A Comparison of the Super-CI and the Newton-Raphson Scheme in the Complete Active Space SCF Method

Siegbahn, Per; Heiberg, Anders; Roos, Björn O.; Levy, Bernard C.

doi:10.1088/0031-8949/21/3-4/014

Cited by 696 publications

(365 citation statements)

References 12 publications

Supporting

Mentioning

363

Contrasting

Order By: Relevance

“…One method that has been used with some success is MRCI, where the most important CSFs of the CAS wave function are used as a multiconfigurational reference in a CI expansion that includes additional CSFs generated by single, double, or higher excitations from occupied orbitals of the reference to virtual orbitals. 19 This method, however, quickly demands enormous computational resources for systems with many electrons and also fails to be size extensive, although this latter problem can be corrected for in an approximate fashion. 20 In a single configuration approach, the preferred choices for including dynamical correlation derive from sizeextensive many-body theory and include coupled cluster ͑CC͒ methods or, if the system is too large, less demanding approximations such as second-order perturbation theory.…”

Section: A the Casscf/caspt2 Modelmentioning

confidence: 99%

The restricted active space followed by second-order perturbation theory method: Theory and application to the study of CuO2 and Cu2O2 systems

Malmqvist

Pierloot

Shahi

et al. 2008

The Journal of Chemical Physics

490

532

View full text Add to dashboard Cite

A multireference second-order perturbation theory using a restricted active space self-consistent field wave function as reference ͑RASPT2/RASSCF͒ is described. This model is particularly effective for cases where a chemical system requires a balanced orbital active space that is too large to be addressed by the complete active space self-consistent field model with or without second-order perturbation theory ͑CASPT2 or CASSCF, respectively͒. Rather than permitting all possible electronic configurations of the electrons in the active space to appear in the reference wave function, certain orbitals are sequestered into two subspaces that permit a maximum number of occupations or holes, respectively, in any given configuration, thereby reducing the total number of possible configurations. Subsequent second-order perturbation theory captures additional dynamical correlation effects. Applications of the theory to the electronic structure of complexes involved in the activation of molecular oxygen by mono-and binuclear copper complexes are presented. In the mononuclear case, RASPT2 and CASPT2 provide very similar results. In the binuclear cases, however, only RASPT2 proves quantitatively useful, owing to the very large size of the necessary active space.

show abstract

Section: A the Casscf/caspt2 Modelmentioning

confidence: 99%

The restricted active space followed by second-order perturbation theory method: Theory and application to the study of CuO2 and Cu2O2 systems

Malmqvist

Pierloot

Shahi

et al. 2008

The Journal of Chemical Physics

490

532

View full text Add to dashboard Cite

show abstract

“…For the excited states calculations, two methods were employed in this study: one is the single-excitation configuration interactions ͑CI-singles͒, 22 and the other one is the complete active space multiconfiguration self-consistent field ͑CASSCF͒. 23 The CIsingles method was used for numerical calculations along the reaction pathway and to obtain the potential energy surfaces, while the CASSCF method was used to refine the key points and calculate the vertical excitation energy. More importantly, CASSCF ground and excited states calculations serve as the bridges, which link the ground energy surface and the excited energy surface.…”

Section: Computational Detailsmentioning

confidence: 99%

A computational study of photoisomerization in Al3O3− clusters

Cui

Morrison

Han

2002

The Journal of Chemical Physics

View full text Add to dashboard Cite

Ab initio calculations are employed to understand the photoisomerization process in small Al 3 O 3 Ϫ clusters. This process is the first example of a photoinduced isomerization observed in an anion cluster gas-phase system. Potential energy surfaces for the ground state and the excited state ͑S 1 and T 1 ͒ are explored by means of B3LYP, MP2, CI-singles, and CASSCF methods. We demonstrate that the isomerization process occurs between the global minimum singlet state Book structure (C 2v , 1 A 1 ) and the triplet state Ring structure (C 2v , 3 B 2 ). The calculated vertical excitation energy is 3.62 eV at the CASSCF level of approximation, in good agreement with the experimental value ͑3.49 eV͒. A nonplanar conical intersection, which hosts the intersystem crossing between the S 1 and T 1 surfaces is identified at the region of around R(1,6)ϭ2.4 Å. Beyond the experimental results, we predict, that this isomerization is reversible upon absorption of a phonon with energy of 1.92 eV. Our results describe a unique system, whose structure depends on its spin multiplicity; it exists as the Book structure on singlet states and as the Ring structure on triplet states.

show abstract

“…The latter start with complete-active-space self-consistent-field (CASSCF) [13][14][15] calculations, which provide optimum (occupied and virtual) orbitals for the embedded clusters in their ground and excited states, and continue with second-order many-body perturbation theory (CASPT2) [16][17][18][19] calculations, which provide the ground and excited state energies of the embedded clusters involved in the Ce 3+ 4f → 5d optical absorptions.…”

Section: Introductionmentioning

confidence: 99%

Structural effects and4f−5dtransition shifts induced by La codoping in Ce-doped yttrium aluminum garnet: First-principles study

et al. 2010

View full text Add to dashboard Cite

According to first-principles calculations performed on Ce-doped and Ce,La-codoped yttrium aluminum garnet (YAG) Y 3 Al 5 O 12 , the effect of La-codoping on the local structure around Ce defects in Ce:YAG is an anisotropic expansion in overall, in opposition to recent propositions of local lattice compression. Its effect on the lowest Ce 3+ 4f → 5d transition is found to be a red shift, in agreement with experiments. The red shift is the result of a decrease of the difference between the energy centroids of the 5d 1 and 4f 1 configurations and an increase of the effective ligand field on the Ce-5d shell associated with electronic effects of La substituting for Y. These effects are mitigated by the ligand field decrease associated with the local expansion around Ce, which gives a blue shift contribution of a smaller value. The behavior of the energy difference between the centroids of the configurations cannot be anticipated by the usual model for this quantity, in spite of its usefulness to rationalize 5d → 4f luminescences. The second 4f → 5d transition is found to be blue shifted upon La-codoping, also in agreement with experiments.

show abstract

A Comparison of the Super-CI and the Newton-Raphson Scheme in the Complete Active Space SCF Method

Abstract: A comparison of the super-C I and the Newton-Raphson scheme in the ccimpletc. active space SCF method.

Cited by 696 publications

References 12 publications

The restricted active space followed by second-order perturbation theory method: Theory and application to the study of CuO2 and Cu2O2 systems

The restricted active space followed by second-order perturbation theory method: Theory and application to the study of CuO2 and Cu2O2 systems

A computational study of photoisomerization in Al3O3− clusters

Structural effects and4f−5dtransition shifts induced by La codoping in Ce-doped yttrium aluminum garnet: First-principles study

Contact Info

Product

Resources

About