Alkali co-adsorption enhances bond-resolved N2 activation at O2-active MoS2 single-atom sites

Supporting data archive.

Scope

This archive contains optimized structures, representative VASP input files, selected LOBSTER input and output files, and numerical source data supporting the manuscript and Supplemental Information.

This is a curated supporting-data package rather than a complete recomputation package. Large intermediate files and selected post-processing materials are retained by the lead contact and are available on request.

For the alkali-added systems, bond-resolved COHP analysis is restricted to the side-on activation-capable branch. The corresponding alkali-added end-on structures are included as thermodynamic and structural controls through their relative energies, optimized geometries, N-N bond lengths, and magnetic moments. Alkali-added end-on COHP data are not part of the reported analysis.

Directory layout

• Structures/
  • 16 optimized geometries distributed as POSCAR
  • Each file is the relaxed VASP CONTCAR renamed as POSCAR
  • Includes baseline, alkali-added, and gas-phase structures
• Structures/OUTCAR_excerpts/
  • Final energy, magnetization, force, and convergence excerpts
  • The 12 N2-adsorbed slab entries correspond to the final force-converged verification relaxations used for the deposited structures and relative energies
• Inputs/
  • Representative INCAR files organized by calculation type
  • K-point meshes are documented below and can be reconstructed
• LOBSTER/
  • 12 selected LOBSTER datasets containing lobsterin, ICOHPLIST.lobster, and COHPCAR.lobster
  • Baseline side-on and end-on N2 on Fe and Co
  • Side-on O2 references on Fe and Co
  • Alkali-added side-on N2: Fe+K, Fe+Na, Co+K, and Co+Na
  • Isolated gas-phase N2 and O2
• Figure_Source_Data/COHP/
  • Baseline side-on/end-on N-N -pCOHP source curves used for Figure S2
  • Figure_S3_Fe_NN_pCOHP.csv and Figure_S3_Co_NN_pCOHP.csv, containing the baseline, K-added, and Na-added side-on N-N -pCOHP curves used for Figure S3
• Figure_Source_Data/PDOS/
  • Side-on adsorbate-p and transition-metal-d PDOS source data for Figure 2
• Figure_Source_Data/EField/
  • Field-dependent PDOS source data at 0.0 and +/-0.5 V/A for Figure 4
• Figure_Source_Data/Alkali/
  • fig5b_elongation.csv
  • fig5c_delta_icohp.csv
  • fig5d_bond_partition.csv
  • alkali_cohp_summary.csv
  • Table_S5_normalizations.csv
    • Complementary molecule-normalized bond-weakening metrics reported in Table S5
• summary.csv
  • Final VASP energies, relative energies, magnetic moments, convergence information, and file provenance
• MANIFEST.csv
  • Figure and table provenance within this archive

Tables 1 and S1-S5 are not duplicated as separate formatted tables. Their source values can be recovered from Structures/, summary.csv, LOBSTER/ICOHPLIST.lobster, and Figure_Source_Data/Alkali/. The manuscript and Supplemental Information remain the reference for the reported, rounded tabulated values.

Calculation settings

• Code and method: VASP, PAW, GGA-PBE
• Plane-wave cutoff: 500 eV
• Aspherical PAW corrections: LASPH = .TRUE.
• Electronic convergence: EDIFF = 1E-6 eV
• Ionic convergence: EDIFFG = -0.01 eV/A
• Geometry-optimization smearing: Gaussian, 0.05 eV
• K-point meshes:
  • Geometry optimization: 2 x 2 x 1
  • PDOS: 6 x 6 x 1
  • COHP: 3 x 3 x 1
  • Gas-phase references: Gamma point
• LOBSTER wavefunctions: separate spin-polarized static SCF calculations with 560 bands
• Supercell: orthorhombic 5 x 3 x 1 cell with 90 framework atoms and approximately 15 A of vacuum
• POTCAR files are not distributed
• PAW datasets used: Mo_sv, S, Fe, Co, O, N, K_sv, and Na_pv

Gas references and numerical consistency

• Gas-phase cells use the same in-plane lattice vectors as the unstrained slab supercell and Gamma-point sampling.
• N2 was evaluated in the singlet state, and O2 in the spin-polarized triplet state.
• The deposited gas-phase O2 ICOHPLIST.lobster gives a spin-summed -ICOHP of 16.24723 eV, reported as 16.247 eV.
• The deposited gas-phase N2 ICOHPLIST.lobster gives a spin-summed -ICOHP of 22.99983 eV, reported as 23.000 eV.
• The manuscript, Supplemental Information, and numerical source tables use these deposited gas-phase reference values. The values in Figure_Source_Data/Alkali/alkali_cohp_summary.csv were generated using the same N2 reference of 22.99983 eV.

Percent bond elongations use fixed standard bond lengths of 1.098 A for N2 and 1.210 A for O2 rather than the computed gas-phase bond lengths. Percentages were evaluated from unrounded optimized coordinates. Therefore, recalculation from bond lengths rounded to three decimals can differ in the last reported digit.

• The deposited N2 geometry-optimization structures and energy summaries were updated from final verification relaxations. Existing PDOS and LOBSTER datasets were retained because the verified structures preserve the same adsorption motifs and differ only slightly from the geometries used for the static analyses.
• In Figure_Source_Data/Alkali/alkali_cohp_summary.csv, dNN_poscar and elong_pct refer to the final deposited POSCAR geometry, whereas dNN_lobster refers to the geometry associated with the retained LOBSTER dataset.

O2 structures

The side-on O2 structures were adopted from the authors' previous work and the companion manuscript. All O2 electronic-structure and COHP analyses reported in the present study were performed anew using those structures.

Conventions

• Positive -pCOHP values indicate bonding contributions.

• Negative -pCOHP values indicate antibonding contributions.

• Delta(-ICOHP) = (-ICOHP)_gas - (-ICOHP)_ads

• The O2-normalized N-N bond-weakening ratio is

  100 x Delta(-ICOHP)_N2 / Delta(-ICOHP)_O2

  evaluated against the side-on O2 reference for the same metal.

• The O2-normalized ratio quantifies relative gas-referenced intramolecular bond weakening. It is not a geometric activation fraction or an overall catalytic-activity metric.

• The complementary molecule-normalized fractional weakening is

  f_weak(X2) = 100 x Delta(-ICOHP)_X2 / (-ICOHP)_X2,gas

  where X2 denotes N2 or O2.

• The O2-relative fractional response is

  R_frac = 100 x f_weak(N2) / f_weak(O2)

  evaluated against the side-on O2 reference for the same metal.

• Table_S5_normalizations.csv reports these complementary quantities using the deposited gas-phase reference values of 22.99983 eV for N2 and 16.24723 eV for O2. The values in the Supplemental Information are rounded from this source table.

• The absolute-loss O2-normalized ratio and the molecule-normalized fractional response address different aspects of adsorption-induced bond weakening. Neither should be interpreted as an overall catalytic-activity metric.

• Empty alkali-N entries for baseline systems indicate that no alkali atom is present; they are not zero values.

• N1 and N2 follow the fixed atom labels of the optimized structures and are used consistently for TM-N and alkali-N pair analyses.

• Spin channels are summed unless otherwise stated.

• The absolute charge spilling was below 2% for all LOBSTER calculations reported in the manuscript. The corresponding lobsterout files are not included in this archive.

• M_tot in summary.csv is signed; the manuscript generally discusses its absolute magnitude.

• TOTEN_eV is the final VASP free-energy TOTEN.

• deltaE_sigma0_meV was calculated from energy(sigma->0).

• Relative energies are referenced to the corresponding side-on N2 branch.

• paper4_descriptors.dat files in Figure_Source_Data/PDOS/ contain supplementary PDOS-derived quantities generated during analysis. They are provided for completeness and are not used in the manuscript.

Files not included

The following files are retained by the lead contact and are available on request:

• Full OUTCAR
• DOSCAR, CHGCAR, WAVECAR, and POTCAR
• Initial structures and unselected alkali candidate structures
• Per-system INCAR files
• LOBSTER lobsterout logs
• Post-processing scripts