An official website of the United States government
Unusually large energy errors of semi-local density functional approximations (DFAs) for molecules are often strongly reduced by using the Hartree–Fock (HF) electron density instead of the self-consistent DFA density. For reaction barriers and water clusters, some of us earlier found that HF-density functional theory (DFT) succeeds not because the HF density is accurate but due to the cancellation of negative functional-driven error (FE) by positive density-driven error (DE). Since DE, as defined here, is biased toward the self-consistent DFA density and against the HF density, the DE of the HF density is referred to as non-variational density over-localization (NVDO). In this work, we show that interaction energy errors in halogen- and chalcogen-bonded complexes in the B30 dataset are not dominated by density-driven error. Instead, HF-DFT again succeeds through FE–NVDO cancellation. Benchmark Kohn–Sham inversions of coupled-cluster densities for NH3⋯ClF, Cl−⋯ClF, Cl−⋯SF2, Cl−⋯SCF2, and Cl−⋯PF3 provide strong evidence for this cancellation. For additional complexes, we employ the long-range-corrected hybrid LCωPBE as a proxy for electron-transfer errors in the exact density. We also examine several self-interaction correction (SIC) methods and find significant improvement from FLOSIC. We point out common features of the density errors in the NH3⋯ClF and Cl−⋯ClF complexes and three transition states, arguing that significant density-driven errors of energy arise only from electron-transfer errors. We also highlight a common feature in our present and previous work: long bonds can lead to non-negligible functional-driven self-interaction error of the energy from otherwise accurate semi-local functionals in transition states, water clusters, and halogen or chalcogen bonds.
more »
« less
Density Sensitivity of Empirical Functionals
Empirical fitting of parameters in approximate density functionals is common. Such fits conflate errors in the self-consistent density with errors in the energy functional, but density-corrected DFT (DC-DFT) separates these two. We illustrate with catastrophic failures of a toy functional applied to H2+ at varying bond lengths, where the standard fitting procedure misses the exact functional; Grimme’s D3 fit to noncovalent interactions, which can be contaminated by large density errors such as in the WATER27 and B30 data sets; and double-hybrids trained on self-consistent densities, which can perform poorly on systems with density-driven errors. In these cases, more accurate results are found at no additional cost by using Hartree–Fock (HF) densities instead of self-consistent densities. For binding energies of small water clusters, errors are greatly reduced. Range-separated hybrids with 100% HF at large distances suffer much less from this effect.
more »
« less
- Award ID(s):
- 1856165
- PAR ID:
- 10220445
- Date Published:
- Journal Name:
- The journal of physical chemistry letters
- Issue:
- 0
- ISSN:
- 1948-7185
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Kohn–Sham (KS) inversion, that is, the finding of the exact KS potential for a given density, is difficult in localized basis sets. We study the precision and reliability of several inversion schemes, finding estimates of density-driven errors at a useful level of accuracy. In typical cases of substantial density-driven errors, Hartree–Fock density functional theory (HF-DFT) is almost as accurate as DFT evaluated on CCSD(T) densities. A simple approximation in practical HF-DFT also makes errors much smaller than the density-driven errors being calculated. Two paradigm examples, stretched NaCl and the HO·Cl– radical, illustrate just how accurate HF-DFT is.more » « less
-
Most torsional barriers are predicted with high accuracies (about 1 kJ/mol) by standard semilocal functionals, but a small subset was found to have much larger errors. We created a database of almost 300 carbon–carbon torsional barriers, including 12 poorly behaved barriers, that stem from the Y═C—X group, where Y is O or S and X is a halide. Functionals with enhanced exchange mixing (about 50%) worked well for all barriers. We found that poor actors have delocalization errors caused by hyperconjugation. These problematic calculations are density-sensitive (i.e., DFT predictions change noticeably with the density), and using HF densities (HF-DFT) fixes these issues. For example, conventional B3LYP performs as accurately as exchange-enhanced functionals if the HF density is used. For long-chain conjugated molecules, HF-DFT can be much better than exchange-enhanced functionals. We suggest that HF-PBE0 has the best overall performance.more » « less
-
Standard approximations for the exchange–correlation functional in Kohn–Sham density functional theory (KS-DFT) typically lead to unacceptably large errors when applied to strongly correlated electronic systems. Partition-DFT (PDFT) is a formally exact reformulation of KS-DFT in which the ground-state density and energy of a system are obtained through self-consistent calculations on isolated fragments, with a partition energy representing inter-fragment interactions. Here, we show how typical errors of the local density approximation (LDA) in KS-DFT can be largely suppressed through a simple approximation, the multi-fragment overlap approximation (MFOA), for the partition energy in PDFT. Our method is illustrated on simple models of one-dimensional strongly correlated linear hydrogen chains. The MFOA, when used in combination with the LDA for the fragments, improves LDA dissociation curves of hydrogen chains and produces results that are comparable to those of spin-unrestricted LDA, but without breaking the spin symmetry. MFOA also induces a correction to the LDA electron density that partially captures the correct density dimerization in strongly correlated hydrogen chains. Moreover, with an additional correction to the partition energy that is specific to the one-dimensional LDA, the approximation is shown to produce dissociation energies in quantitative agreement with calculations based on the density matrix renormalization group method.more » « less
-
For cyclic conjugated structures, erratic computational results have been obtained with Hartree–Fock (HF) molecular orbital (MO) methods as well as density functional theory (DFT) with large HF-exchange contributions. In this work, the reasons for this unreliability are explored. Extensive computations on [18]annulene and related compounds highlight the pitfalls to be avoided and the due diligence required for such computational investigations. In particular, a careful examination of the MO singlet-stability eigenvalues is recommended. The appearance of negative eigenvalues is not (necessarily) problematic, but near-zero (positive or negative) eigenvalues can lead to dramatic errors in vibrational frequencies and related properties. DFT approaches with a lower HF admixture generally appear more robust in this regard for the description of benzenoid structures, although they may exaggerate the tendency toward planarity and C–C bond-equalization. For the iconic [18]annulene, the results support a nonplanar equilibrium structure. The density-fitted frozen natural orbital coupled-cluster singles and doubles with perturbative triples [DF-FNO CCSD(T)] method of electron correlation with an aug-pVQZ/aug-pVTZ basis set places the C2 global minimum 1.1 kcal mol–1 below the D6h stationary point.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Journal Article:
- https://doi.org/doi.10.1021/acs.jpclett.0c03545
