Fe-based superconductors (FeSC) exhibit all the properties of systems that allow the formation of a superconducting phase with oscillating order parameter, called the Fulde-Ferrell-Larkin-Ovchinnikov (FFLO) phase. By the analysis of the Cooper pair susceptibility in two-band FeSC, such systems are shown to support the existence of a FFLO phase, regardless of the exhibited order parameter symmetry. We also show the state with nonzero Cooper pair momentum, in superconducting FeSC with ∼ cos(k x ) · cos(k y ) symmetry, to be the ground state of the system in a certain parameter range.

Obtaining a thermodynamically accurate phase diagram through numerical calculations is a computationally expensive problem that is crucially important to understanding the complex phenomena of solid state physics, such as superconductivity. In this work we show how this type of analysis can be significantly accelerated through the use of modern GPUs. We illustrate this with a concrete example of free energy calculation in multi-band iron-based superconductors, known to exhibit a superconducting state with oscillating order parameter (OP). Our approach can also be used for classical BCS-type superconductors. With a customized algorithm and compiler tuning we are able to achieve a 19x speedup compared to the CPU (119x compared to a single CPU core), reducing calculation time from minutes to mere seconds, enabling the analysis of larger systems and the elimination of finite size effects.Keywords: FFLO, pnictides, NVIDIA CUDA, PGI CUDA Fortran, superconductivity PROGRAM SUMMARYManuscript Title: GPU-based acceleration of free energy calculations in solid state physics Authors: Micha l Januszewski, Andrzej Ptok, Dawid Crivelli, Bart lomiej Gardas Journal Reference: Catalogue identifier: Licensing provisions: LGPLv3 Programming language: Fortran, CUDA C Computer: any with a CUDA-compliant GPU Operating system: no limits (tested on Linux) RAM: Typically tens of megabytes. Keywords: superconductivity, FFLO, CUDA, OpenMP, OpenACC, free energy Classification: 7, 6.5 Nature of problem: GPU-accelerated free energy calculations in multi-band iron-based Email addresses: michalj@gmail.com (Micha l Januszewski), aptok@mmj.pl (Andrzej Ptok) February 4, 2015 superconductor models. Solution method: Parallel parameter space search for a global minimum of free energy. Preprint submitted to Computer Physics Communications Unusual features:The same core algorithm is implemented in Fortran with OpenMP and OpenACC compiler annotations, as well as in CUDA C. The original Fortran implementation targets the CPU architecture, while the CUDA C version is hand-optimized for modern GPUs.Running time: problem-dependent, up to several seconds for a single value of momentum and a linear lattice size on the order of 10 3 .

We study a subsystem of an isolated one-dimensional correlated metal when it is driven by a steady electric field or when it relaxes after driving. We obtain numerically exact reduced density matrix ρ for subsystems which are sufficiently large to give significant eigenvalue statistics and spectra of log(ρ). We show that both for generic as well as for the integrable model, the statistics follows the universality of Gaussian unitary and orthogonal ensembles for driven and equilibrium systems, respectively. Moreover, the spectra of modestly driven subsystems are well described by the Gibbs thermal distribution with the entropy determined by the time-dependent energy only.