1: \begin{abstract}

2: The polarizability operator plays a central role in density functional

3: perturbation theory and other perturbative treatment of first principle

4: electronic structure theories.  The cost of computing the polarizability

5: operator generally scales as $\Or(N_{e}^4)$ where $N_e$ is the number of

6: electrons in the system. The recently developed adaptively compressed

7: polarizability operator (ACP) formulation [L. Lin, Z. Xu and L. Ying,

8: Multiscale Model. Simul. 2017] reduces such complexity to

9: $\Or(N_{e}^3)$ in the context of phonon calculations with a large basis

10: set for the first time, and demonstrates its effectiveness for model

11: problems. In this paper, we improve the performance of the ACP

12: formulation by splitting the polarizability into a near singular

13: component that is statically compressed, and a smooth component that is

14: adaptively compressed. The new split representation maintains the

15: $\Or(N_e^3)$ complexity, and accelerates nearly all components of

16: the ACP formulation, including Chebyshev interpolation of energy levels,

17: iterative solution of Sternheimer equations, and convergence of the Dyson equations. For simulation of real materials, we discuss how to incorporate nonlocal pseudopotentials and finite temperature effects.

18: We demonstrate the effectiveness of our method using one-dimensional model problem in insulating and metallic regimes, as

19: well as its accuracy for real molecules and solids.

20: \end{abstract}

21: