1: \begin{abstract}

2:   We describe the use of coupled-cluster theory as an impurity solver 

3: %% to calculate the interacting

4: %%   one-particle Green's function of the impurity problem arising

5:   in dynamical

6: mean-field theory (DMFT) and its cluster extensions.  We present numerical results

7: at the level of coupled-cluster theory with single and double excitations

8: (CCSD)

9: %% which has a computational cost that scales as $N^6$, where $N$ is number

10: %% of impurity and bath sites. %% I commented out since we don't use the locality of the interaction.

11: for the density of states and self-energies of cluster impurity

12: problems in the one- and two-dimensional Hubbard models.

13: Comparison to exact diagonalization

14: shows that CCSD produces accurate density of states and self-energies

15: at a variety of

16: values of $U/t$ and filling fractions.

17: However, the low cost allows for the use of many bath

18: sites, which we define by a discretization of the hybridization directly on

19: the real frequency axis.  We observe convergence of dynamical quantities using approximately 30 bath

20: sites per impurity site, with our largest 4-site cluster DMFT calculation using 120 bath sites. We

21: suggest coupled cluster impurity solvers will be attractive in ab initio

22: formulations of dynamical mean-field theory.

23: \end{abstract}

24: