2026-10-14T10:42:19Zhttps://www.radar-service.eu/oai/OAIHandler

10.35097/pqmqcrq4h47eucz52026-10-09T10:57:27Zradar4kit

https://dx.doi.org/10.35097/pqmqcrq4h47eucz5 Grüger, Tilman Studt, Felix MC-Cluster: a Monte Carlo simulation for nanoparticle Karlsruhe Institute of Technology 2025 Engineering dataset Software info:eu-repo/semantics/openAccess Other info:eu-repo/semantics/openAccess GNU Affero General Public License. Sharapa, Dmitry I. Sireci, Enrico Plessow, Philipp N. This Monte-Carlo simulation simulates nanoparticles at different temperatures. By applying a simulated annealing protocol, the user can find the most stable, entropically realistic structure of a nanoparticle. Simple energy model The energy input is a simple mapping of coordination number to energy in eV. The input can be a direct mapping of every coordination number to a distinct energy. Alternatively, a linear energy increase with coordination number can be used, which speeds up the simulation. High performance By using a simple energy model, the simulation can perform up to 10^11 iterations in one day. With a linear coordination number to energy mapping it can reach up to 10^12 iterations in one day. This allows for the simulation of larger nanoparticles. Multithreaded The code allows for as many parallel simulations as cores available on the computing node. This enables the user to obtain a statistically relevant ensemble of low-energy structures. Adding parallel simulations has negligible increase in required RAM. Track the Simulation using the .xyz-format Throughout the simulation it is possible to track the energy and save snapshots of the particle. Furthermore the particle with the lowest energy will be saved as an .xyz-file. Support The simulation can simulate the effect a support can have on the particle. A monometallic support can be created by giving a vector that is orthogonal to the support pane. Usage Requirements Rust Python used version 3.11.10 ASE used version: ase-3.22.1 Install from git git clone git@github.com:T-136/MC-Cluster.git Build the binary Build the program with: cargo build -r After the build step is complete the compiled program can be found in your project folder under "./target/release/mc". Run the simulation ./target/release/MC-Cluster -a Pt,1000 --support Al,1,1,1 -t 1000 -i 1e7 -r 0-1 --e-cn ./example_data/cn_input_example.json -o 9/10 -g ./example_data/303030-grid --support-e 0 --xyz-trajectory Use "-h" or "--help" to see the available flags and how to use them. -s, --start-cluster <START_CLUSTER> -a, --atoms <ATOMS> Atom name and number of that atom seperated by a comma. "Pt,4000" -s, --support <SUPPORT> When creating a new particle using the atoms flag, write the Atom name and a vector orthogonal to the support surface Al,1,1,1. When starting from a xyz file only the support atom is required -f, --folder <FOLDER> Output folder [default: ./sim/] -i, --iterations <ITERATIONS> Determines the length of the simulation -b, --begin-temperature <BEGIN_TEMPERATURE> Temperature where the annealing process starts [default: 5000] -t, --temperature <TEMPERATURE> Temperature at which the annealing process stops [default: 300] -o, --optimization-cut-off-fraction <OPTIMIZATION_CUT_OFF_FRACTION> Fraction of the simulation after which the annealing process is completed. After that, the temperature remains constant [default: 1 2] --support-e <SUPPORT_E> Support energy --e-l-cn <E_L_CN> File path or string containing a JSON-formatted list of energies --e-cn <E_CN> File path or string containing JSON-formatted energies -r, --repetition <REPETITION> How many times the same simulation is run. Multiple runs allow for convergence tests. The number will be part of the simulation folder name. After running `-r 0-1`, you can run `-r 1-2` and the previous simulation will not be overwritten [default: 0 1] -g, --grid-folder <GRID_FOLDER> Folder containing the setup files like neighbor sites. It can be created using the Python script `create_sites.py` [default: ../303030-pair] -x, --xyz-trajectory Track the simulation by taking snapshots of the cluster throughout the simulation --heat-map Generate a heat map -h, --help Print help -V, --version Print version Monte Carlo Nanoparticles DFT Coordination number Catalysis Simulated annealing Sharapa, Dmitry I. Sireci, Enrico Plessow, Philipp N. https://publikationen.bibliothek.kit.edu/1000181075 application/x-tar