Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics

Purification of correlated reduced density matrices

David A. Mazziotti

Published Jan 18, 2002• 77 citations
Publisher

Abstract

The notion of purification is generalized to treat correlated reduced density matrices. Traditionally, purification denotes the process by which a one-particle reduced density matrix (1-RDM) is made idempotent, that is, its eigenvalues are mapped to either 0 or 1. Purification of correlated RDMs is defined as the iterative process by which an arbitrary RDM is forced to satisfy several necessary N-representability conditions. Using the unitary decomposition of RDMs and the positivity conditions, we develop an algorithm to purify the 2-RDM. The algorithm is applied within the solution of the contracted Schrödinger equation CSE for the 2-RDM [D. A. Mazziotti, Phys. Rev. A 57, 4219 (1998)]. Previous attempts to solve the CSE by powerlike methods have frequently produced divergent energies, but we show that the purification process eliminates the divergent behavior for systematic and reliable convergence of the contracted power method to the N-particle energy.

DOI
10.1103/physreve.65.026704

Related papers

Solution of the 1,3-contracted Schrödinger equation through positivity conditions on the two-particle reduced density matrix

David A. Mazziotti

Correlation energies and reduced density matrices (RDMs) of atoms and molecules are directly computed by solving the 1,3-contracted Schr\"odinger equation (1,3-CSE). The solution of the 1,3-CSE synthesizes two optimization strategies recently employed for the direct determination of the 2-RDM: (i) variational minimization of the energy with respect to a 2-RDM constrained by positivity conditions [D. A. Mazziotti, Phys. Rev. A $65,$ 062511 (2002)] and (ii) the contracted power method for solving the 2,4-CSE [D. A. Mazziotti, J. Chem. Phys. $116,$ 1239 (2002)]. While both the 3- and the 4-RDMs in the 2,4-CSE are reconstructed from the 2-RDM by cumulant expansions, similar techniques cannot be directly applied to the 1,3-CSE because constructing the 2-RDM from the 1-RDM with cumulant theory does not improve upon the mean-field approximation. We, however, establish a unique mapping from the 1-RDM to the 2-RDM by searching for the 2-RDM, constrained by contraction and N-representability conditions, which minimizes the energy. The 2-RDM constrained search is practically implemented through recent advances in positive semidefinite programming. With the variational reconstruction of the 2-RDM and a cumulant reconstruction of the 3-RDM, the 1,3-CSE may be solved via a contracted power method for the ground-state energy and RDMs. The initial RDMs, it is shown, need not be N representable for the contracted power method to converge; this allows us to choose the original RDMs from a variational calculation with approximate N-representability conditions on the 2-RDM. Application of the 1,3-CSE algorithm to atoms and molecules yields highly accurate correlation energies both near and far from equilibrium geometries.

Physical Review A•Dec 12, 2002•48 citations

Variational method for solving the contracted Schrödinger equation through a projection of the N-particle power method onto the two-particle space

David A. Mazziotti

The power method for solving N-particle eigenvalue equations is contracted onto the two-particle space to produce a reduced “variational” method for solving the contracted Schrödinger equation (CSE), also known as the density equation. In contrast to the methods which solve a system of approximate nonlinear equations to determine the two-particle reduced density matrix (2-RDM) nonvariationally, the contracted power method updates the 2-RDM iteratively through a “gradient” of the N-particle energy. After each power iteration we modify the 2-RDM to satisfy certain N-representability conditions through an extension of purification to correlated RDMs. The contracted power method is illustrated with a variety of molecules. Significant features of the present calculations include (i) accurate results for both first- and second-order functionals for building the 3- and the 4-RDM’s from the 2-RDM’s; (ii) the first molecular implementation of the Mazziotti correction within the CSE [Mazziotti, Phys. Rev. A 60, 3618 (1999)]; (iii) a spin–orbital formulation; (iv) the treatment of both core and valence orbitals as active; and; (v) a reduction of the CSE computational scaling through fast summation and the natural-orbital transformation.

The Journal of Chemical Physics•Jan 22, 2002•94 citations

Linear scaling and the 1,2-contracted Schrödinger equation

David A. Mazziotti

A contracted Schrödinger equation (1,2-CSE) is derived for the class of Hamiltonians without explicit interactions including those from Hartree–Fock and density functional theories. With cumulant reconstruction of the two-particle reduced density matrix (2-RDM) from the one-particle-RDM (1-RDM), the 1,2-CSE may be expressed solely in terms of the 1-RDM. We prove that a 1-RDM satisfies the 1,2-CSE if and only if it is an eigenstate of the N-particle Schrödinger equation. The 1,2-CSE is solved through the development and implementation of a reduced, linear-scaling analog of the ordinary power method for finding matrix eigenvalues. The power formula for updating the 1-RDM requires fewer matrix operations than the gradient procedure derived by Li et al. [Phys. Rev. B 47, 10891 (1993)] and Daw [Phys. Rev. B 47, 10895 (1993)]. Convergence of the contracted power method with purification is illustrated with several molecules. While providing a new tool for semiempirical, Hartree–Fock, and density functional calculations, the 1,2-CSE also represents an initial step toward a linear-scaling algorithm for solving higher CSEs which explicitly treat electron correlation.

The Journal of Chemical Physics•Nov 8, 2001•40 citations

Towards idempotent reduced density matrices via particle-hole duality: McWeeny’s purification and beyond

David A. Mazziotti

Generalizations of McWeeny's purification formula are developed within the formalism of the particle-hole duality from the theory of reduced density matrices. Each of the generalized purification formulas is expressed as a sum of the one-particle reduced density matrix (1-RDM) and a finite series in the product of the one-particle and the one-hole RDMs, a product which vanishes in the limit that the 1-RDM is idempotent. Two categories of purification formulas are explored: (i) formulas which treat the "occupied" and the "virtual" occupation numbers equivalently and (ii) formulas which treat these occupation numbers differently. The latter category includes and extends the purification formulas derived in the context of the 1,2-contracted Schrödinger equation [D. A. Mazziotti, J. Chem. Phys. 115, 8305 (2001)]. While the McWeeny purification minimizes the absolute error in the occupation numbers quadratically, the generalized purification formulas exhibit faster than quadratic convergence of the 1-RDM towards idempotency. Application of these purification formulas in existing algorithms for linear scaling will be explored and discussed including illustrative calculations on sodium wires of length 10, 20, 30, and 40 atoms.

Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics•Dec 3, 2003•44 citations

Projective purification of correlated reduced density matrices

Elias Pescoller, Marie Eder, Iva Březinová

In the search for accurate approximate solutions of the many-body Schrödinger equation, reduced density matrices play an important role, as they allow one to formulate approximate methods with polynomial scaling in the number of particles. However, these methods frequently encounter the issue of N -representability, whereby in self-consistent applications of the methods, the reduced density matrices become unphysical. A number of algorithms have been proposed in the past to restore a given set of N -representability conditions once the reduced density matrices become defective. However, these purification algorithms either have ignored symmetries of the Hamiltonian related to conserved quantities or have not incorporated them in an efficient way, thereby modifying the reduced density matrix to a greater extent than is necessary. In this paper, we present an algorithm capable of efficiently performing all of the following tasks in the least invasive manner: restoring a given set of N -representability conditions, maintaining contraction consistency between successive orders of reduced density matrices, and preserving all conserved quantities. We demonstrate the superiority of the present purification algorithm over previous ones in the context of the time-dependent two-particle reduced density matrix method applied to the quench dynamics of the Fermi-Hubbard model.

arXiv:2412.13566•Physical Review Research•Feb 26, 2025•4 citations

Variational reduced-density-matrix calculation of the one-dimensional Hubbard model

Jeff R. Hammond, David A. Mazziotti

Variational reduced-density-matrix theory is applied to calculating the ground-state energy and two-electron reduced density matrices (2-RDMs) of the one-dimensional Hubbard model for a range of interaction strengths. The 2-RDM is constrained to represent an $N$-particle wave function by two sets of $N$-representability conditions, known as the 2- and partial 3-positivity conditions. Variational optimization of the energy with the 2-RDM constrained by $N$-representability conditions is performed using a first-order semidefinite-programming algorithm that was developed for treating atoms and molecules [D. A. Mazziotti, Phys. Rev. Lett. 93, 213001 (2004)]. Accurate energies for a broad range of interaction strengths indicate that the variational 2-RDM method is a valuable tool for studying strongly correlated electrons.

Physical Review A•Jun 13, 2006•41 citations
The notion of purification is generalized to treat correlated reduced density matrices. Traditionally, purification denotes the process by which a one-particle reduced density matrix (1-RDM) is made idempotent, that is, its eigenvalues are mapped to either 0 or 1. Purification of correlated RDMs is defined as the iterative process by which an arbitrary RDM is forced to satisfy several necessary N-representability conditions. Using the unitary decomposition of RDMs and the positivity conditions, we develop an algorithm to purify the 2-RDM. The algorithm is applied within the solution of the contracted Schrödinger equation CSE for the 2-RDM [D. A. Mazziotti, Phys. Rev. A 57, 4219 (1998)]. Previous attempts to solve the CSE by powerlike methods have frequently produced divergent energies, but we show that the purification process eliminates the divergent behavior for systematic and reliable convergence of the contracted power method to the N-particle energy.
Correlation energies and reduced density matrices (RDMs) of atoms and molecules are directly computed by solving the 1,3-contracted Schr\"odinger equation (1,3-CSE). The solution of the 1,3-CSE synthesizes two optimization strategies recently employed for the direct determination of the 2-RDM: (i) variational minimization of the energy with respect to a 2-RDM constrained by positivity conditions [D. A. Mazziotti, Phys. Rev. A 65,65,65, 062511 (2002)] and (ii) the contracted power method for solving the 2,4-CSE [D. A. Mazziotti, J. Chem. Phys. 116,116,116, 1239 (2002)]. While both the 3- and the 4-RDMs in the 2,4-CSE are reconstructed from the 2-RDM by cumulant expansions, similar techniques cannot be directly applied to the 1,3-CSE because constructing the 2-RDM from the 1-RDM with cumulant theory does not improve upon the mean-field approximation. We, however, establish a unique mapping from the 1-RDM to the 2-RDM by searching for the 2-RDM, constrained by contraction and N-representability conditions, which minimizes the energy. The 2-RDM constrained search is practically implemented through recent advances in positive semidefinite programming. With the variational reconstruction of the 2-RDM and a cumulant reconstruction of the 3-RDM, the 1,3-CSE may be solved via a contracted power method for the ground-state energy and RDMs. The initial RDMs, it is shown, need not be N representable for the contracted power method to converge; this allows us to choose the original RDMs from a variational calculation with approximate N-representability conditions on the 2-RDM. Application of the 1,3-CSE algorithm to atoms and molecules yields highly accurate correlation energies both near and far from equilibrium geometries.
The power method for solving N-particle eigenvalue equations is contracted onto the two-particle space to produce a reduced “variational” method for solving the contracted Schrödinger equation (CSE), also known as the density equation. In contrast to the methods which solve a system of approximate nonlinear equations to determine the two-particle reduced density matrix (2-RDM) nonvariationally, the contracted power method updates the 2-RDM iteratively through a “gradient” of the N-particle energy. After each power iteration we modify the 2-RDM to satisfy certain N-representability conditions through an extension of purification to correlated RDMs. The contracted power method is illustrated with a variety of molecules. Significant features of the present calculations include (i) accurate results for both first- and second-order functionals for building the 3- and the 4-RDM’s from the 2-RDM’s; (ii) the first molecular implementation of the Mazziotti correction within the CSE [Mazziotti, Phys. Rev. A 60, 3618 (1999)]; (iii) a spin–orbital formulation; (iv) the treatment of both core and valence orbitals as active; and; (v) a reduction of the CSE computational scaling through fast summation and the natural-orbital transformation.
