Application of second-order density functional methods to the calculation of the BeFH potential energy surface

Abstract

ABSTRACTŽ Second-order density functional methods where the correlation energy depends on the . second-order density matrix and on a density functional are used to introduce the Ž . electron correlation in two-configuration direct minimization TCDM ab initio electronic Ž . energy calculations of three-dimensional potential energy surfaces PES . We analyzed the behavior of these methods in PES calculations by applying them to the Be q FH reaction. This system was studied also by the usual techniques, allowing a full comparison for the lowest 1 A X adiabatic state. In particular, we compared the results obtained using the TCDM and multiple reference single and double excitations configuration interaction Ž . Ž . MRDCI methods with the corresponding results obtained using the Colle᎐Salvetti CS Ž . and Moscardo᎐San Fabian MSF procedures, within the correlation factor method, usinǵá s starting point the TCDM results. We found that the CS and MSF results are in a good overall agreement with the more accurate ab initio results, including the heights of the saddle points and the transition state.

Renormalized Coupled-Cluster Calculations of Reactive Potential Energy Surfaces: The BeFH System

“…Clearly, the overall description of the product channel by the CCSD(T) method is not correct. For example, the endothermicity of the Be + HF f BeF + H reaction of -0.009 eV, obtained with the CCSD(T) method, has the wrong sign when compared to the MRCI(Q) endothermicity of 0.140 eV or the MRCI value reported by Aguado et al 2 CCSD(T) result for the endothermicity of the Be + HF f BeF + H reaction is also in very good agreement with the experimentally derived value of 0.193 eV, obtained using the experimental data for the binding energies of HF 47 and BeF. 48 The product (BeF + H) valley and the Be + F + H asymptotic region of the CR-CCSD(T) PES obtained with the cc-pVTZ basis set are shaped in almost exactly the same way as the product valley and the Be + F + H region of the MRCI(Q) PES.…”

“…Clearly, the overall description of the product channel by the CCSD(T) method is not correct. For example, the endothermicity of the Be + HF → BeF + H reaction of −0.009 eV, obtained with the CCSD(T) method, has the wrong sign when compared to the MRCI(Q) endothermicity of 0.140 eV or the MRCI value reported by Aguado et al of 0.26 eV. The CR-CCSD(T) endothermicity value of 0.284 eV, although somewhat above the MRCI(Q) value, retains the correct sign and is in excellent agreement with the MRCI result reported in ref .…”

“…The Be + HF reaction has been the subject of several theoretical studies, including calculations of the ground-state potential energy surfaces (PES) for the BeFH system with diatomics-in-molecules, density functional theory, and the configuration interaction (CI) approach, − as well as fitting the PES to different functional forms − and calculations of reaction dynamics. , Being the lightest member of an important family of exchange reactions involving alkaline earth metals and halides, − the Be + HF reaction serves as an excellent test case for new theoretical methods, particularly new electronic structure methods that are aimed at describing PESs involving breaking and making of chemical bonds.…”

“…First, the use of a small basis set may cause the small errors in the results of the CR-CCSD(T) calculations relative to full CI to be artificially low because of the unsatisfactory description of the relevant dynamic correlation effects by a small number of virtual orbitals in the basis set. Second, it is important to know if the small errors observed in the CR-CCSD(T) calculations for the collinear Be + HF → BeF + H reaction, reported in ref , remain equally small if we examine other reaction channels, including the insertion of Be into the H−F bond (see, for example, refs , , and 7), the case of the Be atom approaching HF from the hydrogen side, and other angles of approach of the HF molecule by Be.…”

Renormalized coupled-cluster calculations of reactive potential energy surfaces: A comparison of the CCSD(T), renormalized CCSD(T), and full configuration interaction results for the collinear BeFH system

“…In order to test the R-CCSD͑T͒ and CR-CCSD͑T͒ methods, we chose a prototypical case of the BeϩHF→BeFϩH reaction, which has been studied by many authors using a variety of theoretical techniques, ranging from the ab initio, [21][22][23] density-functional, 23 and diatomics-inmolecules 24 calculations of the ground-state PES to fitting the PES to a variety of functional forms 21,22,[25][26][27] and dynamical calculations. 21,26 In the absence of the independent experimental results for the BeϩHF→BeFϩH reaction that would allow us to make a critical evaluation of the existing theoretical PESs, in this pilot study, we focus on assessing the performance of the R-CCSD͑T͒ and CR-CCSD͑T͒ methods by comparing the R-CCSD͑T͒ and CR-CCSD͑T͒ results with the exact PES for the BeFH system, obtained with the full configuration interaction ͑CI͒ approach.…”