pith. sign in

arxiv: 2606.17761 · v1 · pith:WRJDX23Unew · submitted 2026-06-16 · ⚛️ physics.chem-ph

Constrained Optimization Algorithms for Orbital Optimization in Quantum Chemistry

Junzhe Zhang , Shuoyi Hu , Bing Gu This is my paper

Pith reviewed 2026-06-26 22:13 UTC · model grok-4.3

classification ⚛️ physics.chem-ph
keywords constrained orbital optimizationCASSCFreduced density matricesStiefel manifoldquantum chemistryorbital rotationpotential energy curves
0
0 comments X

The pith

A modular framework using only reduced density matrices allows orbital optimization that recovers lower-energy solutions than conventional CASSCF iterations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a constrained optimization method for updating molecular orbitals that works with any electronic structure solver providing one- and two-particle reduced density matrices. This separation lets the same orbital optimizer serve MP2, CASCI, and DMRG calculations. In cases where standard CASSCF orbital rotations converge to higher local energy minima, the new approach finds lower-energy stationary points. Applications to molecules such as LiF, water, and pyrazine demonstrate lower total energies and smoother potential energy curves compared to fixed-orbital references.

Core claim

The constrained orbital optimization (CO-CAS) framework separates the solver from the optimizer, updating orbitals on the Stiefel manifold via implicit steepest descent driven solely by reduced density matrices, which enables recovery of lower-energy stationary solutions when conventional CASSCF iterations become trapped in local minima.

What carries the argument

The orthonormality-constrained optimization on the Stiefel manifold with an implicit steepest-descent algorithm that takes reduced density matrices as input.

If this is right

  • Orbital optimization with this method lowers energies relative to fixed-orbital MP2, CASCI, and DMRG calculations.
  • It improves convergence behavior and the smoothness of potential-energy curves.
  • The same interface works for multiple solvers including MP2, CASCI, and DMRG.
  • A modified direct inversion in the iterative subspace accelerates macro-iteration convergence.
  • Dynamical weighting improves state-averaged excited-state calculations.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This approach might extend to other correlated methods beyond those tested by supplying appropriate density matrices.
  • Improved orbital optimization could lead to more reliable excited-state calculations in larger molecules where local minima are common.
  • Since it avoids direct Hamiltonian access, it may integrate more easily with approximate solvers that only output density matrices.

Load-bearing premise

That reduced density matrices from the solver contain enough information to guide the orbital optimizer to better stationary points without additional wavefunction details or Hamiltonian elements.

What would settle it

Running both conventional CASSCF and the constrained optimizer on the same molecular system and finding that the constrained method does not yield a lower energy stationary point than the conventional one.

Figures

Figures reproduced from arXiv: 2606.17761 by Bing Gu, Junzhe Zhang, Shuoyi Hu.

Figure 1
Figure 1. Figure 1: LiF ground-state potential energy curves calculated with CO-MP2 and MP2 using the 6-31G basis set. 4.2 CO-CAS: CASSCF Bench￾mark, DIIS Acceleration, and Excited States We next combine constrained orbital optimiza￾tion with a CASCI solver, denoted CO-CAS, and apply it to the LiF ground-state poten￾tial energy curve [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: LiF ground-state PECs calculated with CO-CAS(6,6) and CASSCF(6,6) using the 6-31G and aug-cc-pVDZ basis sets. Next, we calculate the water molecule at its equilibrium geometry, with an OH bond length of 1.84345a0 and an HOH bond angle of 110.6◦ . Using 10 active electrons, we vary the number of active orbitals and compare the ground-state energies obtained from CO-CAS and CASSCF with the cc-pVDZ and cc-pVQ… view at source ↗
Figure 4
Figure 4. Figure 4: LiF ground- and first-excited-state PECs calculated with different methods using the 6-31G basis set and a CAS(6,6) subspace. DW-CO-CAS denotes dynamically weighted constrained optimization configuration interaction, SA-CO-CAS denotes state-averaged constrained optimization configuration interaction, and SA-CASSCF denotes state-averaged complete active space self-consistent field. calculation. The energy l… view at source ↗
Figure 3
Figure 3. Figure 3: Convergence of LiF CO-CAS ground-state energy calculations with and without DIIS for different basis sets. The maximum number of macro-iterations is 200. CAS yields lower energies than SA-CASSCF for both states: the ground-state energy is lowered by approximately 10 mEh, while the first ex￾cited state is lowered by approximately 1 mEh. These results indicate that dynamical weight￾ing provides a more flexib… view at source ↗
Figure 5
Figure 5. Figure 5: LiF ground-state energies as a function of bond length calculated with CO-DMRG and DMRG in a CAS(6,6) active space with the 6-31G basis set. the STO-6G basis set, the calculation converges in two macro-iterations with or without DIIS. DIIS slightly accelerates convergence for the cc-pVDZ and aug-cc-pVDZ basis sets, but it slows convergence for cc-pVTZ in the present tests. Therefore, DIIS is clearly benefi… view at source ↗
Figure 7
Figure 7. Figure 7: Convergence of LiF CO-DMRG ground-state energy calculations with and without DIIS for different basis sets. tensor-network wavefunctions. These results demonstrate that ISD-based constrained orbital optimization is a useful and general strategy for correlated electronic￾structure methods. The method is not re￾stricted to a specific wavefunction ansatz; rather, it can be combined with different elec￾tronic … view at source ↗
read the original abstract

