ase

Files (7)

• SKILL.md
• ase_optimize.cpython-313.pyc · scripts/__pycache__/
• ase_properties.cpython-313.pyc · scripts/__pycache__/
• demo.cpython-313.pyc · scripts/__pycache__/
• ase_optimize.py · scripts/
• ase_properties.py · scripts/
• demo.py · scripts/

SKILL.md (first lines)

# Atomic Simulation Environment (ASE)

## Overview

ASE is a Python library for working with atoms and atomic structures. This skill provides computational design capabilities for materials science, including DFT geometry optimization, electronic structure calculations, phonon analysis, and molecular dynamics with classical force fields. ASE interfaces with multiple computational backends (MOPAC, Quantum ESPRESSO, VASP) and is excellent for designing novel materials and predicting their properties computationally.

## Core Capabilities

### 1. Structure Optimization

**Geometry Optimization:**

Optimize atomic structures to find stable configurations:

```python
from ase import Atoms
from ase.optimize import BFGS
from ase.calculators.mopac import MOPAC

# Create structure
atoms = Atoms('H2O', positions=[[0, 0, 0], [1, 0, 0], [0, 1, 0]])

# Set calculator (semi-empirical quantum chemistry)
atoms.calc = MOPAC(method='PM6')

# Optimize geometry
dyn = BFGS(atoms)
dyn.run(fmax=0.01)

# Get optimized coordinates and energy
energy = atoms.get_potential_energy()
forces = atoms.get_forces()
```

**Key Parameters:**
- `fmax`: Force convergence criterion (eV/Å)
- `steps`: Maximum optimization steps
- `trajectory`: File to save optimization trajectory

### 2. Electronic Structure Calculations

**Band Structure:**

Compute electronic band structures for periodic systems:

```python
from ase.build import bulk
from ase.calculators.mopac import MOPAC

# Create periodic structure (bulk silicon)
atoms = bulk('Si', 'diamond', a=5.4)

# Calculate band structure at high-symmetry k-points
atoms.calc = MOPAC(method='PM6-D3H4X')
```

**Density of States:**

Compute electronic density of states:

```python
# Get DOS at different energy levels
from ase.dft.band_structure import calculate_band_structure
```

### 3. Molecular Properties

**Predict from Structure:**

Calculate molecular properties computationally:

```python
# Geometry-optimized properties
- Dipole moment
- Polarizability
- Band gap (for semiconductors)
- Formation energy (for compounds)
- Cohesive energy (for crystals)
```

### 4. Phonon Analysis

**Vibrational Properties:**

Compute phonon frequencies for material stability:

```python
from ase.phonons import Phonons

# Create phonon object
phonons = Phonons(atoms, MOPAC_calc, supercell=(2, 2, 2))
phonons.run()

# Get phonon frequencies and DOS
phonon_frequencies = phonons.get_frequencies()
```

### 5. Molecular Dynamics with Classical Force Fields

**NVT/NPT Ensemble Simulation:**

Run classical MD with force fields (using EMT or custom potentials):

```python
from ase.md.verlet import VelocityVerlet
from ase.md.langevin import Langevin
from ase import units

# NVT ensemble (constant T)
dyn = Langevin(atoms, timestep=1*units.fs, temperature_K=300, friction=0.02)

# Run for specified timesteps
for i in range(1000):
    dyn.run(1)
```

## Available Calculators

### MOPAC (Semi-empirical QM)

From lamm-mit/scienceclaw, LGPL-3.0.

More from lamm-mit/scienceclaw