This is the submitted version of the article: Li J., Zuo Y., Liu J., Wang X., Yu X., Du R., Zhang T., Infante-Carrió M.F., Tang P., Arbiol J., Llorca J., Luo Z., Cabot A.. Superior methanol electrooxidation performance of (110)-faceted nickel polyhedral nanocrystals. Journal of Materials Chemistry A, (2019). 7. : 22036 - . 10.1039/c9ta07066d. Available at: https://dx.doi.org/10.1039/c9ta07066d Please do not adjust margins ARTICLE Superior Methanol Electrooxidation Performance of (110)-Faceted Nickel Polyhedral Nanocrystals Received 00th January 20xx, Accepted 00th January 20xx Junshan Li,a,b Yong Zuo,a,b Junfeng Liu,a,b Xiang Wang,a,b Xiaoting Yu,a,b Ruifeng Du,a,b Ting Zhang,c Maria F. Infante-Carrió,c Pengyi Tang,c Jordi Arbiol,c,d Jordi Llorca,e Zhishan Luo,a* Andreu Cabota,d,* DOI: 10.1039/x0xx00000x We present the synthesis of (110)-faceted nickel polyhedral nanocrystals (NCs) and their characterization as electrocatalysts for the methanol oxidation reaction (MOR). Ni NCs were produced at 180 C through the reduction in solution of a Ni salt. They were combined with carbon black and Nafion and deposited over glassy carbon to study their electrocatalytic properties. Electrodes based on (110)-faceted Ni NCs displayed a first order reaction with KOH in the concentration range from 0.1 M to 1.0 M. These electrodes were characterized by higher coverages of active species, but lower diffusion coefficients of the species limiting the reaction rate when compared with electrodes prepared from spherical Ni NCs. Overall, electrodes based on faceted Ni NCs displayed excellent performance with very high current densities, up to 61 mA cm-2, and unprecedented mass activities, up to 2 A mg-1, at 0.6 V vs. Hg/HgO in 1.0 M KOH containing 1.0 M methanol. These electrodes also displayed a notable stability. While they suffered an activity loss of ca. 30% during the first 10000 s of operation, afterward activity stabilized at very high current densities, 35 mA cm-2, and mass activities, 1.2 A mg-1, with only a 0.5% decrease during operation from 20000 to 30000 s. excellent performance towards MOR in alkaline media. 10–12 To optimize its performance, Ni has been alloyed with different Introduction elements and it has been combined or grown on a variety of conductive supports.13–16 Additionally, Ni particles with sizes in Direct methanol fuel cells (DMFCs) are characterized by low the nanometer range have been produced to maximize the operating temperatures, a relatively simple architecture, high catalyst surface area and thus the density of surface site. 17–20 energy efficiencies, a high fuel energy density and convenient Besides composition and size, another fundamental parameter fuel storage and transportation.1–5 In spite of their advantages that strongly influence electrocatalytic activity is the catalyst over competing fuel cell technologies, the requirement of Pt- exposed surface facets.21–23 In related Pt catalyst, which has based electrocatalysts restricts their commercialization to been more thoroughly studied, single crystal Pt(110) niche markets, where cost-efficiency is not a major concern.5–8 electrodes with more open atomic structures generally exhibit The widespread implementation of DMFCs requires the higher electrocatalytic activities than other facets.24–27 development of cost-effective electrocatalysts for the However, (110) facets of single crystals are also considered methanol oxidation reaction (MOR).9 In this direction, Ni is more prone to poisoning and thus to provide lower stabilities. considered a main alternative to noble metals, exhibiting We hypothesize that using small and highly faceted NCs, one can exploit the high activity of a preferential facet, while at the same time take advantage of the large density of interfaces a Catalonia Institute for Energy Research - IREC, Sant Adrià de Besòs, Barcelona, and vertices to catalytically remove poisoning species. 08930, Spain. We produced polyhedral Ni NCs with predominant (110) facets b Department of Electronics and Biomedical Engineering, Universitat de Barcelona, 08028 Barcelona, Spain with the aim to take advantage of this a priori high activity c Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST, facet. After removing the insulating ligands, these Ni NCs were Campus UAB, Bellaterra, 08193 Barcelona, Spain d ICREA, Pg. Lluís Companys 23, 08010 Barcelona, Spain mixed with carbon black and Nafion and deposited on a glassy e Institute of Energy Technologies, Department of Chemical Engineering and carbon support. The activity and also the stability of such Barcelona Research Center in Multiscale Science and Engineering. Universitat electrocatalysts toward MOR were subsequently analyzed by Politècnica de Catalunya, EEBE, 08019 Barcelona, Spain Z. Luo current address: Department of Chemistry, Southern University of Science means of cyclic voltammetry (CV) and chronoamperometry and Technology (SUSTech), Shenzhen, Guangdong 518055, P. R. China. (CA). Electronic