This is the submitted version of the journal article: Li, Junshan; Wang, Xiang; Xing, Congcong; [et al.]. «Electrochemical reforming of ethanol with acetate Co-Production on nickel cobalt selenide nanoparti- cles». Chemical engineering journal, Vol. 440 (Jul.2022), art. 135817. DOI 10.1016/j.cej.2022.135817 This version is available at https://ddd.uab.cat/record/270833 under the terms of the license 0DQXVFULSW &OLFNKHUHWRYLHZOLQNHG5HIHUHQFHV 1 2 1 Electrochemical Reforming of Ethanol with Acetate Co-Production on 3 4 5 2 Nickel Cobalt Selenide Nanoparticles 6 7 8 3 Junshan Li,ab Xiang Wang,c Congcong Xing,c Luming Li,a Shijia Mu,b Xu Han,d Ren He,c Zhifu 9 10 4 Liang,c,d Paulina Martinez,c Yunan Yi,b Qianbao Wu,b Huiyan Pan,e Jordi Arbiol,d,f Chunhua Cui,b* Yu 11 12 5 Zhang,c* Andreu Cabotc,f* 13 14 6 a Institute of Advanced Study, Chengdu University, 610106, Chengdu, China 15 7 b Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of 16 17 8 China, 610054, Chengdu, China. 18 9 c Catalonia Institute for Energy Research - IREC, Sant Adrià de Besòs, 08930, Barcelona, Catalonia, 19 20 10 Spain. 21 22 11 d Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST, Campus UAB, 23 12 Bellaterra, 08193, Barcelona, Catalonia, Spain. 24 25 13 e School of Biological and Chemical Engineering, Nanyang Institute of Science and Technology, 26 14 Nanyang, 473004, China 27 28 15 f ICREA, Pg. Lluís Companys 23, 08010, Barcelona, Catalonia, Spain. 29 30 31 16 Abstract: The energy efficiency of water electrolysis is limited by the sluggish reaction kinetics of the 32 33 34 17 anodic oxygen evolution reaction (OER). To overcome this limitation, OER can be replaced by a less 35 36 37 18 demanding oxidation reaction, which in the ideal scenario could be even used to generate additional 38 39 19 valuable chemicals. Herein, we focus on the electrochemical reforming of ethanol in alkaline media to 40 41 42 20 generate hydrogen at a Pt cathode and acetate as a co-product at a Ni1-xCoxSe2 anode. We first detail 43 44 45 21 the solution synthesis of a series of Ni1-xCoxSe2 electrocatalysts. By adjusting the Ni/Co ratio, the 46 47 48 22 electrocatalytic activity and selectivity for the production of acetate from ethanol are optimized. Best 49 50 23 performances are obtained at low substitutions of Ni by Co in the cubic NiSe2 phase. Density function 51 52 53 24 theory reveals that the Co substitution can effectively enhance the ethanol adsorption and decrease the 54 55 56 25 energy barrier for its first step dehydrogenation during its conversion to acetate. However, we 57 58 59 26 experimentally observe that too large amounts of Co decrease the Faradaic efficiency from values 60 61 62 1 63 64 65 1 27 above 90% to just 50 %. At the optimized composition, the Ni0.75Co0.25Se2 electrode delivers a stable 2 3 28 chronoamperometry current density of up to 45 mA cm-2, corresponding to 1.2 A g-1, in a 1 M KOH + 4 5 6 29 1 M ethanol solution, with a high Faradaic efficiency of 82.2% at a relatively low potential, 1.50 V vs. 7 8 9 30 RHE, and with an acetate production rate of 0.34 mmol cm-2 h-1. 10 11 12 13 31 Keywords: Electrocatalysis; ethanol reforming; hydrogen production; selenide nanoparticle; acetate. 14 15 16 17 32 18 19 20 33 21 22 23 34 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 2 63 64 65 1 35 1. Introduction 2 3 4 36 Molecular hydrogen is a convenient carbon-free energy carrier and a key component in the chemical 5 6 7 37 industry. While molecular hydrogen does not exist in nature, it can be produced by water electrolysis 8 9 10 38 using renewable energy sources such as wind, hydropower, and solar.[1] However, while water 11 12 39 electrolysis systems have been commercialized for several years, their high cost has prevented 13 14 15 40 extensive deployment, and most molecular hydrogen is still being produced by the partial oxidation 16 17 18 41 and steam reforming of natural gas and coal gasification.[2] To turn this trend around, cost-effective 19 20 21 42 strategies for the production of hydrogen from renewable sources need to be developed. In this context, 22 23 43 biomass-derived products provide an excellent alternative source of hydrogen owing to their 24 25 26 44 renewable character and their net-zero CO2 footprint.[3] Besides, the partial oxidation of an organic 27 28 29 45 molecule offers a less energy demanding anodic oxidation reaction, which can contribute to 30 31 32 46 overcoming the energy efficiency limitation of water electrolysis associated with the sluggish reaction 33 34 47 kinetics of the oxygen evolution reaction (OER).[4] An additional advantage of the production of 35 36 37 48 hydrogen from the reforming of waste or biomass-derived organics is the potential co-generation of 38 39 40 49 valuable organic chemicals, which can both improve process economics and further diminish the 41 42 43 50 dependence on fossil resources that are currently used to produce them.[5–7] On top of these 44 45 51 advantages, the replacement of OER by an organic oxidation reaction enables the implementation of 46 47 48 52 coupled cost-effective waste abatement processes. Thus, overall, the clean electrochemical conversion 49 50 51 53 of organic waste and biomass-derived products into value-added chemicals driven by renewable 52 53 54 54 energy has both high fundamental interest and enormous potential for socio-economic and 55 56 57 55 environmental impact.[8–11] 58 59 60 56 In this scenario, (bio)ethanol is a paradigmatic and particularly interesting hydrogen source. Ethanol is 61 62 3 63 64 65 1 57 a key commodity in the chemical industry, used as a precursor or building block for the synthesis of a 2 3 58 plethora of chemicals, including formaldehyde, acetic acid, and plastics.[12,13] Ethanol can be 4 5 6 59 renewably generated from the reduction of CO2 and several biomass-derived feedstocks and organic 7 8 9 60 residues such as sewage sludge.[12-14] As a liquid, ethanol can be easily stored and transported. 10 11 12 61 Besides, bioethanol aqueous solutions can be directly used in electrocatalytic processes, without the 13 14 62 need for purification. Compared with water splitting, the production of hydrogen from ethanol is 15 16 17 63 thermodynamically advantageous ('G0 = +237 kJ·mol-1 for water oxidation vs. 'G0 = +9.6 kJ·mol-1 18 19 20 64 for ethanol oxidation to acetate), which decreases the energy input required.[14] The reforming of 21 22 23 65 ethanol also prevents the H2 and O2 back reaction. Besides, compared with the complete ethanol 24 25 26 66 reforming down to CO2, the production of H2 and acetate could have higher economical profitability 27 28 67 associated with the high economic value of acetate as a side product.[15] 29 30 31 32 68 In terms of catalysts, while OER generally requires high-cost noble metal electrocatalysts, the 33 34 69 electrooxidation of organic molecules, such as methanol or ethanol, in alkaline media can be activated 35 36 37 70 at lower cost using abundant noble-metal-free transition metal electrocatalysts, e.g. nickel, iron and 38 39 40 71 cobalt.[12,16,17] Among them, nickel chalcogenides have demonstrated particularly high activity and 41 42 43 72 selectivity for the conversion of methanol to formate,[18–23] thus we hypothesize them to be excellent 44 45 73 ethanol oxidation catalysts. 46 47 48 49 74 Herein, we detail the preparation of bimetallic Ni-Co selenides nanoparticles (NPs) over a full range of 50 51 52 75 compositions using a solution-based strategy. This set of materials is tested for the partial 53 54 76 electrooxidation of ethanol in an alkaline electrolyte. Finally, the material is optimized to minimize the 55 56 57 77 overpotential and maximize the activity and Faradaic efficiency for ethanol to acetate conversion. 58 59 60 61 62 4 63 64 65 1 78 2. Experimental section 2 3 4 79 2.1. Chemicals 5 6 7 80 All chemicals, including cobalt(II) acetylacetonate (Co(acac)2, 97%, Sigma Aldrich), selenium powder 8 9 10 81 (Se, 200 mesh, 99.5%, Acros Organics), nickel(II) acetylacetonate (Ni(acac)2, 96%), oleylamine (OAm, 11 12 82 C18H37N, 80-90%, Acros Organics), potassium hydroxide (KOH, 85%, Sigma Aldrich), potassium 13 14 15 83 carbonate (K2CO3, 99.5%, Sigma Aldrich), potassium bicarbonate (KHCO3, 99.7%, Sigma Aldrich), 16 17 18 84 1,2-ethanedithiol (EDT, HSCH2CH2SH, 98%, Sigma Aldrich), oleic acid (OAc, C18H34O2, 99%, 19 20 21 85 Sigma Aldrich), carbon black (CB, Vulcan XC72, Sigma Aldrich), and Nafion (10 wt.%, 22 23 86 perfluorinated ion-exchange resin, dispersion in water, Sigma Aldrich), were used as received, without 24 25 26 87 further treatment. Analytical grade chloroform and ethanol for the washing process were obtained 27 28 29 88 from various sources. MilliQ water (18.2 MΩ·cm) was used for electrolyte and ink preparation, and 30 31 32 89 ion chromatography (IC) measurements. An argon-filled glove box was used to handle sensitive 33 34 90 chemicals and for ink preparation. 35 36 37 38 91 2.2. Solution-based synthesis of Ni1-xCoxSe2 NPs 39 40 41 92 Bimetallic selenide NPs were produced from a two-step process involving the preparation of a reactive 42 43 93 ink and its decomposition. First, the precursor ink was prepared inside an argon-filled glove box by 44 45 46 94 dissolving (1-x) mM of Ni(acac)2, x mM of Co(acac)2, and 8 mM of Se powder inside a vial with 30 47 48 49 95 mL of OAm and 3 mL of EDT under vigorous stirring for 1h at 50 °C. The obtained solution was then 50 51 52 96 filtered through a 0.2 μm filter. Then, 36 mL of the prepared precursor together with 60 mL of OAm 53 54 97 and 30 mL OAc were placed in a 250 mL three-neck flask and vacuumed at 60 °C for 30 min followed 55 56 57 98 by heating to 220 °C under Ar at a rate of 5°C/min. After reacting for 20 min, the solution was rapidly 58 59 60 99 cooled down to room temperature in a water bath. Next, the crude solution was mixed with 60 mL of 61 62 5 63 64 65 1 100 chloroform and then centrifuged at 8000 rpm for 5 min. The washing process was repeated by 2 3 101 dispersion in chloroform/ethanol and centrifugation several times. Finally, the obtained product was 4 5 6 102 stored in chloroform and stored in the glovebox. 7 8 9 10 103 2.3. Material characterization 11 12 104 Laboratory X-ray diffraction (XRD) was performed on a Bruker AXS D8 Advance (Cu K radiation: λ 13 14 15 105 = 1.5106 Å) to study the crystal structure. Scanning electron microscopy (SEM) with energy 16 17 18 106 dispersive spectroscopy (EDS) measurements were conducted on a Gemini 300 field emission 19 20 21 107 scanning electron microscope (Zeiss, Germany) equipped with an AZtecOne UltimMax40 energy 22 23 108 spectrometer. High-resolution transition electron microscopy (HRTEM) and scanning TEM (STEM) 24 25 26 109 investigation were performed on a field emission gun FEI Tecnai F20 microscope. High angle annular 27 28 29 110 dark-field (HAADF) STEM was combined with electron energy loss spectroscopy (EELS) in the 30 31 32 111 Tecnai microscope using a Gatan Quantum filter. X-ray photoelectron spectroscopy (XPS) analyses 33 34 112 were conducted on a Specs system. 35 36 37 38 113 2.4. Electrochemical characterization 39 40 41 114 Electrochemical tests were carried out on a Bio-logic SP-200 potentiostat using a conventional 42 43 115 three-electrode system including a Pt wire as the counter electrode, a Hg/HgO reference electrode, and 44 45 46 116 the Ni1-xCoxSe2 NPs supported on CB on top of a 5 mm diameter glassy carbon (GC) as working 47 48 49 117 electrode (WE). To prepare the WE, a catalytic ink was prepared by dispersing 3 mg of dried NPs and 50 51 52 118 6 mg of CB in a vial containing 1.8 ml of MilliQ water/ethanol (v/v = 1:1) and 100 μL of a 10 wt% 53 54 119 Nafion solution, followed by half-an-hour of vigorous sonication. Afterwards, 5 μL of this ink were 55 56 57 120 drop-casted on a carefully polished GC electrode and dried naturally in air. Before the measurement, 58 59 60 121 the alkaline electrolyte was bubbled with high-purity N2 gas for half an hour. Cyclic voltammetry (CV) 61 62 6 63 64 65 1 122 and chronoamperometry (CA) were applied to study the electrocatalytic performance. To analyze the 2 3 123 product of the oxidation reaction at the anode and calculate the faradaic efficiency, 0.5 mL of the 4 5 6 124 solution was collected, diluted 20-fold with MilliQ water, and then analyzed by IC. A freshly prepared 7 8 9 125 4.5 mM KCO3 and 0.8 mM KHCO3 solution was used as leachate solution for the measurement. The 10 11 12 126 amount of product was quantified with a standard curve obtained from known acetate concentrations. 13 14 127 At the cathode, the generated hydrogen was determined by gas chromatography (GC, 8890, Agilent 15 16 17 128 Technologies) connected to the electrocatalytic cell. Before measurement, a stable flow rate of 20 cm3 18 19 20 129 min-1 Ar (99.999%) was bubbled for 30 mins at the Pt wire in an H-type two-compartment cell, which 21 22 23 130 is separated by a Nafion-115 proton membrane. The GC product analysis was conducted every 12 24 25 26 131 mins. The produced gas was identified and quantified by a calibration curve using known 27 28 132 concentrations of standard H2 gas. All obtained potentials (vs. Hg/HgO) were converted to the 29 30 31 133 reversible hydrogen electrode (RHE) reference according to the Nernst equation: ERHE = EHg/HgO + 32 33 34 134 0.059x pH + EϴHg/HgO, where EHg/HgO is the measured potential, EϴHg/HgO is the reference potential of 35 36 37 135 0.098 V, and pH is 13.6 as it corresponds to the 1.0 M KOH solution. All current densities were 38 39 136 referred to the geometrical area of the GC electrode, i.e. 0.196 cm-2. 40 41 42 43 137 2.5 Computational method 44 45 46 138 The Vienna ab initio package (VASP) was used to perform all the density functional theory (DFT) 47 48 49 139 calculations within the generalized gradient approximation (GGA) using the PBE formulation. [24–26] 50 51 52 140 The projected augmented wave (PAW) potential was used to describe the ionic cores and valence 53 54 141 electrons were taken into account using a plane wave basis set with a kinetic energy cutoff of 400 55 56 57 142 eV.[27,28] Partial occupancies of the Kohn−Sham orbitals were allowed using the Gaussian smearing 58 59 60 143 method and a width of 0.05 eV. The electronic energy was considered self-consistent when the energy 61 62 7 63 64 65 1 144 change was smaller than 10−5 eV. A geometry optimization was considered convergent when the force 2 3 145 change was smaller than 0.02 eV/Å. Grimme’s DFT-D3 methodology was used to describe the 4 5 6 146 dispersion interactions.[29] 7 8 9 10 147 Using a 7×11×9 Monkhorst-Pack k-point grid for Brillouin zone sampling, the equilibrium lattice 11 12 148 constants of the monoclinic NiOOH unit cell were optimized to be a=5.168 Å, b=2.847 Å, c=4.516 Å, 13 14 15 149 at α=90°, β=107.1°, γ=90°. They were used to construct a NiOOH(001) surface model (model 1) with 16 17 18 150 p(2×3) periodicity in the x and y directions and 1 stoichiometric layer in the z direction separated by a 19 20 21 151 vacuum layer in the depth of 15 Å to separate the surface slab from its periodic duplicates. In another 22 23 152 model (model 2), three Co atoms were doped into model 1 by replacing three Ni atoms. One O atom 24 25 26 153 was removed from both models to create an oxygen vacancy. During structural optimizations, the 27 28 29 154 gamma point in the Brillouin zone was used for k-point sampling, and all atoms were allowed to relax. 30 31 32 155 The free energy of a gas phase molecule or an adsorbate on the surface was calculated by the equation 33 34 35 156 G = E + ZPE − TS, where E is the total energy, ZPE is the zero-point energy, T is the temperature in 36 37 38 157 kelvin (298.15 K), and S is the entropy. 39 40 41 158 3. Results and discussion 42 43 44 159 3.1. Characterization of catalysts 45 46 47 160 Ni1-xCoxSe2 NPs were produced through a two-step process involving the preparation of a reactive ink 48 49 50 161 from the dissolution of Se and the proper amount of Ni and Co precursors in EDT and OAm and its 51 52 162 posterior reaction (see the experimental section for details and Fig. 1a). Fig. 1b displays the XRD 53 54 55 163 patterns of the obtained Ni1-xCoxSe2, with 0dxd1. For x= 0 and x=1, the crystal structures of NiSe2 and 56 57 58 164 CoSe2 could be properly index with the cubic pa-3 (JCPDS No. 01 088 1771) and the orthorhombic 59 60 61 62 8 63 64 65 1 165 pnnm (JCPDS No. 00 053 0449) phases, respectively. With increasing amounts of Co within the NiSe2 2 3 166 structure, a continuous transition from the cubic NiSe2 phase to the orthorhombic CoSe2 structure was 4 5 6 167 observed. Within the experimental error, SEM-EDS analyses showed the composition of the obtained 7 8 9 168 materials to be consistent with the nominal Ni/Co ratios (Fig. S1). Thus, in the following, we refer to 10 11 12 169 the different samples using the nominal values. 13 14 15 16 a b CoSe2 17 xCo=0.25 18 Intensity (a. u.) xCo=0.50 19 20 xCo=0.25 21 22 EDT OAm filtered NiSe2 23 Co(acac)2 CoSe2 JCPDS NO. 00-053-0449 24 Ni(acac)2 36 mL ink NiSe2 JCPDS NO. 01-088-1711 25 Se powder 60 mL OAm 26 30 mL OAc 27 ink preparation 20 30 40 50 60 70 80 28 2 theta (q) 170 29 30 171 Fig. 1. (a) Schematic drawing of the solution-based process used to produce Ni1-xCoxSe2 NPs. (b) 31 32 172 XRD pattern of the as-synthesized product. 33 34 35 173 Fig. 2a displays a representative TEM micrograph of the Ni0.75Co0.25Se2 sample. It consists of NPs 36 37 38 174 having an average diameter of around 16 nm. When varying the Co composition over the entire Ni/Co 39 40 175 range, the NPs size remained unchanged (Fig. S2). HRTEM micrographs show the NPs to be highly 41 42 43 176 crystalline (Fig. 2b). The lattice parameter of the Ni0.75Co0.25Se2 NPs is a=b=c=5.840 Å, which is in 44 45 46 177 between that of the NiSe2 cubic phase (a=b=c=5.973 Å) and the CoSe2 cubic phase (a=b=c=5.681 Å), 47 48 49 178 in good agreement with a lattice containing a mixture of both cations NixCo1-xSe2. According to 50 51 179 Vegard’s law, and considering a linear approximation, the amount of Co present in the structure is 52 53 54 180 estimated at x=0.46 for the Ni0.75Co0.25Se2 sample. The composition deviation obtained from Vegard’s 55 56 57 181 law can be attributed to the different bonding forces of Ni and Co with selenium within the 58 59 60 182 selenide.[30] As can be seen in the EELS compositional maps (Fig. 2c), the Ni0.75Co0.25Se2 NPs display 61 62 9 63 64 65 1 183 a homogeneous distribution of the three elements, with an average ration Ni/Co=7/3, consistent with 2 3 184 the nominal composition and EDS data. 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 185 27 28 186 Fig. 2. TEM and XPS characterization of Ni0.75Co0.25Se2 NPs. (a) TEM image and size distribution 29 187 histogram. (b) HRTEM micrograph and indexed power spectra. (c) HAADF STEM image and 30 31 188 EELS-STEM compositional maps obtained from HAADF STEM shown region. Individual Ni L2,3-edges 32 189 at 855 eV (red), Co L2,3-edges at 779 eV (red), and Se M1-edges at 231 eV (blue) composition maps. 33 34 190 Colour gradient maps display the relative compositions of Ni and Co. The colour scale is the same for both 35 36 191 panels (Ni and Co). (d) Ni 2p3/2, Co 2p3/2, and Se 3d high resolution XPS spectra. 37 38 192 XPS analysis was used to characterize the surface composition and element oxidation state (Fig. 2d 39 40 41 193 and Fig. S3). Figure 2d displays the high resolution Ni 2p3/2, Co 2p3/2, and Se 3d XPS spectra of 42 43 44 194 Ni0.75Co0.25Se2 NPs. Each of the three elements presents two chemical states that we associate with a 45 46 47 195 metal selenide and a metal oxide environment, in good agreement with previous publications including 48 49 196 an XPS characterization of nickel- and cobalt- selenides.[31,32] Specifically, aside from a satellite 50 51 52 197 peak at 859.8 eV, the Ni 2p3/2 region shows the presence of two chemical states that we associate with 53 54 55 198 Ni ions within Ni-Se (852.6 eV) and within Ni-O (854.7 eV) chemical environments.[33] The Co 2p3/2 56 57 58 199 region also shows two chemical environments associated with Co-Se (778.0 eV) and Co-O (779.4 59 60 61 62 10 63 64 65 1 200 eV).[34] Finally, the Se 3d XPS spectrum shows a component at 53.4 eV associated with a metal 2 3 201 selenide environment and a component at 58.4 eV associated with a selenium oxide environment.[35] 4 5 6 202 The surface oxide component was attributed to the partial oxidation of the NPs surface during XPS 7 8 9 203 sample preparation and transportation.[36] 10 11 12 204 3.2. Electrocatalytic performance 13 14 15 16 205 Working electrodes were prepared by coating a glassy carbon with a catalyst ink prepared by mixing 17 18 19 206 dried NPs and CB in an ethanol/water solution containing Nafion. The electrochemical performance 20 21 207 was explored in a conventional three-electrode electrochemical cell using a Pt wire as the counter 22 23 24 208 electrode and a Hg/HgO reference electrode (see the experimental section for details). 25 26 27 NiSe2 NPs CoSe2 NPs 80 80 a b 28 1 M KOH 1 M KOH Current density (mA cm ) Current density (mA cm ) 29 -2 -2 1 M KOH + 1 M Ethanol 1 M KOH + 1 M Ethanol 30 60 60 2 -2 31 Co3O4 mA cm CoOOH 32 40 40 0 Co3O4 33 NiOOH 34 20 20 -2 0.9 1.2 1.5 35 Ni(OH)2 E (V vs. RHE) 36 0 0 37 38 -20 -20 39 1.0 1.2 1.4 1.6 1.8 1.0 1.2 1.4 1.6 1.8 40 E (V vs. RHE) E (V vs. RHE) 41 209 42 210 Fig. 3. Electrocatalytic performance for (a) NiSe2 NP-based electrodes, and (b) CoSe2 NP-based 43 44 211 electrodes electrolyte containing a 1 M KOH (red curve) and a 1 M KOH + 1 M ethanol (black curve). 45 46 47 212 First, NiSe2- and CoSe2-based electrodes were tested in the potential window 0.9-1.7 V vs. RHE with a 48 49 213 scan rate of 50 mV s-1 in 1 M KOH electrolyte solution. In a basic solution, a hydroxide is formed on 50 51 52 214 the surface of Ni oxides and Ni chalcogenides.[37] As shown in Fig. 3a, during the forward scan, the 53 54 55 215 first current density peak measured from NiSe2–based electrodes (black curve) was attributed to the 56 57 58 216 oxidation of surface Ni(OH)2 to NiOOH at around 1.392 V vs. RHE. At higher applied potential, 59 60 217 above ca. 1.65 V vs RHE, a rise in the current density was associated with the OER. During the 61 62 11 63 64 65 1 218 reverse scan, NiSe2-based electrodes displayed a NiOOH to Ni(OH)2 reduction peak at around 1.380 V 2 3 219 vs. RHE. 4 5 6 50 120 120 7 a Ni1-xCoxSe2 b Ni1-xCoxSe2 c 1 M KOH [@1.55 V] Current density (mA cm ) Current density (mA cm ) Current density (mA cm ) -2 -2 -2 40 1 M KOH + 1 M Ethanol 8 0.75 90 90 9 30 0.25 0.50 60 10 20 0.75 60 11 0.25 30 0.50 10 12 30 13 0 0 [1.0 M KOH] [1.0 M KOH + 1.0 M Ethanol] 14 -10 0 15 1.0 1.2 1.4 1.6 1.8 1.0 1.2 1.4 1.6 1.8 NiSe2 0.25 0.50 0.75 CoSe2 E (V vs. RHE) E (V vs. RHE) x (Ni1-xCoxSe2) 16 0.06 0.06 6 17 d e f Current density (mA cm ) -2 18 0.03 0.03 4 Current (mA) 19 Current (mA) -4 iE0.98=2.71u10 Q+0.01 2 20 0.00 0.00 21 0 -1 22 -0.03 10 mV s -1 -0.03 10 mV s 23 -4 iE0.92=-2.97u10 Q-0.001 -2 -1 100 mV s -1 24 -0.06 -0.06 -4 100 mV s 25 0.90 0.93 0.96 0.99 0 20 40 60 -1 80 100 1.0 1.2 1.4 1.6 E (V vs. RHE) Q (mV s ) 26 E (V vs. RHE) 6 12 6 g 6 Peak current density (mA cm ) 27 Peak current density (mA cm ) h i -2 -2 28 4 4 Γ*(u 10 mol cm ) -2 29 jpa=0.040Q+0.264 D (u cm s ) 8 4 30 2 2 1/2 jpa=0.461Q -0.999 31  -8 0 0 1/2 32 jpc=-0.334Q +0.746 4 2 2 33 -1 -2 -2 jpc=-0.029Q-0.156 34 -4 -4 0 0 35 10 20 30 40 50 7.7 8.4 9.1 9.8 10.5 NiSe2 0.25 0.50 0.75 36 -1 Q(mV s )  -1 1/2 Q (mV s ) x (Ni1-xCoxSe2) 37 220 38 39 221 Fig. 4. (a) CV curves for the Ni1-xCoxSe2 (x=0.25; 0.50; 0.75) electrodes in 1 M KOH electrolyte in the 40 222 potential range from 0.9 V to 1.7 V with a scan rate of 50 mV s-1. (b) CV curves in 1M KOH + 1M 41 42 223 ethanol electrolyte. (c) Current densities in 1 M KOH (green) and 1M KOH + 1M ethanol electrolyte 43 224 (blue) at 1.55 V vs. RHE. (d) CV curves of Ni1-xCoxSe2 (x=0.25) based electrode in 1.0 M KOH in the 44 45 225 non-Faradaic potential range (0.9-1.0 V) with a scan rate of 10-100 mV s-1. (e) Linear fit of the 46 47 226 forward current at 0.98 V and backward current at 0.92 V as a function of scan rate, from 10 to 100 48 227 mV s-1. (f) CV curves of a Ni1-xCoxSe2 (x=0.25) electrode in the potential range from 1.0 V to 1.60 V 49 50 228 with a variable scan rate, from 10-100 mV s-1. (g) Linear fit of the peak current and scan rate from 10 51 229 to 50 mV s-1. (h) Liner fit of the peak current and square root of scan rate 60 to 100 mV s-1. (i) 52 53 230 Comparison of the surface coverage of redox species (*∗) and diffusion coefficient (D) for the 54 55 231 Ni-containing Ni1-xCoxSe2 NPs based electrode in 1 M KOH solution. 56 57 232 For the CoSe2–based electrode in 1M KOH solution (black curve in Fig. 3b), two small current density 58 59 60 233 peaks in the forward scan were observed. In an alkaline medium, the surface of Co-based 61 62 12 63 64 65 1 234 chalcogenides is generally oxidized.[38–40] During the forward scan, the surface cobalt oxide is 2 3 235 further oxidized probably to CoOOH at 1.11 V vs. RHE. At ca. 1.439 V vs RHE, CoOOH is even 4 5 6 236 further oxidized, possibly to CoO2. These phases are reduced during the reverse scan. At ca. 1.60 V vs. 7 8 9 237 RHE, the OER is activated on the CoSe2-based electrode, showing a significantly higher OER activity 10 11 12 238 than the NiSe2–based electrode. Upon adding 1M ethanol to the electrolyte solution, a sharp rise in the 13 14 239 current density (red curves in Fig. 3) and corresponding to the ethanol oxidation reaction (EOR) was 15 16 17 240 measured at ca. 1.35 V for both NiSe2- and CoSe2–based electrodes. Above the OER onset potential, 18 19 20 241 both reactions, OER and EOR, compete with each other [41–43]. 21 22 23 242 Figure 4 displays the results from the electrochemical characterization of the electrodes based on 24 25 26 243 ternary chalcogenides, Ni1-xCoxSe2. In a 1M KOH solution, we observed the OER activity to increase 27 28 29 244 with the amount of Co. Specifically, as shown in Fig. 4a, the potential required to generate 10 mA cm-2 30 31 32 245 decreased from 1.698 V for NiSe2, to 1.672 V for x=0.25, 1.638 V for x=0.50, and 1.620 V for x=0.75. 33 34 246 These values were still above that of CoSe2 electrodes, 1.599 V (Table S1). 35 36 37 38 247 When adding 1M ethanol to the electrolyte, a rapid current density increase at lower potentials was 39 40 41 248 observed. In contrast with the OER, the overpotential for EOR at 10 mA cm-2 did not follow a 42 43 249 monotonous trend with the Co introduction, but we observed the addition of a small amount of Co to 44 45 46 250 result in the highest improvement of the results obtained for NiSe2 (Fig. 4b and Table S1). Figure 4c 47 48 49 251 compares the current density of the different electrodes at a fixed potential, 1.55 V, significantly below 50 51 52 252 the OER onset potential. After subtracting the current density measured in 1 M KOH solution, the net 53 54 253 EOR current density increased from 33.3 mA cm-2 for NiSe2, to 81.6 mA cm-2 for Ni0.75Co0.25Se2, and 55 56 57 254 decreased at higher Co concentrations. If considering the loading mass of NPs, the highest mass 58 59 60 255 current density at 1.55 V was also obtained for the Ni0.75Co0.25Se2 electrode, 2.1 A mg-1, and it was 61 62 13 63 64 65 1 256 significantly above that of recently reported Ni NPs/C,[44] Co3O4 nanosheets,[45] CoNi hydroxide 2 3 257 nanosheets,[46] and fluorine-modified FeOOH,[47] and comparable to that of phosphine doped 4 5 6 258 CoS2,[48] Ni-Fe phosphide and Co3S4 nanosheets (Table S2).[49,50] 7 8 9 10 259 As shown in Fig. 4d, a double-layer pattern for Ni0.75Co0.25Se2 NPs was obtained in 1 M KOH in the 11 12 260 potential window of 0.9 V-1.0 V from 10 mV s-1 to 100 mV s-1. The electrochemical double-layer 13 14 15 261 capacitance (Cdl) in the non-faradaic region was calculated from the linear fit of the current (i) with the 16 17 18 262 scan rate (ν): 19 20 21 263 = 22 23 264 From this capacitance, the electrochemical active area (ECSA) was determined using the following 24 25 26 265 equation: 27 28 29 266 = ⁄ 30 31 32 267 where Cs is 0.04 mF cm-2, according to the values reported for Ni-based electrodes in alkaline 33 34 268 solution.[51] By averaging the slope obtained from the forward scan at 0.98 V and backward scan at 35 36 37 269 0.92 V (Fig. 4e), the calculated ECSA for Ni0.75Co0.25Se2 was 90.1 cm2 g-1, well above the values 38 39 40 270 obtained for NiSe2 (75.7 cm2 g-1), Ni0.50Co0.50Se2 (71.0 cm2 g-1), and Ni0.25Co0.75Se2 (49.7 cm2 g-1) 41 42 43 271 electrodes (Fig. S4 and Table S1). 44 45 46 272 The CV curves of the bimetallic selenides displayed just one peak in the forward scan and one peak in 47 48 49 273 the reverse scan (Fig. S5). The peak current density for the bimetallic selenides was found at lower 50 51 52 274 potentials than for NiSe2 (1.381 V) and the peak potential decreased with the amount of Co introduced, 53 54 275 from 1.371 for NiSe2 to 1.209 V for Ni0.25Co0.75Se2. On the other hand, the voltage of the peak current 55 56 57 276 density in the backward scan did not change monotonically with the amount of Co introduced, but it 58 59 60 277 increased from NiSe2 (1.219 V) to Ni0.25Co0.75Se2 (1.249 V) and decreased at higher Co concentrations. 61 62 14 63 64 65 1 278 Associating these peaks to the Ni(OH)2QNiOOH redox reaction, we observed the potential difference 2 3 279 ΔEp between the oxidation and reduction peaks decreased from 162 mV for NiSe2 to 79, 40, and 20 4 5 6 280 mV with an increasing amount of Co. The smaller potential differences are related to the improved 7 8 9 281 electron transfer kinetics between the electrode surface and the active centres, associated with the 10 11 12 282 modified electronic structure of the bimetallic selenides.[52] 13 14 15 283 The surface coverage of Ni(OH)2/NiOOH redox species was determined from the dependence of the 16 17 18 284 peak current density on the scan rate. Fig. 4h shows CV profiles from the Ni0.75Co0.25Se2 electrode in 1 19 20 21 285 M KOH solution at 10-100 mV s-1. The redox peak associated with the Ni(OH)2QNiOOH redox 22 23 286 reaction can be observed in the positive and negative scans. From the linear relationship between the 24 25 26 287 peak current density and the scan rate from 10-50 mV s-1, the surface coverage of Ni(OH)2/NiOOH 27 28 29 288 redox species (*∗ ) was determined using the following equation: 30 31 32 289 = *∗ 33 4 34 35 36 290 where n is the number of transferred electrons (assumed to be 1 for Ni(OH)2/NiOOH), F is the Faraday 37 38 39 291 constant (96,845 C mol-1), R is the gas constant (8.314 J K-1 mol-1), T is the absolute temperature (295 40 41 42 292 K) and A is the geometric surface area of the GC electrode (0.196 cm2). *∗ was calculated 43 44 293 considering the average of the fitted slopes for the anodic and cathodic peaks. As expected, the surface 45 46 47 294 coverage of redox species, *∗ , measured for the Ni0.75Co0.25Se2 electrode, 3.7 × 10-8 mol cm-2, was 48 49 50 295 well below that of NiSe2, 9.2 × 10-8 mol cm-2, but slightly above the values obtained from materials 51 52 53 296 containing higher amounts of Co (Fig. 4i and S6). These values were consistent with those previously 54 55 297 obtained from Ni-based chalcogenide.[18,53] 56 57 58 59 298 The proton diffusion is generally regarded as a rate-limiting step that controls the Ni(OH)2QNiOOH 60 61 62 15 63 64 65 1 299 redox reaction in Ni-based electrodes.[54] In the scan rate range 60-100 mV s-1 (Figure 4j), both the 2 3 300 anodic and cathodic peak current (ip) were found to be a linear function of the square root of scan rate 4 5 6 301 (v1/2), inferring a diffusion-limited redox reaction within the NiSe2 and the bimetallic selenides. Thus, 7 8 9 302 according to the Randles–Sevcik equation, we investigated the diffusion coefficient (D) of the redox 10 11 12 303 limiting species within the electrodes using the following equation:[55] 13 14 15 ⁄ ⁄ ⁄ 304 = 2.69 × 10 16 17 18 19 305 where n and C are the electron transfer number (assumed to be 1) and the initial concentration of redox 20 21 22 306 species, respectively. The concentration of Ni(II) was found to be 3.97 g cm-3, thus we estimated the C 23 24 307 value to be 0.043 mol cm-3.[42,53] Based on this equation, the diffusion coefficient of the redox 25 26 27 308 limiting species in the Ni0.75Co0.25Se2 electrode was estimated at 1.2×10-9 cm2 s-1, an order of 28 29 30 309 magnitude above that of the other two Ni-Co-based selenide electrodes (Fig. 4i and S6), but well 31 32 33 310 below that of NiSe2, 5.0×10-9 cm2 s-1. 34 35 36 311 The mechanism of the partial ethanol oxidation to acetate involves two main steps: (1) Ni(OH)2 37 38 39 312 activated to NiOOH, and (2) the subsequent oxidation of the adsorbed intermediate:[56–59] 40 41 42 313 (1) OH- + Ni(OH)2 o NiOOH + H2O +e- 43 44 45 314 (2) CH3CH2OH + 3NiOOH + OH- o CH3CHOO- + 3Ni(OH)2 46 47 48 315 The Ni(OH)2 oxidation to NiOOH is generally considered to be the limiting step.[60] The partial 49 50 51 316 substitution of Ni by Co within the Ni2Se certainly implies a slight reduction of the amount of surface 52 53 54 317 NiOOH, but much higher EOR current densities were obtained for Ni0.75Co0.25Se2 when compared with 55 56 57 318 NiSe2. This could be directly related to a higher surface area, better kinetics of NiOOH formation, 58 59 319 and/or more effective ethanol adsorption by the introduction of Co in the NiSe2 structure.[56] 60 61 62 16 63 64 65 50 30 1.2 100 1 a b CH3COO - c Current density (mA cm ) -2 40 1.7 V 29 80 CCH3COO (mmol/L) Faradaic Efficiency (%) 2 0.9 No CH3COO- detected 1.5 V 1.6 V Potential (mV) - 3 30 Cl 28 60 4 0.6 - 20 1.4 V 27 4.4 4.8 40 5 - - 6 10 26 F HCOO 0.3 20 7 0 1.3 V Water 8 25 0.0 0 0 2 4 6 8 10 0 2 4 6 8 10 1.3 1.4 1.5 1.6 1.7 9 3 Time (u10 s) Time (min) E (V vs. Hg/HgO) 10 100 120 120 d e f Current density change (%) Current density (mA cm ) 11 -2 Faradaic Efficiency (%) 80 100 100 12 60 80 80 13 0.25 40 60 60 14 15 20 NiSe2 40 40 16 0.50 20 20 0 0.75 17 CoSe2 0 0 18 0 2 4 6 4 8 10 NiSe2 0.25 0.50 0.75 CoSe2 NiSe2 0.25 0.50 0.75 CoSe2 19 Time (u10 s) x (Ni1-xCoxSe2) x (Ni1-xCoxSe2) 320 20 21 321 Fig. 5. Electrochemical performance of the bimetallic selenide electrode in 1M KOH + 1M ethanol. (a) 22 23 322 CA curves of continuous 2,000 s operation for Ni0.75Co0.25Se2 NPs at each potential of 1.3, 1.4, 1.5, 1.6, 24 25 323 and 1.7 V. (b) IC profile at the end of each 2,000 s intervals. Inset shows the peaks at 4.6 min obtained 26 324 at each potential. Black, red, blue, pink, and green curves were recorded at the 2000th, 4000th, 6000th, 27 28 325 8000th, and 10000th s respectively. (c) Calculated acetate concentration and Faradaic efficiency after 29 326 2,000 s testing at each potential for Ni0.75Co0.25Se2. (d) CA curves for the selenide over the entire 30 31 327 composition range at 1.50 V for 10,000 s testing. (e,f) Corresponding current density change and 32 33 328 faradaic efficiency. 34 35 329 Fig. 5a displays the CA test for Ni0.75Co0.25Se2 electrode in 1M KOH + 1M ethanol at 1.3, 1.4, 1.5, 1.6, 36 37 38 330 and 1.7 V. After reacting for 2000 s, a small portion of the electrolyte was analyzed by IC (see details 39 40 41 331 in the experimental section). As shown in Fig. 5b, the peak at 4.6 min in the IC profile was associated 42 43 44 332 with acetate, which was the only identified product of the ethanol electrooxidation on Ni0.75Co0.25Se2 45 46 333 electrodes. No acetate was detected at 1.3 V, where EOR was not activated on NiOOH. The 47 48 49 334 cumulative acetate concentration in solution was quantified at 0.29, 1.08, 2.01, 2.77 mmol L-1, 50 51 52 335 corresponding to 9.2, 33.5, 61.1, and 82.6 μmol acetate produced at 1.4, 1.5, 1.6, and 1.7 V, 53 54 55 336 respectively. If only considering the acetate production at each potential, the acetate concentration was 56 57 337 found to be 0.29, 0.79, 0.93, and 0.75 mM, equaling to 9.2, 24.2, 27.7, and 21.4 μmol (Fig. 5c and 58 59 60 338 Table S3). The acetate production increased with the applied potential, up to 1.6 V. Above this voltage, 61 62 17 63 64 65 1 339 competition with OER decreased the amount of acetate produced. 2 3 4 340 The Faradaic efficiency (FE) was calculated using the following equation: 5 6 7 × × 8 341 (%) = × 100% ℎ 9 10 11 342 where n is the electron transfer number (assumed to be 4 for the ethanol to acetate reaction), and F is 12 13 14 343 the Faradaic constant (96485 C mol-1). 15 16 17 344 For Ni0.75Co0.25Se2 electrodes, the lowest Faradaic efficiency for the ethanol to acetate was achieved at 18 19 345 1.4 V, and it was found to increase above 80% at 1.6 V, and then decrease at higher potentials due to 20 21 22 346 the activation of the competing OER. More quantitatively, a total charge of 6.9, 12.2, 12.9, and 13.4 C 23 24 25 347 passed through the electrode at 1.4 V, 1.5 V, 1.6 V and 1.7 V, which translated into a Faradaic 26 27 28 348 efficiency of 51.9%, 76.6%, 82.9% and 61.7%, respectively (Table S3). 29 30 31 349 The long-term stability of the electrocatalyst was tested using CA at 1.5 V. As can be seen in Fig. 5d, 32 33 34 350 almost all the electrodes suffered a rapid decay in current density for the first several minutes. Around 35 36 37 351 71.3%, 56.4, 90.6%, 90.4, and 64.6% of the initial current densities were lost after 1x104 s operation 38 39 352 (Fig. 5e). The activity loss is mainly associated with the accumulation of strongly adsorbed 40 41 42 353 intermediate species poisoning the active surface sites. But changes in the surface chemistry of the 43 44 45 354 electrode, including element rearrangement, surface aggregation of the elements and phase evolution 46 47 48 355 in alkaline media at a high external potential could also play an important role.[61–63] 49 50 51 356 For the Ni0.75Co0.25Se2 electrode, a stable current density of 45 mA cm-2, corresponding to 1.2 A g-1, 52 53 54 357 was sustained over the entire test after the initial rapid decay. IC analysis of the solution after 1x104 s 55 56 57 358 operation (Fig. S8) showed the produced acetate amount to reach 182.3 μmol for the Ni0.75Co0.25Se2 58 59 359 electrode, well above the values obtained for NiSe2 (80.2 μmol) and the other Ni-Co selenides (Table 60 61 62 18 63 64 65 1 360 S3). The rate of acetate production on the Ni0.75Co0.25Se2 electrode was 0.34 mmol cm-2 h-1. In terms of 2 3 361 Faradaic efficiency, NiSe2 reached 91.3%, and this value decreased with the incorporation of Co 4 5 6 362 (Table S4). 7 8 9 10 363 CA at 1.5 V in 1M KOH showed the contribution of current not associated to EOR to increase with the 11 12 364 Co addition, from 1.9% for NiSe2 and 1.8% for Ni0.75Co0.25Se2, to values well above 10% for materials 13 14 15 365 containing higher amounts of Co (Fig. S7b and Table S4). Thus the decrease of Faradaic efficiency 16 17 18 366 with the increase of the Co content was associated with the enhanced OER performance. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 367 45 46 368 Fig. 6. (a) DFT optimized adsorbed intermediates on the Co-containing NiOOH surface of 47 48 369 Ni0.75Co0.25Se2 electrode, grey ball for Ni, blue for Co, red for O, orange for H, and wine for C. (b) 49 370 Gibbs free energy diagrams for the ethanol conversion on NiOOH and Co-containing Ni–OOH 50 51 371 surfaces. 52 53 54 372 The EOR electrocatalytic mechanism and the role of Co in the NiSe2 electrode were systematically 55 56 57 373 investigated by DFT. Since NiOOH is the active surface for EOR in alkaline solution, we built two 58 59 374 slab models, NiOOH, and Co-substituted NiOOH. Fig. S9 and 6a show the adsorbed intermediates on 60 61 62 19 63 64 65 1 375 the NiOOH and Co-substituted NiOOH surface, respectively. Fig. 6b compares the corresponding 2 3 376 Gibbs free energy diagram. The ethanol molecule adsorption energy was calculated to be -0.14 eV and 4 5 6 377 -0.30 eV on the NiOOH and Co-substituted NiOOH surface, indicating stronger ethanol adsorption in 7 8 9 378 the presence of cobalt. The presence of cobalt decreases the formation energy of the intermediate 10 11 12 379 CH3CH2O through a dehydrogenation process, from 0.82 eV for NiOOH, to 0.69 eV for 13 14 380 Co-substituted NiOOH. The conversion of CH3CH2O to CH3CHO is energetically downwards for both 15 16 17 381 catalysts, with a free energy change of -0.45 eV for and -0.32 eV for Co-substituted NiOOH. Both 18 19 20 382 catalysts show a similar barrier (0.46 eV for NiOOH and 0.42 eV for Co-NiOOH) for the third 21 22 23 383 dehydrogenation step, which results in adsorbed CH3CO. This dehydrogenation process is generally 24 25 26 384 accompanied by the reduction of NiOOH to Ni(OH)2. Last, adsorbed CH3CO is released when 27 28 385 interacting with OH- ions to form CH3COOH. 29 30 31 120 32 a b th c Signal (a. u.) 5 operation Faradaic efficiency (%) 33 100 34 Signal (a. u.) 35 Cathode H2 Anode st 1 operation 1.0 1.1 1.2 80 36 Time (min) 37 60 38 39 Ar 40 40 HER EOR st nd rd th th 0 1 2 3 4 5 1 2 3 4 5 41 Time (min) GC operation 42 386 43 44 387 Fig. 7. (a) Setup for the H-type cell with two compartments. (b) GC curves within 1 hour GC sampling 45 46 388 at the cathode. (c) Calculated Faradaic efficiency for HER. 47 48 49 389 To detect and quantify the gas product generated at the cathode, an H-type cell with two compartments 50 51 390 separated by a proton exchange membrane was used (Fig. 7a). The cell contained 1 M KOH and 1 M 52 53 54 391 ethanol in both compartments. At the cathode side, high-purity Ar was used to carry the produced gas 55 56 57 392 to the GC. As can be seen in Fig. 7b, two peaks were located at ca. 1.1 and 1.5 min, which were 58 59 60 393 associated to molecular hydrogen and argon. By comparing with known standard hydrogen 61 62 20 63 64 65 1 394 concentrations, the calculated Faradaic efficiency was determined to be nearly 100% (Fig. 7c), 2 3 395 indicating that only hydrogen was produced on the cathode. In the absence and presence of ethanol in 4 5 6 396 the electrolyte, a total charge of 1.6 C and 85.6 C passed through the Ni 0.75Co0.25Se2 electrode and the 7 8 9 397 Pt wire, respectively (Table S4). In the presence of ethanol, over 50-fold higher hydrogen was 10 11 12 398 produced during 10,000 s operation. 13 14 15 399 4. Conclusion 16 17 18 19 400 In summary, we developed a solution-based method to produce Ni1-xCoxSe2NPs with tuned metal 20 21 401 ratios. The electrochemical performance of the materials was tested in 1 M KOH and 1 M KOH + 1 M 22 23 24 402 ethanol aqueous electrolytes. CoSe2 showed enhanced OER activity and NiSe2 demonstrated a more 25 26 27 403 efficient EOR. The incorporation of small amounts of Co to the NiSe2 structure resulted in the highest 28 29 30 404 EOR activities. DFT calculations showed that the presence of Co improved ethanol adsorption and 31 32 405 decreased the barrier for ethanol dehydrogenation. A current density up to 82.3 mA cm-2 was achieved 33 34 35 406 for Ni0.75Co0.25Se2 in 1 M KOH + 1 M ethanol electrolyte at 1.55 V, which is twofold higher than that 36 37 38 407 obtained for NiSe2. Over 10,000s testing, the optimized electrocatalyst delivered an average acetate 39 40 41 408 production rate of 0.34 mmol cm-2 h-1. Additionally, the ethanol to acetate faradaic efficiency was 42 43 409 determined as 82.3%. This work provides a cost-effective approach to the high Faradaic efficiency 44 45 46 410 electrochemical reforming of ethanol with acetate co-production. This work is also an excellent 47 48 49 411 example for other small molecules oxidation and conversion into valuable chemicals. 50 51 52 412 Acknowledgements 53 54 55 56 413 This work was supported by the UESTC start-up funding and the Recruitment Program of Thousand 57 58 59 414 Youth Talents. It was also supported by the European Regional Development Funds and by the 60 61 62 21 63 64 65 1 415 Spanish Ministerio de Economía y Competitividad through the project SEHTOP 2 3 416 (ENE2016-77798-C4-3-R) and NANOGEN (PID2020-116093RB-C43). X. Wang, C. Xing, X. Han, R. 4 5 6 417 He, Z. Liang, and Y. Zhang are grateful for the scholarship from China Scholarship Council (CSC). X. 7 8 9 418 Han and J. Arbiol acknowledge funding from Generalitat de Catalunya 2017 SGR 327. ICN2 10 11 12 419 acknowledges support from the Severo Ochoa Programme (MINECO, Grant no. SEV-2013-0295). 13 14 420 IREC and ICN2 are funded by the CERCA Programme / Generalitat de Catalunya. Part of the present 15 16 17 421 work has been performed in the framework of Universitat Autònoma de Barcelona Materials Science 18 19 20 422 PhD program. 21 22 23 423 Conflict of interest 24 25 26 27 424 The authors declare no competing financial interest. 28 29 30 425 Reference 31 32 426 [1] J.A. Turner, Sustainable hydrogen production, Science (80-. ). 305 (2004) 972–974. 33 427 [2] J.R. McKone, S.C. Marinescu, B.S. Brunschwig, J.R. Winkler, H.B. Gray, Earth-abundant 34 428 hydrogen evolution electrocatalysts, Chem. Sci. 5 (2014) 865–878. 35 36 429 [3] M. Besson, P. Gallezot, C. Pinel, Conversion of biomass into chemicals over metal catalysts, 37 430 Chem. Rev. 114 (2014) 1827–1870. 38 39 431 [4] B. You, Y. Sun, Innovative Strategies for Electrocatalytic Water Splitting, Acc. Chem. Res. 51 40 432 (2018) 1571–1580. 41 433 [5] P.G. Levi, J.M. Cullen, Mapping Global Flows of Chemicals: From Fossil Fuel Feedstocks to 42 43 434 Chemical Products, Environ. Sci. Technol. 52 (2018) 1725–1734. 44 435 [6] V.G. Yadav, G.D. Yadav, S.C. Patankar, The production of fuels and chemicals in the new 45 436 world: critical analysis of the choice between crude oil and biomass vis-à-vis sustainability and 46 47 437 the environment, Clean Technol. Environ. Policy. 22 (2020) 1757–1774. 48 438 [7] J. Garcia-Martinez, Chemistry 2030: A Roadmap for a New Decade, Angew. Chemie - Int. Ed. 49 50 439 60 (2021) 4956–4960. 51 440 [8] L.T. Mika, E. Cséfalvay, Á. Németh, Catalytic Conversion of Carbohydrates to Initial Platform 52 441 Chemicals: Chemistry and Sustainability, Chem. Rev. 118 (2018) 505–613. 53 54 442 [9] Z. Zhang, G.W. Huber, Catalytic oxidation of carbohydrates into organic acids and furan 55 443 chemicals, Chem. Soc. Rev. 47 (2018) 1351–1390. 56 57 444 [10] B. Garlyyev, S. Xue, J. Fichtner, A.S. Bandarenka, C. Andronescu, Prospects of Value-Added 58 445 Chemicals and Hydrogen via Electrolysis, ChemSusChem. 13 (2020) 2513–2521. 59 446 [11] W.J. Liu, Z. Xu, D. Zhao, X.Q. Pan, H.C. Li, X. Hu, Z.Y. Fan, W.K. Wang, G.H. Zhao, S. Jin, 60 61 62 22 63 64 65 447 G.W. Huber, H.Q. Yu, Efficient electrochemical production of glucaric acid and H2 via 1 448 glucose electrolysis, Nat. Commun. 11 (2020) 1–11. 2 3 449 [12] L. Yaqoob, T. Noor, N. Iqbal, A comprehensive and critical review of the recent progress in 4 450 electrocatalysts for the ethanol oxidation reaction, RSC Adv. 11 (2021) 16768–16804. 5 6 451 [13] R.A. Dagle, A.D. Winkelman, K.K. Ramasamy, V. Lebarbier Dagle, R.S. Weber, Ethanol as a 7 452 Renewable Building Block for Fuels and Chemicals, Ind. Eng. Chem. Res. 59 (2020) 4843– 8 453 4853. 9 10 454 [14] J. Dolfing, The microbial logic behind the prevalence of incomplete oxidation of organic 11 455 compounds by acetogenic bacteria in methanogenic environments, Microb. Ecol. 41 (2001) 12 456 83–89. 13 14 457 [15] B. You, X. Liu, X. Liu, Y. Sun, Efficient H2 Evolution Coupled with Oxidative Refining of 15 458 Alcohols via A Hierarchically Porous Nickel Bifunctional Electrocatalyst, ACS Catal. 7 (2017) 16 17 459 4564–4570. 18 460 [16] Y. Tong, X. Yan, J. Liang, S.X. Dou, Metal-Based Electrocatalysts for Methanol 19 461 Electro-Oxidation: Progress, Opportunities, and Challenges, Small. (2019) 1904126. 20 21 462 [17] Y. Xu, M. Liu, S. Wang, K. Ren, M. Wang, Z. Wang, X. Li, L. Wang, H. Wang, Integrating 22 463 Electrocatalytic Hydrogen Generation with Selective Oxidation of Glycerol to Formate over 23 464 Bifunctional Nitrogen-Doped Carbon Coated Nickel-Molybdenum-Nitrogen Nanowire Arrays, 24 25 465 Appl. Catal. B Environ. 298 (2021) 120493. 26 466 [18] J. Li, C. Xing, Y. Zhang, T. Zhang, M.C. Spadaro, Q. Wu, Y. Yi, S. He, J. Llorca, J. Arbiol, A. 27 28 467 Cabot, C. Cui, Nickel Iron Diselenide for Highly Efficient and Selective Electrocatalytic 29 468 Conversion of Methanol to Formate, Small. 17 (2021) 2006623. 30 469 [19] B. Zhao, J. Liu, C. Xu, R. Feng, P. Sui, L. Wang, J. Zhang, J.L. Luo, X.Z. Fu, Hollow NiSe 31 32 470 Nanocrystals Heterogenized with Carbon Nanotubes for Efficient Electrocatalytic Methanol 33 471 Upgrading to Boost Hydrogen Co-Production, Adv. Funct. Mater. 31 (2021) 2008812. 34 472 [20] Q. Luo, M. Peng, X. Sun, A.M. Asiri, In situ growth of nickel selenide nanowire arrays on 35 36 473 nickel foil for methanol electro-oxidation in alkaline media, RSC Adv. 5 (2015) 87051–87054. 37 474 [21] B. Zhao, J. Liu, C. Xu, R. Feng, P. Sui, J.X. Luo, L. Wang, J. Zhang, J.L. Luo, X.Z. Fu, 38 39 475 Interfacial engineering of Cu2Se/Co3Se4 multivalent hetero-nanocrystals for energy-efficient 40 476 electrocatalytic co-generation of value-added chemicals and hydrogen, Appl. Catal. B Environ. 41 477 285 (2021) 119800. 42 43 478 [22] H. Mao, Z. Cao, X. Guo, M. Liu, D. Sun, Z. Sun, H. Ge, Y. Zhang, X.M. Song, Enhanced 44 479 electrocatalytic performance for the oxidation of methanol by hierarchical NiS/Ni(OH) 2 45 480 @polypyrrole/graphene oxide nanosheets, Appl. Surf. Sci. 471 (2019) 355–367. 46 47 481 [23] Y. Yi, J. Li, C. Cui, Trimetallic FeCoNi disulfide nanosheets for CO2-emission-free methanol 48 482 conversion, Chinese Chem. Lett. (2021). https://doi.org/10.1016/j.cclet.2021.07.005 49 50 483 [24] G. Kresse, J. Furthmüller, Efficiency of ab-initio total energy calculations for metals and 51 484 semiconductors using a plane-wave basis set, Comput. Mater. Sci. 6 (1996) 15–50. 52 485 [25] G. Kresse, J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations 53 54 486 using a plane-wave basis set, Phys. Rev. B - Condens. Matter Mater. Phys. 54 (1996) 11169– 55 487 11186. 56 57 488 [26] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made simple, Phys. 58 489 Rev. Lett. 77 (1996) 3865–3868. 59 490 [27] D. Joubert, From ultrasoft pseudopotentials to the projector augmented-wave method, Phys. 60 61 62 23 63 64 65 491 Rev. B - Condens. Matter Mater. Phys. 59 (1999) 1758–1775. 1 492 [28] P.E. Blöchl, Projector augmented-wave method, Phys. Rev. B. 50 (1994) 17953–17979. 2 3 493 [29] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, A consistent and accurate ab initio parametrization 4 494 of density functional dispersion correction (DFT-D) for the 94 elements H-Pu, J. Chem. Phys. 5 6 495 132 (2010) 154104. 7 496 [30] M. Castellanos, A.R. West, Deviations from Vegard’s law in oxide solid solutions. The 8 497 systems Li2TiO3-MgO and Li2TiO3-Na2TiO3, J. Chem. Soc. Faraday Trans. 1 Phys. Chem. 9 10 498 Condens. Phases. 76 (1980) 2159–2169. 11 499 [31] Y. Zhang, C. Zhang, Y. Guo, D. Liu, Y. Yu, B. Zhang, Selenium vacancy-rich CoSe2 ultrathin 12 500 nanomeshes with abundant active sites for electrocatalytic oxygen evolution, J. Mater. Chem. 13 14 501 A. 7 (2019) 2536–2540. 15 502 [32] C. Zhang, J.J. Biendicho, T. Zhang, R. Du, J. Li, X. Yang, J. Arbiol, Y. Zhou, J.R. Morante, A. 16 17 503 Cabot, Combined High Catalytic Activity and Efficient Polar Tubular Nanostructure in 18 504 Urchin-Like Metallic NiCo2Se4 for High-Performance Lithium–Sulfur Batteries, Adv. Funct. 19 505 Mater. 29 (2019) 1903842. 20 21 506 [33] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray photoelectron 22 507 spectroscopy: a reference book of standard spectra for identification and interpretation of XPS 23 508 data, Perkin- Elmer Corp., Minnesota, 1992. 84-85. 24 25 509 [34] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray photoelectron 26 510 spectroscopy: a reference book of standard spectra for identification and interpretation of XPS 27 28 511 data, 1992. 82-83. 29 512 [35] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray photoelectron 30 513 spectroscopy: a reference book of standard spectra for identification and interpretation of XPS 31 32 514 data, Minnesota, 1992. 96-97. 33 515 [36] L. Nguyen, F.F. Tao, Y. Tang, J. Dou, X.J. Bao, Understanding Catalyst Surfaces during 34 516 Catalysis through Near Ambient Pressure X-ray Photoelectron Spectroscopy, Chem. Rev. 119 35 36 517 (2019) 6822–6905. 