Chapter 4. Multidirectional charge transfer in perovskite oxide and transition metal dichalcogenides
4.4 Experimental details
500 mg of bulk MoSe2 (< 2 μm, purity >99%, Alfa Aesar) and 200 mg of metallic potassium (stored in oil, purity > 99.95%, high purity, Korea) were added for the production of K-MoSe2. A glass tube under inert conditions of the glove box. The tube was sealed and treated at 400 °C for 1 hour. The prepared potassium intercalation MoSe2 was washed with deionized water and ethanol to remove potassium ion residues. The produced powder was dried at 80 °C for 24 hours.
Synthesis of LSC
La, Sr and Co nitrite (La(NO3)3∙6H2O, Sr(NO3)2, Co(NO3)2∙6H2O (Alfa Aesar) was dissolved to synthesize LSC perovskite oxide, and acid (C6H8O7) in citrate deionized water including a stoichiometric amount. After solvent evaporation, the produced wet-gel was calcined at 900 °C for 2 hours and calcined at 950 °C for 10 hours to remove organic fraction. Finally, the resulting reaction product was self-breathed to homogenize LSC.
Synthesis of LSC/K-MoSe2
９７
LSC and K-MoSe2 prepared to synthesize LSC/K-MoSe2 heterogeneous structures were high-energy- milled with 10 wt.% of Ketjen Black (KB) (EC-600JD, Lion Specials Co. Ltd.) through a planetary ball mill systems (PM-100, Resch). In order to find an optimized weight ratio of LSC to K-MoSe2 for electrochemical performance, LSC, K-MoSe2, and KB were prepared with various ratio (LSC:K- MoSe2:KB 8:1:1, 7:2:1, 6:3:1, and 5:4:1). The total weight of the mixture was maintained at 600 mg.
With ethanol, LSC, K-MoSe2 and KB were sealed in steel bottles and ball milled at 500 rpm for 2 hours.
The results of the ball-milling process were completely dried and collected.
Material characteristics
Catalyst form and EDS element mapping were characterized using HR-TEM (JEM-2100F, JEOL) with an acceleration voltage of 200 kV. The crystallographic information of the catalyst was analyzed through high power XRD (D/MAX2500V/PC, Rigaku) at 40 kV and 200 mA at a scanning rate of 1°
min-1 in a diffraction range of 10°-80°. The chemical state and work function were investigated through XPS (ESCALAB 250XI, Thermo Fisher Scientific) with monochromatic Al-Kα radiation. TGA analysis was performed to investigate the surface adsorption capacity of the catalyst at a ramp temperature rate of 10 °C min-1 using a thermogravimetric analyzer (Q500, TA). The surface area and pore size of the catalyst were evaluated using the N2 desorption/adsorption isotherm of the catalyst using a physical adsorption analyzer (ASAP 2420). The optical properties of the catalyst were collected using a UV-Vis- NIR spectrometer (Cary 5000, Agilent). The Raman spectrum was recorded using confocal Raman spectroscopy (Alpha300R, WITHEC) equipped with a 532 nm laser. The electrode form before and after total water electrolysis was obtained through cold FE-SEM (S-4800, HITACHI).
Electrochemical measurement
Half-cell electrochemical measurements are electrochemical workstations (CHI 760E, CH Instruments Co., Ltd.). Catalyst ink with 9 mg of catalyst and 1 mg of KB containing LSC/K-MoSe2, K-MoSe2, and LSC has a 5% by weight Nafion solution (Sigma-Aldrich), ethanol and isopropyl alcohol, and then bath ultrasound treatment. Catalyst ink of Pt/C and IrO2 was similarly prepared except for KB.
The working electrode was prepared by dropping 5 μL of prepared catalyst ink into a 0.071 cm2 glassy carbon disk electrode. HER and OER LSV polarization curves were obtained at scan rates of 5 mV s-1 in 0 to -1.0 V (vs. reversible hydrogen electrodes (RHE)) and N2 saturated electrolytes measured at 1.0 to 2.0 V (vs. RHE), respectively. The Tafel slope value was derived from the LSV curve by plotting the
９８
overpotential for the current density at the log scale from 1 to 10 mA cm-2. All potentials of this operation were measured for Ag/AgCl reference electrodes and converted from 1 M KOH (pH 14) to RHE scale using the following formula. E(vs. RHE) = E(vs. Ag ∕ AgCl) + EAg∕AgCl (= 0.197 V) + 0.0592 pH = E(vs. Ag/AgCl) + 1.0258 V. The EIS measurements were performed at overpotentials of -0.2 and 0.7 V (vs. RHE), respectively, for HER and OER in a frequency range of 100 kHz to 0.01 Hz at an amplitude of 10 mV at 1M KOH. All half-cell polarization curves were corrected for resistance loss by the following equation. E = E(RHE) – 𝑖𝑅s , where E is the potential after iR-calibration, E(RHE) is the measured potential for RHE (before iR calibration), i is the measured current, and Rs is the uncompensated resistance obtained from EIS analysis. To measure the double layer capacitance (Cdl), the potential window of the cyclic voltage current diagram was circulated at various scan rates of 20-160 mVs-1 in a non-Faraday region of 0.03 to 0.33V (vs. RHE). The Cdl value was derived by plotting the difference in charging current density (j = (ja − jc) / 2) at 0.18 V. The entire water electrolysis test was performed using a two-electrode configuration containing catalyst ink electrosprayed onto Ni foam (including catalyst). The results of the Chronopotentiometry stability test of total water electrolysis at a current density of 100 mA cm-2 were obtained using electrochemical workstations (ZIVE BP2C, Wonatech Co., Ltd.).
Calculation details
We performed the initial calculation of spin polarization ab using the Vienna AB Initial Simulation Package (VASP) [61] within the projector augmented wave (PAW) method [62] and Perdew-Burke- Ernzerhof (PBE) [63] exchange and correlation functions. The DFT+U method based on Dudarev's approach [64] was adopted as U = 4.3 and J = 1.0 eV (Ueff = 3.3 eV) for Co-3d and Ueff = 4.0 eV for Mo-4d, as used in previous studies [65, 66].
First, the LSC slab structure was prepared based on 2√2 3√2 2 supercells with > 20 Å vacuum space. A 1 1 1 Monkhorst-pack k-points mesh was adopted with 2 10−2 eV/Å for force criterion in the ionic relaxations and a 400 eV energy cutoff for the plane-wave basis set was used with valance electron configurations of 5s25p65d16s2 (La), 4s24p65s2 (Sr_sv), 3d84s1 (Co), 2s22p4 (O), 4s24p64d55s1(Mo_sv), 4s24p4(Se), and 3s23p64s1 (K_sv) orbitals for La, Sr, Co, O, Mo, Se, and K, respectively. The LSC/MoSe2 and LSC/K-MoSe2 structures were modelled using 2 5 supercell of 2H- or 1T-MoSe2 monolayer placed on the LSC (001) surfaces with <3 % lattice mismatch for LSC/K- MoSe2 (the length of the cell-vectors: a = 11.20 Å, b = 16.57 Å, and c = 35.00 Å),and K atoms relax to simulate the top two atomic layers heterogeneous structures of the 2 5 MoSe2 supercell surface. The
９９
charge transfer of the LSC/K-MoSe2 structure was analyzed using Bader population analysis. Second, the free energy of the HER reaction was calculated until the residual force component was within 5 10−3 eV/Å using the equation 𝐺𝐻= 𝐸(𝐻) + 0.24 𝑒𝑉. where E(H) is the adsorption energy of the H atom calculated for 1 2⁄ H2 at pH = 0 and p(H2) = 1 bar, and 0.24 eV correction is for the difference between zero energy and entropy [67, 68]. Free energy of each sub-reaction to OER was evaluated using equations 𝐺 =𝐸 +𝑍𝑃𝐸 − 𝑇𝑆 at pH = 0, T = 298 K, and zero applied potential (0 V vs.
RHE), where ΔG, ΔE, ΔZPE, T, and ΔS are free energy change amounts, total energy difference change amounts of change in zero temperature, entropy energy, respectively. Free energy O2 gas was calculated by fixing the change in free energy in the entire reaction (H2O → 1 2⁄ O2 + H2) is 2.46 eV, an experimentally measured value adopted in previously reported OER calculation [69-73]. LSC(001) surfaces were exposed on the LSC surface to ensure sufficient space for OER adsorbents on the LSC surface using 2 1.5 MoSe2 supercells cut to model the OER on the LSC/K-MoSe2. A 4 4 1 Monkeyhorst-pack k-point mesh) was used to analyze the DOS.