The Origin of High Activity of Amorphous MoS2 in the Hydrogen Evolution Reaction

Abstract Molybdenum disulfide (MoS2) and related transition metal chalcogenides can replace expensive precious metal catalysts such as Pt for the hydrogen evolution reaction (HER). The relations between the nanoscale properties and HER activity of well‐controlled 2H and Li‐promoted 1T phases of MoS2, as well as an amorphous MoS2 phase, have been investigated and a detailed comparison is made on Mo−S and Mo−Mo bond analysis under operando HER conditions, which reveals a similar bond structure in 1T and amorphous MoS2 phases as a key feature in explaining their increased HER activity. Whereas the distinct bond structure in 1T phase MoS2 is caused by Li+ intercalation and disappears under harsh HER conditions, amorphous MoS2 maintains its intrinsic short Mo−Mo bond feature and, with that, its high HER activity. Quantum‐chemical calculations indicate similar electronic structures of small MoS2 clusters serving as models for amorphous MoS2 and the 1T phase MoS2, showing similar Gibbs free energies for hydrogen adsorption (ΔG H*) and metallic character.

Part II: Ex-situ Grazing Incidence XANES, EXAFS, XPS, and ICP analysis 7 Figure S1. Mo-K edge EXAFS spectra of MoS2 films plotted as χ (k) with k-weight of 3 7 Table S1. Mo-K edge EXAFS fitting parameters of MoS2 films 8 Figure S2. Mo-K edge XANES spectra of experimental data and simulations for 1T-MoS2 8 Figure S3. XPS spectrum for MoS2 spent films 9 Table S2. Summary of Mo 3d5/2 binding energy for different 9 Table S3. Summary for the atomic ratio of sulfur species based on S 2p spectra 9 Table S4. ICP analysis results of pristine MoS2 films and those after stability tests 10  Table S8. Summary of Mo 3d5/2 binding energy in in-situ XAS samples 17 Table S9. Summary for the atomic ratio of sulfur species based on S 2p spectra 17 Figure S14. XPS of 2H, 1T and Am-MoS2 after in-situ XAS measurements 18 Table S10. ICP analysis results of MoS2 films after in-situ XAS tests 19  Å. Plotted spectra have a k-weight of 3 and were not phase-corrected.  SiO2. For the cross-sectional imaging, the MoS2 film was coated with a SiOx film stack as a protective layer and subsequently prepared using a standard FIB lift-out TEM sample preparation scheme. [1] X-ray photoelectron spectroscopy (XPS) was performed on a ThermoScientific K-Alpha instrument equipped with a monochromatic X-ray source (E(Al Kα) = 1486.6 eV). Energy calibration was performed by using the C 1s peak of sp 3 carbon at 284.6 eV as a reference and the spectra were fitted by CasaXPS software.  simulations have been performed using the FDMNES program package. [4] According to the code, simulated spectra are obtained via the multiple scattering theory in Green formalism based on the muffin-tin approximation on the potential shape. [4] In the simulated signals, the muffin-tin radii have been tuned to have a 10% overlap between the different spherical potentials. Moreover, to correct for inelastic losses, the Hedin-Lundqvist exchange potential has been used. [4] The approximation of non-excited absorbing atoms, which better reproduces the experimental data, has been adopted. The simulated XANES spectra were calculated considering that all the atoms are surrounding the absorber Mo within a 7 Å radius sphere.

Scanning electron microscopy (SEM
In addition, in the simulation of the XANES spectra, the tabulated core-hole broadening together with an energy resolution of 1.9 eV have been set. The effect of the structural disorder has not been considered in the calculation of the theoretical signal. The XANES spectra have been energy calibrated and then were compared with the calculated spectra. Figure S1. Ex-situ grazing incidence Mo-K edge EXAFS spectra of MoS2 films plotted as χ (k) with k-weight of 3. The black curves represent experimental data and the red curves show fitted spectra.  Figure S2. Mo K-edge XANES of 1T-MoS2 (black curve) and calculated simulation (red, blue curve) based on rhombohedral structure (inset, purple, yellow and green balls corresponds to Mo, S and Li atoms respectively), the red curve represents the ideal simulated spectrum while the blue curve represents the simulated spectrum with consideration of broadening by core-hole lifetime. Figure S2 presents the simulated Mo K-edge XANES spectra based on a rhombohedral MoS2 structure upon Li intercalation. However, the calculated spectra differ from the experimental data in their main features, which means that the proposed rhombohedral structure is not suitable. Therefore, it is assumed that the as-prepared 1T-MoS2 has a monoclinic symmetry ( Figure 2k) with distorted octahedral Mo coordination. Figure S3. XPS spectrum of Mo 3d (a) and S 2p (b) for MoS2 spent films; c) valence band spectra of corresponding samples, the Fermi edge energy was determined by the cross of two linear parts of the VB spectra.  Table S4. ICP-OES analysis results of pristine MoS2 films and those after 24 h HER stability tests.    Operando XANES spectra shown in Figure S6 were plotted with a merge of 10 scans while the spectra for spot 1 and 2 in Figure S7 were plotted with only one quick scan. As shown in Figure S7, there is no obvious difference in XANES features between in-situ and spot 1, 2, which indicates that beam damage can be neglected. Figure S8. Mo-K edge EXAFS spectra of 2H-MoS2 films plotted as χ (k) with k-weight of 2. The black curves represent experimental data and the red curves show fitted spectra.  Figure S9. Mo-K edge EXAFS spectra of 1T-MoS2 films plotted as χ (k) with k-weight of 2.