37 518 [37] D.M. MacArthur, The Hydrated Nickel Hydroxide Electrode Potential Sweep Experiments, 38 39 519 The Electrochemical Society, 1970.117.4. 40 520 [38] J. Li, Z. Luo, F. He, Y. Zuo, C. Zhang, J. Liu, R. Du, X. Yu, T. Zhang, P. Tang, M.F.I. Carrió, 41 521 J. Arbiol, J. Llorca, A. Cabot, Colloidal Ni-Co-Sn Nanoparticles as Efficient Electrocatalysts 42 43 522 for the Methanol Oxidation Reaction, J. Mater. Chem. A. 6 (2018) 22915–22924. 44 523 [39] X. Cui, W. Guo, M. Zhou, Y. Yang, Y. Li, P. Xiao, Y. Zhang, X. Zhang, Promoting effect of 45 524 Co in Ni m Con (m + n = 4) bimetallic electrocatalysts for methanol oxidation reaction, ACS 46 47 525 Appl. Mater. Interfaces. 7 (2015) 493–503. 48 526 [40] S. Rezaee, S. Shahrokhian, Facile synthesis of petal-like NiCo/NiO-CoO/nanoporous carbon 49 50 527 composite based on mixed-metallic MOFs and their application for electrocatalytic oxidation 51 528 of methanol, Appl. Catal. B Environ. 244 (2019) 802–813. 52 529 [41] D. Wu, W. Zhang, D. Cheng, Facile Synthesis of Cu/NiCu Electrocatalysts Integrating Alloy, 53 54 530 Core–Shell, and One-Dimensional Structures for Efficient Methanol Oxidation Reaction, ACS 55 531 Appl. Mater. Interfaces. 9 (2017) 19843–19851. 56 57 532 [42] J. Li, Z. Luo, Y. Zuo, J. Liu, T. Zhang, P. Tang, J. Arbiol, J. Llorca, A. Cabot, NiSn bimetallic 58 533 nanoparticles as stable electrocatalysts for methanol oxidation reaction, Appl. Catal. B Environ. 59 534 234 (2018) 10–18. 60 61 62 24 63 64 65 535 [43] J. Li, Y. Zuo, J. Liu, X. Wang, X. Yu, R. Du, T. Zhang, M.F. Infante-Carrió, P. Tang, J. Arbiol, 1 536 J. Llorca, Z. Luo, A. Cabot, Superior methanol electrooxidation performance of (110)-faceted 2 3 537 nickel polyhedral nanocrystals, J. Mater. Chem. A. 7 (2019) 22036–22043. 4 538 [44] A. Cuña, C. Reyes Plascencia, E.L. da Silva, J. Marcuzzo, S. Khan, N. Tancredi, M.R. Baldan, 5 6 539 C. de Fraga Malfatti, Electrochemical and spectroelectrochemical analyses of hydrothermal 7 540 carbon supported nickel electrocatalyst for ethanol electro-oxidation in alkaline medium, Appl. 8 541 Catal. B Environ. 202 (2017) 95–103. 9 10 542 [45] L. Dai, Q. Qin, X. Zhao, C. Xu, C. Hu, S. Mo, Y.O. Wang, S. Lin, Z. Tang, N. Zheng, 11 543 Electrochemical partial reforming of ethanol into ethyl acetate using ultrathin Co3O4 12 544 nanosheets as a highly selective anode catalyst, ACS Cent. Sci. 2 (2016) 538–544. 13 14 545 [46] W. Wang, Y.B. Zhu, Q. Wen, Y. Wang, J. Xia, C. Li, M.W. Chen, Y. Liu, H. Li, H.A. Wu, T. 15 546 Zhai, Modulation of Molecular Spatial Distribution and Chemisorption with Perforated 16 17 547 Nanosheets for Ethanol Electro-oxidation, Adv. Mater. 31 (2019) 1900528. 18 548 [47] G.F. Chen, Y. Luo, L.X. Ding, H. Wang, Low-Voltage Electrolytic Hydrogen Production 19 549 Derived from Efficient Water and Ethanol Oxidation on Fluorine-Modified FeOOH Anode, 20 21 550 ACS Catal. 8 (2018) 526–530. 22 551 [48] S. Sheng, K. Ye, L. Sha, K. Zhu, Y. Gao, J. Yan, G. Wang, D. Cao, Rational design of Co-S-P 23 552 nanosheet arrays as bifunctional electrocatalysts for both ethanol oxidation reaction and 24 25 553 hydrogen evolution reaction, Inorg. Chem. Front. 7 (2020) 4498–4506. 26 554 [49] S. Sheng, Y. Song, L. Sha, K. Ye, K. Zhu, Y. Gao, J. Yan, G. Wang, D. Cao, Simultaneous 27 28 555 hydrogen evolution and ethanol oxidation in alkaline medium via a self-supported bifunctional 29 556 electrocatalyst of Ni-Fe phosphide/Ni foam, Appl. Surf. Sci. 561 (2021) 150080. 30 557 [50] Y. Ding, Q. Xue, Q.L. Hong, F.M. Li, Y.C. Jiang, S.N. Li, Y. Chen, Hydrogen and Potassium 31 32 558 Acetate Co-Production from Electrochemical Reforming of Ethanol at Ultrathin Cobalt Sulfide 33 559 Nanosheets on Nickel Foam, ACS Appl. Mater. Interfaces. 13 (2021) 4026–4033. 34 560 [51] C.C.L. McCrory, S. Jung, J.C. Peters, T.F. Jaramillo, Benchmarking Heterogeneous 35 36 561 Electrocatalysts for the Oxygen Evolution Reaction, J. Am. Chem. Soc. 135 (2013) 16977– 37 562 16987. 38 39 563 [52] D. Chen, S.D. Minteer, Mechanistic study of nickel based catalysts for oxygen evolution and 40 564 methanol oxidation in alkaline medium, J. Power Sources. 284 (2015) 27–37. 41 565 [53] X. Cui, P. Xiao, J. Wang, M. Zhou, W. Guo, Y. Yang, Y. He, Z. Wang, Y. Yang, Y. Zhang, Z. 42 43 566 Lin, Highly Branched Metal Alloy Networks with Superior Activities for the Methanol 44 567 Oxidation Reaction, Angew. Chemie - Int. Ed. 56 (2017) 4488–4493. 45 568 [54] L. Xiao, J.T. Lu, P.F. Liu, L. Zhuang, J. Yan, Y. Hu, B. Mao, C. Lin, Proton diffusion 46 47 569 determination and dual structure model for nickel hydroxide based on potential step 48 570 measurements on single spherical beads, J. Phys. Chem. B. 109 (2005) 3860–3867. 49 50 571 [55] A.J. Bard, L.R. Faulkner, N. York, C. @bullet, W. Brisbane, S.E. Toronto, Electrochemical 51 572 methods: fundamentals and applications, 2nd editio, John Wiley & Sons, Inc., New York, 52 573 Chichester, Weinheim, Brisbane, Singapore, Toronto, 2001. 53 54 574 [56] A. Kowal, S.N. Port, R.J. Nichols, Nickel hydroxide electrocatalysts for alcohol oxidation 55 575 reactions: An evaluation by infrared spectroscopy and electrochemical methods, Catal. Today. 56 57 576 38 (1997) 483–492. 58 577 [57] M. Fleischmann, K. Korinek, D. Pletcher, The oxidation of organic compounds at a nickel 59 578 anode in alkaline solution, J. Electroanal. Chem. 31 (1971) 39–49. 60 61 62 25 63 64 65 579 [58] M. Fleischmann, K. Korinek, D. Pletcher, The kinetics and mechanism of the oxidation of 1 580 amines and alcohols at oxide-covered nickel, silver, copper, and cobalt electrodes, J. Chem. 2 3 581 Soc. Perkin Trans. 2. (1972) 1396–1403. 4 582 [59] M.T. Bender, Y.C. Lam, S. Hammes-Schiffer, K.S. Choi, Unraveling Two Pathways for 5 6 583 Electrochemical Alcohol and Aldehyde Oxidation on NiOOH, J. Am. Chem. Soc. 142 (2021) 7 584 21538–21547. 8 585 [60] M.A.A. Rahim, R.M.A. Hameed, M.W. Khalil, Nickel as a catalyst for the electro-oxidation of 9 10 586 methanol in alkaline medium, J. Power Sources. 134 (2004) 160–169. 11 587 [61] A.A. Dubale, Y. Zheng, H. Wang, R. Hübner, Y. Li, J. Yang, J. Zhang, N.K. Sethi, L. He, Z. 12 588 Zheng, W. Liu, High-Performance Bismuth-Doped Nickel Aerogel Electrocatalyst for the 13 14 589 Methanol Oxidation Reaction, Angew. Chemie - Int. Ed. 59 (2020) 13891–13899. 15 590 [62] X. Yu, Z. Luo, T. Zhang, P. Tang, J. Li, X. Wang, J. Llorca, J. Arbiol, J. Liu, A. Cabot, 16 17 591 Stability of Pd3Pb Nanocubes during Electrocatalytic Ethanol Oxidation, Chem. Mater. 32 18 592 (2020) 2044–2052. 19 593 [63] Y. Zuo, Y. Liu, J. Li, R. Du, X. Han, T. Zhang, J. Arbiol, N.J. Divins, J. Llorca, N. Guijarro, K. 20 21 594 Sivula, A. Cabot, In Situ Electrochemical Oxidation of Cu2S into CuO Nanowires as a 22 595 Durable and Efficient Electrocatalyst for Oxygen Evolution Reaction , Chem. Mater. 31 (2019) 23 596 7732–7743. 24 25 597 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 26 63 64 65 6XSSOHPHQWDU\0DWHULDO &OLFNKHUHWRDFFHVVGRZQORDG 6XSSOHPHQWDU\0DWHULDO 6,B-XQVKDQGRF[ 'HFODUDWLRQRI,QWHUHVW6WDWHPHQW Advanced Materials Research Area Functional Nanomaterials Group Jardi de les Dones de Negre 1 08930, Sant Adria del Besos, Barcelona (Spain) Tel. +34 93 403 9154 Fax +34 93 402 11 48 Chemical Engineering Journal Dec 16th, 2021 To whom this may concern: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Thank you for your consideration of our potential contribution. Sincerely, Prof. Andreu Cabot Advanced Material Research Department Catalonia Institute for Energy Research - IREC Phone: +34 625615115 E-mail: acabot@irec.cat