Anisotropic intermolecular interactions in van der Waals and hydrogen-bonded complexes: What can we get from density functional calculations?

Abstract

Articles you may be interested inPerformance of dispersion-corrected double hybrid density functional theory: A computational study of OCShydrocarbon van der Waals complexes Computational and experimental investigation of intermolecular states and forces in the benzene-helium van der Waals complex Comment on "Anisotropic intermolecular interactions in van der Waals and hydrogen-bonded complexes: What can we get from density-functional calculations?" [J.The applicability of various density functional theory ͑DFT͒ methods to describe the anisotropy of the intermolecular potential energy surfaces of hydrogen-bonded ͓OH Ϫ -H 2 O, (H 2 O͒ 2 ͔ and van der Waals ͓CO-H 2 O, He-CO 2 ] complexes has been tested by comparison with supermolecule CCSD͑T͒ ͑coupled-cluster method restricted to single, double, and noniterative triple excitations͒ and perturbational SAPT ͑symmetry-adapted perturbation theory͒ results computed for the same geometries and with the same basis sets. It is shown that for strongly bound ionic hydrogen-bonded complexes, like OH Ϫ -H 2 O, hybrid approaches provide accurate results. For other systems, including the water dimer, the DFT calculations fail to reproduce the correct angular dependence of the potential surfaces. It is also shown that a hybrid functional adjusted to reproduce the CCSD͑T͒ value of the binding energy for the water dimer produces results worse than the standard hybrid functionals for OH Ϫ -H 2 O, and fails to describe the correct anisotropy of the CO-H 2 O interaction.

Interaction energies in hydrogen-bonded systems: A testing ground for subsystem formulation of density-functional theory

“…1 and 2 for GGA and LDA functionals, respectively. Similar comparisons of KSCED ͑GGA and LDA͒ results with ab initio reference data 25 were made for seven other structures of the water dimer. Opposite to the previous set of structures, these geometries do not correspond to stationary points on the potential-energy surface but are the structures generated via rotation ͑ = −52°, −22°, −8°, 38°, 68°, 98°, and 128°͒ of one of the water molecules keeping the inter- The reference data used for two water dimer sets is of a different quality.…”

Influence of tert-amine groups on the solubility of polymers in CO2

“…Moreover, they reported that the CO 2 /pyridine complex was more strongly bound by 5.6 kJ/mol than the CO 2 /triethylamine complex [35]. However, the calculations performed by Meredith et al may be in error because DFT is known not to be accurate for weakly bound systems because the electron correlation treatment in B3LYP has the wrong long-range behavior, and therefore cannot capture the correct van der Waals dispersion interactions [36]. Hence, for supramolecular systems in which the binding is dominated by dispersive interactions, both the geometries and binding energies may be in error.…”

“…They concluded that steric repulsion of the ethyl groups in triethylamine causes the binding energy of the CO 2 -triethylamine complex to be weaker than that of the CO 2 -pyridine complex, whereas the planar structure of pyridine leads to less steric hindrance, allowing the nitrogen atom to be closer to the carbon of CO 2 . The density functional theory formalism used by Meredith et al to compute the CO 2 -amine interactions [35] lacks an accurate description of electron correlation effects that give rise to van der Waals interactions [36], which are expected to dominate the CO 2 -amine interaction; therefore the binding energies and perhaps even the geometries they report may be substantially in error. Furthermore, contrary to the results by Meredith et al, our ab initio calculations, reported in this paper, indicate that interactions between CO 2 and tert-amines are highly favorabledeven stronger than CO 2 -carbonyl interactions.…”

Long-range correlation energies from frequency-dependent weighted exchange-hole dipole polarisabilities

“…The scaling factor s r, 6 for the cutoff radii in the damping function used in conjunction with this functional has been set to a value of s r, 6 = 1.0.…”

“…1) that are based on the local density approximation or the generalised gradient approximation (GGA) are unable to describe the important long-range correlation interactions between molecular systems that are not bound by covalent bonds. [2][3][4][5][6][7][8][9] The reason for this stems from the fact that these functionals contain exchange-correlation energy densities which are only locally dependent on the electron density and its derivatives and therefore lack the capability to describe interactions of molecular regions that are spatially separated. More recent (semi-local) hybrid meta GGA's (Refs.…”