Supplementary Information (ESI) available: EELS and XPS for polyhedral Ni NCs, FTIR, XRD and HRTEM for spherical Ni NCs and Additional electrochemistry data. See DOI: 10.1039/x0xx00000x Please do not adjust margins Please do not adjust margins ARTICLE Journal Name Experimental Section performed on a ZEISS Auriga SEM with an energy dispersive X- ray spectroscopy (EDS) detector at 20 kV. Transmission Chemicals: Nickel(II) acetylacetonate (Ni(acac)2·xH2O (x2), electron microscopy (TEM) analyses were carried out on a 95%, Sigma-Aldrich), tri-n-octylphosphine (TOP, 97%, Strem), ZEISS LIBRA 120, operating at 120 kV, using a 200 mesh oleylamine (OAm, 80-90%, TCI), borane tert-butylamine Carbon-coated grid from Ted-Pella as substrate. High- complex (TBAB, 97%, Sigma-Aldrich), oleic acid (OAc, Sigma- resolution TEM (HRTEM) and scanning TEM (STEM) studies Aldrich), Nafion (10 wt. %, perfluorinated ion-exchange resin, were carried out using a field emission gun FEI Tecnai F20 dispersion in water), methanol (anhydrous, 99.8%, Sigma- microscope at 200 kV with a point-to-point resolution of 0.19 Aldrich), carbon black (CB, Vulcan XC72), potassium hydroxide nm. High angle annular dark-field (HAADF) STEM was (KOH, 85%, Sigma-Aldrich), trioctylphosphine oxide (TOPO, combined with electron energy loss spectroscopy (EELS) in the 99%, Sigma-Aldrich) and ammonium thiocyanate (NH4SCN, ACS Tecnai microscope by using a GATAN QUANTUM filter. From reagent, ≥ 97.5%, Sigma-Aldrich) were used as received. HRTEM images we obtained the necessary structural info to Hexane, acetone and ethanol were of analytical grade and create the 3D atomic models of the observed Ni polyhedral purchased from various sources. An argon-filled glove-box was nanocrystals by using the Rhodius software package. 29,30 X-ray used for storing and manipulating sensitive chemicals. photoelectron spectroscopy (XPS) analyses were performed on a SPECS system equipped with an Al anode XR50 source Ni polyhedral NCs: All syntheses were conducted using operating at 150 mW and a Phoibos 150 MCD-9 detector. The standard airless techniques. To prepare Ni polyhedral NCs, a pressure in the analysis chamber was kept below 10-7 Pa. The solution of 10 mL OAm and 0.5 mL OAc together with 0.4 area analyzed was about 2 mm x 2 mm. The pass energy of the mmol Ni(acac)2·xH2O were loaded into a 25 mL three-necked hemispherical analyzer was set at 25 eV and the energy step flask connected a vacuum/dry argon gas Schlenk line and was maintained at 0.1 eV. Data processing was performed with containing a magnetic bar. The reaction was strongly stirred the Casa XPS program (Casa Software Ltd., UK). Binding and degassed under vacuum at 80 °C for 1 hour to remove energies were shifted according to the reference C 1s peak water, air, and other low-boiling point impurities. Then, a that was located at 284.8 eV. Fourier transformed infrared gentle flow of argon was introduced. Meanwhile, 0.5 mmol (FTIR) spectroscopy data were recorded on an Alpha Bruker TBAB was dissolved in 0.5 mL OAm and then the mixture was spectrometer before and after ligand exchange. sonicated for half an hour and degassed for an additional hour. Subsequently, the reaction flask was heated to 180 °C within a Ligand exchange: Native organic ligands were displaced from ramp of 5 °C/min. At this temperature, the prepared reductant the NC surface using a NH4SCN solution according to mixture was injected. Upon injection, a visible color change, previously published reports.31 In a typical procedure, 5 mL of from deep green to black was immediately observed. The a 0.13 M (1.0 g NH4SCN in 100 mL acetone) NH4SCN solution in reaction was maintained at this temperature for 1 hour, acetone was added to 5 mL as-synthesized NCs in hexanes and followed by a rapid cool down to room temperature using a the resulting solution agitated for 2 min and maintained water bath. The content of the reaction mixture were unperturbed for another 10 min. The, the solution was centrifuged at 6000 rpm for 3 min with help of acetone as centrifuged at low speed and the supernatant was discarded. polar solvent. The isolated powder was suspended using Further washing process was repeated by centrifuging with hexane and acetone and then centrifuged again. This entire equal volume of acetone and hexane. Finally, the precipitated process was repeated twice. Finally, the NCs were suspended NCs were dried under vacuum overnight. in 5 mL hexane in a vial for further use. Electrochemical characterization: Electrochemical Ni spherical NCs: 13 nm fcc-Ni NCs were prepared following measurements were conducted at room temperature on the procedure described by Y. Chen et al.28 In a typical AutoLab and Metrohm workstations using conventional three- synthesis, 1 mmol Ni(acac)2, 7 ml OAm, 0.4 mmol TOP and 0.25 electrode systems: a counter electrode (Pt mesh), a working mmol TOPO were loaded in a three-neck flask and stirred electrode (glassy carbon) and a reference