1: \begin{abstract}

2: 

3: We present  a many-body $GW$  formalism for quantum subsystems embedded in  discrete polarizable environments containing up to several hundred thousand atoms described at a fully \textit{ab initio} random phase approximation level.

4: Our approach is based on a fragment approximation in the construction of the Green's function and independent-electron susceptibilities.  

5: Further, the environing fragments susceptibility matrices  are reduced  to a minimal but accurate representation preserving  low order polarizability tensors through a constrained minimization scheme. 

6: This approach dramatically reduces the cost associated with inverting the Dyson equation for the screened Coulomb potential $W$, 

7: % (\textbf{for the interacting susceptibility $\chi$ ?}) XB: il y a GW dans le titre donc plutôt mettre W en avant .. c'est pareil ... 

8: while preserving  the description of short to long-range screening effects. 

9: The efficiency and accuracy of the present   scheme is exemplified in the paradigmatic cases of 

10: %oligoacene  molecule and  

11: fullerene bulk, surface, subsurface, and slabs with varying number of layers.  

12: %The present scheme paves the way to  QM/QM' calculations in very large systems at the fully \textit{ ab initio} many-body level. 

13: %% together with a paradigmatic pentacene/fullerene interface.

14: %%[We show in particular that the screening-induced polarization energy  scales as the inverse number of molecular layers away from the surface or interface. 

15: %%Finally, as an intermediate scheme, we present a QM/QM/PCM implementation showing that embedding our QM/QM caculation in a continuous olarizable environment significantly speed-up the convergence with respect to the size of the atomistic environment.]

16: \end{abstract}

17: