many platinum particles with diameters greater than 10 nm were found, which is undesirable for using such VACNT arrays in catalytic electrodes. For the platinum nanodots on the VACNT arrays grown in the one-step process at the same 40 wt.-% platinum loading, an average size of 3 nm was found with a narrow size distribution. These results relate to a volumetric mass of 0.08 g / cm for the VACNTs grown in the two-step process. 3 , and the volume mass of VACNT grown in a one-step process is about 0.25 g / cm 3 ; The volume mass values ​​reported in this specification refer only to the mass of the nanotubes and do not include the growth substrate.

[0065] Regarding the bulk mass, according to an advantageous embodiment of the present invention, the bulk mass of the VACNT array is higher than 0.15 g / cm 3 , preferably higher than 0.20 g / cm 3 , and still more preferably at least 0.25 g / cm 3 . Preferably not more than 0.70g / cm 3 The value is preferably not more than 0.5 g / cm 3 Value; at 0.25g / cm 3 Up to 0.3g / cm 3 In an advantageous embodiment, for VACNTs grown in a one-step process, the volume mass of the VACNTs is about 0.25 g / cm 3 .

[0066] The inventors have also found that there are significant differences between catalytic electrodes using a platinum nanodot vapor deposition process with Pt(PF3)4 according to the present invention compared to the same platinum nanodot deposition process using a different platinum precursor, even when the underlying VACNT array is identical, and especially when the underlying VACNT array has been grown in a one-step process. In addition, significant differences were found when commercial carbon supports were used. These differences can be described in different ways.

[0067] It was found that the platinum utilization efficiency during the first vapor deposition on a commercial carbon support (reference Ketjen Black "KBEC300J") using the platinum nanodot vapor deposition process with Pt(PF3)4 according to the present invention was low, which will be described in more detail below; in 8 deposition cycles, 13.2 wt.-% platinum deposition on KBEC300J (0.35 g) was achieved, compared to 24.4 wt.-% deposition achieved on a VACNT array grown in a one-step process, indicating that the nucleation and growth behavior of Pt(PF3)4 on KBEC300J was much lower. Similarly, on yet another different commercial carbon support reference ("Vulcan XC72", 1 g), 48 cycles were required to achieve a loading of 31 wt.-%, compared to 18 cycles required to achieve a loading of 38 wt.-% on a VACNT array grown in a one-step process (0.35 g). The coverage was uneven, with many agglomerates and uncovered areas. The use of different precursor gases for platinum deposition, in particular Pt(MeCp)Me3, does not allow reaching 40 wt.-% platinum loading on VACNT arrays grown in a one-step process. Pt(MeCp)Me3 requires an O2-based ALD process at high temperature (250°C), which induces thermal degradation of the VACNTs during the Pt nanodot vapor deposition process.

[0068] We will now describe the so-called first vapor deposition process in more detail. In an advantageous embodiment of the method according to the invention, platinum nanodots are deposited on the VACNT array by sequentially injecting Pt precursors and co-reactants in a pulsed CVD process (also called ALD, atomic layer deposition) in a reactor vessel provided with a vacuum device and a heating device. The VACNT array is first dried under an inert gas (e.g. nitrogen) stream, preferably at a temperature above 150°C for a duration of at least 1 h. After drying, the Pt precursor and the reactant (respectively Pt(PF3)4 and H2 gas) are introduced alternately into the container in a sequence having a duration between tens of seconds and hundreds of seconds, the two subsequent sequences being separated by a purge using an inert gas. These sequences can be repeated several times (usually several tens of times) until the desired platinum loading is reached. The entire vapor deposition process is carried out at a temperature below 300°C, preferably between 25°C and 275°C, and even more preferably between 50°C and 250°C.

[0069] In a favorable embodiment, each Pt(PF3)4 pulse has a duration between about 40 seconds and about 800 seconds, preferably between about 100 seconds and about 400 seconds; each H2 pulse has a duration between about 50 seconds and about 1000 seconds, preferably about 100 seconds and about 750 seconds. They are separated by an inert gas purge of about 200 seconds to about 1000 seconds. As an example, each sequence consists of a Pt(PF3)4 pulse of about 150 seconds to 250 seconds, a nitrogen purge of 500 seconds to 700 seconds, and a H2 pulse of about 500 seconds to about 700 seconds. The sequence is typically repeated 10 to 15 times until the desired platinum loading is reached.

[0070] In the first vapor deposition process, which is an ALD process for platinum nanodot deposition, the following co-reactants may be used as a substitute for H gas: H2O, O2, O3, NO2, oxygen radicals and mixtures thereof, NH3, SiH4, Si2H6, Si3H8, SiH2Me2, SiH2Et2, N(SiH3)3, SiH2(NEt2)2, other Si-H containing reactants, hydrogen radicals, hydrazine, methylhydrazine, borane, amines, NO, N2O, and mixtures thereof.