electrode (Hg/HgO). under a gentle flow of argon. Temperature was raised to 130 The catalyst ink was prepared by mixing 2 mg of NCs together °C and kept for 20 min. Then, the solution was quickly heated with 4 mg of CB in 1.6 mL MilliQ water/ethanol (v/v = 1:1) and to 215 °C and maintained at this temperature for 45 min. 100 μL of a 10 wt% Nafion solution. The working electrode was Subsequently, the flask was cooled down to room temperature prepared by a drop-casting method. Specifically, 5 μL of the using a water bath. The black precipitate was separated catalyst ink was pipetted onto the polished and carefully through centrifugation after adding ethanol. NCs were re- washed glassy carbon electrode, and then it was allowed to dispersed and precipitated three times using hexane and dry naturally in open air. The Hg/HgO reference electrode was ethanol as solvent and non-solvent, respectively. The product placed in a salt bridge of 1.0 M KOH. Prior to each experiment, was finally dispersed in hexane. the alkaline electrolyte was purged with high-purity N2 gas for 30 min. All potential values presented in this paper were Characterization: Powder X-ray diffraction (XRD) was referred to the reference electrode, vs. Hg/HgO. Cyclic measured on a Bruker AXS D8 Advance X-ray diffractometer voltammetry (CV) and chronoamperometry (CA) with Cu K radiation (λ = 1.5106 Å) operating at 40 kV and 40 measurements were performed to investigate the activity and mA. Scanning electron microscopy (SEM) analyses were stability for MOR. The current densities were modified by the 2 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Journal Name ARTICLE geometric surface area of the GC electrode (0.196 cm -2) or the EELS chemical composition maps showed the Ni NCs to have a metal mass loading (~ 5.9 μg NCs). Electrochemical impedance homogeneous composition, although a thin oxygen-rich shell spectroscopy (EIS) studies were carried out in 1.0 M KOH with could be discerned on its surface (Figure S1). This shell was 1.0 M methanol. EIS data was obtained in open circuit voltage related both to the presence of oxygen and oxygen-containing conditions with an AC voltage amplitude of 10 mV and within species bond to the NC surface and to a slight material the frequency range 10 Hz – 1 × 106 Hz. Data was fitted using oxidation during manipulation before TEM analysis. EIS Spectrum Analyzer 1.0.32 XPS analysis showed that the surface of Ni NCs exposed to air presented no less than two Ni chemical states: Ni0 and at least Results and Discussion one oxidized Ni phase which accounted for 70% of the Ni observed (Figure S2). Taking into account the reduced Ni NCs were produced at 180 °C from the reaction of nickel thickness probed by XPS, we estimate the oxidation layer on acetylacetonate with TBAB in the presence of OAm and OAc the surface of the air-exposed NCs to extend around 1-2 (see experimental section for details). Figure 1a shows a nm.33,34 representative TEM micrograph of the produced material. Ni 40 40 NCs displayed highly faceted polyhedral geometries and had a Ni polyhedral NCs b Ni spherical NCs an average size of 16 ± 2 nm, as shown in Figure 1b. XRD 30 30 analysis proved the Ni NCs to have the fcc crystal phase and 1 M KOH + 0.2 M MeOH displayed no peaks corresponding to any additional crystalline I (mA) 20 20 I (mA) phase (Figure 1c). 1 M KOH + 0.2 MeOH 10 10 a b 16 2 nm 0 0 Counts 1 M KOH 1 M KOH 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 E (V vs. Hg/HgO) E (V vs. Hg/HgO) 8 12 16 20 24 c Figure 2. Cyclic voltammograms of Ni electrodes in 1.0 M KOH Size (nm) Ni 01-070-0989 solution in the absence and presences of 0.2 M methanol at a Intensities (a.u.) (111) scan rate of 50 mV s-1: (a) Polyhedral NCs (b) spherical NCs. (200) (220) While the formation of faceted Ni NCs relied on the use of 20 40 2 theta (degrees) 60 80 organic ligands to adjust surface energy and provide suitable crystal growth environment, the presence of these organic d molecules could strongly limit their electrical properties and ability to interact with the media. Therefore, before applying the Ni NCs as electrocatalysts, organic ligands were removed using a 0.13 M NH4SCN solution (see experimental section for [001] 1 n m details).31 After ligand removal, NCs could not be re-dispersed (2-20) (1-1-1) (1-11) in organic solvents such as hexane. Additionally, IR absorption (002) bands corresponding to the C–H vibration modes (2851-2923 cm-1) completely disappeared from the FTIR spectrum (Figure [110] [111] S3). These experimental observations proved the effective 2 0 n m [110] Ni Fm3m removal of organics from the Ni NC surface.35 Figure 1. (a) Representative TEM micrograph of Ni polyhedral NCs. (b) Subsequently, NCs were mixed with carbon black and Nafion in Size distribution histogram. (c) XRD pattern of Ni NCs including the a mixture of water/isopropanol (1:1). The obtained ink was JCPDS 01-070-0989 reference corresponding to the Ni fcc-phase. (d) drop-casted