Al2Pt for Oxygen Evolution in Water Splitting: A Strategy for Creating Multifunctionality in Electrocatalysis

Abstract The production of hydrogen via water electrolysis is feasible only if effective and stable catalysts for the oxygen evolution reaction (OER) are available. Intermetallic compounds with well‐defined crystal and electronic structures as well as particular chemical bonding features are suggested here to act as precursors for new composite materials with attractive catalytic properties. Al2Pt combines a characteristic inorganic crystal structure (anti‐fluorite type) and a strongly polar chemical bonding with the advantage of elemental platinum in terms of stability against dissolution under OER conditions. We describe here the unforeseen performance of a surface nanocomposite architecture resulting from the self‐organized transformation of the bulk intermetallic precursor Al2Pt in OER.

1 Supporting information S1. Materials and methods

S1.1. Sample preparation
Platinum granules (Chempur, 99.99 %, 1-6 mm) and pieces of aluminum rod (Alfa Aesar, 99.9965 %) were weighed in atomic Al : Pt ratio equal to 32 : 68. The constituent components were arc melted at least two times with mass losses not exceeding 0.2 %. Obtained ingots were placed into alumina crucibles and sealed under argon into tantalum containers. For oxidation protection at high temperatures, tantalum containers were inserted into quartz glass ampoules and sealed under vacuum. The samples were annealed in a resistance furnace at 1000 o C for 670 hours with subsequent quenching of quartz ampoules in ice water.
Material discs (necessary for electrochemical experiments, diameter -8 mm, height -3-4 mm) were manufactured via Spark Plasma Sintering (SPS) technique. The ground powder (agate mortar, argon-filled glovebox) was filled into SPS press-form, which was exposed to the heating by pulsed direct electrical current with low voltage (Tmax = 1000 o C at 80 MPa for 10 min).

S1.2. Characterization
The synthesized samples were characterized via powder X-ray diffraction (PXRD) using Huber Imaging Plate Guinier Camera G670 (CuKα1 radiation, λ = 1.54059 Å, LaB6 with a = 4.1569 Å as internal standard). Quantitative phase analysis of synthesized samples was carried out via comparison of the experimental patterns with the theoretically calculated ones (program WinXPow [1] ). WinCSD software package [2] was used for the indexing of diffraction patterns and determination of lattice parameters.
Scanning electron microscopy (SEM) was used for the examination of sample homogeneity as well as for the precise determination of the composition. The sample was embedded into conductive polymer and polished with SiC papers and diamond powders with different grain sizes (ending with 1/4 μm diamond powder in slurry).
Initially, optical microscopy (Axioplan 2, Zeiss) with bright-field, dark-field, polarized light and differential interference contrast was carried out. Afterwards, elemental analysis of metallographic cross sections was performed with JEOL 7800F with an attached EDXS system (Quantax 400, Bruker, Silicon-Drift-Detector (SDD), FEG cathode, acceleration voltage: 0.1 kV-30 kV). For accurate quantitative analysis, 2 wavelength-dispersive X-ray spectroscopy (WDXS) as an option of the electron microprobe Cameca SX100 (tungsten cathode, acceleration voltage: 1 kV-30 kV) was applied. X-ray intensities were measured at 20 kV using elemental Al (100 %) and Pt (99.999 %) as reference probes for the intensities of Al Kα and Pt Lα lines. The PAP matrix correction mode [3] was used for chemical composition calculations.

S1.3. Electrochemical measurements
The electrochemical performance of Al2Pt was investigated in a combined electrochemical flow cell (EFC)/inductively coupled plasma-optical emission spectrometry (ICP-OES) setup. The setup design as well as details on the benchmarking protocol used for the corresponding electrochemical performance evaluation are described elsewhere. [5] A coil-shaped platinized platinum wire (PT-5W, 125 μm diameter, 99.99 %, Science Products GmbH), placed along the flow channel following the electrolyte outlet flow, was used as the counter electrode (CE), while the reference electrodes (RE) (SCE, CH Instruments Inc., CHI150, reference potential +241 mV vs. NHE) was inserted perpendicular to the electrolyte inlet channel to avoid possible Pt corrosion and re-deposition on the WE during the electrochemical testing. was performed in potential range from open circuit potential up to 2.0 VRHE with a scan rate of 5 mV s -1 . Current density of 10 mA cm -2 was chosen for comparison of the data because it is a benchmarking value for fundamental studies and testing of electrocatalyst materials for OER. [5,6] Since the initial LSV reached a maximal current density of 90 mA cm -2 at 2.0 VRHE, this value was taken for the harsh stability measurement. The stability was thus judged using chronopotentiometry (CP) technique at a current density of 90 mA cm -2 for 456 h. To monitor the activity of the sample during the CP measurement, LSVs were recorded every 24 h. In order to release the system, open circuit potential (OCP) was measured for 5 min before and after LSV scan. In total, there were 19 cycles. The depth of material changes after 456 h long-term experiment is 400 m, which is less than 10 % from the specimen thickness. Due to a good conductivity of Al2Pt material, it works as a current collector. From our experiments, the thickness of the pellet may impact only technical characteristics and tightness of the EC cell, but not the measured overpotential of the material.

S1.4. Theoretical calculations
Electronic structure calculations were performed for the ordered model Al2Pt (space group Fm3 ̅ m, a = 5.9190 Å) on the fully-relativistic level by using the all-electron, fullpotential local orbital method (FPLO) [7] and the Fritz-Haber Institute ab initio molecular simulations method (FHI-aims). [8] The local density approximation to the density functional theory as parameterized by Perdew-Wang [9] was employed to account for the exchange-correlation effects. The Brillouin zone was sampled with a mesh of 24×24×24 k points. The electron density (ED) and the electron localizability indicator (ELI) in its ELI-D representation (ΥD) [10,11] were computed employing the interface implemented in the FHI-aims package. [12] The topological analysis of ED and ELI-D was carried out by the program DGrid. [13] The projected densities of states (DOS) corresponding to the states occupied in the free atom electronic configurations were convoluted by a Lorentzian function with a broadening of 0.5 eV. These were then multiplied by the photoionization cross sections [4] and their sum was compared to the valence band spectrum obtained by XPS.
The core level shifts were calculated using the delta SCF method. [14] Calculations were performed using PBE exchange [15] and correlation potential using the Quantum ESPRESSO package [16] with norm-conserving pseudopotentials with the kinetic energy cutoff of 40 Ry and k-point mesh of 10×10×10 and a Marzari-Vanderbilt smearing [17] with the parameter of 0.02 Ry.

S2. The phase Al2Pt
According to our studies, the Al2Pt phase possesses small homogeneity region with compositions deviating slightly from stoichiometry 2:1. There is ongoing project dedicated to the structural features and nature of such compositional changes of Al2Pt, which will be the topic of a separate publication. Concerning the sensitivity of catalytic properties (particularly OER activity) towards the composition, at least initial OER activity for all compositions is the same within the sensitivity range of applied methods.