spinmc 0.2.2

spinmc

SpinMC

SpinMC is an open-source Monte Carlo simulation package for classical spin models, designed for studying magnetic systems and phase transitions in statistical physics.

✨ Features

• Multiple Spin Models:

  • Ising model
  • XY model
  • Heisenberg model

• Support Interactions:

  • Exchange coupling (isotropic)
  • Automatic exchange neighbor generation by coordination shell or distance range
  • Single-ion anisotropy
  • [Planned] Anisotropic exchange coupling
  • [Planned] External magnetic field

• Support Algorithms:

  • Metropolis
  • Wolff cluster algorithm
  • [Planned] Parallel Tempering (Replica Exchange)

• Simulation Capabilities:

  • Energy, Heat capacity
  • Magnetization, Susceptibility
  • Absolute magnetization, absolute susceptibility
  • Group-wise (sublattice) magnetization & susceptibility
  • Group-wise (sublattice) Absolute magnetization & susceptibility
  • [planned] magnetic hysteresis loop
  • [planned] binder cumulant (u4)

📦 Installation

Requirements

• Python 3.9+ (for pip installation)
• Rust 1.88+ (for source compilation)

From PyPI

pip install spinmc

From Pre-built Releases

1. Visit the Releases page
2. Download the package for your OS
3. Unpack the archive and run the executable

From source:

git clone https://github.com/mxmf/spinmc.git
cd spinmc
cargo build --release

🚀 Quick Start

1. Create a configuration file (e.g., ising.toml):

[grid]
dimensions = [50, 50, 1]
sublattices = 1
spin_magnitudes = [1.0]
periodic_boundary = [true, true, false]

[simulation]
initial_state = "random"
boltzmann_constant = 1
model = "ising"
equilibration_steps = 10000
measurement_steps = 100000
algorithm = "wolff"
num_threads = 10
temperature_range = [
  { start = 1, end = 3, step = 0.1 },
]

[output]
outfile = "result.txt"
energy = true
heat_capacity = true
magnetization = true
susceptibility = true
magnetization_abs = true
susceptibility_abs = true

[[exchange]]
from_sublattice = 0
to_sublattice = 0
offsets = [[0, -1, 0], [0, 1, 0], [-1, 0, 0], [1, 0, 0]]
strength = 1.0

Exchange interactions can also be generated from the [structure] section instead of listing every offset manually. The structure can be written directly with cell and positions, or loaded from an external structure file:

[structure]
file = "POSCAR"
# format = "poscar"          # Optional; POSCAR/CONTCAR and .vasp files are detected automatically.
# magnetic_indices = [0, 2]  # Optional; select magnetic atoms when the file also contains non-magnetic atoms.
tolerance = 0.0001

Use neighbor_order = 1 for the first coordination shell, or distance_range = [min, max] to include every pair whose distance falls within the given range in Å. Specify both from_sublattice and to_sublattice to target one pair, only from_sublattice to target all destinations from one source sublattice, or neither to apply the rule to all sublattice pairs.

See examples/heisenberg_2d_cri3_poscar for a CrI3 monolayer example loaded from a POSCAR file.

2. Run the simulation:

spinmc run -i ising.toml

3. Results will be saved in result.txt with the requested measurements.

4. If spinmc is installed via Python, you can plot the results with:

spinmc plot -i result.txt

📚 See more examples in the examples folder.