A contracted Schrödinger equation (1,2-CSE) is derived for the class of Hamiltonians without explicit interactions including those from Hartree–Fock and density functional theories. With cumulant reconstruction of the two-particle reduced density matrix (2-RDM) from the one-particle-RDM (1-RDM), the 1,2-CSE may be expressed solely in terms of the 1-RDM. We prove that a 1-RDM satisfies the 1,2-CSE if and only if it is an eigenstate of the N-particle Schrödinger equation. The 1,2-CSE is solved through the development and implementation of a reduced, linear-scaling analog of the ordinary power method for finding matrix eigenvalues. The power formula for updating the 1-RDM requires fewer matrix operations than the gradient procedure derived by Li et al. [Phys. Rev. B 47, 10891 (1993)] and Daw [Phys. Rev. B 47, 10895 (1993)]. Convergence of the contracted power method with purification is illustrated with several molecules. While providing a new tool for semiempirical, Hartree–Fock, and density functional calculations, the 1,2-CSE also represents an initial step toward a linear-scaling algorithm for solving higher CSEs which explicitly treat electron correlation.
Generalizations of McWeeny's purification formula are developed within the formalism of the particle-hole duality from the theory of reduced density matrices. Each of the generalized purification formulas is expressed as a sum of the one-particle reduced density matrix (1-RDM) and a finite series in the product of the one-particle and the one-hole RDMs, a product which vanishes in the limit that the 1-RDM is idempotent. Two categories of purification formulas are explored: (i) formulas which treat the "occupied" and the "virtual" occupation numbers equivalently and (ii) formulas which treat these occupation numbers differently. The latter category includes and extends the purification formulas derived in the context of the 1,2-contracted Schrödinger equation [D. A. Mazziotti, J. Chem. Phys. 115, 8305 (2001)]. While the McWeeny purification minimizes the absolute error in the occupation numbers quadratically, the generalized purification formulas exhibit faster than quadratic convergence of the 1-RDM towards idempotency. Application of these purification formulas in existing algorithms for linear scaling will be explored and discussed including illustrative calculations on sodium wires of length 10, 20, 30, and 40 atoms.
In the search for accurate approximate solutions of the many-body Schrödinger equation, reduced density matrices play an important role, as they allow one to formulate approximate methods with polynomial scaling in the number of particles. However, these methods frequently encounter the issue of N -representability, whereby in self-consistent applications of the methods, the reduced density matrices become unphysical. A number of algorithms have been proposed in the past to restore a given set of N -representability conditions once the reduced density matrices become defective. However, these purification algorithms either have ignored symmetries of the Hamiltonian related to conserved quantities or have not incorporated them in an efficient way, thereby modifying the reduced density matrix to a greater extent than is necessary. In this paper, we present an algorithm capable of efficiently performing all of the following tasks in the least invasive manner: restoring a given set of N -representability conditions, maintaining contraction consistency between successive orders of reduced density matrices, and preserving all conserved quantities. We demonstrate the superiority of the present purification algorithm over previous ones in the context of the time-dependent two-particle reduced density matrix method applied to the quench dynamics of the Fermi-Hubbard model.
Variational reduced-density-matrix theory is applied to calculating the ground-state energy and two-electron reduced density matrices (2-RDMs) of the one-dimensional Hubbard model for a range of interaction strengths. The 2-RDM is constrained to represent an NNN-particle wave function by two sets of NNN-representability conditions, known as the 2- and partial 3-positivity conditions. Variational optimization of the energy with the 2-RDM constrained by NNN-representability conditions is performed using a first-order semidefinite-programming algorithm that was developed for treating atoms and molecules [D. A. Mazziotti, Phys. Rev. Lett. 93, 213001 (2004)]. Accurate energies for a broad range of interaction strengths indicate that the variational 2-RDM method is a valuable tool for studying strongly correlated electrons.