Thermal DMRG for quasi one-dimensional magnetic molecules and chains

Abstract: Density matrix renormalization group methods or tensor network methods in general can not only be used to determine ground states but also to evaluate thermal equilibrium properties. Although convergence is superior for one-dimensional quantum spin systems with nearest neighbor exchange and open boundary conditions, the hope is that these methods can as well be employed to approximate magnetic observables of magnetic molecules with more complex interaction patterns. Here, we study the accuracy that can realistically be achieved for several archetypical quasi one-dimensional structures using the TenPy suite. In our study, we aim at systems that are too large for exact diagonalization or Krylov space methods.