[0071] According to a basic feature of the present invention, platinum nanodots are deposited by ALD using Pt(PF3)4 on a VACNT array that is grown in a one-step process using a floating catalyst generated from a precursor that is continuously or quasi-continuously added to a carbon source gas. The inventors believe (but do not wish to be bound by this theory) that Pt(PF3)4 can preferentially dissociate at local defects on the surface of the carbon nanotubes, and that these local defects are more frequent on VACNT arrays prepared by a one-step process than on VACNT arrays prepared by a two-step process. Although these local defects have not yet been identified, this initial explanation is reasonable when considering the inventors' following findings.

[0072] It is known that Pt(PF3)4 is a highly symmetrical molecule with a radius of gyration of about 3.04 angstroms. The PF3 entity is very stable compared to the Pt-P bond: in the Pt(PF3)4 molecule, the PF bond energy is 488 kJ / mol, while the Pt-P bond energy is 83 kJ / mol. In the Pt(PF3)3 molecule, the corresponding bond energies are 483 kJ / mol and 95 kJ / mol. Calculations performed by the inventors show that the adsorption of Pt(PF3)3 on graphene is more favorable than the adsorption of Pt(PF3)4. Although the complete dissociation of PF3 from Pt(PF3)4 is energetically unfavorable at 80.6 kJ / mol, the formation of the more stable intermediate with a Pt-P bond length of 3.56 angstroms only requires an activation energy of 59.9 kJ / mol (equivalent to 24 kT at 298 K). Since the adsorption of Pt(PF3)3 on the graphene surface is more favorable than that of Pt(PF3)4, this may shift the equilibrium towards the dissociation direction of Pt(PF3)4.

[0073] This so-called “dissociative physisorption” mechanism can also be qualitatively understood by the geometrical rearrangement of the molecule upon detachment of one PF3 unit: in Pt(PF3)4, the coordination of Pt is tetrahedral, whereas in Pt(PF3)3, it is coplanar, which provides a shorter C-Pt bond length.

[0074] As shown in calculations performed by the inventors, the adsorption energy of Pt(PF3)3 on graphene is higher (72 kJ / mol) and the Pt-C bond length is shorter (3.46 angstroms) compared to Pt(PF3)4 (52 kJ / mol and 4.78 angstroms, respectively). In addition, the energy difference between Pt(PF3)3 on top of the carbon atom and Pt(PF3)3 on top of the ring center is lower than the value of kT, which allows Pt(PF3)3 species to migrate easily on the surface. This can at least qualitatively explain that these Pt(PF3)3 species are able to diffuse on the graphene surface until they find local defects where the energetics are most favorable for the bonding of the molecule and the subsequent dissociation to release atomic platinum. This theory is supported by the following findings: For Pt(PF3)4 adsorption, the C-Pt bond length is 4.8 Å on a perfect graphene plane, while it is 4.6 Å on a 5-7 type ring defect, 4.1 Å on a pendant CH3 group, 3.9 Å on a pendant CH2 group, and 4.2 Å on a vertical edge of a carbon nanotube. Therefore, Pt(PF3)4 is adsorbed on the substrate in the form of Pt(PF3)3–-C+PF3(g). Pt(PF3)3 molecules are relatively mobile on the carbon surface and migrate to one of the hypothesized defects during the purge time of the pulsed CVD or ALD sequence. Pt(PF3)3 is more strongly bound at the defects, so once fixed, the Pt(PF3)3 molecules do not migrate and allow the nucleation of platinum nanodots.

[0075] It is known that in catalytic applications, the size of platinum nanodots tends to increase, thereby reducing the useful life of the platinum nanodots in the catalytic electrode. This can be prevented by applying a thin porous protective inorganic coating to all parts of the substrate between the platinum nanodots; this process, known as "nanocaging" of the platinum dots, is known, as will be explained in detail below. This thin porous inorganic coating can be deposited by a vapor deposition process (e.g. ALD). This vapor deposition process for preparing the porous inorganic coating is referred to herein as the "second" vapor deposition process.

[0076] We will now describe the so-called second vapor deposition process in more detail. Preferred porous inorganic coatings for forming metal oxide nanocages are selected from (but not limited to) the following list: niobium oxide, tantalum oxide, vanadium oxide, tungsten oxide, molybdenum oxide, titanium oxide, zirconium oxide, hafnium oxide, cobalt oxide, nickel oxide, yttrium oxide, cerium oxide, lanthanum oxide and other elements from the rare earth / lanthanum series. Niobium (V) oxide is a particularly preferred oxide for the inorganic coating.

[0077] Examples of metal oxide precursors for the second vapor deposition process include, but are not limited to:

[0078] Nb(=NtBu)(nEt2)3、Nb(=NtBu)(nMe2)3、Nb(=NtBu)Cp(nEt2)2、Nb(=NtBu)Cp(nMe2)2、Nb(=NbtBu)(MeCp)(nEt2)2、Nb(=NtBu)(MeCp)(nMe2)2 tBu)(OtBu)2(nEt2)、Nb(=NtBu)(OtBu)2(nEt2)、Ta(=NtBu)(nEt2)3、Ta(=NtBu)(nMe2)3、Ta(=NtBu)Cp(nEt2)2、Ta(=NtBu)Cp(nMe2)2、Ta(=NtBu)(nMe2)3 eCp)(nEt2)2、Ta(=NtBu)(MeCp)(nMe2)2、Ta(=NtBu)(OtBu)2(nEt2)、Ta(=NtBu)(OtBu)2(nEt2)、Zr(nEt2)4、Zr(nMe2)4、ZrCp(nMe2)3 rMeCp(nMe2)3、Zr(EtCp)(nMe2)3、Zr(nPrCp)(nMe2)3、Zr(tBuCp)(nMe2)3、Zr(nBuCp)(nMe2)3、Zr(iBuCp)(nMe2)3、Hf(nMe2)4、Hf(nMe2)4、Hf(nMe2)4 e2)3、HfMeCp(nMe2)3、Hf(EtCp)(nMe2)3、Hf(nPrCp)(nMe2)3、Hf(tBuCp)(nMe2)3、Hf(nBuCp)(nMe2)3、Hf(iBuCp)(nMe2)3、Ti(nMe2)4、Ti(nEt2)4 iCp(nMe2)3、Ti(Me5Cp)(nMe2)3、Ti(Me5Cp)(oMe)3、Ti(Me5Cp)(OiPr)3、Ti(OiPr)4、V(nMe2)4、V(nEtMe)4、W(CO)6、W(MeCp)(CO)3、W(Me3Cd)(C) O)3、W(=NtBu)2(nMe2)2、W(=NtBu)2(nHtBu)2、W(=NtBu)2(OtBu)2、WH2(iPrCp)2、Mo(CO)6、Mo(MeCp)(CO)3(NO)、Mo(Me3CHD)(CO)3、Mo(=NtBu)2(nMe2) )2、Mo(=NtBu)2(nHtBu)2、Mo(=NtBu)2(OtBu)2、MoH2(iPrCp)2、CoCp2、Co(MeCp)2、Co(EtCp)2、CoCp(CO)2、Co(iPr2-amd)2、Co(CO)3(NO), CoCp(CO)2Co(iPr2-amd)2, Co(CO)3(NO), NiCp2, Ni(MeCp)2, Ni(EtCp)2, Ni(iPr2-amd)2, Ni(tBu2-amd)2, Ni(a llyl)(iPr2-amd), Ni(PF3)4, Y(MeCp)3, Y(EtCp)3, Y(iPrCp)3, Y(nBuCp)3, Y(EtCp)(iPr2-amd)2, Y( MeCp)(iPr2-amd)2, Ce(iPrCp)3, Ce(nBuCp)3, Ce(EtCp)(iPr2-amd)2, Ce(iPrCp)(iPr2-amd)2, La(iPrCp)3, La(nBuCp)3, La(MeCp)(iPr2-amd)2, La(EtCp)(iPr2-amd)2, La(iPrCp)(iPr2-amd)2 or mixtures thereof.

[0079] It is desirable not to coat the platinum nanodots themselves, as this may reduce their catalytic activity, but only to coat the carbon nanotube substrate between adjacent platinum nanodots. This can be achieved by protecting the platinum nanodots with a capping agent before depositing the inorganic coating. Zirconium dioxide is deposited onto platinum nanodots protected by a capping agent by ALD, which is known in the publication (N. Cheng et al., "Extremely Stable Platinum Nanoparticles Encapsulated in a Zirconia Nanocage by Area-Selective Atomic Layer Deposition for the Oxygen Reduction Reaction", Advanced Materials, Vol. 27, No. 2, January 14, 2015, pp. 277-281) (so-called "nanocage").

[0080] The capping agent may be a surfactant, a thiol, an acid or an amine, such as oleylamine ((9Z)octadec-9-en-1-amine, CAS No.: 112-90-3). A detailed description of the nanodot cage formation by niobium oxide is given below in conjunction with Example 3; similar methods may also be used for other oxides. Examples of capping agents include, but are not limited to, C1-C 30 Alkylamines (e.g. oleylamine, dodecylamine, hexadecylamine, octadecylamine, C1-C 30 Allylamine), C1-C 30 Alkyl mercaptans (eg 1-octanethiol, dodecanethiol, hexadecylthiol, octanethiol, octadecylthiol) and acids (eg linoleic acid, capric acid, lauric acid, oleic acid, stearic acid, tetradecylphosphonic acid) and mixtures thereof.

