Materials Today Energy 27 (2022) 101023 Contents lists available at ScienceDirect Materials Today Energy journal homepage: www.journals.elsevier.com/materials-today-energy/ Oxygen reduction reaction and proton exchange membrane fuel cell performance of pulse electrodeposited PteNi and PteNieMo(O) nanoparticles Konrad Eiler a, *, 1, Live Mølmen b, c, **, 1, Lars Fast b, Peter Leisner c, Jordi Sort a, d, Eva Pellicer a, *** a Department de Física, Universitat Autonoma de Barcelona, 08193 Bellaterra, Spain b Department of Electrification and Reliability, RISE Research Institutes of Sweden, Brinellgatan 4, 50462 Borås, Sweden c Department of Materials and Manufacturing, School of Engineering, Jo €nko€ping University, Gjuterigatan 5, 55111 Jo €nko €ping, Sweden d  Catalana de Recerca I Estudis Avançats (ICREA), Pg. Lluís Companys 23, 08010 Barcelona, Spain Institucio a r t i c l e i n f o a b s t r a c t Article history: Proton exchange membrane fuel cells (PEMFCs) are an important alternative to fossil fuels and a com- Received 14 December 2021 plement to batteries for the electrification of vehicles. However, their high cost obstructs commerciali- Received in revised form zation, and the catalyst material, including its synthesis, constitutes one of the major cost components. In 14 April 2022 this work, PteNi and PteNieMo(O) nanoparticles (NPs) of varying composition have been synthesized in Accepted 21 April 2022 a single step by pulse electrodeposition onto a PEMFC's gas diffusion layer. The proposed synthesis route Available online 5 May 2022 combines NP synthesis and their fixation onto the microporous carbon layer in a single step. Both PteNi and PteNieMo(O) catalysts exhibit extremely high mass activities at oxygen reduction reaction (ORR) Keywords: Pulse electrodeposition with very low Pt loadings of around 4 mg/cm2 due to the favorable distribution of NPs in contact with the PEM fuel cell proton exchange membrane. Particle sizes of 40e50 nm and 40e80 nm were obtained for PteNi and Pt Electrocatalysis eNieMo(O) systems, respectively. The highest ORR mass activities were found for Pt67Ni33 and Pt66Ni32 Hydrogen energy eMoOx NPs. The feasibility of a single-step electrodeposition of PteNieMo(O) NPs was successfully Electrosynthesis demonstrated; however, the ternary NPs are of more amorphous nature in contrast to the crystalline, binary PteNi particles, due to the oxidized state of Mo. Nevertheless, despite their heterogeneous nature, the ternary NPs show homogeneous behavior even on a microscopic scale. © 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 1. Introduction components, leading to an enormous research interest in the reduction of Pt content. Among the various strategies towards this The market for proton exchange membrane fuel cells (PEMFCs) goal, the alloying of Pt is one of the most promising and advanced is growing as the EU has highlighted hydrogen as a key factor in the approaches [2]. energy market in their Green Deal plan [1]. One of the ways in Along with the strive for cost reduction, the durability of the which hydrogen shall be used is to power heavy-duty vehicles, electrocatalysts is a major issue to be tackled. The widely used, where large storage capacity and fast refueling are crucial. To in- commercially available Pt/C catalysts show a number of disadvan- crease the competitivity of such fuel cell electric vehicles, the tages with respect to oxygen reduction reaction (ORR) performance production cost of fuel cells needs to be reduced. Platinum-based at the cathode of the PEMFC. Apart from their poor efficiency with electrocatalysts are one of the main cost-determining respect to the Pt loading, their limited durability is a major draw- back for reliable long-term application and sustainability [3]. Pt itself, however, is crucial for ORR especially in acidic environments * Corresponding author. such as the PEMFC. The most active transition metals at ORR and ** Corresponding author. therefore worth considering for alloying with Pt are Cu, Ni, and Co *** Corresponding author. [4]. The most investigated and most promising Pt-based nano- E-mail addresses: konrad.eiler@uab.cat (K. Eiler), live.molmen@ri.se particles (NPs) tackling durability issues are PteCo and PteNi NPs, (L. Mølmen), eva.pellicer@uab.cat (E. Pellicer). 1 These authors contributed equally to this work. often in the form of core-shell NPs with a Pt shell [5,6]. Further https://doi.org/10.1016/j.mtener.2022.101023 2468-6069/© 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/ ). K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 effort to simultaneously improve both ORR activity and durability usually achieved using sodium molybdate [31], may lead to the are sought by investigation of ternary and quaternary systems [7,8]. formation of intermediate Mo oxide species due to partial reduc- Apart from the composition of the electrocatalyst, a high number of tion of Mo(VI) [32]. However, even NPs containing Mo oxide show active sites is targeted to achieve high catalytic efficiency [9]. For improved electrocatalytic properties [8,33]. The coordination of Mo this reason, the incorporation of NPs and even single-metal atom with electronegative elements like oxygen or nitrogen can move catalysts in carbonaceous matrices is currently a subject of intense the d-band center of Mo so that its binding capacity with reaction investigation [10,11]. intermediates (O*, OH*, and OOH*) increases, thus making Mo One of the challenges, and an important factor influencing the atoms moderately active towards ORR [34]. Therefore, the intro- fuel cell performance, is the synthesis route of the catalyst, which duction of oxidized Mo atoms to PteNi should not be, in principle, also adds up to the production cost. In most approaches, the cata- deleterious for the ORR performance. lyst NPs are synthesized on a carbon support, mixed with a binder Considering the above results, it seems that (i) PteNi is both an and ionomer, and sprayed onto the membrane, which leads to NPs adequate candidate for ORR and (ii) the incorporation of a third being deposited in locations where they are not accessible for the alloying element can improve electrocatalytic performance. Yet, the catalytic reactions [12]. Using direct electrodeposition of Pt alloy expected improvement is not always attained. For example, Sorsa NPs onto the gas diffusion layer (GDL) of the PEMFC, the NPs are et al. electrodeposited PteNi from liquid crystalline solution onto deposited on the most active sites of the carbon support because GDL substrates, but they did not observe any significant improve- the metal ions should occupy the locations where the local electric ment of the electrodeposited PteNi over commercial Pt/C [35]. field is highest during the electrodepositiondlocations on the Hence, further investigation on this topic is still to be performed. substrate where the path is short and the local charge density is In this work, pulse electrodeposition is used to deposit PteNi high. Furthermore, this approach combines both synthesis and and PteNieMo(O) particles directly onto the microporous layer of fixation (or distribution) of NPs into a single step, leading to a cost a commercial GDL. The GDL is composed of a woven carbon cloth reduction of the synthesis route [13]. with a microporous, PTFE-coated carbon layer, where the catalysts Fundamental research on the mechanisms of NP formation by shall be applied. For such a complex three-dimensional substrate, electrodeposition has been carried out by Ustarroz et al., including electrodeposition is an especially suitable method [36]. In the the electrodeposition of Pt NPs for potential application at ORR deposition process, all particles are deposited on the most active [14e16]. The electrodeposition of Pt alloy NPs has also been shown sites of the carbon support's surface where they guarantee excel- promising for alkaline ORR and other catalytic reactions [17,18]. lent contact with both the support and the proton exchange A common synthesis route for metallic NPs is by pulse electro- membrane (PEM) in the fuel cell. Although the use of PteNi NPs for deposition. An extensive study on the pulse electrodeposition of Pt ORR, and the electrodeposition of PteNi alloys are rather devel- NPs for ORR was carried out by Huang et al. [19]. A pulse reverse oped, the combination of both, in addition to the direct deposition process was employed by Sriwannaboot et al. for the codeposition onto the GDL, has been reported very scarcely. The PteNi and of PteCo alloy onto carbon cloth substrates [20]. Egetenmeyer et al. PteNieMo(O) NPs with different compositions are chemically and thoroughly investigated the effects of pulse electrodeposition pa- structurally characterized, and ORR in acidic media is investigated. rameters for Pt, PteNi, and PteCo NPs deposited onto GDLs, and Finally, fuel cell performance and durability tests in a PEMFC pro- determined optimum electrodeposition parameters for each alloy totype are carried out after hot-pressing the obtained cathodes [21]. Santiago et al. were able to reduce the rather high Pt NP size with the PEM and a commercial Pt/C electrode as anode. commonly obtained by electrodeposition down to below 10 nm by deposition onto a rotating disk electrode (RDE) [22]. Wang et al. 2. Experimental achieved Pt particle sizes of 3e10 nm via the use of a complexing agent [23]. Remarkably, pulse electrodeposition offers several ad- 2.1. NP synthesis vantages compared with other NP fabrication schemes. First, syn- thesis and anchoring to a substrate such as GDL is achieved in one The synthesis of PteNi and PteNieMo(O) NPs was performed by step. Second, NPs nucleate onto the most active sites of the GDL, as pulse electrodeposition from aqueous solution in a three-electrode aforementioned, as opposed to other post-synthesis approaches for electrochemical cell. An Autolab PGSTAT204 potentiostat/galvano- which the catalyst is distributed on the GDL irrespective of local stat was used with a Pt wire as counter electrode (CE), an Ag|AgCl electrical properties and depth, so that many NPs may be located at (3 M KCl) reference electrode (RE) and a working electrode (WE). inaccessible or unfavorable locations. The main disadvantage of The WE consisted of a 2 cm by 2 cm GDL, supplied by Freudenberg, pulse electrodeposition compared with conventional liquid-phase mounted on a Cu support to ensure electrical connection and a synthesis and high-temperature calcination [24] is that an in- homogeneous charge distribution over the entire area of the GDL crease in the catalyst loading is achieved at the expense of during electrodeposition. The excessive area of the Cu support was increasing particle size, which is often undesirable for catalytic isolated with polyimide tape (Fig. 1). purposes. The aqueous electrolytes were loaded with nickel chloride, so- Interestingly, Liu et al. synthesized a PteNi alloy with low Pt dium hexachloroplatinate, boric acid, and ammonium chloride, content via a carbothermal shock method, which was shown to be based on an earlier study [28]. For the deposition of PteNieMo(O), effective at ORR in acidic media and PEMFC testing [25]. This is in sodium molybdate (as Mo precursor) as well as citric acid as contrast to most ORR electrocatalysts investigated for PEMFC, complexing agent were added (Table 1). All chemicals for electro- which usually rely on high-Pt content alloy NPs. lyte preparation were of analytical grade and had been supplied by Although the synthesis of PteNi NPs for ORR [26,27], as well as Merck. The pH was adjusted to 2.7 by addition of sulfuric acid the electrodeposition of PteNi alloys in general [28,29] is well solution. advanced, Mo-containing Pt alloys have been scarcely investigated. The electrodeposition was carried out in stagnant conditions at Huang et al. showed that the doping of Pt3Ni NPs with Mo led to 30  C with pulses of varying current density to obtain different extraordinary ORR performance, making the ternary PteNieMo compositions. The choice of pulse deposition parameters was system an interesting candidate for ORR studies [30]. These NPs initially orientated on the study on the electrodeposition of PteNi were obtained by decomposition of acetonate and hexacarbonyl NPs carried out by Egetenmeyer et al. [21] and then altered by precursors. The electrochemical co-deposition of Mo, which is empirically optimizing pulse deposition parameters with respect to 2 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 that higher current densities were needed for the ternary NPs to obtain the desired particle size and composition. 2.2. Electrochemical characterization The same three-electrode set-up used for NP electrodeposition was used for cyclic voltammetry (CV) studies of the NP/C assem- blies in 0.5 M H2SO4 to activate the catalysts and to determine their electrochemically active surface area (ECSA). To this end, 30 cycles at 200 mV/s and 5 cycles at 50 mV/s were recorded in a potential window between 0 and 1.3 V versus reversible hydrogen electrode (RHE). ORR was studied by CV in O2-saturated 0.1 M HClO4 after recording a background CV in N2-saturated electrolyte, both at 10 mV/s between 0 and 1.1 V versus RHE and in static conditions. Potentials applied against Ag|AgCl were converted to RHE scale. The curves were corrected for Ohmic drop (iR-correction) after deter- mination of the instrumentation resistance by electrochemical impedance spectroscopy (EIS) [38]. A Pt/C GDE with a Pt loading of 0.3 mg/cm2 was tested as a reference. It should be noted that the procedure for ORR measurements employed here differs from the usual approach using RDE since the electrocatalysts are directly deposited onto the GDL substrates and cannot be deposited directly Fig. 1. Set-up of the three-electrode electrochemical cell with the Pt counter electrode onto an RDE. As a result, a quantitative determination of ORR pa- (CE), Ag|AgCl reference electrode (RE), and the GDL mounted on a Cu plate at the rameters, such as the half-wave potential, is not feasible. Never- working electrode (WE). theless, the measurements give qualitative information to show trends between catalysts of different compositions, in addition to Table 1 the comparison with a commercial GDE. Electrolyte composition for the deposition of PteNi and PteNieMo(O) nanoparticles. PteNi PteNieMo(O) 2.3. SEM 200 mM NiCl2 NiCl2 5 mM Na2PtCl6,6H2O Na2PtCl6,6H2O All deposited NPs, as well as the unloaded GDL substrate, were 200 mM H3BO3 H3BO3 analyzed by scanning electron microscopy (SEM) on a Zeiss Merlin 25 mM NH4Cl NH4Cl electron microscope to evaluate particle size, distribution, and 50 mM Na2MoO4,2H2O loading of the substrates, using an acceleration voltage of 1 kV. For 100 mM citric acid energy-dispersive X-ray spectroscopy (EDX), an acceleration voltage of 20 kV was used. However, quantification of the electro- particle size, catalyst loading, and composition. It was observed catalyst NPs was not feasible by EDX due to the very low loading of that lower particle sizes were obtained at the expense of the NPs. For this reason, a chemical analysis method was employed (see catalyst loading, so that finally the process parameters were chosen below). Particle sizes were determined by image analysis of the as a compromise between particle size and catalyst loading. Pulse SEM micrographs using ImageJ, taking the average of 100e200 on-time and off-time were kept constant at 5 ms and 70 ms, particles per sample. SEM was also used for post analysis after respectively, whereas the number of cycles was changed with the PEMFC testing on a TESCAN LYRA3, after the cathode was delami- current density to obtain identical deposited charges and, assuming nated from the membrane electrode assembly (MEA). For cross- similar Faradaic current efficiencies for each electrolyte, similar sectional observation, one MEA was cut, embedded in epoxy resin, catalyst loading (Table 2). The total charge was increased for the mechanically ground and polished and subsequently ion milled by PteNieMo(O) deposition due to the addition of citric acid, which Ar ions on a Gatan PECS II. was observed to compromise the current efficiency. This is in accordance with previous studies, where the addition of citrate was 2.4. TEM found to increase Mo content in the deposits while lowering the current efficiency [37]. It was also observed during initial studies Transmission electron microscopy (TEM) was conducted on a JEOL JEM-2011 in high resolution and diffraction mode to study the crystalline structure of the NPs. Samples for TEM were prepared by Table 2 scratching the catalyst NPs, together with part of the carbon sup- Conditions for pulse electrodeposition of PteNi and PteNieMo(O) NPs. For the sake port, off the electrodes, dispersing the samples in ethanol, and of comparison, the atomic composition of the ternary NPs is given as if Mo were fully dropping them onto the carbon film of a TEM copper grid. Electron reduced. energy loss spectroscopy (EELS) was performed on a FEI Tecnai G2 Cathodic peak current density Number of cycles Composition F20 STEM. Both TEM were working at an acceleration voltage of [mA/cm2] 200 kV. 31 240 Pt79Ni21 37 200 Pt67Ni33 2.5. Chemical analysis 43 170 Pt80Ni20 49 150 Pt73Ni27 54 920 Pt78Ni21Mo1 For determination of both composition and loading of the sub- 77 640 Pt66Ni32Mo2 strates, NPs were dissolved in aqua regia, consisting of hydrochloric 100 490 Pt5Ni71Mo24 acid and nitric acid in a ratio of 3:1 in volume, and the concentra- 123 440 Pt5Ni74Mo21 tion of Pt, Ni, and Mo in the solution was determined by inductively 3 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 with a Pt loading of 0.3 mg/cm2 as anode using a pressure of 0.5 MPa at 110  C for 3 min to form the MEA. All MEAs were tested in a single-cell fuel cell tester using coated stainless steel flow plates. The active area of the fuel cell was 2.9 cm2. A Greenlight G20 test station was used for PEMFC tests, with a temperature of 80  C for both gases and a dewpoint tem- perature of 65  C. The flow of H2 and air were 0.042 and 0.10 ln/min (normal liter per minute), respectively. The cells were activated by cycling between 0.9 V and 0.6 V for 2 000 cycles before a polarization curve was obtained. For dura- bility testing, an accelerated stress test (AST) consisting of 20 000 additional cycles were performed, with polarization curves ob- tained after 10 000 cycles and at end of test (EoT). 3. Results and discussion 3.1. Electrodeposition process For all NPs, the amplitude of the deposition potential increases with the amplitude of the current density (Fig. 2). During pulse deposition of PteNi NPs, the amplitude ranges initially between 1.4 and 2.0 V versus Ag|AgCl depending on the applied current density. This potential drops towards the end of the depo- sition to within 1.0 to 1.3 V. The potential established during off- time changes from just below 0 V to slightly positive potential. In Fig. 2. Initial and final pulses for pulse deposition of PteNi (a) and PteNieMo(O) (b) turn, the potential amplitudes for the deposition of ternary nanoparticles. Currents are normalized by the geometric area of the gas diffusion layer. PteNieMo(O) NPs ranges from 2.0 V to 4.0 V due to the higher current densities applied. Those latter potentials do seem uncom- monly high, however they are based on an empirical determination coupled plasma mass spectrometry (ICP-MS) on an Agilent 7500ce of the process parameters. They may thus be the result of a signifi- spectrometer for each variation of electrodeposition parameters. In cant resistivity of the GDL substrate. The potential does not change addition, the Faradaic current efficiency was determined by relating significantly until the end of deposition, whereas the potential the total charge converted during electrodeposition to the absolute during off-time changes to 0.6 V towards the end, showing a trend masses of Pt, Ni, and Mo and assuming that all metals had been which is inverse to the one observed for binary NPs. fully reduced from their initial oxidation state. 3.2. Characterization of the as-deposited nanoparticles 2.6. X-ray photoelectron spectroscopy 3.2.1. Chemical analysis X-ray photoelectron spectroscopy (XPS) was performed on a PHI The results of the ICP-MS shows that all four depositions of 5500 spectrometer to study the surface state of the NPs, recording PteNi resulted in a Pt content of 67e80 at% (Table 3) and a Pt the core spectra for C1s, Ni2p, Pt4f, and Mo2p, before and after Ar loading of 3.7e4.1 mg/cm2. Assuming that the metal is completely ion sputtering for 30 s, as well as on the disassembled MEAs after reduced, which is fairly true for this case, the current efficiency is PEMFC testing. Peak fitting of the spectra was done by the software ca. 25% for all PteNi depositions. However, PteNieMo(O) appeared XPSPEAK. Energy calibration was done by positioning the main C1s to pass a threshold as the current density increased from 77 to peak at 284.5 eV. 100 mA/cm2, where the two lower current densities resulted in similar composition to the PteNi with 78 and 66 at% Pt and low 2.7. Fuel cell testing amounts of 1 and 2 at% Mo, respectively. The higher current den- sities on the other hand resulted in very low Pt content of only 5 at% For the PEMFC testing, the cathodes containing the electro- while Ni became the main constituent. The Mo content increased to deposited PteNi and PteNieMo(O) NPs were hot-pressed together 24 and 21 at%. Note that the composition of the ternary NPs is given with a Nafion 212 membrane (Chemours) and a commercial GDE in atomic percentage disregarding the oxygen content. The Table 3 Summary of the determined parameters for PteNi and PteNieMo(O) nanoparticles (NPs). Current efficiency of deposition, NP size at beginning of test (BoT) and end of test (EoT), total metal catalyst and Pt loadings, electrochemically active surface area (ECSA) calculated with respect to mass of Pt, and Tafel slope b. Composition Current efficiency Mean ø BoT Mean ø EoT Catalyst loading Pt loading ECSA b % nm nm mg/cm2 mg/cm2 m2/gPt mV Pt79Ni21 23 50 ± 13 46 ± 19 4.0 3.7 36 66 Pt67Ni33 26 47 ± 12 44 ± 20 4.6 4.0 19 65 Pt80Ni20 25 40 ± 8 46 ± 20 4.4 4.1 40 65 Pt73Ni27 25 49 ± 9 38 ± 17 4.4 4.0 34 66 Pt78Ni21Mo1 7 83 ± 30 45 ± 24 8.1 7.4 28 66 Pt66Ni32Mo2 2 49 ± 14 26 ± 14 2.5 2.2 18 62 Pt5Ni71Mo24 12 41 ± 19 24 ± 12 7.2 0.9 25 72 Pt5Ni74Mo21 8 49 ± 30 17 ± 9 5.6 0.8 88 82 4 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 Fig. 3. SEM micrographs of gas diffusion layer (GDL) substrate (a), electrodeposited binary Pt79Ni21 (b), Pt67Ni33 (c), Pt80Ni20 (d), Pt73Ni27 (e), and ternary Pt78Ni21Mo1 (f), Pt66Ni32Mo2 (g), Pt5Ni71Mo24 (h), Pt5Ni74Mo21 (i) nanoparticles on the GDL. In the micrographs, the NPs appear brighter than the microporous carbon. occurrence of oxygen in the Mo-containing NPs originating from region of the microporous layer, which appears darker in the SEM incomplete reduction of Mo(VI) precursor was deduced by XPS, as images compared with the deposited NPs (Fig. 3). All PteNi NPs shown later on. The total catalyst loading was less uniform for the show a uniform, spherical shape. The particle size is spread be- PteNieMo(O), ranging from 2.5 to 8.1 mg/cm2. The current effi- tween 20 nm and 80 nm, with an average particle size between ciency was also significantly reduced when comparing the ternary 40 nm and 50 nm independent of composition or process param- PteNieMo system to the binary PteNi, with current efficiencies as eters (Table 3), and are agglomerated in very few cases (Fig. 3bee). low as 2e12%. The PteNieMo(O) NPs have the tendency to a spherical structure, although less defined and with higher roughness than the PteNi 3.2.2. Microstructural analysis NPs. While the particle size of Pt78Ni21Mo1 is around 80 nm (Fig. 3f), The GDL substrates show the typical homogeneous microporous the other ternary NPs lie between 40 nm and 50 nm. For layer, consisting of aggregates of globular carbon particles (Fig. 3a). Pt5Ni74Mo21, a mixture of large (>100 nm) and small (<50 nm) NPs Fluorine emissions in EDX confirmed the presence of PTFE (not is appreciated (Fig. 3i). This observation is in agreement with the shown). large distribution in particle size determined for this composition, After the pulse electrodeposition, PteNi and PteNieMo(O) NPs while all other compositions show relatively narrow size distribu- are homogeneously distributed on the surface and near-surface tions (Fig. 4aeh). While commercial Pt/C has significantly smaller 5 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 Fig. 4. (aeh) Particle size distribution of Pt79Ni21 (a), Pt67Ni33 (b), Pt80Ni20 (c), Pt73Ni27 (d), Pt78Ni21Mo1 (e), Pt66Ni32Mo2 (f), Pt5Ni71Mo24 (g), and Pt5Ni74Mo21 (h) NPs. (iep) TEM micrographs of C-supported Pt67Ni33 (top, iel) and Pt78Ni21Mo1 (bottom, mep) NPs (i,j,m,n), in high resolution (k,o), and SAED pattern for a single NP (l,p). (qeu) XPS detail spectra of Pt4f (q) and Ni2p (r) of Pt73Ni27 NPs and of Pt4f (s), Ni2p (t) and Mo3d (u) of Pt5Ni71Mo24 NPs before (top) and after sputtering (bottom). 6 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 by EELS to confirm the distribution of Pt, Ni, and Mo. However, even for the NPs with the highest Mo content (Pt5Ni74Mo21), Mo was not detectable by EELS due to the high delay of the Mo signal, and the high amount of carbon causing a large back- ground signal (and resulting in a very low Mo content with respect to carbon). In contrast, Pt and Ni are clearly distributed evenly over the NPs (Fig. 5). 3.2.3. XPS The Pt4f spectra of the surfaces of the deposited PteNi NPs, Fig. 5. Dark field TEM image (a) and element distribution of Pt (b) and Ni (c) deter- exemplary shown for Pt73Ni27, show emissions of metallic Pt at mined by electron energy loss spectroscopy on a Pt5Ni74Mo21 nanoparticle. 71.4 eV and of Pt(I) at 72.3 eV, which can be assigned to platinum hydroxide [41]. In the Ni2p spectra, no contribution of metallic Ni is observed and most superficial Ni is bound in Ni(OH)2 (Fig. 4q,r). After Ar ion sputtering, the XPS spectra show a contribution of metallic Ni, as well as a higher fraction of metallic Pt with respect to the hydroxide species. It must be noted that, in contrast to bulk material, the contribution of the NPs' surface cannot be eliminated by sputtering. This means that after sputtering, there is still a sig- nificant contribution of the surface state, and the actual bulk state at the NPs’ cores is assumed to contain even higher fractions of metallic and less oxidized species than represented in the XPS spectra after Ar ion sputtering. For the ternary PteNieMo(O) NPs, very similar observations are made with respect to the Pt4f and Ni2p emissions (Fig. 4s and t). The Mo3d detail spectra do not show any evidence of metallic Mo at 228.0 eV [41]. The initial surface state shows the occurrence of Fig. 6. Representative cyclic voltammetry of gas diffusion layeresupported Pt73Ni27 in Mo(VI) exclusively, which points to the presence of MoO3. After 0.5 M H2SO4 for the determination of electrochemically active surface area. The inte- sputtering, the lower oxidation state Mo(IV) is found, which may grated area of the hydrogen desorption peak is displayed in gray. correspond to MoO2 (Fig. 4u). Therefore, Mo(VI) was not fully reduced to Mo(0) in spite of the presence of Ni(II) and citrate in the electrolyte. It can be argued that sufficient complexing of molyb- particle size, Fouda-Onana et al. electrodeposited Pt particles with a date species with citrate, and therefore the full discharge of Mo(VI) size of 50 nm and found that despite the larger particle size, the to Mo(0), does not occur at the rather low pH of the electrolyte of utilization rate was high due to all particles having triple phase 2.7. At this pH, the citrate mostly exists in its protonated form, thus boundaries (TPBs) [39]. hindering complexation of metal ions. In TEM observations, the PteNi and PteNieMo(O) NPs are For both binary and ternary NPs, a drastic reduction of O1s clearly distinguished from the C support by phase contrast (in emissions (not shown) were observed after sputtering, indicating bright field TEM, the NPs appear darker than the carbon-based that the NPs exhibit an oxidized surface and a metallic core. Due to support; Fig. 4i,m). The binary NPs are of well-defined, globular the fact that contributions of the NPs' surfaces are present even form, whereas the ternary NPs are more irregular (Fig. 4j,n). The after sputtering, O1s emissions are not completely suppressed, nor high-resolution mode reveals the nanocrystalline structure of can it be completely ruled out that there remain low amounts of PteNi NPs, indicated by diffraction planes with different orienta- residual oxygen in the NPs’ cores. tions (Fig. 4k). This nanocrystallinity is also clearly observed in the Combining all knowledge obtained on the ternary PteNieMo selected area electron diffraction (SAED) pattern of a single NP with system, the following conclusions can be drawn. Taking into ac- a diameter (ø) <50 nm, where almost any diffraction direction is count the existence of Pt(0), Ni(0), Mo(IV), and Mo(VI) shown by present, confirmed by the quasi-continuous rings observed in the XPS, the absence of concrete diffraction rings for the PteNi fcc pattern (Fig. 4l). The corresponding d-spacings match the face- phase in SAED, the very small (<5 nm) crystalline regions in high centered cubic (fcc) phase for a PteNi alloy. For Pt67Ni33, the resolution TEM, and the homogeneous distribution of Pt and Ni average cell parameter determined from SAED was 3.88 Å, which throughout the NPs as evidenced by EELS, it is concluded that PteNi corresponds well with the d(111) spacing of 2.2 Å measured in the crystals and molybdenum oxide species are coexistent and high-resolution image. randomly distributed among the NPs. Thus, the PteNieMo(O) NPs In contrast to the PteNi NPs, the PteNieMo(O) NPs show more consist of a heterogeneous compound of PteNi and molybdenum diffuse diffraction rings, which are more characteristic of an oxide, which due to their nanosized nature and random amorphous material, even when the Mo content is very low (Fig. distribution, appear as a homogeneous compound even on the 4p). Some additional diffraction spots indicate the presence of Mo nanoscale. oxide species such as MoO3 or other non-stoichiometric oxides [40]. Diffraction planes in high-resolution TEM are only appreciated in extremely few occasions at the NPs’ surfaces, showing crystals 3.3. Half-cell electrochemical tests smaller than 5 nm (Fig. 4o). Other than that, diffraction planes are not observed, leading to the conclusion that most PteNi crystals are The GDL substrates exhibit a rather low electric conductivity, too small to diffract and that they are completely mixed in with the leading to a relatively high double-layer capacitance observed in molybdenum oxide species in a sort of composite. CV. Nevertheless, the hydrogen adsorption (Hads) and desorption From SAED data, the existence of a homogeneous alloy is clear peaks (Hdes) used for the determination of ECSA are well defined, as for the PteNi NPs, the ternary PteNieMo(O) NPs were analyzed shown exemplarily for Pt73Ni27 (Fig. 6). 7 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 The ECSA of both the PteNi and PteNieMo(O) is comparable successfully in PEMFC with only little performance loss over 30 h of with what is found in literature (Table 3) [2], showing that despite constant operation [25]. Among the Pt-rich ternary NPs, the electrodeposited catalysts having a large particle size, the Pt Pt66Ni32Mo2 shows the highest Pt mass activity, corresponding to utilization is high. the Ni/Pt ratio of the binary Pt67Ni33. Zalitis et al. showed that ul- The electrodeposited PteNi and PteNieMo(O) NPs all show trathin catalyst layers of 200 nm thickness can be employed to lower reduction potentials at ORR compared with the Pt/C (Fig. overcome limitations by internal resistances such as experienced 7a). Among them, the Mo-containing NPs are inferior to the bi- here with the use of the GDL. In this way, the mass activity can be nary alloy NPs with comparable Pt/Ni ratio. Interestingly, the best further increased by one order of magnitude at comparable Pt performance among the electrodeposited NPs is observed for loading [42]. Pt67Ni33. A diffusion-controlled region is not observed for any of An increase in total load of electrodeposited NPs on the GDL is the samples; this is related to the fact that kinetics is limited expected to increase the half-wave potential at ORR. However, the because the measurements were carried out in stagnant condi- electrodeposition of higher amounts of catalyst could easily lead to tions, and also due to the GDL substrates which are not perfect NP growth rather than nucleation of more particles. conductors. Another way of improving the ORR of the electrodeposited NPs The advantage of electrodeposited PteNi and PteNieMo(O) NPs is by addition of an ionomer such as Nafion, which can lead to a becomes apparent when the Pt mass activity at ORR is considered significant increase in ECSA and ORR performance by improving the (Fig. 7b). Pt67Ni33 shows the highest Pt mass activity among the wettability of the GDL [43]. binary NPs. In addition, the Ni-rich ternary NPs, which also have In terms of kinetics, the determined Tafel slopes (b) show a significantly higher amounts of Mo, show the highest mass activity significant improvement with respect to the commercial Pt/C with respect to Pt content. This shows that the addition of Mo does electrode, for which a Tafel slope of 103 mV is obtained. Interest- not compromise the catalytic activity and may indeed lead to ingly, all binary PteNi binary alloy NPs exhibit an almost identical improvement, however, the stability at ORR of these low-Pt content Tafel slope of 65e66 mV (cf. Table 3). The ternary PteNieMo(O) NPs in acidic conditions may be compromised. Liu et al. recently NPs show a higher spread in b, where Pt66Ni32Mo2 has the lowest found that a Ni-rich NiePt alloy electrocatalyst can be employed Tafel slope of 62 mV (Fig. 7c). The measured Tafel slopes correspond Fig. 7. Oxygen reduction reaction of PteNi and PteNieMo(O) nanoparticles (NPs) deposited on gas diffusion layer (GDL) in comparison with Pt/C, by geometric current density (a) and by Pt mass activity (b). The inset in (b) shows the mass activity for the Pt/C electrode. Tafel slopes for the GDL-supported PteNi and PteNieMo(O) NPs, and for commercial Pt/C (c). The inset in (c) shows a zoom for the closely-placed PteNi NP electrocatalysts. Back-scattered electron micrograph of a cross section of hot-pressed Pt66Ni32Mo2 membrane electrode assembly prior to proton exchange membrane fuel cell testing (d). 8 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 Fig. 8. Peak power density of PteNi (a) and PteNieMo(O) catalysts (b) after activation, 10 000, and 20 000 accelerated stress test (AST) cycles. SEM micrographs of Pt79Ni21 (c), Pt67Ni33 (d), Pt80Ni20 (e), Pt73Ni27 (f), Pt78Ni21Mo1 (g), Pt66Ni32Mo2 (h), Pt5Ni71Mo24 (i), and Pt5Ni74Mo21 (j) nanoparticles on the gas diffusion layer after AST in the proton exchange membrane fuel cell. well with the one determined by Fortunato et al. on pure Pt NPs dissolution of unstable catalytic sites, such as surface Ni and Mo in synthesized by electrodeposition [44]. the initial half of the AST, leaving the particles more stable in the second half of the AST. No decrease in activity could be seen for the 3.4. Proton exchange membrane fuel cell testing commercial Pt/C, which after 20 000 AST cycles show a peak power density of 0.377 W/cm2. The cross-sectional image of a hot-pressed MEA prior to testing In general, the SEM micrographs of the NPs after AST in the in the fuel cell shows that the catalyst layer (CL) formed by the PEMFC show that the particle size is reduced (cf. Table 3), related to electrodeposited NPs is homogeneous and well adhered to the a loss of material by dissolution in the acidic environment (Fig. 8). membrane, ensuring proton conductivity (Fig. 7d). The exception is Pt80Ni20, where the particle size is slightly The polarization curves show that the low catalyst loading re- increased (Table 3). This increase in particle size can be explained sults in low current density (not shown). However, the catalysts by Pt dissolution from smaller, less stable particles, and re-depo- activity is similar independent of composition, both initially and sition onto larger particles. It should be noted that a non-negligible after AST. Interestingly, the Pt67Ni33 and Pt66Ni32Mo2 again show amount of particles remained attached to the membrane rather the highest activity. The major drop in PEMFC performance takes than the GDL after disassembling the MEA for post-analysis. The place during the first 10 000 cycles, whereas there is not as sig- PteNi particles have kept their spherical morphology, while the nificant drop from 10 000 to 20 000 AST cycles (Fig. 8). This is in PteNieMo(O) NPs have increased roughness after testing. It is also accordance with literature [2] and can be explained by the clear that the two Ni-rich samples with very low Pt content 9 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 4. Conclusions Binary PteNi and ternary PteNieMo(O) NPs were successfully synthesized by pulse electrodeposition from aqueous media directly onto the GDL of a PEMFC. The so-prepared carbon-sup- ported catalysts show high specific activities at ORR and the applicability in PEMFC was demonstrated in a single-cell, with Pt67Ni33 and Pt66Ni32Mo2 showing the highest activity both in ORR measurements and in the PEMFC. NP particle sizes range around 50 nm, and Tafel slopes at ORR of around 65 mV are achieved at low catalyst loadings of 4 mg/cm2. Most importantly, very high ORR mass activities up to 10 mA/mgPt are reached at half-cell electrochemical tests in 0.1 M HClO4, owing to the favorable distribution of electrocatalyst NPs along the cata- lyst layer as a result of the electrodeposition process. In the MEAs produced for PEMFC testing, almost all NPs are expected to contribute to the ORR reaction due to their direct contact with both carbon support and PEM, resulting in extremely high efficiency with respect to Pt utilization. For industrial application in PEMFC, the total power output of the fuel cell would need to be increased. To this respect, a higher total amount of catalyst may be needed, achieved by either increasing the Pt load in electrodeposition or by increasing the fuel cell's active area. An increase of loading by electrodeposition is easily achievable. However, by simply increasing the deposition Fig. 9. XPS detail spectra of Pt4f (a) and Ni2p (b) of Pt73Ni27 nanoparticles (NPs) and of time, additional deposition will occur on already deposited mate- Pt4f (c), Ni2p (d) and Mo3d (e) of Pt5Ni71Mo24 NPs after proton exchange membrane rial and lead to particle growth. This would most likely improve fuel cell tests. half-wave potential and obtainable currents, but adversely affect mass activity. Further optimization is possible by tuning the NP experience the most severe dissolution of catalyst material. How- particle size, or by optimizing the electrical properties by the use of ever, the PEMFC performance at EoT is similar to the other samples. thin carbon layers adapted to electrodeposition, to reduce intrinsic This may be due to the high surface area endowed by the rough resistances. morphology of these NPs. With respect to the ternary system NPs, an obvious advantage The XPS spectra obtained from the catalyst layers after PEMFC over binary PteNi NPs is not observed. Since Mo was mostly found in testing are compromised by the aforementioned loss of material an oxidized state and the crystallinity of the ternary NPs as observed due to particles remaining on the membrane; however, it was by SAED was lower, the question remains open as to whether a true, generally observed that in comparison with the initial surface state, metallic ternary alloy would yield superior ORR activity. both Pt(0) and Ni(0) were present in higher fractions with respect Overall, the demonstrated electrodeposition process provides a to their oxidized forms (Fig. 9). For the Mo-containing catalysts, promising alternative to the conventional methods of ORR elec- only Mo(VI) was detected after the PEMFC tests. Contrarily to what trocatalyst synthesis. In addition to the facile synthesis which ap- might be expected, Ni2p emission levels are close to their initial plies the catalyst NPs directly onto the GDL, the utilization of the values (cf. Fig. 4qeu) while Pt emissions are significantly lower. metal electrocatalyst can be seen as close to 100%. Although electrodeposition is a very common and widely used synthesis method, its employment in the synthesis of ORR elec- trocatalysts for PEMFC is rather unexplored. Few studies have CRediT author statement investigated this topic in recent years, and the processes need to be further optimized in view of the electrocatalysts’ properties and Konrad Eiler: conceptualization, validation, formal analysis, performance, and especially in terms of actual testing in PEMFC investigation, data curation, writingdoriginal draft, writingdre- (Table 4). view & editing, visualization. Live Mølmen: conceptualization, Table 4 Comparison of electrodeposited catalysts for oxygen reduction reaction (ORR) in proton exchange membrane fuel cell (PEMFC) reported in literature. Material Particle size Pt loading ECSA b PEMFC operation Reference nm mg/cm2 m2/gPt mV PteNi 47 4.0 19 65 Successful This work PteNieMo(O) 49 2.2 18 62 Successful This work PteNi Thin film 27 27 e [29] PteNieCo ca. 100 52 153 e [45] Pt ca. 50 6.2 155 63 e [44] Pt 89 1.6 46 e [16] Pt 3e10 9.5 e [23] Pt ca. 1 000 <100 Successful [19] Pt 90 23 Successful [13] PteCo 300 (Pt þ Co) Successful [13] PteCo ca. 1 000 500 In DMFCa [35] PteCo 6.5 mC/cm2 e [20] a Direct methanol fuel cell. 10 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 software, validation, formal analysis, investigation, data curation, [12] J. Hou, M. Yang, C. Ke, G. Wei, C. Priest, Z. Qiao, G. Wu, J. Zhang, Platinum- group-metal catalysts for proton exchange membrane fuel cells: from catalyst writingdoriginal draft, writingdreview & editing, visualization. design to electrode structure optimization, Energy Chem. 2 (2020) 100023, Lars Fast: conceptualization, methodology, validation, resources, https://doi.org/10.1016/j.enchem.2019.100023. supervision, project administration. Peter Leisner: conceptualiza- [13] A. Egetenmeyer, I. Radev, D. Durneata, M. Baumga €rtner, V. Peinecke, H. Natter, tion, resources, supervision, project administration, funding R. Hempelmann, Pulse electrodeposited cathode catalyst layers for PEM fuel cells, Int. J. Hydrog. Energy 42 (2017) 13649, https://doi.org/10.1016/ acquisition. Jordi Sort: conceptualization, resources, writingdre- j.ijhydene.2017.01.212. view & editing, supervision, funding acquisition. Eva Pellicer: [14] J. Ustarroz, T. Altantzis, J.A. Hammons, A. Hubin, S. Bals, H. Terryn, The role of conceptualization, validation, resources, writingdreview & editing, nanocluster aggregation, coalescence, and recrystallization in the electro- chemical deposition of platinum nanostructures, Chem. Mater. 26 (2014) supervision, project administration, funding acquisition. 2396, https://doi.org/10.1021/cm403178b. [15] J. Ustarroz, B. Geboes, H. Vanrompay, K. Sentosun, S. Bals, T. Breugelmans, A. Hubin, Electrodeposition of highly porous Pt nanoparticles studied by Data availability quantitative 3D electron tomography: influence of growth mechanisms and potential cycling on the active surface area, ACS Appl. Mater. Interfaces 9 The raw/processed data required to reproduce these findings (2017) 16168, https://doi.org/10.1021/acsami.7b01619. [16] B. Geboes, J. Ustarroz, K. Sentosun, H. Vanrompay, A. Hubin, S. Bals, cannot be shared at this time due to technical or time limitations. T. Breugelmans, Electrochemical behavior of electrodeposited nanoporous Pt Specific data can be obtained from the corresponding authors on catalysts for the oxygen reduction reaction, ACS Catal. 6 (2016) 5856, https:// request. doi.org/10.1021/acscatal.6b00668. [17] J.H. Park, H.S. Ahn, Electrochemical synthesis of multimetallic nanoparticles and their application in alkaline oxygen reduction catalysis, Appl. Surf. Sci. Declaration of competing interest 504 (2020) 144517, https://doi.org/10.1016/j.apsusc.2019.144517.  [18] F. Alcaide, G. Alvarez, P.L. Cabot, R.V. Genova-Koleva, H.eJ. Grande, M.V. Martínez-Huerta, O. Miguel, Supporting PtRh alloy nanoparticle catalysts The authors declare that they have no known competing finan- by electrodeposition on carbon paper for the ethanol electrooxidation in cial interests or personal relationships that could have appeared to acidic medium, J. Electroanal. Chem. 861 (2020) 113960, https://doi.org/ 10.1016/j.jelechem.2020.113960. influence the work reported in this paper. [19] C. Huang, C.B. Odetola, M. Rodgers, Nanoparticle seeded pulse electrodepo- sition for preparing high performance Pt/C electrocatalysts, Appl. Catal. A 499 (2015) 55, https://doi.org/10.1016/j.apcata.2015.03.043. Acknowledgement [20] J. Sriwannaboot, A. Kannan, N. Tantavichet, Pulse-reverse electrodeposition of PteCo bimetallic catalysts for oxygen reduction reaction in acidic medium, Int. J. This work has received funding from the European Union's Ho- Hydrog. Energy 45 (2020) 7025, https://doi.org/10.1016/j.ijhydene.2019.12.191. [21] A. Egetenmeyer, M. Baumga €rtner, T. Linckh, D. Durneata, H. Natter, rizon 2020 research and innovation programme under the Marie R. Hempelmann, I. Radev, V. Peinecke, Pulse electrodeposition of catalyst Skłodowska-Curie grant agreement No 764977, the Generalitat de nanoparticles for application in PEM fuel cells, Trans. IMF 95 (2017) 9, https:// Catalunya under project 2017-SGR-292, and the Spanish govern- doi.org/10.1080/00202967.2017.1253661. ment under project PID2020-116844RB-C21. The authors want to [22] D. Santiago, G.G. Rodríguez-Calero, H. Rivera, D.A. Tryk, M.A. Scibioh, C.R. Cabrera, Platinum electrodeposition at high surface area carbon Vulcan- express their thanks to Freudenberg, Germany, who gladly sup- XC-72R material using a rotating disk-slurry electrode technique, ported the GDL material for this study. J. Electrochem. Soc. 157 (2010) F189, https://doi.org/10.1149/1.3489948. [23] D. Wang, G. Hu, P. Yang, X. Pan, H. Xu, L. Liu, J. Zhang, M. An, Using DMH as a complexing agent for pulse electrodeposition of platinum nanoparticles to- References wards oxygen reduction reaction, Ionics 26 (2020) 3473, https://doi.org/ 10.1007/s11581-019-03427-1. [1] European Commission, A Hydrogen Strategy for a Climate-Neutral Europe, [24] C. Qiu, S. Wang, R. Gao, J. Qin, W. Li, X. Wang, Z. Zhai, D. Tian, Y. Song, Low- 2020. COM(2020) 301 final. temperature synthesis of PdO-CeO2/C toward efficient oxygen reduction re- [2] L. Mølmen, K. Eiler, L. Fast, P. Leisner, E. Pellicer, Recent advances in catalyst action, Mater, Today Energy 18 (2020) 100557, https://doi.org/10.1016/ materials for proton exchange membrane fuel cells, APL Mater. 9 (2021), j.mtener.2020.100557. 040702, https://doi.org/10.1063/5.0045801. [25] X. Liu, X. Wan, X. Tan, H. Yang, Y. Yang, J. Shui, X. Wang, Nanoscale Pt5Ni36 [3] L. Mølmen, A. Alexandersson, P. Leisner, Surface technology should improve design and synthesis for efficient oxygen reduction reaction in proton ex- PEM fuel cell performance, Trans. IMF 97 (2019) 112, https://doi.org/10.1080/ change membrane fuel cells, J. Mater. Chem. 9 (2021) 21051, https://doi.org/ 00202967.2019.1596573. 10.1039/d1ta05189j. [4] J.K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J.R. Kitchin, T. Bligaard, [26] E. Antolini, The oxygen reduction on Pt-Ni and Pt-Ni-M catalysts for low- H. Jonsson, Origin of the overpotential for oxygen reduction at a fuel-cell temperature acidic fuel cells: a review, Int. J. Energy Res. 42 (2018) 3747, cathode, J. Phys. Chem. B 108 (2004) 17886, https://doi.org/10.1021/jp047349j. https://doi.org/10.1002/er.4134. [5] K. Kodama, T. Nagai, A. Kuwaki, R. Jinnouchi, Y. Morimoto, Challenges in [27] X. Tian, X. Zhao, Y.eQ. Su, L. Wang, H. Wang, D. Dang, B. Chi, H. Liu, applying highly active Pt-based nanostructured catalysts for oxygen reduction E.J.M. Hensen, X.W. Lou, B.Y. Xia, Engineering bunched Pt-Ni alloy nanocages reactions to fuel cell vehicles, Nat. Nanotechnol. 16 (2021) 140, https:// for efficient oxygen reduction in practical fuel cells, Science 366 (2019) 850, doi.org/10.1038/s41565-020-00824-w. https://doi.org/10.1126/science.aaw7493. [6] Y. Yang, C. Tan, Y. Yang, L. Zhang, B.-W. Zhang, K.eH. Wu, S. Zhao, Pt3Co@Pt ~ ach, J. Sort, E. Pellicer, Mesoporous Ni-rich Ni-Pt thin films: [28] K. Eiler, S. Surin core@shell nanoparticles as efficient oxygen reduction electrocatalysts in electrodeposition, characterization and performance toward hydrogen evo- direct methanol fuel cell, ChemCatChem 13 (2021) 1587, https://doi.org/ lution reaction in acidic media, Appl. Catal., B 265 (2020) 118597, https:// 10.1002/cctc.202001868. doi.org/10.1016/j.apcatb.2020.118597. [7] Z. Wang, X. Yao, Y. Kang, D. Xia, L. Gan, Rational development of structurally [29] Y. Liu, C.M. Hangarter, U. Bertocci, T.P. Moffat, Oxygen reduction reaction on ordered platinum ternary intermetallic electrocatalysts for oxygen reduction electrodeposited Pt100xNix: influence of alloy composition and dealloying, reaction, Catalysts 9 (2019) 569, https://doi.org/10.3390/catal9070569. J. Phys. Chem. C 116 (2012) 7848, https://doi.org/10.1021/jp300672h. [8] L. Huang, M. Wei, N. Hu, P. Tsiakaras, P. Kang Shen, Molybdenum-modified [30] X. Huang, Z. Zhao, L. Cao, Y. Chen, E. Zhu, Z. Lin, M. Li, A. Yan, A. Zettl, and vertex-reinforced quaternary hexapod nano-skeletons as efficient elec- Y.M. Wang, X. Duan, T. Mueller, Y. Huang, High-performance transition metal- trocatalysts for methanol oxidation and oxygen reduction reaction, Appl. doped Pt3Ni octahedra for oxygen reduction reaction, Science 348 (2015) Catal., B 258 (2019) 117974, https://doi.org/10.1016/j.apcatb.2019.117974. 1230, https://doi.org/10.1126/science.aaa8765. [9] S. Zhao, Y. Yang, Z. Tang, Insight into structural evolution, active sites, and [31] E. Chassaing, N. Portail, A.eF. Levy, G. Wang, Characterisation of electro- stability of heterogeneous electrocatalysts, Angew. Chem. Int. Ed. 61 (2022), deposited nanocrystalline NieMo alloys, J. Appl. Electrochem. 34 (2004) 1085, e202110186, https://doi.org/10.1002/anie.202110186. https://doi.org/10.1007/s10800-004-2460-z. [10] X. Wang, Y. Li, C. Yang, J. Lu, L. Cao, H.eP. Liang, Surfactant-assisted implan-  mez, E. Pellicer, E. Valle [32] E. Go s, Intermediate molybdenum oxides involved in tation strategy for facile construction of Pt-based hybrid electrocatalyst to binary and ternary induced electrodeposition, J. Electroanal. Chem. 580 (2005) accelerate oxygen reduction reaction, Mater. Today Energy 24 (2022) 100919, 238, https://doi.org/10.1016/j.jelechem.2005.03.031. https://doi.org/10.1016/j.mtener.2021.100919. [11] P. Yin, B. You, Atom migration-trapping toward single-atom catalysts for [33] Z. Liu, J.E. Hu, Q. Wang, K. Gaskell, A.I. Frenkel, G.S. Jackson, B. Eichhorn, PtMo alloy energy electrocatalysis, Mater. Today Energy 19 (2021) 100586, https:// and MoOx@Pt core-shell nanoparticles as highly CO-tolerant electrocatalysts, doi.org/10.1016/j.mtener.2020.100586. J. Am. Chem. Soc. 131 (2009) 6924, https://doi.org/10.1021/ja901303d. 11 K. Eiler, L. Mølmen, L. Fast et al. Materials Today Energy 27 (2022) 101023 [34] C. Wang, D. Wang, S. Liu, P. Jiang, Z. Lin, P. Xu, K. Yang, J. Lu, H. Tong, L. Hu,  noma de Barcelona. Prior to her Italy, and Universitat Auto W. Zhang, Q. Chena, Engineering the coordination environment enables mo- PhD studies, she obtained an MSc in Materials Technology lybdenum single-atom catalyst for efficient oxygen reduction reaction, J. Catal. from the Norwegian University of Science and Technology 389 (2020) 150, https://doi.org/10.1016/j.jcat.2020.05.034. (NTNU). [35] O. Sorsa, H. Romar, U. Lassi, T. Kallio, Co-electrodeposited mesoporous PtM (M¼Co, Ni, Cu) as an active catalyst for oxygen reduction reaction in a poly- mer electrolyte membrane fuel cell, Electrochim. Acta 230 (2017) 49, https:// Lars Fast, senior researcher at RISE AB since 2001. doi.org/10.1016/j.electacta.2017.01.158. Received his PhD from Uppsala University in the field of [36] S.A. Lee, J.W. Yang, S. Choi, H.W. Jang, Nanoscale electrodeposition: Dimension computational material science (1998). He worked as a control and 3D conformality, Exploration 1 (2021) 20210012, https://doi.org/ post-doc in Würzburg, Germany, between 1998 and 2001, 10.1002/exp.20210012. in developing and implementing density functional theory [37] E.J. Podlaha, D. Landolt, Induced codeposition: I. An experimental investiga- for superconductors. At RISE, he has worked mainly with tion of Ni-Mo alloys, J. Electrochem. Soc. 143 (1996) 885, https://doi.org/ batteries, fuel cells and electrical contacts, including the 10.1149/1.1836553. env iro nmental d urab ilit y qu es tions a nd lif et im e [38] C. Wei, R.R. Rao, J. Peng, B. Huang, I.E.L. Stephens, M. Risch, Z.J. Xu, Y. Shao- assessments. Horn, Recommendation practices and benchmark activity for hydrogen and oxygen electrocatalysis in water splitting and fuel cells, Adv. Mater. 31 (2019) 1806296, https://doi.org/10.1002/adma.201806296. [39] F. Fouda-Onana, N. Guillet, A.M. Almayouf, Modified pulse electrodeposition of Pt nanocatalyst as high-performance electrode for PEMFC, J. Power Sources 271 (2014) 401, https://doi.org/10.1016/j.jpowsour.2014.08.031. [40] A. Quintana, A. Varea, M. Guerrero, S. Surin ~ ach, M.D. Baro  , J. Sort, E. Pellicer, Structurally and mechanically tunable molybdenum oxide films and Prof. Peter Leisner, head of Materials and Manufacturing patterned submicrometer structures by electrodeposition, Electrochim. Acta Department at School of Engineering, Jo € nko €ping Univer- 173 (2015) 705, https://doi.org/10.1016/j.electacta.2015.05.112. sity. MSc in electrochemistry (1989) and PhD in pulse [41] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-Ray plating (1992), Technical University of Denmark. 30 years Photoelectron Spectroscopy, Physical Electronics, Inc., MI, 1995. of experience in applied research in micro-technology, sur- [42] C.M. Zalitis, D. Kramer, A.R. Kucernak, Electrocatalytic performance of fuel cell face technology, and reliability. reactions at low catalyst loading and high mass transport, Phys. Chem. Chem. Phys. 15 (2013) 4329, https://doi.org/10.1039/c3cp44431g. [43] P. Mardle, G. Thirunavukkarasu, S. Guan, Y.eL. Chiu, S. Du, Comparative study of PtNi nanowire array electrodes toward oxygen reduction reaction by half- cell measurement and PEMFC test, ACS Appl. Mater. Interfaces 12 (2020) 42832, https://doi.org/10.1021/acsami.0c11531. [44] G.V. Fortunato, F. de Lima, G. Maia, Oxygen-reduction reaction strongly electrocatalyzed by Pt electrodeposited onto graphene or graphene nano- ribbons, J. Power Sources 302 (2016) 247, https://doi.org/10.1016/ j.jpowsour.2015.10.069. Prof. Jordi Sort received his PhD Degree in Materials Sci- [45] M. Li, Y. Lei, N. Sheng, T. Ohtsuka, Preparation of low-platinum-content platinum- ence from Universitat Auto noma de Barcelona (UAB) in nickel, platinum-cobalt binary alloy and platinum-nickel-cobalt ternary alloy 2002. He worked for 2 years as Postdoctoral Researcher at catalysts for oxygen reduction reaction in polymer electrolyte fuel cells, J. Power the SPINTEC Laboratory (Grenoble) and subsequently Sources 294 (2015) 420, https://doi.org/10.1016/j.jpowsour.2015.06.084. stayed 6 months at Argonne National Laboratory (USA). He also performed long-term secondments at the Grenoble High Magnetic Fields Laboratory and at Los Alamos Na- Konrad Eiler obtained his PhD in Materials Science from the tional Laboratory. At present, Prof. Sort leads the “Group Universitat Autonoma de Barcelona in 2021, focusing on the of Smart Nanoengineered Materials, Nanomechanics and electrodeposition of nickel-based materials for electrocatal- Nanomagnetism (Gnm3)” at UAB, which focuses its ysis, and included long-term stays at RISE Research Insti- research activities on the synthesis of a wide variety of tutes of Sweden and AGH University of Science and functional materials and the study of their structural, me- Technology, Krako  w. Previously, he completed his B.Sc. in chanical, magnetic, and thermal properties. Materials Science at TU Darmstadt and his M.Sc. in the same field at TU Berlin in 2014, including an internship at Volkswagen's R&D department in Wolfsburg. Prior to his PhD, Konrad Eiler has been working at the European Organi- Eva Pellicer is currently an associate professor at Uni- zation for Nuclear Research (CERN), Geneva, for over 2 years. noma de Barcelona (UAB), Spain. She received versitat Auto her BSc in Chemistry in 2000 and the PhD in Materials Technology in 2005, both from Universitat de Barcelona (UB). So far, Dr. Pellicer has published more than 200 pa- pers in journals like Advanced Functional Materials, Live Mølmen is a PhD student at Jo €nko€ ping University in Applied Catalysis B: Environmental, ACS Nano, etc. Her the Department of Materials and Manufacturing, with current research interests are the electrodeposition and employment at RISE Research Institutes of Sweden AB. Her characterization of nanostructured and porous alloys for a PhD focuses on materials reliability in PEM fuel cells and variety of applications, ranging from electrocatalysis to has included secondments to both the University of Udine, magneto-electric applications. 12