From 43e1ed2be90161a21bbcf83b3eb2887f44c60de0 Mon Sep 17 00:00:00 2001 From: Sam Dareska <37879103+samwaseda@users.noreply.github.com> Date: Wed, 1 Nov 2023 11:46:27 +0100 Subject: [PATCH] Update paper.md --- paper.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/paper.md b/paper.md index a971308..f4d17ce 100644 --- a/paper.md +++ b/paper.md @@ -34,6 +34,8 @@ Magnetic interactions account for a significant portion of free energy in certai `mamonca` is a C++-based python software package for the computation of magnetic interactions in solid materials. It has a clear interface consisting of setters (methods starting with `set_`) for inputs and getters (methods starting with `get_`). It has also an excellent interactivity, as the parameters can be changed on-the-fly, as well as the outputs can be retrieved in the same way. In order to have full flexibility in terms of multi-component systems and inclusion of defects, `mamonca` outsourced the structure definition, so that the user can analyse the interactions externally using a software package like pyiron [@janssen2019pyiron] or Atomic Structure Environment (ASE) [@larsen2017atomic]. `mamonca` has also high flexibility in defining the Hamiltonian, as it allows the user to define not only the classical Heisenberg model, but higher order components including the longitudinal variation can be defined, as it has been employed for Fe-Mn systems [@schneider2021ab]. In addition to the classical Monte Carlo and spin-dynamics, `mamonca` allows also for an addition of Metadynamics [@theodoropoulos2000coarse] and magnetic thermodynamic integration [@frenkel2023understanding], which can deliver the free energy variation. +Internally, each atom points to the magnetic moment of the interacting neighboring atoms, which allows for an efficient computation of pairwise interactions both in terms of speed and memory. Moreover, while it performs an on-the-fly computation of average magnetization and average energy, it does not allocate new memory for any property on-the-fly, which makes it highly memory-efficient. + # Acknowledgements We acknowledge contributions from Brigitta Sipocz, Syrtis Major, and Semyeong