Optimization of Capping Potentials for Spectroscopic Parameters in Hybrid Quantum Mechanical/Mechanical Modeling Calculations

Komin, Sittipong; Sebastiani, Daniel

doi:10.1021/ct800525u

Cited by 19 publications

(24 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In ONIOM, covalent bonding between two regions leads to open valences in the model system that are saturated with link atoms ͑typically hydrogens͒. Other strategies can be followed to cap open valences in hybrid methods, for example, using effective potentials, 25,26 but the link atom is the simplest and most general, thus in line with the ONIOM approach. For ground state calculations, there are simple guidelines that should be followed in the definition of the model system, 13 for example, the cuts should be made on single, nonpolar, and nonconstrained bonds; an atom in the low level layer should not be bonded to two or more atoms in the high level layer.…”

Section: Introductionmentioning

confidence: 99%

Link atom bond length effect in ONIOM excited state calculations

Caricato¹,

Vreven²,

Trucks³

et al. 2010

The Journal of Chemical Physics

View full text Add to dashboard Cite

Articles you may be interested inAccurate rotational constant and bond lengths of hexafluorobenzene by femtosecond rotational Raman coherence spectroscopy and ab initio calculations J. Chem. Phys. 141, 194303 (2014) We investigate how the choice of the link atom bond length affects an electronic transition energy calculation with the so-called our own N-layer integrated molecular orbital molecular mechanics ͑ONIOM͒ hybrid method. This follows our previous paper ͓M. Caricato et al., J. Chem. Phys. 131, 134105 ͑2009͔͒, where we showed that ONIOM is able to accurately approximate electronic transition energies computed at a high level of theory such as the equation of motion coupled cluster singles and doubles ͑EOM-CCSD͒ method. In this study we show that the same guidelines used in ONIOM ground state calculations can also be followed in excited state calculations, and that the link atom bond length has little effect on the ONIOM energy when a sensible model system is chosen. We also suggest further guidelines for excited state calculations which can help in checking the effectiveness of the definition of the model system and controlling the noise in the calculation.

show abstract

Section: Introductionmentioning

confidence: 99%

Link atom bond length effect in ONIOM excited state calculations

Caricato¹,

Vreven²,

Trucks³

et al. 2010

The Journal of Chemical Physics

View full text Add to dashboard Cite

show abstract

“…There exist a variety of approaches to describe that frontier bond, for example capping potentials [22,26], generalized hybrid orbitals [16] or localized self-consisted field [3,33,40], but the simplest, yet very accurate [2] and therefore also highly popular [27,36] method is hydrogen capping [39].…”

mentioning

confidence: 99%

Optimizing link atom parameters for DNA QM/MM simulations

et al. 2016

View full text Add to dashboard Cite

utilize popular force fields like AMBER [9], CHARMM [8], OPLS [23] or GROMOS [35] is the ability to describe effects stemming from the intra-or intermolecular shifts of electron density for any desired chemical system. Due to this, phenomena such as charge transfers, the cleavage or formation of covalent bonds and polarization effects can be studied in-depth. Additionally, the influence of different spin states, Jahn Teller distortion and even electronic excitations can be investigated without having to resort to specifically tailored force fields.One of the major challenges for that kind of investigation is a proper description of the interface between the QM and the MM region, especially if covalent bonds are reaching through the boundary between the two regions [27]. There exist a variety of approaches to describe that frontier bond, for example capping potentials [22,26], generalized hybrid orbitals [16] or localized self-consisted field [3,33,40], but the simplest, yet very accurate [2] and therefore also highly popular [27,36] method is hydrogen capping [39].A big question regarding hydrogen-capping approaches is the proper placement of the inserted atom, so that its influence on the behavior of the surrounding atoms is minimized as much as possible. In a previous work [18], this was discussed in detail for amino acids, with the recommendation to use separate parameters for each type of sidechain to considerably improve the description of the link bond. In this work, the focus lies on the QM/MM separation of DNA nucleosides, with the C-N bond between the deoxyribose and the nucleobase acting as the link bond. Abstract This work optimizes the link bond description of the quantum mechanical/molecular mechanical separation of deoxynucleosides. The nucleosides are separated at the C-N bond between the nucleobase and the deoxyribose, with the former acting as the quantum mechanically described species. By using a flexible link atom-ansatz plus a harmonic potential to correct the energy deviation from a full quantum mechanical description, the potential energy well of the bond's stretching motion is mimicked with very high accuracy.

show abstract

“…These are generated so that they approximate the orbital energies and the electron densities as close as possible. For pseudobond ECPs this can be achieved by optimizing the differences in the eigenvalues of the modified Fock operator and the original Fock operator or the difference of the electron density in the QM part as done in recent publications 29, 30, 41. The author also included the differences in density matrix elements and the ADMA energy in the objective function.…”

Section: Discussionmentioning

confidence: 99%

Critical investigation on the pseudobond approach for QM/MM and fragment‐based QM methods

Exner

2011

Int J of Quantum Chemistry

View full text Add to dashboard Cite

In mixed quantum mechanical/molecular mechanical and fragmentbased quantum chemical calculations an accurate treatment of bonds, which are cut to build the fragments, is greatly desired so that the structural and electronic properties of the fragment resemble the properties of the original molecule as accurately as possible. Many different methods have been proposed in the literature to deal with bond fragmentation. One of these methods, the pseudobond approach of Zhang et al. (J Chem Phys, 1999, 110, 46), is analyzed in this work. Reparameterizations were conducted for different training sets and diviations in various molecular properties were used as the objective function of the optimazation process. Most of these parameterizations, including the original one, show good performance in reproducing the structures of five substituted ethane molecules and 17 amino acids, as well as in the energy calculation of an oligopeptide based on our fragment-based quantum chemical approach: the field-adapted adjustable density matrix assembler. Nevertheless, none of the parameterizations shows significant and consistent improvement over simple capping hydrogen atoms, despite the simplicity of that approach.

show abstract

Optimization of Capping Potentials for Spectroscopic Parameters in Hybrid Quantum Mechanical/Mechanical Modeling Calculations

Cited by 19 publications

References 67 publications

Link atom bond length effect in ONIOM excited state calculations

Link atom bond length effect in ONIOM excited state calculations

Optimizing link atom parameters for DNA QM/MM simulations

Critical investigation on the pseudobond approach for QM/MM and fragment‐based QM methods

Contact Info

Product

Resources

About