Introduction

Introduction#

The aim of the SCCE package is to enable the study of interacting tight-binding or spin models, which, for simplicity, we just call lattice models, see Physics. Such models consist of a discrete set of sites representing the corresponding orbitals or spin degrees of freedom, respectively. A lattice can be viewed as a regular tiling of space by a primitive cell, where, in the context of condensed matter physics, the primitive cell is usually called unit cell. The lattice is then constructed by repeating the unit cell in the directions of the lattice vectors. The number of linear independent lattice vectors defines the number of dimensions of the system.

Within SCCE we have added another layer in the construction of the lattice system, which allows us to define an embedding, see section on cluster embedding. We start with the unit cell as the primitive unit and build a cluster as described above. We can then repeat the cluster to construct a full system. In this way we are able to describe various embedding schemes, described in the Physics section. If the number of repetitions in every direction along a lattice vector is one, which corresponds to system_size = [1, 1, 1] in the input, the system consists of a single cluster ony, in which case no embedding is employed.

It is perfectly fine to define a system consisting of just a single cluster with a large single unit cell which enables the study of amorphous systems.

Currently, our emphasis is on adding functionality for the system_size = [1, 1, 1] mode. At the moment embedding schemes have only been implemented for very simple lattice models.

Note for the DLR release Using a single cluster is the recommended usage for the current DLR release. Embedding schemes beyond the standard CPT are planned for the 2025 DLR release.

We are always eager to learn which features are most useful to our user base, so any feedback is highly encouraged. Depending on this user feedback, we envisage to implement the following features in the future:

  • Optional coarse-graining of the bath.

  • Expansion of available embedding schemes: Dynamical Mean Field Theory (DMFT) will be added as options for embedding.

Please keep in mind that the current release offers a limited feature set only, but we have plenty of ideas for extensions to the platform.