Abstract
Ising models in nanosystems are studied in the presence of a magnetic field. For a one-dimensional (1-D) array of spins interacting via nearest-neighbour and next-nearest-neighbour interactions we calculate the heat capacity, the surface energy, the finite-size free energy and the bulk free energy per site. The heat capacity versus temperature exhibits a common wide peak for systems of any size. A small peak also appears at lower temperatures for small arrays when the ratio of magnetic field spin interaction energy over the nearest-neighbour spin–spin interaction energy, f, is within . The peak becomes smaller for longer array and eventually vanishes for long arrays, disappearing when the number of spins, N, is greater than 25 when only nearest-neighbour interactions are taken into account, and more than 14 when next-nearest-neighbour interactions are included as well. Ising models in which the nearest-neighbour interactions are ferromagnetic, while the next-nearest-neighbour interactions are either ferromagnetic or antiferromagnetic, are compared, and it is found that the reduced free energy in the former case exhibits a larger deviation from the bulk value.
Acknowledgements
Financial support from Sharif University of Technology and the SEPON project within the ERC Advanced Grants is gratefully acknowledged.