Fermi resonance in CO2: Exploring the vibrational structure of the carbon dioxide dimer
Olaseni Sode,1 Murat Keçeli,2 and Samuel Maystrovsky1
1. The University of Tampa, Department of Chemistry, Biochemistry and Physics, Tampa, FL, 33606.
2. Argonne National Laboratory, Computational Science Division, Argonne, IL, 60439.
The vibrational structure of the carbon dioxide monomer and dimer are explored using a flexible-monomer two-body potential energy function. This recently developed potential [O. Sode and J. N. Cherry, J. Comput. Chem., (2017)] is fit to the electronic energies at the CCSD(T)-F12b/aug-cc-pVTZ level of theory and integrated into vibrational structure programs, such as MaVi, NITROGEN and SINDO, to determine anharmonic corrections to the harmonic frequencies. Vibrational correlation approaches were employed, and the vibrational configuration interaction (VCI) results agree to within a few wavenumbers of the experimentally determined peaks, especially in the intramolecular region. While the dimer intermolecular vibrations in THz region are difficult to explore experimentally, we identify fundamental, overtone and combination bands with our computational approach. Our results suggest Fermi resonance in this low-frequency region due to mixing between the overtone and fundamental bands..
The mechanism and kinetics of oxidation of cyclopentadienyl radical in combustion flames: A theoretical view
A. R. Ghildina,1 A. D. Oleinikov, G. R. Galimova,1 V. N. Azyazov,1 and A. M. Mebel2
1 - Samara National Research University, Samara 443086, Russian Federation
2 - Florida International University, Miami, Florida 33199, USA
We will report a detailed theoretical investigation of the oxidation mechanism of the cyclopentadienyl radical with the main oxidants present in combustion flames, including O, OH, and O2. Ab initio calculations of potential energy surfaces in conjunction with the RRKM-Master Equation theoretical approach have been employed to evaluate temperature- and pressure-dependent total and product specific rate constants and product branching ratios for these and related reactions. The C5H5 + O reaction is shown to proceed by barrierless oxygen addition to the ring followed by fast H migration, ring opening, and dissociation to C4H5 + CO. The C5H5 + O rate constant is calculated to be close to 1.0x10-10 cm3 molecule-1 s-1 and to be pressure-independent and nearly independent of temperature. The C5H5 + OH reaction is shown to proceed either by well-skipping pathways without stabilization of C5H6O intermediates leading to the bimolecular products ortho-C5H5O + H, C5H4OH (hydroxycyclopentadienyl) + H, and C4H6 (1,3-butadiene) + CO, or via stabilization of the C5H6O intermediates, which then undergo unimolecular thermal decomposition to ortho-C5H5O + H and C4H6 + CO. The well-skipping and stabilization/dissociation pathways compete depending on the reaction conditions; higher pressures favor the stabilization/dissociation and higher temperature favor the well-skipping channels. For the C5H5 + O2 reaction, the results show that at low temperatures from 500 to 800-1250 K (depending on pressure), the reaction predominantly forms a collisionally-stabilized C5H5-OO complex and then, the thermalized complex rapidly decomposes back to the reactants establishing a C5H5 + O2/C5H5-OO equilibrium. At higher temperatures, typically above 1000 K, the mechanism is different and the C5H5 + O2 reaction proceeds to form various bimolecular products. Cyclopentadienone C5H4O + OH are predicted to be the predominant product, whereas relatively minor products include H2CCHCHC(H)O + CO, vinylketene + HCO, and highly endothermic C5H5O + O produced directly by the O-O bond cleavage in the initial complex. Overall, the rate constant of the C5H5 + O2 reaction at combustion-relevant temperatures is predicted to be very slow, 10-16-10-15 cm3 molecule-1 s-1, that is typically ~5 orders of magnitude lower than those for the oxidation reactions of cyclopentadienyl with OH and O(3P).
Development of Complex v2RDM Driven Relativistic CASSCF Methods
Run R. Li and A. Eugene DePrince III
Florida State University
In order to use ab initio methods to correctly predict the properties of chemical systems containing heavy elements, one must account for both correlation and relativistic effects, including spin-orbit coupling. The complete active space self-consistent-field method (CASSCF) provides a reliable description of nondynamical correlation effects, but the steep scaling of configuration interaction (CI)-based CASSCF precludes its application to large systems. The variational two-electron reduced-density-matrix (v2RDM) driven methods provide a computationally efficient alternative to CI-based approaches. Both scalar relativistic and spin-orbit coupling effects can be captured within an exact two component (X2C) extension of v2RDM-driven CASSCF. We have developed a complex generalized implementation of the v2RDM approach for X2C-v2RDM-CASSCF computations. Our preliminary investigations indicate that the complex generalized v2RDM approach has some surprising numerical properties. For example, in atomic systems, constraints on the expectation values of the square of the total orbital angular momentum and its projection on the z-axis yield quite different results for a given L and different ml values. It appears that v2RDM computations corresponding to the maximal orbital angular momentum projection provides the best agreement with CI-based computations.
New Approaches for Overcoming Ensemble Mismatches in QM/MM Free Energy Simulations
Phillip S. Hudson, Fiona L. Kearns, Stefan Boresch, H. Lee Woodcock
University of South Florida
University of Vienna
The hybrid quantum mechanical / molecular mechanical (QM/MM) framework is the current tool of choice when accurate computations of macromolecular systems are essential. However, when carrying out alchemical free energy simulations (FES) with QM/MM, technical and computational challenges necessitate taking an "indirect" approach, i.e., introducing a thermodynamic cycle and using a "low" level of theory (typically MM) to perform the alchemical transformation. This leaves computing the free energy between low and high levels of theory as the main challenge. While this may seem like a straight-forward task, it is fraught with problems. The chief amongst these is the fact that low and high levels of theory often lead to vastly different ensembles; methods to connect these via FES, in an affordable, robust way are essentially non-existent. Herein, we will present new methods that greatly improve both the accuracy and efficiency of computing free energies between MM and QM levels of theory. These include the development, implementation, and validation of both more robust FES techniques as well as new schemes for modifying "low level" potentials to facilitate overcoming disparate ensembles.
Embedded cluster density approximation for exchange-correlation energy
Department of Scientific Computing, Florida State University
Accurate, large-scale electronic structure simulations are essential for understanding the properties of materials and molecules at the nanoscale. Kohn-Sham density functional theory (KS-DFT) is widely used for large-scale material simulations, however, its accuracy is limited by the accuracy of the exchange-correlation (XC) functionals. As we are improving the accuracy of XC functionals, KS-DFT starts to lose its computational efficiency. In this presentation, we discuss our effort on developing the embedded cluster density approximation (ECDA) method for scaling up high-level KS-DFT simulations in large systems. In ECDA, for each atom (called central atom) its nearby atoms are selected as its buffer atoms. The central atom and its buffer atoms form the cluster. The rest atoms define the environment. The system’s electron density is partitioned among the cluster and its environment based on the finite-temperature density functional embedding theory. The obtained cluster is embedded in the system and is a small Kohn-Sham system whose XC energy density is calculated using a high-level XC functional. System’s XC energy is constructed by patching these locally computed, high-level XC energy densities over the entire system in an atom-by-atom manner. A key step is to efficiently compute the system's XC potential for the patched XC energy. We directly take the functional derivative of the patched XC energy with respect to the electron density without invoking the system's unoccupied orbitals, therefore making ECDA computationally efficient for investigating large systems. Since ECDA is a variational method, forces can be efficiently calculated based on the Hellmann-Feynman theorem. The accuracy of ECDA is investigated by patching the random phase approximation correlation energy in one-dimensional hydrogen chains, and by patching the exact exchange energy in molecules.
Triumphs and Tribulations in the Molecular Modeling of Porous Materials
University of South Florida
Highly accurate molecular models for gas sorption in MOF's have been extensively applied to both hydrogen and carbon dioxide – critical comparisons between models are considered. Calculated observables such as isosteric heats, sorption isotherms and compressibilities are objectively compared with experimental measurements and found to be in excellent agreement. A series of MOF structures have been examined from non-polar to polar and open to confined to assess the what topologies and associated potential energy interactions are responsible for increased sorption. Polarization interactions are shown to be essentially many body in nature and non-negligible considering MOF”s that are promising soprtion candiates for multiple applications as the interactions are tunable. Widely used models for other sorbates are found to be unpredictable in fidelity and unreliable, yet continue to see wide acceptance. This has troubling implications for biological simulations as well, where poorly constructed force fields are ramapant.