We present a modular constrained-orbital-optimization framework for quantum chemistry. The formulation separates the correlated electronic-structure solver from the orbital optimizer: the solver supplies one- and two-particle reduced density matrices, while the molecular orbitals are updated on the orthonormality-constrained Stiefel manifold with an implicit steepest-descent algorithm. Because the orbital optimizer only requires reduced density matrices, MP2, CASCI, and DMRG can be treated within the same interface. For CASCI solvers, the approach is closely related to optimal-orbital full configuration interaction and CASSCF\cite{helgaker_MulticonfigurationalSelfConsistentField_2000a}, but uses a solver-independent constrained-optimization update rather than CAS-specific orbital-rotation equations. When conventional CASSCF orbital-rotation iterations converge to higher-energy local solutions, CO-CAS can recover lower-energy stationary solutions. We also introduce a modified direct inversion in the iterative subspace procedure to accelerate macro-iteration convergence and a dynamical-weighting scheme to improve state-averaged excited-state calculations. Applications to LiF, H$_2$O, and pyrazine show that orbital optimization lowers energies relative to fixed-orbital MP2, CASCI, and DMRG references while improving convergence and potential-energy-curve smoothness.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript presents a modular constrained-orbital-optimization (CO) framework that decouples the correlated electronic-structure solver from the orbital optimizer. The solver supplies only the one- and two-particle reduced density matrices while the orbitals are updated on the Stiefel manifold via an implicit steepest-descent algorithm; a modified DIIS accelerator and dynamical-weighting scheme are also introduced. The approach is applied to MP2, CASCI, and DMRG solvers and is claimed to recover lower-energy stationary points than conventional CASSCF orbital-rotation iterations on LiF, H_{2}O, and pyrazine while producing smoother potential-energy curves.

Significance. If the modularity and performance claims are substantiated, the framework would supply a uniform, solver-agnostic route to orbital optimization that can escape local minima encountered by standard CASSCF and improve convergence across multiple correlated methods. The explicit separation of RDM provision from the manifold-constrained update is a potentially reusable design pattern in multireference quantum chemistry.

major comments (1)
  1. [Abstract] Abstract: the assertion that 'the orbital optimizer only requires reduced density matrices' is inconsistent with the stated algorithm. Any first-order method on the Stiefel manifold must evaluate either the energy E = Tr(hD) + ½ Tr(gΓ) or its Riemannian gradient; both contractions require the AO one- and two-electron integrals (h, g) in addition to the supplied RDMs. This directly weakens the central claim of a fully solver-independent interface that needs 'no direct access' to the Hamiltonian.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for identifying an important point of clarification regarding the solver-orbital optimizer interface. We address the single major comment below and will make the requested revision to improve accuracy.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the assertion that 'the orbital optimizer only requires reduced density matrices' is inconsistent with the stated algorithm. Any first-order method on the Stiefel manifold must evaluate either the energy E = Tr(hD) + ½ Tr(gΓ) or its Riemannian gradient; both contractions require the AO one- and two-electron integrals (h, g) in addition to the supplied RDMs. This directly weakens the central claim of a fully solver-independent interface that needs 'no direct access' to the Hamiltonian.

    Authors: We agree that the orbital optimizer requires the AO one- and two-electron integrals (h, g) in addition to the RDMs supplied by the solver, since the energy or Riemannian gradient on the Stiefel manifold is obtained via the contractions E = Tr(hD) + ½ Tr(gΓ). The modularity claim refers specifically to the fact that the correlated solver (MP2, CASCI, or DMRG) is completely decoupled from the orbital-optimization loop: it only furnishes the RDMs and has no knowledge of the manifold constraints or the orbital-update algorithm. The integrals remain inside the orbital-optimizer module and are therefore solver-independent. Nevertheless, the abstract phrasing that the optimizer 'only requires reduced density matrices' is imprecise. We will revise the abstract (and the corresponding sentence in the introduction) to state explicitly that the optimizer requires the RDMs from the solver together with the one- and two-electron integrals, thereby removing any implication of a Hamiltonian-free interface. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper introduces a modular orbital optimizer on the Stiefel manifold driven by supplied 1- and 2-RDMs, with the update equations and convergence claims resting on standard constrained-optimization machinery and an external citation to Helgaker et al. for conventional CASSCF. No load-bearing step reduces by construction to a fitted parameter, self-citation chain, or definitional equivalence; the reported improvements on LiF, H2O, and pyrazine are presented as empirical outcomes of the algorithm rather than tautological restatements of its inputs. The derivation chain is therefore self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review; no explicit free parameters, new axioms, or invented entities are stated. The framework rests on standard concepts of reduced density matrices and the Stiefel manifold from prior literature.

pith-pipeline@v0.9.1-grok · 5753 in / 1183 out tokens · 35061 ms · 2026-06-26T22:13:39.284772+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

71 extracted references · 17 canonical work pages · 1 internal anchor

  1. [1]

    The Complete Active Space

    Roos, Bj. The Complete Active Space. International Journal of Quantum Chemistry , volume =

  2. [2]

    Gradients and Non-Adiabatic Derivative Coupling Terms for Spin-Orbit Wavefunctions -

  3. [3]

    Second-Order Perturbation Theory with a Complete Active Space Self-Consistent Field Reference Function , author =. J. Chem. Phys. , volume =

  4. [4]

    The Density Matrix Renormalization Group in Chemistry and Molecular Physics:

    Baiardi, Alberto and Reiher, Markus , year = 2020, journal =. The Density Matrix Renormalization Group in Chemistry and Molecular Physics:

    2020

  5. [5]

    , year = 1988, journal =

    Barzilai, Jonathan and Borwein, Jonathan M. , year = 1988, journal =. Two-

    1988

  6. [6]

    Gradients and

    Belcher, Lachlan , year = 2011, pages =. Gradients and

    2011

  7. [7]

    Quadratically Convergent Algorithm for Orbital Optimization in the Orbital-Optimized Coupled-Cluster Doubles Method and in Orbital-Optimized Second-Order

    Bozkaya, U. Quadratically Convergent Algorithm for Orbital Optimization in the Orbital-Optimized Coupled-Cluster Doubles Method and in Orbital-Optimized Second-Order. J. Chem. Phys. , volume =

  8. [8]

    Chan, Garnet Kin-Lic and Sharma, Sandeep , year = 2011, journal =. The

    2011

  9. [9]

    M ller--

    Cremer, Dieter , year = 2011, journal =. M ller--. doi:10.1002/wcms.58 , copyright =

    work page doi:10.1002/wcms.58 2011

  10. [10]

    Analytical Gradients and Derivative Couplings for Dynamically Weighted Complete Active Space Self-Consistent Field , author =. J. Chem. Phys. , volume =

  11. [11]

    Glover, W. J. , year = 2014, journal =. Communication:

    2014

  12. [12]

    and Mukamel, Shaul , year = 2023, journal =

    Gu, Bing and Gu, Yonghao and Chernyak, Vladimir Y. and Mukamel, Shaul , year = 2023, journal =. Cavity. doi:10.1021/acs.accounts.3c00280 , copyright =

    work page doi:10.1021/acs.accounts.3c00280 2023

  13. [13]

    Gu, Bing , year = 2025, journal =. Toward. doi:10.1021/acs.jpclett.4c02896 , copyright =

    work page doi:10.1021/acs.jpclett.4c02896 2025

  14. [14]

    Nonadiabatic

    Gu, Bing , year = 2024, journal =. Nonadiabatic. doi:10.1021/acs.jctc.3c01317 , copyright =

    work page doi:10.1021/acs.jctc.3c01317 2024

  15. [15]

    Gu, Bing and Ren, Jiajun and Zhang, Junzhe , year = 2025, journal =. Quantum. doi:10.1021/acs.jctc.5c00029 , copyright =

    work page doi:10.1021/acs.jctc.5c00029 2025

  16. [16]

    and Knowles, Peter J

    Kreplin, David A. and Knowles, Peter J. and Werner, Hans-Joachim , year = 2020, journal =

    2020

  17. [17]

    and Knowles, Peter J

    Kreplin, David A. and Knowles, Peter J. and Werner, Hans-Joachim , year = 2019, journal =. Second-Order

    2019

  18. [18]

    Regularized

    Lee, Joonho and. Regularized. J. Chem. Theory Comput. , volume =

  19. [19]

    Li, Yingzhou and Lu, Jianfeng , year = 2020, journal =. Optimal. doi:10.1021/acs.jctc.0c00613 , copyright =

    work page doi:10.1021/acs.jctc.0c00613 2020

  20. [20]

    Lochan, Rohini C. and. Orbital-Optimized Opposite-Spin Scaled Second-Order Correlation:. J. Chem. Phys. , volume =

  21. [21]

    Optim Lett , volume =

    Implicit Steepest Descent Algorithm for Optimization with Orthogonality Constraints , author =. Optim Lett , volume =

  22. [22]

    the Case of Scf Iteration , author =

    Convergence Acceleration of Iterative Sequences. the Case of Scf Iteration , author =. Chemical Physics Letters , volume =

  23. [23]

    A General Automatic Method for Optimal Construction of Matrix Product Operators Using Bipartite Graph Theory , author =. J. Chem. Phys. , volume =

  24. [24]

    A Simple Method for the Evaluation of the Second-Order-Perturbation Energy from External Double-Excitations with a

    Roos, Bj. A Simple Method for the Evaluation of the Second-Order-Perturbation Energy from External Double-Excitations with a. Chemical Physics , volume =

  25. [25]

    Annals of Physics , series =

    The Density-Matrix Renormalization Group in the Age of Matrix Product States , author =. Annals of Physics , series =

  26. [26]

    Spin-Adapted Density Matrix Renormalization Group Algorithms for Quantum Chemistry , author =. J. Chem. Phys. , volume =

  27. [27]

    Pyrazine Excited States Revisited Using the Extended Multi-State Complete Active Space Second-Order Perturbation Method , author =. Phys. Chem. Chem. Phys. , volume =

  28. [28]

    Siegbahn, Per E. M. and Alml. The Complete Active Space. J. Chem. Phys. , volume =

  29. [29]

    Siegbahn, Per E. M. , year = 1984, journal =. A New Direct

    1984

  30. [30]

    and Blunt, Nick S

    Sun, Qiming and Berkelbach, Timothy C. and Blunt, Nick S. and Booth, George H. and Guo, Sheng and Li, Zhendong and Liu, Junzi and McClain, James D. and Sayfutyarova, Elvira R. and Sharma, Sandeep and Wouters, Sebastian and Chan, Garnet Kin-Lic , year = 2018, journal =

    2018

  31. [31]

    Nondirect-

    Xie, Yujuan and Yang, Yukun and Zhu, Xiaotong and Chen, Ahai and Gu, Bing , year = 2024, journal =. Nondirect-. doi:10.1021/acs.jctc.4c00673 , copyright =

    work page doi:10.1021/acs.jctc.4c00673 2024

  32. [32]

    doi:10.48550/arXiv.2501.05003 , archiveprefix =

    Path-Ordered Linked Product Approximation to the Global Electronic Overlap Matrix , author =. doi:10.48550/arXiv.2501.05003 , archiveprefix =. 2501.05003 , primaryclass =

    work page doi:10.48550/arxiv.2501.05003

  33. [33]

    The Density Matrix Renormalization Group Self-Consistent Field Method:

    Zgid, Dominika and Nooijen, Marcel , year = 2008, journal =. The Density Matrix Renormalization Group Self-Consistent Field Method:

    2008

  34. [34]

    Obtaining the Two-Body Density Matrix in the Density Matrix Renormalization Group Method , author =. J. Chem. Phys. , volume =

  35. [35]

    Zhu, Xiaotong and Gu, Bing , year = 2024, journal =. Making. doi:10.1021/acs.jpclett.4c01688 , copyright =

    work page doi:10.1021/acs.jpclett.4c01688 2024

  36. [36]

    doi:10.1063/1674-0068/cjcp2510161 , copyright =

    Xie, Yujuan and Zhu, Xiaotong and Gu, Bing , year = 2026, journal =. doi:10.1063/1674-0068/cjcp2510161 , copyright =

    work page doi:10.1063/1674-0068/cjcp2510161 2026

  37. [37]

    Edelman, Alan and Arias, T. A. and Smith, Steven T. , year = 1998, number =. The. doi:10.48550/arXiv.physics/9806030 , archiveprefix =. physics/9806030 , publisher =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.physics/9806030 1998

  38. [38]

    A Framework of Constraint Preserving Update Schemes for Optimization on

    Jiang, Bo and Dai, Yu-Hong , year = 2015, journal =. A Framework of Constraint Preserving Update Schemes for Optimization on

    2015

  39. [39]

    A Feasible Method for Optimization with Orthogonality Constraints , author =. Math. Program. , volume =

  40. [40]

    Density Matrix Formulation for Quantum Renormalization Groups , author =. Phys. Rev. Lett. , volume =. doi:10.1103/PhysRevLett.69.2863 , copyright =

    work page doi:10.1103/physrevlett.69.2863

  41. [41]

    Density-Matrix Algorithms for Quantum Renormalization Groups , author =. Phys. Rev. B , volume =

  42. [42]

    Weser, Oskar and Liebermann, Niklas and Kats, Daniel and Alavi, Ali and Li Manni, Giovanni , year = 2022, journal =. Spin. doi:10.1021/acs.jpca.2c01338 , copyright =

    work page doi:10.1021/acs.jpca.2c01338 2022

  43. [43]

    2012 , publisher=

    Modern quantum chemistry: introduction to advanced electronic structure theory , author=. 2012 , publisher=

    2012

  44. [44]

    SIAM Journal on Optimization , volume=

    A Riemannian BFGS method without differentiated retraction for nonconvex optimization problems , author=. SIAM Journal on Optimization , volume=. 2018 , publisher=

    2018

  45. [45]

    2008 , publisher=

    Optimization algorithms on matrix manifolds , author=. 2008 , publisher=

    2008

  46. [46]

    Angeli, Celestino and Cimiraglia, Renzo and Malrieu, Jean-Paul , year = 2001, journal =

    2001

  47. [47]

    A Quasidegenerate Formulation of the Second Order N-Electron Valence State Perturbation Theory Approach , author =. J. Chem. Phys. , volume =

  48. [48]

    Quantum-Information Analysis of Electronic States of Different Molecular Structures , author =. Phys. Rev. A , volume =

  49. [49]

    Analytic Energy Gradients for the Orbital-Optimized Second-Order

    Bozkaya, U. Analytic Energy Gradients for the Orbital-Optimized Second-Order. J. Chem. Phys. , volume =

  50. [50]

    Orbital-Optimized

    Bozkaya, U. Orbital-Optimized. J. Chem. Phys. , volume =

  51. [51]

    Orbital-

    Bozkaya, U. Orbital-. J. Chem. Theory Comput. , volume =

  52. [52]

    An Algorithm for Large Scale Density Matrix Renormalization Group Calculations , author =. J. Chem. Phys. , volume =

  53. [53]

    Highly Correlated Calculations with a Polynomial Cost Algorithm:

    Chan, Garnet Kin-Lic and. Highly Correlated Calculations with a Polynomial Cost Algorithm:. J. Chem. Phys. , volume =

  54. [54]

    Matrix Product Operators, Matrix Product States, and Ab Initio Density Matrix Renormalization Group Algorithms , author =. J. Chem. Phys. , volume =

  55. [55]

    The Multi-State

    Finley, James and Malmqvist, Per-. The Multi-State. Chemical Physics Letters , volume =

  56. [56]

    Orbital Optimization in the Density Matrix Renormalization Group, with Applications to Polyenes and

    Ghosh, Debashree and Hachmann, Johannes and Yanai, Takeshi and Chan, Garnet Kin-Lic , year = 2008, journal =. Orbital Optimization in the Density Matrix Renormalization Group, with Applications to Polyenes and

    2008

  57. [57]

    Hartree--Fock Theory , booktitle =

    Helgaker, Trygve and J. Hartree-. Molecular. doi:10.1002/9781119019572.ch10 , chapter =

    work page doi:10.1002/9781119019572.ch10

  58. [58]

    Multiconfigurational

    Helgaker, Trygve and J. Multiconfigurational. Molecular. doi:10.1002/9781119019572.ch12 , chapter =

    work page doi:10.1002/9781119019572.ch12

  59. [59]

    Chemical Physics Letters , volume =

    A New Determinant-Based Full Configuration Interaction Method , author =. Chemical Physics Letters , volume =

  60. [60]

    Correlated Ab

    Kossmann, Simone and Neese, Frank , year = 2010, journal =. Correlated Ab

    2010

  61. [61]

    and Scuseria, Gustavo E

    Kudin, Konstantin N. and Scuseria, Gustavo E. and Canc. A Black-Box Self-Consistent Field Convergence Algorithm:. J. Chem. Phys. , volume =

  62. [62]

    Numerical

    Conjugate. Numerical. doi:10.1007/0-387-22742-3_5 , isbn =

    work page doi:10.1007/0-387-22742-3_5

  63. [63]

    Determinant Based Configuration Interaction Algorithms for Complete and Restricted Configuration Interaction Spaces , author =. J. Chem. Phys. , volume =

  64. [64]

    , title =

    Pulay, P. , year = 1982, journal =. Improved. doi:10.1002/jcc.540030413 , copyright =

    work page doi:10.1002/jcc.540030413 1982

  65. [65]

    A Complete Active Space

    Roos, Bj. A Complete Active Space. Chemical Physics , volume =

  66. [66]

    Regularized Orbital-Optimized Second-Order Perturbation Theory , author =. J. Chem. Phys. , volume =

  67. [67]

    and Bogdanov, Nikolay A

    Sun, Qiming and Zhang, Xing and Banerjee, Samragni and Bao, Peng and Barbry, Marc and Blunt, Nick S. and Bogdanov, Nikolay A. and Booth, George H. and Chen, Jia and Cui, Zhi-Hao and Eriksen, Janus J. and Gao, Yang and Guo, Sheng and Hermann, Jan and Hermes, Matthew R. and Koh, Kevin and Koval, Peter and Lehtola, Susi and Li, Zhendong and Liu, Junzi and Ma...

    2020

  68. [68]

    A Quadratically Convergent

    Werner, Hans-Joachim and Meyer, Wilfried , year = 1981, journal =. A Quadratically Convergent

    1981

  69. [69]

    A Quadratically Convergent Multiconfiguration--Self-consistent Field Method with Simultaneous Optimization of Orbitals and

    Werner, Hans-Joachim and Meyer, Wilfried , year = 1980, journal =. A Quadratically Convergent Multiconfiguration--Self-consistent Field Method with Simultaneous Optimization of Orbitals and

    1980

  70. [70]

    The Density Matrix Renormalization Group for Ab Initio Quantum Chemistry , author =. Eur. Phys. J. D , volume =

  71. [71]

    and Lee, Seunghoon and Cui, Zhi-Hao and Zhu, Tianyu and Sun, Chong and Peng, Linqing and Peng, Ruojing and Liao, Ke and T

    Zhai, Huanchen and Larsson, Henrik R. and Lee, Seunghoon and Cui, Zhi-Hao and Zhu, Tianyu and Sun, Chong and Peng, Linqing and Peng, Ruojing and Liao, Ke and T. Block2:. J. Chem. Phys. , volume =