# Nanothermodynamics of large iron clusters by means of a flat histogram Monte Carlo method

## Abstract

The thermodynamics of iron clusters of various sizes, from 76 to 2452 atoms, typical of the catalyst particles used for carbon nanotubes growth, has been explored by a flat histogram Monte Carlo (MC) algorithm (called the σ-mapping), developed by Soudan et al. [J. Chem. Phys. 135, 144109 (2011), Paper I]. This method provides the classical density of states, g{sub p}(E{sub p}) in the configurational space, in terms of the potential energy of the system, with good and well controlled convergence properties, particularly in the melting phase transition zone which is of interest in this work. To describe the system, an iron potential has been implemented, called “corrected EAM” (cEAM), which approximates the MEAM potential of Lee et al. [Phys. Rev. B 64, 184102 (2001)] with an accuracy better than 3 meV/at, and a five times larger computational speed. The main simplification concerns the angular dependence of the potential, with a small impact on accuracy, while the screening coefficients S{sub ij} are exactly computed with a fast algorithm. With this potential, ergodic explorations of the clusters can be performed efficiently in a reasonable computing time, at least in the upper half of the solid zone and above. Problems of ergodicity existmore »

- Authors:

- Laboratoire Francis Perrin, CNRS-URA 2453, CEA/DSM/IRAMIS/LIDyL, F-91191 Gif-sur-Yvette Cedex (France)

- Publication Date:

- OSTI Identifier:
- 22308367

- Resource Type:
- Journal Article

- Journal Name:
- Journal of Chemical Physics

- Additional Journal Information:
- Journal Volume: 141; Journal Issue: 10; Other Information: (c) 2014 AIP Publishing LLC; Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 0021-9606

- Country of Publication:
- United States

- Language:
- English

- Subject:
- 37 INORGANIC, ORGANIC, PHYSICAL AND ANALYTICAL CHEMISTRY; ACCURACY; ALGORITHMS; ATOMS; CARBON NANOTUBES; CATALYSTS; COMPARATIVE EVALUATIONS; CONVERGENCE; IRON; LIQUIDS; MELTING; MONTE CARLO METHOD; PARTICLES; SOLIDS; TRANSITION TEMPERATURE

### Citation Formats

```
Basire, M., Soudan, J.-M., and Angelié, C., E-mail: christian.angelie@cea.fr.
```*Nanothermodynamics of large iron clusters by means of a flat histogram Monte Carlo method*. United States: N. p., 2014.
Web. doi:10.1063/1.4894488.

```
Basire, M., Soudan, J.-M., & Angelié, C., E-mail: christian.angelie@cea.fr.
```*Nanothermodynamics of large iron clusters by means of a flat histogram Monte Carlo method*. United States. doi:10.1063/1.4894488.

```
Basire, M., Soudan, J.-M., and Angelié, C., E-mail: christian.angelie@cea.fr. Sun .
"Nanothermodynamics of large iron clusters by means of a flat histogram Monte Carlo method". United States. doi:10.1063/1.4894488.
```

```
@article{osti_22308367,
```

title = {Nanothermodynamics of large iron clusters by means of a flat histogram Monte Carlo method},

author = {Basire, M. and Soudan, J.-M. and Angelié, C., E-mail: christian.angelie@cea.fr},

abstractNote = {The thermodynamics of iron clusters of various sizes, from 76 to 2452 atoms, typical of the catalyst particles used for carbon nanotubes growth, has been explored by a flat histogram Monte Carlo (MC) algorithm (called the σ-mapping), developed by Soudan et al. [J. Chem. Phys. 135, 144109 (2011), Paper I]. This method provides the classical density of states, g{sub p}(E{sub p}) in the configurational space, in terms of the potential energy of the system, with good and well controlled convergence properties, particularly in the melting phase transition zone which is of interest in this work. To describe the system, an iron potential has been implemented, called “corrected EAM” (cEAM), which approximates the MEAM potential of Lee et al. [Phys. Rev. B 64, 184102 (2001)] with an accuracy better than 3 meV/at, and a five times larger computational speed. The main simplification concerns the angular dependence of the potential, with a small impact on accuracy, while the screening coefficients S{sub ij} are exactly computed with a fast algorithm. With this potential, ergodic explorations of the clusters can be performed efficiently in a reasonable computing time, at least in the upper half of the solid zone and above. Problems of ergodicity exist in the lower half of the solid zone but routes to overcome them are discussed. The solid-liquid (melting) phase transition temperature T{sub m} is plotted in terms of the cluster atom number N{sub at}. The standard N{sub at}{sup −1/3} linear dependence (Pawlow law) is observed for N{sub at} >300, allowing an extrapolation up to the bulk metal at 1940 ±50 K. For N{sub at} <150, a strong divergence is observed compared to the Pawlow law. The melting transition, which begins at the surface, is stated by a Lindemann-Berry index and an atomic density analysis. Several new features are obtained for the thermodynamics of cEAM clusters, compared to the Rydberg pair potential clusters studied in Paper I.},

doi = {10.1063/1.4894488},

journal = {Journal of Chemical Physics},

issn = {0021-9606},

number = 10,

volume = 141,

place = {United States},

year = {2014},

month = {9}

}