We present the methodology for correlation analysis of chemical bond operators (CACB) on ab initio wave functions. In CACB the wave function is analyzed in a hierarchy of quantities (charge, bond order, bondbond correlation), where each quantity is the expectation value of an operator related to the statistical covariance of the previous quantity. CACB does not require any preconceived notion of which atoms are bonded and should be useful for reasoning about the similarity, stability, and reactivity of molecular systems. CACB does not require any special form of the wave function, but the applications here are for Hartree-Fock (HF) type wave functions. We use CACB to analyze the bonding in a number of molecules including transition states for several reactions. This analysis extracts chemically useful information without using preconcerned notations of bonding.
I. IntroductionProgress in quantum chemical (QC) calculations of organic and inorganic molecules has made it practicable to calculate the structures and wave functions for very large molecules, including the transition states (TS) and reaction intermediates involved in complicated but important reactions. Indeed the structural parameters, activation energies, and other properties are rapidly approaching the point where their reliability is sufficiently high to be trusted in the absence of experimental data. However, the interpretation of wave functions has lagged behind these developments in extending the methods. Thus, QC papers will often discuss only structures, energies, and vibrational frequencies, with no discussion of the wave function and how it can be used to understand these properties. This is unfortunate since the possibility of interpreting the wave functions is the unique attribute of quantum chemistry. The problem is that the wave function has too much information. The difficulty is extracting a few salient parameters that provide chemical intuition useful in qualitative reasoning.One strategy to extracting chemical information is typified by generalized valence bond (GVB) theory. 1 This utilizes a particular way to incorporate electron correlation and then extracts concepts directly from the GVB wave function. These GVB concepts are often closely related to valence bond concepts developed from empirical reasoning, and such GVB interpretations have often been useful for understanding the mechanisms of chemical reactions and the relationships between structure and energetics for various systems. 2 However, many ab initio studies involve very highly correlated wave functions not amenable to such orbital analyses. In addition, some firstprinciples methodology [density functional theory (DFT)] builds the electron correlation effects into a density functional that does not lead simply to GVB-like orbital interpretations. Thus, we wish to find a general way for extracting an interpretation directly from the wave function, without a preference for the particular nature of the wave function.
We present methodology ͑HBFF/SVD͒ for optimizing the form and parameters of force fields ͑FF͒ for molecular dynamics simulations through utilizing information about properties such as the geometry, Hessian, polarizability, stress ͑crystals͒, and elastic constants ͑crystals͒. This method is based on singular value decomposition ͑SVD͒ of the Jacobian describing the partial derivatives in various properties with respect to FF parameters. HBFF/SVD is effective for optimizing the parameters for accurate FFs of organic, inorganic, and transition metal compounds. In addition it provides information on the validity of the functional form of the FF for describing the properties of interest. This method is illustrated by application to organic molecules ͑CH 2 O, C 2 H 4 , C 4 H 6 , C 6 H 8 , C 6 H 6 , and naphthalene͒ and inorganic molecules ͑Cl 2 CrO 2 and Cl 2 MoO 2 ͒.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.