O(N) methods are based on the localization properties of the density matrix in real space, an effect refered to as nearsightedness. We show that, in addition to this real-space localization there is also a localization in Fourier space. Using a basis set with good localization properties in both real and Fourier space such as wavelets, one can exploit both localization properties to obtain a density matrix that exhibits additional sparseness properties compared to the scenario where one has a basis set with real-space localization only. Taking advantage of this hypersparsity, it is possible to represent very large quantum-mechanical systems in a highly compact way. This can be done both for insulating and metallic systems. We expect that hypersparsity will pave the way for highly accurate
O(N) calculations of large systems requiring many basis functions per atom.
Reduced density-matrix functional theory (RDMFT) provides a variational route to electronic correlations beyond conventional density-functional approximations, but explicit evaluations of density-matrix functionals still scale exponentially with the number of active one-particle states. We formulate and assess a real-space-decomposed density-functional plus reduced-density-matrix-functional (DF+RDMF) scheme in which the Coulomb interaction is partitioned locally in real space and the RDMF correction is evaluated only for the strongly correlated part of the interaction. The resulting local density-matrix functionals are further compressed using the adaptive cluster approximation (ACA), which performs a unitary rotation of the bath subspace before truncation and therefore preserves the local interaction while reducing the number of explicitly correlated bath states. As a molecular test case, we consider the bending potential of carbon suboxide, C
3O
2. While semilocal PBE favors a linear molecule, the DF+RDMF/ACA correction stabilizes a bent configuration in qualitative agreement with the quasilinear behavior inferred from spectroscopy. The approach provides a systematic embedding hierarchy for combining density functionals with explicitly correlated density-matrix corrections in extended or spatially inhomogeneous systems.
The solution of the inverse problem for the time-dependent density matrix means finding the time-dependent Hamiltonian operator $\stackrel{^}{H}$ corresponding to a given time-dependent density matrix $\stackrel{^}{\ensuremath{\rho}}$. We point out that the standard requirements for the time-dependent density operator imply a number of compatibility relations. Accordingly, a given $N\ifmmode\times\else\texttimes\fi{}N$ density matrix might result from a whole family of Hamiltonians, characterized by
N free, real parameters. The cases of a 2\ifmmode\times\else\texttimes\fi{}2 and 3\ifmmode\times\else\texttimes\fi{}3 density matrix and of the free-fermion chain are worked out explicitly.