This is the submitted version of the journal article: Han, Lijuan; González-Cobos, Jesús; Sánchez-Molina, Irene; [et al.]. «Cobalt hexacyanoferrate as a selective and high current density formate oxidation electrocatalyst». ACS Applied Energy Materials, Vol. 3, issue 9 (Sep. 2020), p. 9198-9207. DOI 10.1021/acsaem.0c01548 This version is available at https://ddd.uab.cat/record/236020 under the terms of the license Page 1 of 12 ACS Applied Energy Materials 1 2 3 4 5 6 7 Cobalt hexacyanoferrate as a selective and high current density 8 9 formate oxidation electrocatalyst 10 Lijuan Han,† Jesús González-Cobos,† Irene Sánchez-Molina,† Stefano Giancola,† Scott J. 11 12 Folkman,† Pengyi Tang,§ Marc Heggen,♯ Rafal E. Dunin-Borkowski,♯ Jordi Arbiol,§,¶ Sixto 13 Giménez,⊥ and Jose Ramon Galan-Mascaros*,†,¶ 14 †Instituteof Chemical Research of Catalonia (ICIQ), The Barcelona Institute of Science and Technology (BIST), Av. 15 16 Paisos Catalans 16, 43007 Tarragona, Spain. 17 §CatalanInstitute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST, Campus UAB, Bellaterra, 08193 18 Barcelona, Catalonia, Spain. 19 ♯Ernst-RuskaCentre for Microscopy and Spectroscopy with Electrons and Peter Grünberg Institute, 20 Forschungszentrum Jülich GmbH, 52425 Jülich, Germany. 21 ¶ICREA, Pg. Lluís Companys 23, 08010 Barcelona, Spain. 22 23 ⊥Institute of Advanced Materials (INAM), Universitat Jaume I, 12006 Castelló, Spain. 24 ABSTRACT: Herein we report the selectivity, stability, and electrochemical characterization of cobalt hexacyanoferrate, the Co- 25 Fe Prussian Blue derivative (CoFePB), as a formate/formic acid oxidation electrocatalyst in aqueous media. CoFePB is able to 26 quantitatively catalyze (100% Faradaic efficiency within less than 8% standard error at pH 5) the electrochemical oxidation of 27 formate to CO2 over a pH range of 1-13. This quantitative formate elecrooxidation is possible due to the exclusive selectivity of the 28 catalyst in a wide potential window (from c.a. 1.2 to 1.7 V vs. RHE), where no other substrate in aqueous conditions is activated: 29 neither other organic molecules, such as alcohols or acids, nor water itself. CoFePB is one of the first heterogeneous noble-metal- 30 free catalysts reported for the electrooxidation of small hydrocarbon molecules. Importantly, the catalyst showed a very high 31 tolerance against surface poisoning during the reaction, as supported by the cyclic voltammetry and electrochemical impedance 32 spectroscopy data, thereby allowing CoFePB to operate at greater current density than state-of-the-art noble metal catalysts. For 33 example, we observed that CoFePB is able to achieve a formate oxidation current ~10 mA cm–2 at pH 5, 0.4 M formate at 1.4 V vs 34 RHE, whereas a Pt disk and Pd(5%)/C electrodes had a current of 0.4 and 1.4 mA cm–2, respectively, under identical conditions. 35 The remarkable selectivity, stability, and high current density of CoFePB, in contrast to state-of-the-art catalysts based on platinum-group metals (PGM), is an important step in the search for inexpensive earth-abundant materials for oxidation of organic 36 molecules for use in liquid fuel cells or for selective organic molecule sensors. Furthermore, because CoFePB is not poisoned by 37 intermediates and can achieve higher current density than Pt or Pd, improvement of the catalyst onset potential can lead to higher 38 power density formate oxidation fuel cells using earth abundant metals than with Pt or Pd. 39 40 41 42 Introduction flammable organic solvents and leading to safer operation. 43 Regarding the sensing field, selective formate/formic acid The selective electrochemical oxidation of small organic 44 electrooxidation materials are important, for instance, in molecules is getting attention for multiple applications, 45 analysis of pollutants in the atmosphere and natural including sensing and analysis, environmental 46 waters, detection of liquid fuel cell crossover or control of remediation and energy-related systems, such as fuel cells. 47 aerobic and anaerobic fermentation processes.9-12 In the latter sense, liquid hydrogen carriers (alcohols, 48 acids, sugars) have been proposed as potential H2 storage Electrooxidation catalysts for small organic molecules are 49 media, given their high volumetric energy density and ease typically non-selective, and mostly based on noble metals 50 of transport, handling and storage.1-6 Furthermore, direct or platinum-group metals (PGM) like Pt or Pd. 51 liquid fuel cells could be an alternative, more practical Furthermore, these catalysts may be subject to poisoning 52 system compared to hydrogen fuel cells. In particular, by intermediate products or impurities.13,14 Noble-metal- 53 direct formic acid fuel cells or direct formate fuel cells free electrocatalysts towards the anodic oxidation of small 54 present higher theoretical cell voltage than fuel cells hydrocarbon molecules are scarce.1,15 Thus, we are 55 running on methanol or ethanol. Additionally, formate can interested in catalysts made of earth-abundant materials 56 be produced from CO2 reduction by renewable energy for the selective aqueous oxidation of formate/formic acid. 57 sources, thereby forming a carbon-neutral fuel cycle.7,8 There are few studies on HCOO–/HCOOH Formate is also highly soluble in water, thereby avoiding electrooxidation where some transition metals are 58 59 60 ACS Paragon Plus Environment ACS Applied Energy Materials Page 2 of 12 employed in conjunction with Pd or Pt.16-19 To our mL K3Fe(CN)6 aqueous solution and heated at 60 °C for 24 1 knowledge, there is only one work, from Bisht et al.,20 hours. Finally, CoFePB sample was immersed in the 2 where a fully noble-metal-free catalyst is employed. In this concentrated H2SO4 solution (pH 1) for more than 4 hours 3 study, the authors tested a perovskite-type catalyst, to remove any possible remaining traces of oxide, and then 4 La0.8Sr0.2CoO3, in a 0.5 M KNO3 electrolyte solution rinsed with MilliQ water. The average CoFePB mass 5 containing HCOOH (2.1 M) and, interestingly, they loading on the electrodes is 0.3 mg cm–2. 6 observed some formic acid oxidation even under open The structure of the CoFePB electrodes was characterized 7 circuit conditions, which was attributed to the by means of aberration-corrected scanning transmission 8 contribution of lattice oxygen. However, the reaction rate electron microscopy (STEM) in high angle annular dark 9 in these conditions was sluggish and only reached a field (HAADF) mode. From the atomic resolution HAADF 10 maximum of ~2 mA cm–2 at 1.4 V vs. RHE. There are few STEM images acquired we obtained Fast Fourier a 11 more non-noble metal catalysts for formic acid oxidation Transform (FFT) power spectra in selected regions and 12 but they work under very limited conditions. For example, indexed the patterns obtaining that the sample structure 13 dehydrogenase enzymes are very sensitive to temperature, was in good agreement with the Co4[Fe(CN)6]2.67·H2O15.33 14 pH, etc. and lack long-term stability.21,22 The only synthetic crystal phase.25 Additional electron energy loss 15 homogeneous electrocatalysts with high activity towards spectroscopy (EELS) chemical composition maps were also 16 formate oxidation, the family of [Ni(PR2NRO2)2(CH3CN)]2+ obtained, corroborating the above composition (Figures 1, 17 catalysts,23,24 can only work in ultrapure organic solvents S1 and S2). 18 19 and are sensitive to trace amounts of water, air or other impurities. b 20 As plausible oxidation catalyst for organic substrates, we (02-2) 21 investigated cobalt hexacyanoferrate (CoFePB), a versatile (020) 22 and robust oxidation catalyst in a wide pH range, with a 23 proven performance in water oxidation (022) 24 (photo)electrocatalysis.25-28 In the present work we have (002) found that CoFePB offers excellent selectivity and 25 efficiency towards aqueous HCOO– oxidation under a pH 26 range of 1-13. Its performance was evaluated 27 Co4(Fe(CN)6)2.67(H2O)15.33-[100] electrochemically and has been compared with that of 28 c d state-of-the-art catalysts, also providing mechanistic 29 [010] insights on the reaction kinetics. To our knowledge, 30 CoFePB is the first earth abundant heterogeneous 31 electrocatalyst for the selective oxidation of [100] 32 formate/formic acid in water, and unique when compared 33 to the commonly used but expensive Pt-or Pd-based 34 catalysts. [001] 35 36 Experimental Section 37 Materials and chemicals. All chemicals used were 38 analytical reagent grade, and included, cobalt (II) nitrate Figure 1. (a) Atomic resolution HAADF STEM image showing 39 hexahydrate Co(NO3)2 6H2O (Aldrich), urea 98 % CH4NO2 the edge a nanocube. (b) Respective power spectrum (FFT) 40 (Aldrich), potassium hexacyanoferrate K3Fe(CN)6 confirming that the nanocube structure is in agreement with 41 (Aldrich), 98 % formic acid solution (Aldrich), 63-67 % the Co4(Fe(CN)6)2.67(H2O)15.33 (FM3–M, Space group: 225) 42 nitric acid (Aldrich) and 98 % sulfuric acid (Aldrich). All crystal phase in the [100] direction. (c-d) Enlarged image from 43 solutions were prepared using 18.2 MΩ·cm Milli-Q water. the green squared section in both grey and temperature 44 If not specified, all commercially available reagents and colored scales. 