mat-defect-energy-dft

Point-Defect Formation Energy (DFT)

Goal

To calculate the formation energy of point defects (vacancies, substitutions, interstitials) including charged defect states and finite-size corrections using DFT (VASP) via atomate2 workflows. This produces formation energy diagrams showing defect charge transition levels as a function of Fermi energy.

$$E_f[D^q] = E[D^q] - E[\text{bulk}] + \sum_i \Delta n_i \mu_i + q(E_\text{VBM} + \Delta E_F) + E_\text{corr}$$

where $q$ is the charge state, $E_\text{VBM}$ is the valence band maximum, $\Delta E_F$ is the Fermi energy relative to VBM, and $E_\text{corr}$ is the finite-size correction (Freysoldt/FNV).

Instructions

1. Obtain Bulk Structure

Start with a relaxed primitive cell:

mcp_base_search_materials_project_by_formula(formula="MgO", save_to_file="MgO.cif")

2. Generate Defect Structures

Use pymatgen-analysis-defects to generate all symmetry-unique defect supercells with charge states:

# Env: base-agent
python .agents/skills/mat-defect-energy-dft/scripts/generate_defect_structures.py \
    --bulk MgO.cif \
    --supercell_size 3 3 3 \
    --defect_type vacancy \
    --charge_range -2 2 \
    --output dft_defects/

This generates:

• POSCAR files for each defect × charge state
• A defect_index.json mapping defect names to charge states and structures
• Pristine supercell for the bulk reference

3. Run DFT Calculations (atomate2)

Submit calculations via the atomate2 MCP tool:

# Bulk supercell reference
mcp_atomate2_run_atomate2_vasp_calculation(
    structures_path="dft_defects/pristine_supercell.cif",
    output_dir="./dft_bulk/",
    calculation_type="static",
    preset_type="matpes-pbe",
    execution_mode="remote"
)

# All defect structures
mcp_atomate2_run_atomate2_vasp_calculation(
    structures_path="dft_defects/",
    output_dir="./dft_defect_calcs/",
    calculation_type="relaxation",
    preset_type="matpes-pbe",
    execution_mode="remote"
)