Tunable e g Orbital Occupancy in Heusler Compounds for Oxygen Evolution Reaction

Abstract Heusler compounds have potential in electrocatalysis because of their mechanical robustness, metallic conductivity, and wide tunability in the electronic structure and element compositions. This study reports the first application of Co2 YZ‐type Heusler compounds as electrocatalysts for the oxygen evolution reaction (OER). A range of Co2 YZ crystals was synthesized through the arc‐melting method and the e g orbital filling of Co was precisely regulated by varying Y and Z sites of the compound. A correlation between the e g orbital filling of reactive Co sites and OER activity was found for Co2MnZ compounds (Z=Ti, Al, V, and Ga), whereby higher catalytic current was achieved for e g orbital filling approaching unity. A similar trend of e g orbital filling on the reactivity of cobalt sites was also observed for other Heusler compounds (Co2VZ, Z=Sn and Ga). This work demonstrates proof of concept in the application of Heusler compounds as a new class of OER electrocatalysts, and the influence of the manipulation of the spin orbitals on their catalytic performance.


Synthesis of Heusler compounds.
Polycrystalline ingots of Heusler alloys studied in this work were synthesized by arc melting stoichiometric amounts of the constituent high-purity elements (99.999 %) in an arc furnace with a water-cooled Cu hearth under an Ar atmosphere. Deailed synthesis procedure can be found elsewhere. [1] To increase the volume homogeneity, all the ingot were remelted more than two times. The final weight loss of the samples was less than 0.5% of the initial weight.
For the synthesis of bulk single crystals, the induction-melted samples were crushed into fine powders and then packed in a custom-designed sharp-edged alumina tube that was sealed in a tantalum tube. Take the growth of Co2MnGa crystal as an example, the tube was heated to 1523 K and soaked for 10 h to ensure homogeneity of the melt and then slowly cooled to 1023 K.
Irregularly shaped crystals of Heusler compounds for electrochemical studies ( Figure S1) were crushed into fine powders through a ball milling process. Each sample was processed in a planetary ball mill with keeping a rotating speed at 500 rpm for 10 minutes.

Electrochemical Measurements.
Electrochemical measurements were performed in a three-electrode configuration using a rotating disc electrode (Model: AFMSRCE, PINE Research Instrumentation). A hydrogen reference electrode (HydroFlex, Gaskatel) and Pt wire were used as reference electrode and counter electrode, respectively. All measurements were carried out in 1 M KOH electrolyte in a Teflon cell. The temperature of the cell was controlled at 25 o C using a water circulation system. Prior to the electrochemical measurement, argon was purged through the cell for 30 min to remove oxygen from the electrolyte. During all measurements, argon was continuously purged to remove generated oxygen. Working electrodes were fabricated by depositing target materials on glass carbon (GC) electrodes (PINE, 5 mm diameter, 0.196 cm 2 area). Before use, a thorough cleaning was done on the surface of GC electrodes by polishing with Al2O3 suspension (5 and 0.25 μm, Allied High Tech Products, Inc.). Working electrodes were fabricated by drop-casting catalyst ink on GC electrodes. In detail, 4.8 mg of powder sample was first dispersed in a mixed solution containing 0.75 mL of H2O, 0.25 mL of 2-propanol (Aldrich, 99.5 %) and 50 μL of Nafion (5% in a mixture of water and alcohol). Afterward, the mixture solution was immersed in a sonication bath for 30 min to form a homogeneous ink. Finally, 5.25 μL of catalyst ink was dropped onto the GC electrode and dried under argon atmosphere for 30 min. The catalyst loading on GC electrodes was calculated to be around 0.12 mg/cm 2 following this procedure. For the stability test, an electrode was fabricated by dropping 110 μL on a carbon fiber paper (1 cm × 1 cm), with a catalyst loading of around 0.5 mg/cm 2 .
After dipping GC electrodes into KOH electrolyte, the linear sweep voltammetry (LSV) curves were collected by sweeping the potential from 0.7 V to 1.7 VRHE with a scan rate of 10 mV/s. To minimize the effect of the generated oxygen bubble, a rotating disc electrode configuration was kept a rotation speed of 2000 rpm. The IR drop was compensated at 85 % automatically via the potentiostat software (EC-Lab V11.01).
The value of ECSA was determined by measuring the non-Faradaic capacitance current from the scan-rate dependence of CVs. CV scans with increasing scan rates, from 20 to 180 mV/s, were collected in a non-Faradaic region (0.9 -1 VRHE). By plotting the capacitive current (janode -jcathode) against the scan rate and fitting with a linear fit, the value of Cdl can be estimated as half of the slope. The ECSA of each sample was calculated according to this equation: ECSA = Cdl/Cs, where Cs is the specific capacitance. In this work, 0.04 mF/cm 2 was chosen as the reference value for the measurements in 1M KOH solution. [2] Characterization. Powder X-ray diffraction (XRD) patterns of Co2MnX compounds were collected on a Stoe STADI P transmission diffractometer equipped with a primary Ge (111) monochromator (MoKα1) and a position-sensitive detector. For Co2VX compounds, their XRD patterns were collected on a STOE theta/theta diffractometer in Bragg-Brentano geometry (Cu Kα1/2 radiation) with a secondary monochromator. Low-resolution scanning electron microscopy (SEM) images of cuboid crystal were recorded with a Hitachi TM3030. Transmission electron microscopy (TEM) images of powder samples were measured at 100 kV by an H-7100 electron microscope from Hitachi. High-resolution TEM (HR-TEM) and scanning electron microscopy (SEM) images were taken on HF-2000 and Hitachi S-5500 microscopes, respectively. To slice Co2MnGa crystal, Co2MnGa podwer was first embedded in Spurr resin and then trimmed using an EM TRIM milling system (Leica). Thin slices were cut from the resin blocks by microtome with a 35° diamond knife (Reichert Ultra-Cut), dispersed in Milli-Q water, transferred from the water surface on lacy carbon film-coated Cu grids and observed on Hitachi S-5500 (Hitachi) microscope.
X-ray photoelectron spectroscopy (XPS) measurements were conducted on Co2MnGa/carbon fiber paper before and after electrochemical test via a SPECS GmbH spectrometer with a hemispherical analyzer (PHOIBOS 150 1D-DLD). A monochromatized Al Kα X-ray source (E = 1486.6 eV) was employed and operated at 100 W. The base pressure in the analysis chamber was kept at 5 x 10 -10 mbar during the experiment. The binding energy scale was corrected for surface charging by use of the C 1s peak of contaminant carbon as reference at 284.5 eV.
Theoretical calculation. Our spin-polarized calculations were performed by using the VASP with PAW potentials. [3][4] PBE-GGA was applied to deal with the electron exchange-correlation interaction, [5][6][7] and DFT-D3 extension of Grimme was adopted to describe the long-range Van der Waals (VdW) interactions between the adsorbate and the substrate. Herein, the (110) plane of Co2VSn with the co-expose of Co and V atom was constructed for the adsorption of the adsorbate. A 8×6×1 k-mesh in the BZ was used for the geometry optimization and self-consistent calculations, and the kinetic energy cutoff was set to 400 eV. [7] Table S1. The Currie temperature (Tc), magnetization (M), and eg filling of Co-based Heusler alloys (based on theoretical values and our unpublished works). [8][9][10]

Compounds
Tc ( Figure S1. Digital picture of Heusler crystals with a one euro coin as comparison.  Figure S3 shows the existence of V instead of Ti. By taking into account that V and Ti have a similar element radius, it is reasonable to postulate that a separate Co3V phase was formed on Co2MnV compounds during arc-melting process.            (c), and Ga (d) of Co2MnGa on carbon fiber paper after the chronopotentiometry for 12 h. Note: EDX analysis excluded the content of carbon, and significant amount of F was from Nafion as a binder. A small amount of K was due to the residue of KOH on the electrode. Figure S14. XPS survey of Co2MnGa@carbon fiber paper before and after OER stability test. C, S, F, and K were detected on the electrode, which were from Nafion and KOH electrolyte. High-resolution spectra in Co 2p, Mn 2p, and Ga 2p region show spin-orbit splitting into 2p1/2 and 2p3/2 components, as seen in Figure S15a-c. Prior to the electrochemical test, the Co 2p XPS spectrum in Co2MnGa resembles that of Co(OH)2, [11] suggesting the surface oxidation from Co 0+ to Co 2+ . The formation of surface Co(OH)2 could be due to oxidation in contact with air as well as the electrode preparation process where Co2MnGa was treated in an ultrasound bath. Similarly, surface Mn and Ga exhibited oxidized form compared to the metallic state in pristine Co2MnGa compound. After conducting electrochemical test, Co 2p3/2 peak shifted to lower binding (BE) energy in the spectrum of CoO(OH), [11] illustrating surface reconstruction during electrolysis. The formation of surface CoO(OH) has been widely observed for Co-based catalysts in alkaline electrolytes and viewed as the real catalysts during the OER process. [12] On the other hand, no significant change was observed on the Mn 2p spectrum, and Ga was not detected on the surface due to leaching in KOH electrolyte. Furthermore, the O 1s spectra were analyzed to assign the surface oxygen groups ( Figure S15d). In O 1s spectrum of fresh Co2MnGa electrode, the dominant peak at 532.3 eV is assigned to the oxygen in the hydroxyl group, [11] while the peak at higher BE is considered to be molecular water adsorbed on the surface. [13] After electrolysis, an additional peak evolved at a BE of 529.7 eV, corresponding to the oxygen from oxide ions. [11] This suggests the formation of oxyhydroxyl group containing both hydroxyl and oxide ions, in consistency with the change on Co 2p spectra. Figure S16. LSV curves of Co2MnGa/Ti wire electrode and bare Ti wire as electrode. Co2MnGa single crystal was employed as electrode using silver paste to connect with Ti wire, as seen the structure in the inset photo. The oxidative current was mainly from the surface oxidation on silver paste, with the oxidation peaks centered at 1.23 VRHE and 1.58 VRHE corresponding to the formation of surface Ag2O and its further oxidation to AgO, respectively. [14] Figure S17. XRD patterns of (a) Co2VGa and (b) Co2VSn.