on a glassy carbon electrode (see experimental Representative HRTEM micrograph of a Ni NC, detail of the orange section for details) to produce the electrocatalysts. For the squared region and its corresponding indexed power spectrum, and 3D sake of comparison, 13 nm spherical Ni NCs were prepared atomic models of a NC visualized along two different orientations and according to previous works (see experimental section and displaying a multifaceted morphology with predominant (110) facets. Figures S4-5) and treated in the same manner as the Ni The inset of the HRTEM micrograph displays the same 3D atomic polyhedrons to produce the corresponding electrocatalysts. model visualized from its [110] zone axis. The modeled NC is similar to the one observed experimentally. It presents 6{100}, 8{111} and The electrocatalytic activity of the electrodes prepared from 12{110} facets (the expansion relationship of the cell is 36{110} vs polyhedral and spherical Ni NCs was initially investigated by CV 64{200} vs 48{111}) and is composed by 374,769 Ni atoms. within the potential range 0-1.0 V vs. Hg/HgO, in the presence and absence of 0.2 M methanol in a 1.0 M KOH solution Extensive HRTEM characterization confirmed the Ni NCs to (Figure 2). The current density of polyhedral NC-based display a cubic crystal phase (space group = Fm3m) with a = b = electrodes was systematically and significantly higher than that c = 3.5157 Å (Figure 1d). HAADF-STEM characterization and of electrodes prepared from spherical NCs. This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 3 Please do not adjust margins Please do not adjust margins ARTICLE Journal Name 4 10 Ni(OH)2 + OH-  NiOOH + H2O + e- a b 150 mV s-1 1.0 M KOH NiOOH + CH3OH  Ni(OH)2+ products 2 5 being the possible products/intermediates of the methanol I (mA) I (mA) 0.1 M KOH 10 mV s-1 electrooxidation mainly formic acid, carbonate, formaldehyde, 0 0 CO and CO2. At higher applied voltages, the large increase of current corresponding to the OER adds on.37,38 50 mV s-1 -2 -5 0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8 Note that while in MOR conditions the surface of Ni is oxidized E (V vs. Hg/HgO) E (V vs. Hg/HgO) 10 10 to NiOOH, both for spherical and polyhedral NCs, the atomic c I = 0.05 + 0.65 d I = 0.691/2 - 1.69 distribution and packing, and the related active catalytic sites R2 = 0.99 R2 = 0.98 at the oxidized Ni surfaces will still strongly depend on the 5 5 geometry and exposed facets of the former Ni NCs. I (mA) I (mA) I = -0.02 - 0.13 I = -0.351/2 + 1.34 0 When increasing the sweep rate, the position of the anodic R2 = 0.99 0 R2 = 0.99 peak shifted to higher potential values and the cathodic peak moved to lower potentials (Figure 3 and S6). The current -5 -5 20 40 60 80 9 10 11 12 13 densities of both anodic and cathodic peaks rose with  (mV s ) -1 1/2 (mV s-1)1/2 increasing scan rate. In the low scan rate range (𝑣 < 80 0.06 e 150 mV s-1 0.04 f mV s-1), both peak currents were linearly proportional to the 0.03 scan rate. From the average slope of the anodic and cathodic peaks vs. 𝑣, the surface coverage of redox species (*) was I (mA) IC (mA) 0.00 0.02 R2 = 0.999 estimated (Table 1):39 10 mV s-1 𝑛2 𝐹 2 𝐼𝑝 = ( ) 𝐴 ∗ 𝑣 -0.03 4𝑅𝑇 0.00 -0.06 where n, F, R, T and A are the number of transferred electrons 0.00 0.03 0.06 0.09 0 30 60 90 120 150 E (V vs. Hg/HgO)  (mV s ) -1 (assumed to be 1), the Faraday constant (96845 C mol-1), the gas constant (8.314 J K-1 mol-1), temperature and the geometric surface area of the glassy carbon electrodes (0.196 Figure 3. (a) CV curves from electrodes based on polyhedral Ni cm2), respectively. From this equation, the surface coverage of NCs in 0.1 M, 0.3 M, 0.5 M, 0.7 M and 1.0 M KOH solution at a redox species was found to slightly increase with the KOH scan rate of 50 mV s-1. (b) CV curves from electrodes based on concentration, from 5.7 × 10-8 mol cm-2 in 0.1 M KOH to 1.93 × polyhedral Ni NCs in 1.0 M KOH solution at increasingly higher 10-7 mol cm-2 in 1.0 M KOH. potentials sweep rates: 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150 mV s-1. (c) Linear fitting of anodic and In the high scan rate range (80-150 mV s-1), a linear cathodic peak current densities with the scan rate in the low relationship could be fitted to the dependence of the peak scan rate range (10-80 mV s-1) for polyhedral Ni NCs. (d) Linear current density with the square root of the voltage scan rate. fitting of anodic and cathodic peak current densities to the This dependence is generally related with a diffusion-limited square roots of the scan rates in the high scan rate range (80- Ni(OH)2  NiOOH redox reaction, where the proton diffusion 150 mV s-1) for polyhedral Ni NCs. (e) CVs in the double layer within the particle is considered the diffusion process that region for electrodes based on polyhedral Ni NCs at scan rates limits the reaction rate: 39 of 10, 30, 50, 70, 90, 110, 130, 150 mV s -1 in the non-faradaic range of 0-0.1 V