[0081] The method of the invention results in a uniform distribution of platinum dots on the surface of the VACNT array; in particular, the size distribution is narrow and uniform. The average size of the platinum nanodots is preferably between 2 nm and 3 nm. The method of the invention uses a platinum deposition process that avoids heating the surface of the VACNT to temperatures exceeding about 300° C.; temperatures above about 300° C. may degrade the nanotubes to a degree that is no longer acceptable. The method of the invention allows the manufacture of catalytic membranes in which the vertical alignment of the carbon nanotubes is maintained after the VACNTs have been transferred from the native metal substrate to the membrane.

[0082] The catalytic membrane comprises a polymer layer deposited on a fully fabricated VACNT array, with platinum nanodots and possible nanocages. The polymer layer can be an ionomer layer, such as Nafion.

[0083] Example

[0084] Example 1 : Platinum deposition on VACNT powder and carbon powder

[0085] Platinum nanodots were deposited on VACNT powder (provided by Nawa Technologies) and KET Jenblack EC300J (KB, provided by Lion Specialty Chemicals Co., Ltd.) using a homemade fluidized reactor with a vibration motor by sequentially injecting Pt precursors and co-reactants (pulsed CVD process). VACNT or KB powder was loaded into a glass tube reactor. Stainless steel filters were installed at the bottom and top of the reactor. The reactor was evacuated and N2 gas was passed from the bottom of the reactor to form a fluidized bed. In order to dry the powder before deposition, the reactor was heated to up to 200°C in a N2 stream for more than 2 hours. After drying, the reactor was maintained at 150°C, while the gas line was maintained at 40°C to prevent precursor condensation. The Pt precursor and reactant were Pt(PF3)4 and H2 gas, respectively, and N2 was used as a carrier gas and purge gas. By 19 Pt(PF3)4 with a purity of 99% as confirmed by F-NMR was synthesized by Air Liquide. In the Pt deposition process by pulsed CVD, each sequence consisted of a 200-second Pt(PF3)4 pulse, a 600-second N2 purge, a 500-second H2 pulse, and a 600-second N2 purge; these sequences were repeated 2, 4, 8, or 18 times (18 times for VACNT only).

[0086] After a given number of ALD pulses, a sample of the powdered VACNT array according to the invention was subjected to thermogravimetric analysis using a Mettler Toledo apparatus as follows: A sample of about 5 mg of powder was placed in a ceramic crucible. A heating ramp was started from 25°C up to 950°C in a nitrogen-oxygen atmosphere (80%, 20% respectively) with an increase of 10 degrees per minute. This allowed the determination of the mass loss when the carbon was burned off and, ultimately, the residual platinum content of the powder. The results are shown below.

[0087] Figure 6 The Pt(PF3)4 used was measured on a Magritek Spinsolve 60 NMR spectrometer. 19 FNMR spectroscopy; the pure liquid product was placed in a glass NMR tube.

[0088] Figure 7 Pt loading (wt.-%) on VACNT powders is shown ( Figure 7a ) and the utilization efficiency of Pt(PF3)4 precursor ( Figure 7b ) and Pt(PF3)4 / H2 cycles. Pt loading was measured using thermogravimetric analysis (TGA). The loadings were 8.1wt.-% (weight percent), 14.5wt.-%, 24.4wt.-% and 37.1wt.-% for 2, 4, 8 and 18 cycles, respectively. The utilization efficiency of the Pt(PF3)4 precursor is very high, especially for the 2 and 4 cycle tests. Under non-optimized conditions, increasing the number of cycles will reduce the efficiency.

[0089] Figure 8 The corresponding curves for the above KB powder are shown. The loadings are 3.9 wt.-%, 7.5 wt.-% and 13.2 wt.-% by 2, 4 and 8 cycles respectively, which is significantly lower than that deposited on VACNT powder, although the deposition conditions were not changed. The utilization efficiency of Pt(PF3)4 is significantly reduced on KB powder compared to VACNT powder.

[0090] Fig. 9 TGA analysis of an 18 cycle Pt ALD sample (three replicates, curve (b)) compared to pristine VACNT (curve (a)) is shown. The solid line refers to the residual mass (in wt%), the dashed line refers to DTA.

[0091] Fig.10 X-ray spectroscopy (abbreviated as XPS) analysis of Pt ALD samples at different cycle numbers are shown: (a) 18 cycles; (b) 8 cycles; (c) 4 cycles; (d) 2 cycles; (e) original.

[0092] The particle size distribution of Pt nanodots was studied by transmission electron microscopy (TEM). Fig.11 The BF-TEM images and the distribution of nanodot size as a function of Pt(PF3)4 / H2 cycles are shown. (The abbreviation BF stands for pointed field mode). As the number of cycles increases, the particle size also increases, but even through an 18-cycle process, the size can be mainly controlled to be between 1 nm and 3 nm. In addition, the Pt nanodots are well distributed on the VACNTs, and no large aggregation of the Pt nanodots is observed even through an increased number of cycles.

[0093] Example 2 :Deposition of Pt on VACNT / Al sheets

[0094] Pt nanodots were deposited on VACNT / Al sheets (provided by Nawa Technologies) by a pulsed CVD process using a homemade reactor. The VACNT / Al sheets were loaded into a glass tube reactor. In order to dry the VACNT powder before deposition, the reactor was heated to 200°C for more than 2 hours. After drying, the reactor was maintained at 150°C, while the gas line was maintained at 40°C to prevent precursor condensation. The Pt precursor and reactant were Pt(PF3)4 and H2 gas, respectively, and N2 was used as a carrier gas and purge gas. Pt(PF3)4 with a purity of 99% was synthesized by Air Liquide. In the Pt deposition process, a sequence consisting of a 200-second Pt(PF3)4 pulse, a 600-second N2 purge, a 500-second H2 pulse, and a 600-second N2 purge was repeated 18 times.

[0095] To investigate whether Pt is distributed in the VACNT / Al sheet from the surface to the Al substrate, the samples were cut and exposed cross-sectional images were obtained by SEM-EDS analysis (Hitachi SU9000). (The abbreviation EDS refers to energy dispersive spectroscopy). The as-deposited and post-deposition annealed samples (5% H2 atmosphere) were analyzed using EDS on the thickness of the cross-sectional images; these line scans showed that Pt was well distributed from the VACNT surface to the bottom, indicating that Pt(PF3)4 could be a good Pt precursor for roll-to-roll processing on VACNT / Al foil. Impurities (such as F) remain on the samples as deposited, but the levels of these impurities are dramatically reduced by temperature or other chemical treatments after deposition.

[0096] Example 3 :Pt nanodot cage

[0097] In the first sequence of steps (producing sample S3-R210929-3-V50), platinum nanodots were deposited on VACNT / Al sheets (provided by Nawa Technologies) by ALD using a homemade reactor by sequential injection of Pt precursor and co-reactant (pulsed CVD process). The VACNT / Al sheets were loaded into a glass tube reactor. In order to dry the VACNT / Al samples before deposition, the reactor was heated to up to 200°C for more than 2 hours. After drying, the reactor was maintained at 150°C, while the gas line was maintained at 40°C to prevent precursor condensation. The Pt precursor and reactant were Pt(PF3)4 and H2 gas, respectively, and N2 was used as carrier gas and purge gas. Pt(PF3)4 with a purity of 99% was synthesized by Air Liquide Group. In the Pt deposition process, a sequence consisting of a 200-second Pt(PF3)4 pulse, a 600-second N2 purge, a 500-second H2 pulse, and a 600-second N2 purge was repeated 4 times.

[0098] In the second sequence of steps, the area-selective deposition of Nb2O5 was performed on the product obtained in the first sequence of steps by ALD as follows: After depositing Pt on the VACNT / Al sheet using Pt(PF3)4 and H2, the sample was immersed in 5 mM oleylamine in ethanol for 5 hours. After the oleylamine treatment, the sample was rinsed 3 times with ethanol and then dried using a vacuum dryer. Nb2O5 was deposited by ALD in a homemade ALD reactor at 250°C using a niobium amide imide metal organic precursor (Air Liquide Group) and water as the Nb source and co-reactant, respectively (hereinafter referred to as "Nb precursor"). In the deposition process, a sequence consisting of a 30-second Nb precursor pulse, a 300-second N2 purge, a 30-second water pulse, and a 300-second N2 purge was repeated 3 times. After deposition, the oleylamine was removed by heating at 200°C in air.

[0099] The localization of Pt particles on the carbon support and the distribution of Nb2O5 were characterized using scanning transmission electron microscopy and energy dispersive X-ray spectroscopy under ambient conditions. It was observed that Nb2O5 deposition occurred mainly on the carbon support, and most of the Pt nanodots were exposed even after 3 cycles of Nb2O5 ALD.

[0100] In situ environmental TEM was used to examine the stability of 3-cycle Nb2O5 ALD samples on Pt / C catalyst. To monitor the morphological changes of the samples, the ALD samples and the original Pt / C catalyst were first heated at 200°C in vacuum and then heated at 350°C with air injection. The pressure of the sample holder was maintained at 4Pa. For the 3-cycle ALD samples, the STEM (scanning transmission electron microscope) images showed that the Pt nanodots had less morphological changes. However, the images of the original samples showed that the Pt nanodots were severely aggregated, and the size of the Pt nanodots became larger after the in situ environmental TEM test. In addition, the SEM images showed that the carbon surface was severely damaged and holes were generated on the surface of the carbon, indicating that the catalytic reaction between carbon and oxygen occurred on the original Pt / C catalyst.

[0101] Example 4 : Comparison between VACNT arrays obtained by one-step process and by two-step process

[0102] Fig.12a , Figure 12b , Fig.13a and Fig.13b Transmission electron microscopy images are shown showing platinum nanodots that have been deposited onto a VANCT array obtained by one-step pyrolysis using the ALD process as described in Example 2, wherein the catalyst precursor (ferrocene dissolved in toluene) was periodically injected into the VANCT growth reactor; VACNT growth is based on the use of acetylene as a carbon source. The corresponding histogram of the platinum particle size distribution is also shown (see Fig.12c and Fig.13c ); the horizontal axis is scaled in nanometers. Fig.12a , Figure 12b and Fig.12c refers to four ALD cycles as described in Example 2, and Fig.13a , Fig.13b and Fig.13c refers to 18 such ALD cycles.

[0103] Fig.14a , Fig.14b , Fig.15a and Fig.15b Transmission electron microscopy images are shown showing platinum nanodots that have been deposited using the ALD process as described in Example 2 onto a VACNT array obtained by a two-step process, wherein in a first step a VACNT growth catalyst has been deposited onto a VACNT growth substrate and in a second step a VACNT array has been grown using acetylene as a carbon source. The corresponding histogram of the platinum particle size distribution is also shown (see Fig.14c and Fig.15c ); the horizontal axis is scaled in nanometers. Fig.14a , Fig.14b and Fig.14c refers to four ALD cycles as described in Example 2, and Fig.15a , Fig.15b and Fig.15c refers to 18 such ALD cycles.

[0104] The comparison between Figures 12 and 13 on the one hand, and Figures 14 and 15 on the other hand, is striking because they are based on the same ALD process for depositing platinum nanodots, the only difference being the preparation process used to prepare the VACNT arrays. It can be immediately seen that the platinum ALD process results in much coarser platinum particles on the VACNT arrays obtained in the two-step growth process than in the one-step growth process.

[0105] Example 5 : Comparison of mass activity and surface activity of platinum nanodot catalysts deposited on VACNT arrays obtained by a one-step process according to the present invention and platinum nanodot catalysts deposited on carbon powder according to the prior art.

[0106] The electrochemical tests were performed on seven samples, namely three samples according to the present invention with four, eight and eighteen platinum ALD cycles, two commercial platinum catalysts on carbon powder (numbered Pt_C Vulcan 29% and Pt_C_Vulcan 47.2%, the percentage values ​​refer to platinum loading) and two platinum nanodot catalysts prepared by ALD using a Pt(PF3)4 precursor on commercial carbon powder (Pt_XC72_ALD 150C and Pt_XC72_ALD 300C, where XC72 is the number of the powder and the four commercial catalyst samples, and 150C and 300C are the temperatures of ALD deposition, respectively).

[0107] Electrochemical tests were performed in a conventional rotating disk electrode configuration. Ink was prepared using catalyst powder (for this purpose, the VACNT array was reduced to powder); the ink was coated on the tip of a rotating electrode in an electrochemical cell using 0.1 M perchloric acid (HClO4) electrolyte. A potentiostat (Biologic VPM3) was used.

[0108] Fig.16 Results are summarized for two different potentials, namely 0.90 V and 0.95 V, which is known to be very close to the electrochemical conditions experienced by the catalyst in an oxygen-hydrogen fuel cell. Fig.16a The surface activity (SA) is shown in [μA / cm 2 ] is expressed as the current measured at a certain potential divided by the active surface area of ​​the Pt nanodots. Fig.16b The mass activity (MA for short) is shown, expressed in [A / mg] as the current measured at a certain potential divided by the mass of platinum. It can be seen that at a potential of 0.90 V, the activity of the catalyst according to the invention (18 cycles of platinum deposition by ALD) is close to the activity of the catalyst according to the prior art.

[0109] Example 6 : Comparison of mass activity and surface activity of platinum nanodot catalysts deposited on VACNT arrays obtained by a one-step process according to the present invention and platinum nanodot catalysts deposited on carbon powder according to the prior art as gas diffusion electrodes.

[0110] The commercial gas diffusion electrode H23C8 was cut into suitable sizes and the VACNT according to the present invention was transferred to the gas diffusion electrode by using a laboratory calender (TOBJs-300) with a maximum pressure of 500 N / m. The numbered samples have been realized by coating the gas diffusion electrode with an ink prepared using catalyst powder. The ink was then dried at 70°C for five minutes. The electrode was then installed in an electrolytic cell and an electrolyte with 1M perchloric acid (HClO4) was added. The mass activity was determined at different potentials using different catalysts. Fig.17 The results for different catalysts are shown in FIG. 1 , namely two catalysts according to the invention (numbered Pt_VACNT_150C-Nafion and Pt_VACNT_150C-ss Nafion) and three prior art catalysts (numbered Pt-20%_XC72, Pt-30%_XC72 and Pt-47%_TEC).

[0111] Pt_VACNT_150C-ss Nafion corresponds to the as-prepared VACNT electrode. The activity seems lower than the reference catalyst. This is expected, since VACNTs are highly hydrophobic and therefore the oxygen reduction reaction can only occur at the gas-liquid interface, i.e. on top of the VACNT layer. A small amount of Nafion was added to the VACNT electrode (Pt_VACNT_150C-Nafion) by impregnation with droplets of Nafion solution and subsequent drying. This allows for a better utilization of the volume of the electrode by enlarging the gas-liquid interface within the VACNT layer. As expected, the activity of the catalyst rises significantly.

[0112] Example 7 : Platinum deposition on commercial carbon nanotube powders

[0113] Provided by Nanocyl under the trade name NC7000 TMA powder consisting of multi-walled carbon nanotubes is sold. The powder is obtained from multi-walled carbon nanotubes produced by a catalytic carbon vapor deposition process. According to the measurements obtained on the transmission electron microscope (TEM) images, the average diameter of the powder particles is about 9.5nm and the average length is about 1.5μm. According to TGA, the carbon purity is about 90% and the metal oxide content is about 10% (derived from the aluminum growth support, including the iron catalyst). The specific surface area determined by the BET (Brunauer-Emmett-Teller) method is 250m 2 / g to 300m 2 / g. Aluminum contamination of the powder can also be detected by XPS (no such contamination was found on the VACNT samples provided by Nawa Technologies). No amorphous carbon was detected by high-resolution transmission electron microscopy (HRTEM).

[0114] Fig.18 The results of TGA measurements of pristine carbon nanotube powder (curve (a)) and nanotube powder subjected to four (curve (b)), eight (curve (c)) or eighteen (curve (d)) cycles of platinum deposition by ALD with a Pt(PF3)4 precursor are shown, corresponding to platinum loadings of 0 wt.-%, 6.3 wt.-%, 11.9 wt.-% and 18.6 wt.-%, respectively. TGA measurements (synthetic air, temperature ramp of 10°C / min) were performed in pure nitrogen on samples dried at 200°C; a temperature ramp of 10°C / min was used. The residual masses at the end of the TGA tests were 10.2 wt.-%, 16.5 wt.-%, 22.1 wt.-% and 28.8 wt.-%, respectively.

[0115] Fig.19a and Fig.19b The utilization efficiency (expressed as a percentage) and the residual mass (expressed as a weight percentage) of four, eight, and eighteen ALD cycles are shown, respectively. Figure 7a and Figure 7b In contrast, when the commercial carbon nanotube powder is used, the utilization efficiency and residual mass of platinum are much smaller.

[0116] It is possible to deposit platinum on this commercial carbon nanotube powder by ALD using a Pt(PF3)4 precursor, as shown in TEM-BF (TEM in bright field mode) and STEM-HADF (scanning TEM in high angle annular dark field mode) images after four ALD cycles using Pt(PF3)4 / H2 (TEM-BF see Fig.20a , STEM-HAADF see Fig.20b), small nanodots are observed with an uneven redistribution. Some areas show a sparse presence of such nanodots, while agglomerates are observed on other areas. The two images of FIG. 20 show the same area of ​​the sample; they can be superimposed.

[0117] Together with the lower precursor utilization efficiency observed by TGA, lower precursor nucleation was observed on commercial nanotube powders compared to VACNTs.

Claims

1. A method for preparing an array of vertically aligned carbon nanotubes for use in a catalytic electrode of a fuel cell or electrolytic cell, comprising the steps of: - providing an array of vertically aligned carbon nanotubes obtained by a gas phase growth process, wherein a precursor of a carbon nanotube growth catalyst is continuously added to a feed gas, - depositing a plurality of platinum nanodots onto the outer surface of the vertically aligned carbon nanotubes by using a first vapor deposition process. 2 . The method of claim 1 , wherein the first vapor deposition process is atomic layer deposition, chemical vapor deposition, or pulsed chemical vapor deposition.

3. The method according to claim 1 or 2, wherein the first vapor deposition process is performed at a temperature below 300°C, preferably between 25°C and 275°C, even more preferably between 50°C and 250°C.

4. The method according to claim 2 or 3, wherein the first vapor deposition process comprises a sequence of alternating cycles, each cycle comprising an exposure time and a purge time, and wherein during the exposure time, the array is exposed to Pt(PF3)4 gas and a reactive gas, preferably selected from the group formed by: H2, H2O, O2, O3, NO2, oxygen radicals and mixtures thereof, NH3, SiH4, Si2H6, Si3H8, SiH2Me2, SiH2Et2, N(SiH3)3, SiH2(NEt2)2, other Si-H containing reactants, hydrogen radicals, hydrazine, methylhydrazine, amines, NO, N2O, borane, B2H6, CH4, C2H6, CH3I and mixtures thereof.

5. The method according to claim 4, wherein the duration of the exposure time and / or such purge time of each of the alternating cycles is between 0.1 seconds and 60 minutes, preferably between 1 second and 1000 seconds, and more preferably between 10 seconds and 100 seconds.

6. The method according to claim 4 or 5, wherein the number of the sequences is between 2 and 100, preferably between 5 and 35, more preferably between 8 and 30, and most preferably between 10 and 25.

7. The method according to any one of claims 1 to 6, wherein the volume mass of the array of vertically aligned carbon nanotubes is higher than 0.10 g / cm 3 , preferably higher than 0.15 g / cm 3 , more preferably higher than 0.20 g / cm 3 , and still more preferably higher than 0.30 g / cm 3 , regardless of the substrate on which the VACNT array has been deposited.

8. The method according to any one of claims 1 to 7, wherein the volume mass of the array of vertically aligned carbon nanotubes does not exceed 0.70 g / cm 3 , and preferably not more than 0.50 g / cm 3 , regardless of the substrate on which the VACNT array has been deposited.

9. The method according to any one of claims 1 to 8, wherein the volume mass of the array of vertically aligned carbon nanotubes is between 0.10 g / cm 3 Up to 0.45g / cm 3 and preferably between 0.15 g / cm 3 Up to 0.30g / cm 3 In between, the substrate on which the VACNT array has been deposited is not considered.

10. The method according to any one of claims 1 to 9, wherein the total platinum loading of the array of arrays of vertically aligned carbon nanotubes is higher than 10 wt.-%, preferably higher than 20 wt.-%, more preferably higher than 30 wt.-%, and most preferably higher than 40 wt.-%.

11. A method according to any one of claims 1 to 10, comprising the further step of first treating the array with a surfactant capable of selectively adhering to platinum dots, and then depositing the inorganic oxide by using a second vapour deposition process.

12. A process according to claim 11, wherein the surfactant is an alkylamine, an allylamine, such as oleylamine, an alkylthiol or a carboxylic acid, such as oleic acid.

13. A method according to claim 11 or 12, wherein the second vapor deposition process comprises a sequence of alternating cycles, each cycle comprising an exposure time and a purge time, and wherein during the exposure time, the array is exposed to an organometallic precursor of a metal element and a reactive gas to form an inorganic oxide of the metal element, the metal element preferably being selected from the group formed by: zirconium, niobium, tantalum, vanadium, tungsten, molybdenum, titanium, hafnium, cobalt, nickel, yttrium, cerium, lanthanum, rare earth or other elements of the lanthanide series.

14. An array of vertically aligned carbon nanotubes for use in a catalytic electrode of a fuel cell or electrolytic cell, comprising a plurality of platinum nanodots on the outer surface of said nanotubes, characterized in that The carbon nanotubes have graphitic planes that are oriented tilted, angled, or even perpendicularly with respect to the main direction of the nanotube.

15. The array of vertically aligned carbon nanotubes of claim 14, wherein the carbon nanotubes have graphitic planes that form an angle of between about 30° and about 90° with a major axis of the nanotubes.

16. The array of vertically aligned carbon nanotubes according to claim 14 or 15, wherein the average diameter of the platinum nanodots is between 0.7 nm and 5 nm, preferably between 1 nm and 5 nm, more preferably between 1 nm and 4 nm, and most preferably between 1 nm and 3 nm.

17. The array of vertically aligned carbon nanotubes according to any one of claims 14 to 16, characterized in that: The plurality of platinum nanodots include face-centered cubic platinum crystals.

18. Use of an array of vertically aligned carbon nanotubes according to any one of claims 14 to 17 in the preparation of a catalytic electrode for a fuel cell or an electrolytic cell.

Images