45 solvents were used as received without further For comparison purposes, a commercial Pt disk electrode 46 purification. (0.07 cm2 geometric area) and a carbon-supported Pd 47 Fluorine-doped tin oxide (FTO) coated glass slides (12 - 14 catalyst with 5% metal content deposited on FTO (1 × 1 cm2 48 Ω per square surface resistivity) were purchased from geometric area) were also electrochemically tested. The 49 Pilkington NSG TEC 15A 2.2 mm slides with 80.0 - 81.5 % Pd(5%)/C ink was prepared by a procedure adapted from 50 transmittance. Pt foil (99.95 %, 0.1 mm thickness) was the literature29 by mixing 30 mg of commercially available 51 purchased from Goodfellow. FAA-3 anionic binder (5% Pd(5%)/Vulcan carbon with 25 µL of 5% FAA-3 anionic 52 solution) was purchased from Fumatech. binder and 9.75 mL of a 3:1 EtOH/water solution and 53 sonicating for 30 minutes. FTO coated glass slides were cut Electrode preparation. The CoFePB electrodes were 54 into 3x1 cm2 strips, then sonicated in a soap water solution prepared following a typical synthetic procedure25 where a 55 thin pink film of CoOx with a 1 × 1 cm2 geometric surface for 10 min, rinsed thoroughly with distilled water, 56 area was initially grown on a transparent FTO surface by sonicated in ethanol 10 min, rinsed thoroughly with 57 hydrothermal method. The CoOx/FTO sample was ethanol, and finally sonicated in acetone for 10 minutes 58 subsequently placed in the freshly prepared 800 mg / 100 before allowing to air dry. The Pd(5%)/C ink was then 59 sprayed onto the FTO substrate that had been masked with 60 ACS Paragon Plus Environment Page 3 of 12 ACS Applied Energy Materials Kapton tape to expose only a 1x1 cm2 surface area on a hot temperature. For sample preparation, 250 μL of the mother 1 plated heated to 80 °C. The electrode was weighed before anolyte were extracted from electrochemical cell with a 2 and after the coating to obtain the mass of added catalyst microsyringe and placed in a vial. Then, 250 μL of a stock 3 and was typically between 1.5 and 2 mg of Pd(5%)/C per 1 solution of ascorbic acid, 0.4 M in D2O were added, and the 4 cm2 electrode surface area. sample was passed to the NMR tube for measurement. 5 Electrochemical methods. All electrochemical Electrochemical ac impedance spectroscopy (EIS). EIS 6 experiments were conducted under ambient conditions, was performed by means of a typical three electrode cell in 7 with a computer-controlled Bio-Logic VMP3 multichannel the frequency range from 100 kHz to 0.1 Hz with 8 points 8 potentiostat. In order to maintain a constant ionic per decade. The AC perturbation was 5 mV. 9 strength, 1 M KNO3 was added to reduce the impact Characterization Methods. The chemical composition 10 stemmed from mass diffusion and to ensure a good ionic and the structure of the compounds were characterized by 11 conductivity. The pH was adjusted by adding aliquotes of Infrared transmittance spectra (FTIR-ATR Cary 630 TR1 12 concentrated KOH or HNO3. The total electrolyte volume Ph2) and Raman spectroscopy (Renishaw in Via Reflex 13 was 30 mL. Raman confocal microscope, Gloucester- Shire, UK, light 14 For our study, we employed a standard three-electrode source: 514 nm). Scanning electron microscopy (SEM) 15 arrangement with a Pt mesh counter electrode, and a micrographs were acquired using a FEI Quanta 650 FEG 16 saturated calomel electrode (SCE) as reference electrode. ESEM at 20 kV equipped with an Oxford EDX analyser 17 The potentiodynamic polarization curves were recorded at (Oxford Instruments). Atomic resolution AC HAADF 18 a scan rate of 20 mV s–1 without magnetic stirring, while the STEM and further EELS-STEM analyses were conducted at 19 quasi-steady-state polarization curves were measured at a the FEI TITAN G3 50-300 PICO operated at 80 kV. Images 20 sweep rate of 5 mV s–1 under magnetic stirring. have been analyzed using the DigitalMicrograph (Gatan 21 Chronoamperometric measurements (CA) were conducted Inc., USA) software package. 22 at a constant applied potential under vigorous stirring in a 23 H-shaped cell with medium porosity glass frits. Steady- Results and Discussion 24 state current densities for Tafel analyses were collected at Formate electrooxidation on CoFePB 25 a variety of applied potentials within formate The oxidation performace of CoFePB was initially 26 electrocatalytic oxidation windows. Working potential was investigated at two different pH, 5 and 13, via cyclic 27 swept from low to high values with 40 mV increments. At voltammetry (CV) of a 0.4 M formate solution in 1 M KNO3 28 each potential, the current was stabilized for 10 minutes to electrolyte (Figure 2). A characteristic formate oxidation 29 attain a steady-state value with the stirred solution (600 wave appears at c.a. 1.2 V vs. RHE at both pH 5 and 13, 30 rpm). All potentials reported in this paper are expressed vs. respectively. In the first case, this onset potential is below 31 RHE unless otherwise indicated by adding +0.241 V and the reversible CoII/CoIII redox couple in CoFePB, c.a. 1.5 V 32 +0.059pH to the potential vs. SCE. Current densities are vs. RHE (see inset of Figure 2a) and in the latter case both always normalized to the corresponding geometric surface 33 potentials are fairly overlapping, since the CoII/CoIII redox area. 34 potential is c.a. 1.17 V vs. RHE at pH 13.25 At both pH values, 35 CO2 evolution efficiency. The theoretical CO2 amount the onset of formate oxidation is well below the oxygen 36 was calculated from the total amount of charge (Q) passed evolution reaction (OER) onset potential, which is c.a. 1.7 37 through the electrode during CA measurements, taking and 1.57 V vs. RHE at pH 5 and 13, respectively. Therefore, 38 into account that two electrons are required to produce the contribution of OER to the observed current can be 39 one CO2 molecule. The experimental amount of CO2 considered minimal, consistent with our observed Faradaic produced was quantitatively determined by using a efficiency (vide infra). 40 calibrated Mass Spectrometer (MS). Mass-to-charge ratios 41 If we compare the results obtained at both pH values, the (m/z) of 2, 12, 14, 16, 17, 18, 22, 28, 30, 32, 44, 46 and 47 were 42 formate oxidation onset potential decreases with pH at 61 continuously monitored with a quadrupole mass 43 mV/pH. Such pH dependence is expected since spectrometer (Pffeifer OmniStar GSD 301 C) during 44 formic/formate oxidation reaction will release protons as different chronopotentiometries carried out in the three- 45 product, as shown in eq. 1: electrode cell under N2 bubbling (50 NmL min–1). The 46 experimental flows were estimated from the MS signals HCOOH  CO2 + 2H+ + 2e– E0 = – 0.20 V (1) 47 obtained for m/z = 22 and 44 , which were previously In terms of overpotential (η), it can be observed that the 48 calibrated with certified standards (Air Liquide) of CO2 electrocatalytic activity of CoFePB is almost pH 49 (99.999 % purity and 1.5 % in N2) and the N2 flow was independent, with an onset of c.a. 1.4 V vs the 50 controlled by a mass flowmeter (Bronkhorst EL-FLOW). thermodynamic potential of formate oxidation in both pH 51 conditions. Similar conclusions were obtained in A chronoamperometry experiment was also performed 52 without N2 flow through the cell in order to perform static preliminar experiments with 0.4 M formate/formic 53 solution in 1 M KNO3 electrolyte and 0.1 KPi buffer, which analysis of the headspace (volume of 23 NmL) with an 54 were carried out in a pH range from 1 to 13 (Figure S3). The Agilent 7820A gas chromatography (GC) system equipped 55 current observed in the presence of KPi was lower than in with columns Washed Molecular Sieve 5A, 2m x 1/8’’ OD, 56 its absence indicating an inhibitory effect of the phosphate Mesh 60/80 SS and Porapak Q, 4m x 1/8’’ OD, SS. Mesh: 57 anion on the catalysis, as confirmed by control 80/100 SS and a Thermal Conductivity Detector. 1H and 13C 58 experiments in the presence and absence of KPi (Figure –NMR (Nuclear Magnetic Resonance) experiments were 59 S4). We hypothesize that this could be attributed to partial performed in a Bruker 300MHz spectrometer at room 60 ACS Paragon Plus Environment ACS Applied Energy Materials Page 4 of 12 blocking of the catalytic sites and/or to displacement of mV/s indicating pseudo-steady state reaction kinetics and, 1 reactive intermediate species by the weak adsorption of importantly, provides evidence against the catalyst 2 phosphate anions. poisoning by intermediates (Figure S5). This apparent 3 The linear pH dependence of the onset potential suggests absence of hysteresis between positive and negative scans 4 that CoFePB is active towards the oxidation of formate over indicates a negligible inhibition rate or a low tendency 5 a wide pH range of 1-13 (Figure S3). Furthermore, current towards surface poisoning under these conditions (on this 6 densities were consistent in all pH range studied, time scale and potential range), in contrast with the 7 indicating negligible poisoning effect of OH– ions on the poisoning by COads species taking place on Pt-based 8 active electrocatalytic sites. This is particularly relevant electrocatalysts32,34 and, to a lesser extent, on Pd-based 9 since OH– anions are known to hinder the catalytic electrocatalysts.17,35 10 oxidation of small molecules oxidation, for example, in Pt 11 based systems where severe poisoning above pH 11 has 12 been reported.30-33 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Figure 3. Anodic polarization data on CoFePB electrodes in 1 35 M KNO3 aquesous electrolyte at 5 mV/s scan rates. Current 36 density as a function of [HCOO–] at (a) pH 5 (b) and pH 13. 37 Figure 2. Cyclic voltammograms (CV) on CoFePB electrodes Selectivity of CoFePB to formate oxidation 38 in (a) pH 5 and (b) pH 13 aqueous solutions in the presence Regarding product selectivity, we analyzed the distribution 39 (red line) and absence (blue line) of 0.4 M HCOO–. Electrolyte of gaseous products during formate electro-oxidation at 40 1 M KNO3. Scan rate 5 mV s–1. Note, oxidation waves that are 1.49 V vs. RHE at pH 5 (chronoamperometry shown in 41 typical of formate oxidation with Pt or Pd are not observed here because CoFePB has a different electrochemical Figure S6b). By means of gas chromatography, only CO2 42 gas was detected, apart from nitrogen, used to initially mechanism vide infra. 43 deaerate the cell headspace volume, and oxygen amounts 44 We then examined the effect of formate concentration in similar to those found when injecting N2 blank samples. No 45 alkaline, pH 13, and near-neutral, pH 5, conditions where signals from other gaseous species were found. It is 46 maximum currents and minimum overpotentials were particularly significant the absence of CO which is a 47 observed, respectively. Figure 3a shows the anodic common by-product from the competing dehydration 48 polarization voltammograms for CoFePB electrodes in pH process. We did monitor quantitatively the CO2 49 5 KNO3 (1M), as a function of formate concentration (up to production in the cell headspace by mass spectrometry at 50 2 M) with a scan rate of 5 mV/s. The presence of the pH 5, assisted by continuous flow of N2 carrier gas through 51 formate oxidation wave is detected from low the cell. We found a Faradaic efficiency towards CO2 52 concentrations (10 mM) with the appearance of anodic production of 100% (within <8% error) (Table 1). The MS 53 current with c.a. 1.2 V onset potential. The anodic current experimental data are shown in detail in Tables S1 and S2. increases with concentration, reaching saturation at The high solubility of CO2 in alkaline media (pH 13) did not 54 formate concentration above 1 M. Analogous results were allow us to easily quantitify CO2 production under these 55 found at pH 13 (Figure 3b), although higher concentration conditions. We also recorded 1H and 13C-NMR spectra of 56 is needed to reach maximum current (> 2 M). the aqueous solutions during the electrooxidation 57 58 For both pH 5 and 13 the electrochemical current is nearly experiment (Figures S7). No other side products, from 59 identical when comparing anodic and cathodic sweeps at 5 60 ACS Paragon Plus Environment Page 5 of 12 ACS Applied Energy Materials more complex coupling reactions during electrooxidation, different oxidation mechanism. This can be explained 1 were detected in the electrolyte solution. either by the simpler reaction pathway since it only 2 The selectivity of CoFePB towards HCOO– oxidation was involves a 2-electron transfer; whereas methanol oxidation 3 additionally investigated by carrying out analogous to CO2, for example, would involve a four electron process 4 aqueous electrolysis, at neutral pH, in the presence of other along multiple proton removal, and attack by water; or also 5 small organic molecules with standard oxidation potentials to the small size and negative charge of formate, making it 6 in a similar range (eq. 2-5), including methanol, ethanol a good ligand to replace water molecules bound to the 7 and acetic acid (Figure 4). active Co2+ sites, whereas the rest of the substrates are less 8 prone to substitute bound water molecules, as first step in CH3OH  CO2 + 6H+ + 6e– E0 = 0.040 V (2) 9 the catalytic cycle. 10 CH3CH2OH + 3H2O  2CO2 + 12H+ + 12e– E0 = 0.085 V (3) Stability of the CoFePB electrocatalyst 11 CH3COOH + 2H2O  2CO2 + 8H+ + 8e– E0 = 0.123 V (4) We also looked into the stability of the CoFePB anodes. 12 Table 1. Theoretical vs experimental CO2 evolution on The catalytic current density is relatively stable during 13 CoFePB electrodes during different chronoamperometry (CA) at both pH 5 and pH 13 for 14 several hours (Figure S6). However, there is a small 15 chronopotentiometries at pH 5, 1 M KNO3 and 0.4 M Formate. decrease in catalytic current that we hypothesize can be 16 attributed mostly to formate consumption in the finite- 17 volume reaction vessel. For example, at pH 13, with 0.4 M 18 Current density/ mA cm–2 5 10 15 formate the current density decreases from ~17 to ~13 mA 19 Theoretical CO2 flow / 0.037 0.075 0.112 cm–2, a 24% decrease, whereas the formate depletion 20 NmL min–1 (estimated from the average current, multiplied by the full 21 time frame of the experiment, and converted to molar Experimental CO2 flow / 0.037 ± 0.077 ± 0.113 ± 22 equivlents by the 2e-/formate stoichiometry) was NmL min–1 0.003 0.002 0.003 23 approximately 9%. Alternatively, the drop in current could 24 be due to a change in solution pH (or local pH gradient), 25 therefore changing the formal oxidation potential of 26 formate. A drop in pH is posible because the solutions are 27 not buffered and H+ is generated as a byproduct of formate 28 so H+ accumulation is likely. Indeed, assuming that all of 29 the charge passed can be attributed to formate oxidation 30 in the 30 mL cell with 1e–/H+ we estimate that the pH of the 31 solution will change by approximately one pH unit. The 32 other CA experiments in Figure S6 show a similar trend in 33 the decrease in current with small variations in the relative 34 amount of current decrease/substrate consumption— 35 other variations in the CA curves can be related to specific 36 reactions occurring at the counter electrode, among other 37 experimental factors. However, in all cases the substrate 38 Figure 4. Cyclic Voltammograms (20 mV s–1) with CoFePB depletion and current drop is within a factor of ~2 meaning working electrode in the presence of small organic molecules that at least half of the current decrease can be accounted 39 (formic acid, methanol, ethanol and acetic acid) in 1 M KNO3 for solely from the depletion of formate oxidation.Thus, 40 aqueous solution at pH 7. although we observe a decrease in catalytic current in finite 41 volume (30 mL) set ups, we expect to maintain the catalytic 42 Interestingly, CoFePB catalytic activity towards these other current with in flow or by addition of formate when 43 substrates was found to be very poor, comparatively. Only depleted. 44 formate shows a significant anodic current at potentials 45 below 1.6 V vs. RHE. More importantly, only the formate Post-electrocatalytic SEM imaging shows no appreciable electrooxidation starts below the reversible CoFePB morphological changes in the CoFePB nanocubes (Figure 46 oxidation and well below the thermodynamic potential of S8a). Nanocubic shape and microporous structure is 47 water oxidation, with all other molecules showing preserved without obvious signs of wearing and only 48 oxidation waves at higher potentials. The difference showing partial aggregation. We observe small changes in 49 between the electrooxidation of formate and the other the IR and Raman spectra that could be due to changes in 50 organic molecules observed with the CoFePB catalyst is the surface or minor structural changes that do not affect 51 the observed electrocatalysis (Figures S8b-c and S9). comparable with that found in literature with noble metal 52 Importantly, the catalyst has largely unchanged bands for catalysts. For instance, the obtained onset potential of 53 the CN stretching from 2100 to 2000 cm–1 region. CN formic acid oxidation is typically 0.1-0.4 V lower compared 54 stretching is very sensitive to coordination number and to that of methanol or ethanol oxidation with Pt or Pd 55 oxidation state of the metal centers,40,41 and any in situ catalysts,36-39 while this potential difference is of c.a. 0.5 V 56 in our case. transformation would considerably affect the spectra. 57 Hence, although minor changes are observed in the Raman 58 The precise reason for this CoFePB selectivity towards spectra, the SEM, electrochemistry and vibrational formate oxidation (Figure 4) should be related to a 59 spectroscopy are all consistent with the conclusion that the 60 ACS Paragon Plus Environment ACS Applied Energy Materials Page 6 of 12 bulk of the material remains unchanged during extended Figure 5. Cyclic Voltammograms (5 mV s–1) with CoFePB, Pt 1 electrolysis and no new catalytically active phases are and Pd(5%)/C catalysts (a) pH 5 and (b) pH 13 aqueous 2 formed. solutions containing 0.4 M formate and 1 M KNO3 . CVs of the 3 Pt and Pd catalysts can be observed in more detail in Figure Comparison with CoOx and benchmarking with PGM S11. 4 electrocatalysts 5 Another relevant feature from the CVs is the presence of 6 In oxidation electrocatalysis, it is known that many materials synthesized for catalysis actually serve as several peaks in both forward and reverse scans on the Pt 7 and Pd catalysts, in the potential range from 0.1 to 1.5 V vs. precatalysts to other heterogeneous phases formed during 8 RHE at pH 5 and from 0.2 to 1.6 V vs. RHE at pH 13. The the oxidation.42-45 It can, therefore, be difficult to 9 presence of these peaks is consistent with literature on determine the genuine catalytic activity of a given material. 10 PGM-based catalysts1,16,18,19,30-34,47-49 and is related to the For that purpose, we compared the catalytic activity of a 11 formate oxidation via either direct or indirect path (in the CoOx film in the same working conditions (Figure S10), as 12 latter case through oxidation of adsorbed reaction most plausible impurity arising from Co leaching, as 13 observed for several oxidation electocatalysts. No intermediates) and to the oxidation/reduction of the PGM 14 additional oxidation events appeard in the presence of catalysts. Hence, adsorption of poisoning species like CO 15 formate, confirming the negligible activity of CoOx towards or hydroxyl groups on Pd and Pt leads to very low current 16 this organic compound. This precludes any participation of densities at high overpotentials, as one can observe in 17 CoOx traces in the excellent catalytic activity of CoFePB. Figure 5 as well as in Table 2 for η = 1.6 V. 18 Furthermore, we did not observe substantial changes in Table 2. Comparison of electrocatalytic activity of 19 the electrochemistry during prolonged electrolysis, nor Pd(5%)/C, Pt disk and CoFePB electrodes obtained by 20 changes in the bulk structure of the CoFePB observed in CV under the conditions shown in Figure 5, in terms 21 the electron microscopy or vibrational spectroscopy. All of of onset overpotential (ηonset), maximum current 22 this evidence taken together provides a strong case that density (jmax) obtained on the forward scan (at an 23 CoFePB is indeed the authentic formate oxidation catalyst. overpotential ηpeak), and current density (j) obtained 24 To benchmark CoFePB as a formate oxidation catalyst we at given overpotentials (η). 25 tested two noble metal catalysts, a Pt disk electrode and a 26 Catalys ηonset jmax j (mA cm–2) at η (V) carbon-supported palladium catalyst (Pd(5%)/C), by CV 27 t (V) (mA cm–2) under the same operating conditions, i.w., with 0.4 M η =1.2 η=1.4 η=1.6 28 formate at pH 5 and 13 at a scan rate of 5 mV/s (Figure 5). Pd/C 0.3 0.6 0.5 0.7 1.4 29 The overpotential values required for the onset of formate (at 0.8 V) 30 oxidation can be observed in Table 2, and follow the trend (pH 5) 31 of Pd(5%)/C