vs. Hg/HgO. (f) Corresponding linear fit of the 𝐼𝑝 = 2.69 × 105 𝑛3⁄2 𝐴𝐷 1⁄2 𝐶𝑣 1⁄2 capacitive current vs. scan rates to calculate C dl and ECSA. where C is the initial concentration of redox species. This equation is valid for a semi-infinite layer. In the case of 16 nm In alkaline media, Ni is oxidized to Ni(OH) 2. When increasing NCs, the related proton diffusion lengths are comparatively the applied voltage, an anodic peak at ca. 0.43 V vs. Hg/HgO is very small. Still, using the above equation, rate limiting ascribed to the oxidation of Ni(OH) 2 to NiOOH.6 In the absence diffusion coefficients at different KOH concentrations were of methanol, at higher potential values, > 0.75 V vs. Hg/HgO, a estimated at 1.19 × 10-10, 0.45 × 10-10, 0.81 × 10-10, 1.20 × 10-10 rise in current density associated with the oxygen evolution and 2.02 × 10-10 cm2 s-1 in 0.1, 0.3, 0.5, 0.7 and 1.0 M KOH from water (OER) is clearly observed. 36 In the reverse scan, solution, respectively. When calculating these parameters for NiOOH is reduced back to Ni(OH)2 at ca. 0.35 V vs. Hg/HgO. spherical NCs (Figure S7), we realized that the coverage * in spherical Ni NCs was much lower than in polyhedral NCs, 3.19 In the presence of methanol, the oxidation of Ni(OH) 2 to × 10-8 mol cm-2 in 1.0 M KOH. On the other hand, the rate NiOOH triggers the electrocatalytic oxidation of methanol, limiting diffusion coefficients was an order of magnitude fusing the increase of current density of the two chemical higher, 1.45 × 10-9 cm2 s-1 in 1.0 M KOH. reactions together: 4 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Journal Name ARTICLE Table 1. Summary of the electrocatalytic performance of Ni polyhedral NCs based electrodes in variable concentration of KOH. CKOH Ipa Epa Ipc Epc ΔE Γ* D (M) (mA) (V vs. Hg/HgO) (mA) (V vs. Hg/HgO) (V) ( 10-7) mol cm-2 ( 10-9) cm2 s-1 0.1 1.021 0.583 -0.862 0.366 0.217 0.57 0.12 0.3 1.833 0.532 -1.225 0.383 0.149 1.09 0.45 0.5 2.556 0.525 -1.395 0.378 0.146 1.47 0.81 0.7 3.040 0.500 -1.400 0.381 0.120 1.71 1.20 1.0 3.331 0.493 -1.222 0.383 0.110 1.93 2.02 Ipa, Epa, Ipc, Epc and ΔE were measured from CVs at a scan rate of 50 mV s-1 in 0.1 M, 0.3 M, 0.5 M, 0.7 M and 1.0 M KOH solution. S10, a linear fit of the logarithmic plot of the current density We investigated the kinetics of the Ni(OH) 2 oxidation to NiOOH versus the methanol concentration for electrodes based on Ni as a function of the KOH concentration for the polyhedral Ni polyhedral NCs pointed toward an apparent methanol reaction NCs by CV (Figure 4 and Table 2). We observed the anodic order of around 0.3 at 0.6 V. peak current (Ipa) to significantly increase and shift to lower potentials (Epa) with the KOH concentration, while the cathodic 150 peak current (Ipc) just moderately raised at a similar potential a 150 b J (mA cm-2) 1.0 M KOH (Epc). Accordingly, the potential difference between the two 100 redox peaks (ΔE) was strongly reduced when increasing the 100 J (mA cm-2) 50 KOH concentration, indicating enhanced electron transfer 01.0 kinetics between the electrode surface and the electrolyte. 0.8 50 ) (M The linear fit of the log j vs log [OH-] plot gave a OH- reaction 0.6 OH 0.4 0.1 M KOH Me order of 1.2 for the Ni(OH)2 to NiOOH reaction, which was C 0.2 consistent with previous literature reports displaying a 0.0 0.2 0.4 0.6 0.0 0 0.0 0.2 0.4 0.6 0.8 1.0 reaction order of 1.40 E (V vs. Hg/HgO) CMeOH (M) Electrochemically active surface areas (ECSA) were estimated c d 120 [KOH] 4 [MeOH] = 1.0 M from the electrochemical double-layer capacitance (Cdl) on the 1.0 M [MeOH] = 1.0 M Spherical NCs basis of CVs recorded at different scan rates in the non- 3 J (mA cm-2) 0.6 V J (A g-1) faradaic potential range 0-0.1 V vs. Hg/HgO (Figure 3.e). 41 80 0.7 M 0.7 V Spherical NCs 2 Plotting the capacitive current (Ic) vs. the scan rate (𝑣) yielded 1.0 M a straight line with a slope equal to C dl (Figure 3f). ECSA was 40 0.5 M 0.3 M 1 calculated by dividing Cdl by the specific capacitance (Cs):42 0.1 M 0 0 𝐸𝐶𝑆𝐴 = 𝐶𝑑𝑙 ⁄𝐶𝑠 H H H H H H 0.0 0.2 0.4 0.6 0.8 KO KO KO KO KO KO M M M M M M where Cs is 0.04 mF cm-2 based on values reported for metal E (V vs. Hg/HgO) 0. 1 0. 3 0. 5 0. 7 1 . 0 1. 0 electrodes in aqueous NaOH solution.43 For polyhedral Ni NC Figure 4. (a) CVs of an electrode based on polyhedral Ni NCs in electrodes in 1.0 M KOH solution, the calculated value of C dl 1.0 M KOH solution with different methanol concentrations, was 0.25 mF cm-2, and ECSA was 6.5 cm-2. from 0.1 M to 1.0 M, at a scan rate of 50 mV s-1. (b) Current Figures 4a and S8 display CV (50 mV s-1) curves of the Ni density at 0.7 V of an electrode based on Ni polyhedral NCs as polyhedral NCs electrocatalyts in a 1.0 M KOH media a function of methanol concentration, from 0.1 M to 1.0 M. (c) containing different methanol concentrations, from 0.1 M to CVs of an electrode based on Ni polyhedral NCs in 0.1 M, 0.3 1.0 M. In addition, the electrochemical response towards M, 0.5 M, 0.7 M and 1.0 M KOH containing 1.0 M methanol at methanol concentration for spherical Ni NCs in 1.0 M KOH a scan rate of 50 mV s-1. Data for an electrode based on Ni containing 1.0 M methanol is displayed in Figure S9. It can be spherical NCs measured in 1.0 M KOH with 1.0 M methanol is observed that the Ni(OH)2 oxidation peak gradually also plotted for comparison. (d) Mass current density of an disappeared within the large current density increase related electrode based on Ni polyhedral NCs in 0.1 M, 0.3 M, 0.5 M, to the MOR when the methanol concentration increased. 