A generalization of the löwdin orthogonalization
Hiroshi Kashiwagi, Fukashi Sasaki
Abstract
Hiroshi Kashiwagi, Fukashi Sasaki
Abstract
The Löwdin orthonormalized set consists of those unique orthonormal functions which minimize the quantity Σi∫|gi — fi|2 dv, i.e., the sum of the squared distances in the Hilbert space between each initial function ff and a corresponding function gg of the orthonormal set. In this paper, the Löwdin orthogonalization is extended to a more general transformation. The new transformation also minimizes the sum Σi∫|gi — fi|2 dv dv, but the overlap matrix of the resultant functions gg is an arbitrary positive definite matrix in contrast to the unit matrix in the Löwdin orthogonalization. The new orbital set may be useful as a basis set for some many-electron problems in which the smallness of the sum Σi∫|gi — fi|2 dv and a particular form of the overlap matrix are requested at the same time.
OpenAlex reports 32 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The Löwdin orthonormalized set consists of those unique orthonormal functions which minimize the quantity Σi∫|gi — fi|2 dv, i.e., the sum of the squared distances in the Hilbert space between each initial function ff and a corresponding function gg of the orthonormal set. In this paper, the Löwdin orthogonalization is extended to a more general transformation. The new transformation also minimizes the sum Σi∫|gi — fi|2 dv dv, but the overlap matrix of the resultant functions gg is an arbitrary positive definite matrix in contrast to the unit matrix in the Löwdin orthogonalization. The new orbital set may be useful as a basis set for some many-electron problems in which the smallness of the sum Σi∫|gi — fi|2 dv and a particular form of the overlap matrix are requested at the same time.
Key concepts: Orthogonalization, Orthonormality, Orthonormal basis, Matrix (chemical analysis), Generalization, Set (abstract data type), Transformation (genetics), Hilbert space