Sample name Mo mass (μg) S mass (μg) Li mass (μg) Formula (based on atomic ratio)
The black curves represent experimental data and the red curves show fitted spectra.
As shown in Table S5, we assign the peak at ~2 Å to a short Mo-S bond which is very similar to Mo-O bond. However, XPS measurements indicate that molybdenum oxide species account for only 3.6 % of total Mo species at the surface of 1T-phase MoS2. In addition, due to the configuration of the operando electrochemical cell, the collected EXAFS signals reflect bulk information of the samples. Therefore, it is more reasonable to assign this peak to short Mo-S bonds rather than to Mo-O bonds.  Figure S10. Mo-K edge EXAFS spectra of Am-MoS2 films plotted as χ (k) with k-weight of 2. The black curves represent experimental data and the red curves show fitted spectra.
As shown in Table S6, we assign the peak at ~2 Å to a short Mo-S bond, which is very similar to a Mo-O bond. However, XPS measurements indicate that molybdenum oxide species account for 4.6 % of the total Mo species. In addition, due to the configuration of the operando electrochemical cell, the collected EXAFS signals reflect the bulk information of the samples. Therefore, it is more reasonable to assign this peak to short Mo-S bonds rather than to Mo-O bonds. Part IV: Raman, SEM, and XPS analysis Figure S11. XPS spectrum of Li 1s for 1T-MoS2 films before and after operando XAS measurements.
As shown in Figure S11, the surface lithium content was greatly reduced after operando XAS experiments, which means that lithium gradually leaches into the electrolyte during HER.    Figure 3d).
As we use Pt wire as counter electrode in the operando electrochemical cell for XAS measurements, we performed XPS to check whether the films have been contaminated by Pt or not. Figure S14 shows the Pt 4f and survey spectra of 2H, 1T and amorphous MoS2 films after operando XAS experiments. Pt contamination could not be observed in any of the samples. Figure S15. X-ray diffraction patterns with grazing incidence angle of 0.3° for MoS2 films before and after in-situ XAS measurements.
The interference fringes, seen in all GIXRD patterns, can be ascribed to the glassy carbon substrate structure as shown in Figure S15   There exists a redshift of ~2 cm -1 for E 1 2g and A1g peaks of 1T-MoS2 compared to 2H-MoS2, which is characteristic for 1T-MoS2, and the shift stays constant before and after operando XAS measurements. [5] Part V: Electrochemically Active Surface Area Typically, the electrochemical double layer capacitance (Cdl) method is applied to assess the electrochemically active surface area (ECSA). However, the ECSA value is obtained by the formula: where Cdl(sample) represents the Cdl value of the sample and Cdl(single crystal) represents the Cdl of a single crystal reference material with certain orientation. This method assumes that the sample has the same orientation as the single crystal reference which is often not the case, and even single crystals with different orientations can have different Cdl values. [6] Therefore, this method sometimes can over-or underestimate the real ECSA. For instance, as has been shown in Figure S19 and  Figure S20 we compare the electrocatalytic activity based on geometric surface area.

Part VI: Density Functional Theory (DFT) modeling
Density Functional Theory (DFT) calculations were carried out using the Vienna ab-initio simulation (VASP) package. [7] The electronic exchange-correlation potential was described using the Perdew, Burke, and Ernzerhof (PBE) functional. [8] Van der Waals interactions were taken into account with the D3 method of Grimme et al. with Becke-Johnson damping (DFTD3(BJ)). [9] The kinetic wave functions were expanded in a plane wave basis with a high energy cut-off of 600 eV and the convergence criterion was set to 10 −6 eV between two ionic steps for the self-consistency process.