0.7 M, 1.0 M KOH and of an electrode based on Ni spherical Figure 4b summarizes the current density (0.7 V vs. Hg/HgO) as NCs in 1.0 M KOH, both in 1.0 M methanol at 0.6 V and 0.7 V. a function of methanol concentration in 0.1-1.0 M KOH media. Note: the mass current density at 0.7 V was corrected by The current density monotonically increased with the subtracting the OER contribution (obtained from concentration of KOH. At the same time, it rose gradually with measurements in the absence of methanol) to the current the methanol concentration until 0.3 M of methanol was densities measured in the presence of KOH and methanol. reached, followed by a rather lazy variation. Specifically, current densities around 110 mA cm-2 were measured in 1.0 M Figure 4c displays the CVs of electrodes based on polyhedral Ni KOH solution containing 0.5 M methanol. As shown in Figure NCs in 0.1 M, 0.3 M, 0.5 M, 0.7 M and 1.0 M KOH and of an This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 5 Please do not adjust margins Please do not adjust margins ARTICLE Journal Name electrode based on spherical Ni NCs in 1.0 M KOH. For these apparently decreased in around a 35% before the catalyst measurements, methanol concentration was set at 1.0 M. The stabilizes. current density in the vertical axis was normalized by the Electrodes based on Ni polyhedral NCs displayed significantly surface area of the electrode. As expected, MOR performances improved stabilities over those based on Ni spherical NCs, were strongly dependent on the alkaline media concentration. which suffered a 58% loss after 10000 s in 1.0 M KOH with 1.0 At equal KOH and methanol concentrations, polyhedral NCs M methanol at 0.6 V vs. Hg/HgO (Figure S11). This provided close to twofold higher current densities than experimental observation, which was unexpected taking into spherical NCs. account previous reports on Pt(110) single crystals, could be Mass current densities obtained in the presence of variable explained by the different surface sites of polyhedral and KOH concentrations and 1.0 M methanol from Ni polyhedral spherical NCs, but additional work is required to exactly asses NCs based electrodes are listed in Table S1. In the same table, the mechanism of stability improvement in electrodes based our results are compared with literature values. At 1.0 M KOH, on Ni polyhedral NCs with predominant (110) facets.44 mass current densities up to 2016.2 mA mg-1 at 0.6 V and 3162 The final shape and composition of the polyhedral Ni NCs after mA mg-1 and 0.7 V were obtained for electrodes based on Ni MOR operation for 10000 s and 30000s was characterized by polyhedral NCs. These values are twofold higher than those HRTEM and EELS-TEM. Figure S12 displays HRTEM micrographs obtained for spherical NCs and well above values previously of polyhedral Ni NCs after 10,000 s and 30,000 s CA test at 0.6 reported in literature. V vs. Hg/HgO. As can be seen, smaller and still partially faceted a b 120 75 NPs were observed after long term MOR test. A closer 1st 60 J = 23.5% 500th investigation on the crystalline structure of the sample after J (mA cm-2) 80 J (mA cm-2) J = 30% 10,000 s test showed the crystallographic structure of the NCs 45 J = 28.7% J = 0.5% to match that of NiO (Figure S13). This result is consistent our 40 30 1000th previous work on the in situ oxidation of copper chalcogenides 15 to CuO during OER.45 0 0 Finally, electrodes based on polyhedral and spherical Ni NCs 0 1 2 3 0.0 0.2 0.4 0.6 0.8 4 Time ( 10 s) were analyzed by EIS in open circuit condition within a solution E (V vs. Hg/HgO) containing 1.0 M KOH and 1.0 M methanol at 0.6 V vs. Hg/HgO Figure 5. (a) CA response of an electrode based on Ni (Figure S14). Data was precisely fitted using the model circuit polyhedral NCs in 1.0 M KOH and 1.0 M methanol at 0.6 V for shown in figure S14. From these fittings, slightly lower charge 30000 s. (b) 1st, 500th, 1000th CV cycles of the electrode based transfer resistances in open circuit were obtained for the on Ni polyhedral NCs in 1.0 M KOH electrolyte with 1.0 M electrode based on spherical Ni NCs compared with that based methanol at a scan rate of 100 mV s−1 in the potential range 0- on polyhedral NCs (Table S2). 0.7 V. The long-term stability of Ni-based electrodes was evaluated Conclusions by chronoamperometry (CA) in a 1.0 M KOH solution Colloidal Ni polyhedral NCs (16 ± 2 nm) with predominant containing 1.0 M methanol at 0.6 V vs. Hg/HgO for 30,000 s (110) facets were synthesized and tested as electrodes for (Figure 5a). The current density of electrodes based on Ni methanol electrooxidation. Results obtained from electrodes polyhedral NCs suffered a relatively important decrease during based on Ni polyhedral NCs were compared with those the first few hours of operation (30% loss after 10,000 s). A obtained for electrodes based on Ni spherical NCs. similar current density decrease was obtained after 1,000 CV Electrocatalytic properties were first investigated in variable cycles (Figure 5b). After this initial lose, the current density concentrations of KOH. We found the Ni(OH)2NiOOH redox quasi-stabilized at a value of ca. 37 mA cm-2. During the last reaction rate was first order with the KOH concentration. The hours of operation tested, from t=20,000 s to t=30,000 s, a surface coverage of active species was much higher in current density decrease of just a 0.5% was obtained. Notice electrodes based on polyhedral Ni NCs than in spherical ones. that the mass current density at which the electrode stabilized, On the other hand, the rate limiting diffusion coefficient was 1.2 A mg-1, was still significantly higher than the maximum higher in electrodes based on spherical NCs. Electrodes based current densities measured from Ni-based electrodes in on Ni polyhedral NCs displayed impressive current densities previous reports. (59.4 mA cm-2) and mass activities (2016 mA mg-1) at 0.6 V vs. The initial current density decay was probably ascribed to the Hg/HgO in the presence of 1.0 M methanol and 1.0 M KOH, blocking of active sites by reaction products that moderately which corresponded to a twofold increase over electrodes strongly bonded to the Ni surface. We hypothesize that after based on spherical Ni NCs and over most Ni-based some hours of operation, the coverage of these species electrocatalyts previously reported. Such superior catalytic stabilized either because of having reached an equilibrium performance should be ascribed to the proper facets of Ni adsorption/desorption or because of the complete blocking of polyhedral NCs, which provided abundant active sites to certain preferential sites. The density of surface reaction sites promote the oxidation of methanol in alkaline media. 6 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Journal Name ARTICLE Electrodes based on faceted polyhedral NCs displayed a 30% 5 A. Serov and C. Kwak, Appl. Catal. B Environ., 2009, 90, loss of activity during the first few operation hours, but activity 313–320. stabilized to around a 65% of the initial value after ca. 20000 s 6 X. Cui, W. Guo, M. Zhou, Y. Yang, Y. Li, P. Xiao, Y. Zhang and of operation. These results suggest Ni polyhedral NCs with X. Zhang, ACS Appl. Mater. Interfaces, 2015, 7, 493–503. 7 Y. Qin, Y. Liu, F. Liang and L. Wang, ChemSusChem, 2015, 8, predominant (110) facets to be among the best candidate 260–263. electrocatalysts for MOR. 8 Z. Daşdelen, Y. Yıldız, S. Eriş and F. Şen, Appl. Catal. B Environ., 2017, 219, 511–516. 9 X. Cui, P. Xiao, J. Wang, M. Zhou, W. Guo, Y. Yang, Y. He, Z. Conflicts of interest Wang, Y. Yang, Y. Zhang and Z. Lin, Angew. Chemie - Int. There are no conflicts to declare. Ed., 2017, 56, 4488–4493. 10 R. M. Abdel Hameed and R. M. El-Sherif, Appl. Catal. B Environ., 2015, 162, 217–226. Author contributions 11 H. Sun, Y. Ye, J. Liu, Z. Tian, Y. Cai, P. Li and C. Liang, Chem. Commun., 2018, 54, 1563–1566. The manuscript was prepared through the contribution of all 12 G. S. Ferdowsi, S. A. Seyedsadjadi and A. Ghaffarinejad, J. authors. A. Cabot and Z. Luo conceived and guided the project, Nanostructure Chem., 2015, 5, 17–23. and supervised the work. J. Li designed the experiments, 13 J. Li, Z. Luo, Y. Zuo, J. Liu, T. Zhang, P. Tang, J. Arbiol, J. produced the NCs, conducted XRD, TEM and FT-IR Llorca and A. Cabot, Appl. Catal. B Environ., 2018, 234, 10– characterization, electrochemical measurements, and wrote 18. 14 J. Li, Z. Luo, F. He, Y. Zuo, C. Zhang, J. Liu, X. Yu, R. Du, T. the first draft of the manuscript. X. Wang performed the Zhang, M. F. Infante-Carrió, P. Tang, J. Arbiol, J. Llorca and sample after the long-term CA test for HRTEM. Y. Zuo, J. Liu, X. A. Cabot, J. Mater. Chem. A, 2018, 6, 22915–22924. Yu and R. Du significantly contributed to the results discussion. 15 N. A. M. Barakat, M. Motlak, B. S. Kim, A. G. El-Deen, S. S. T. Zhang, M. F. Infante-Carrió, P. Tang and J. Arbiol performed Al-Deyab and A. M. Hamza, J. Mol. Catal. A Chem., 2014, structural and compositional NCs characterization by means of 394, 177–187. HRTEM and EELS, and discussed the results. J. Llorca measured 16 I. S. Pieta, A. Rathi, P. Pieta, R. Nowakowski, M. Hołdynski, and discussed XPS data. The manuscript was corrected and M. Pisarek, A. Kaminska, M. B. Gawande and R. Zboril, Appl. Catal. B Environ., 2019, 244, 272–283. improved by all authors. 17 Y. Yu, Q. Yang, X. Li, M. Guo and J. Hu, Green Chem., 2016, 18, 2827–2833. 18 A. Roy, H. S. Jadhav, G. M. Thorat and J. G. Seo, New J. Acknowledgements Chem., 2017, 41, 9546–9553. This work was supported by the European Regional 19 X. Li, X. Niu, W. Zhang, Y. He, J. Pan, Y. Yan and F. Qiu, Development Funds and by the Spanish Ministerio de ChemSusChem, 2017, 10, 976–983. 20 L. Yang, Y. Tang, S. Luo, C. Liu, H. Song and D. Yan, Economía y Competitividad through the project SEHTOP ChemSusChem, 2014, 7, 2907–2913. (ENE2016-77798-C4-3-R) and VALPEC (ENE2017-85087-C3). J. 21 S.-L. Yau, F.-R. F. Fan, T. P. Moffat and A. J. Bard, J. Phys. Li and T. Zhang thank the China Scholarship Council (CSC) for Chem., 1994, 98, 5493–5499. scholarship support. T. Zhang and J. Arbiol acknowledge 22 Z. Luo, J. Lu, C. Flox, R. Nafria, A. Genç, J. Arbiol, J. Llorca, funding from Generalitat de Catalunya 2017 SGR 327. ICN2 M. Ibáñez, J. R. Morante and A. Cabot, J. Mater. Chem. A, acknowledges support from the Severo Ochoa Programme 2016, 4, 16706–16713. (MINECO, Grant no. SEV-2013-0295). IREC and ICN2 are 23 G. You, J. Jiang, M. Li, L. Li, D. Tang, J. Zhang, X. C. Zeng and funded by the CERCA Programme / Generalitat de Catalunya. R. He, ACS Catal., 2018, 8, 132–143. Part of the present work has been performed in the 24 H. Kita, Y. Gao, S. Ye and K. Shimazu, Bull. Chem. Soc. Jpn., framework of Universitat Autònoma de Barcelona Materials 2006, 66, 2877–2882. 25 J. C. Davies, B. E. Hayden and D. J. Pegg, Surf. Sci., 2000, Science PhD program. JL is a Serra Húnter Fellow and is 467, 118–130. grateful to ICREA Academia program and to MINECO/FEDER 26 J. Clavilier, A. Rodes, K. El Achi and M. Zamakhchari, J. grant ENE2015-63969-R. Chim. Phys., 2017, 88, 1291–1337. 27 C. F. Mai, C. H. Shue, Y. C. Yang, L. Y. O. Yang, S. L. Yau and K. Itaya, Langmuir, 2005, 21, 4964–4970. 28 Y. Chen, X. Luo, H. She, G.-H. Yue and D.-L. Peng, J. Nanosci. Notes and references Nanotechnol., 2009, 9, 5157–5163. 1 R. G. Minet, Science (80-. )., 1967, 157, 1373. 29 J. Arbiol, A. Cirera, F. Peiró, A. Cornet, J. R. Morante, J. J. 2 X. Zhao, M. Yin, L. Ma, L. Liang, C. Liu, J. Liao, T. Lu and W. Delgado and J. J. Calvino, Appl. Phys. Lett., 2002, 80, 329– Xing, Energy Environ. Sci., 2011, 4, 2736. 331. 3 J. N. Tiwari, R. N. Tiwari, G. Singh and K. S. Kim, Nano 30 S. Bernal, F. J. Botana, J. J. Calvino, C. López-Cartes, J. A. Energy, 2013, 2, 553–578. Pérez-Omil and J. M. Rodríguez-Izquierdo, Ultramicroscopy, 4 N. Kakati, J. Maiti, S. H. Lee, S. H. Jee, B. Viswanathan and 1998, 72, 135–164. Y. S. Yoon, Chem. Rev., 2014, 114, 12397–12429. 31 A. T. Fafarman, W. K. Koh, B. T. Diroll, D. K. Kim, D. K. Ko, S. J. Oh, X. Ye, V. Doan-Nguyen, M. R. Crump, D. C. This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 7 Please do not adjust margins Please do not adjust margins ARTICLE Journal Name Reifsnyder, C. B. Murray and C. R. Kagan, J. Am. Chem. Soc., 40 P. Oliva, J. Leonardi, J. F. Laurent, C. Delmas, J. J. 2011, 133, 15753–15761. Braconnier, M. Figlarz, F. Fievet and A. de Guibert, J. Power 32 G. A. Bondarenko, A.S.; Ragoisha, Nova Science Publishers: Sources, 1982, 8, 229–255. New York, New York, 2005, pp. 89–102. 41 J. D. Benck, Z. Chen, L. Y. Kuritzky, A. J. Forman and T. F. 33 J. Li, X. Xu, Z. Luo, C. Zhang, Y. Zuo, T. Zhang, P. Tang, M. F. Jaramillo, ACS Catal., 2012, 2, 1916–1923. Infante-Carrió, J. Arbiol, J. Llorca, J. Liu and A. Cabot, 42 C. C. L. McCrory, S. Jung, I. M. Ferrer, S. M. Chatman, J. C. ChemSusChem, 2019, 12, 1451–1458. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2015, 137, 34 J. Li, X. Xu, Z. Luo, C. Zhang, X. Yu, Y. Zuo, T. Zhang, P. Tang, 4347–4357. J. Arbiol, J. Llorca, J. Liu and A. Cabot, Electrochim. Acta, 43 C. C. L. McCrory, S. Jung, J. C. Peters and T. F. Jaramillo, J. 2019, 304, 246–254. Am. Chem. Soc., 2013, 135, 16977–16987. 35 J. Liu, Z. Luo, J. Li, X. Yu, J. Llorca, D. Nasiou, J. Arbiol, M. 44 C. Cui, L. Gan, M. Heggen, S. Rudi and P. Strasser, Nat. Meyns and A. Cabot, Appl. Catal. B Environ., 2019, 242, Mater., 2013, 12, 765–771. 258–266. 45 Y. Zuo, Y. Liu, J. Li, R. Du, X. Han, T. Zhang, J. Arbiol, N. J. 36 D. Wu, W. Zhang and D. Cheng, ACS Appl. Mater. Divins, J. Llorca, N. Guijarro, K. Sivula and A. Cabot, Chem. Interfaces, 2017, 9, 19843–19851. Mater., 2019, acs.chemmater.9b02790. 37 Z. Liu, H. Tan, D. Liu, X. Liu, J. Xin, J. Xie, M. Zhao, L. Song, L. Dai and H. Liu, Adv. Sci., , DOI:10.1002/advs.201801829. 38 Z. Liu, H. Tan, J. Xin, J. Duan, X. Su, P. Hao, J. Xie, J. Zhan, J. Zhang, J. J. Wang and H. Liu, ACS Appl. Mater. Interfaces, 2018, 10, 3699–3706. 39 A. J. Bard and L. R. Faulkner, Electrochemical Methods: Fundamentals and Applications, 2nd edn., 2001. TOC Methanol oxidation reaction in alkaline media (110)-faceted Ni Methanol 1.0 M KOH 0.0 0.2 0.4 0.6 0.8 E (V vs. Hg/HgO) Colloidal (110)-faceted nickel polyhedral nanocrystals present superior methanol electrooxidation performance in alkaline media. 1 0 0 n m 8 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins