A set of molecular models for carbon monoxide and halogenated hydrocarbons

Stoll, Jürgen; Vrabec, Jadran; Hasse, Hans

doi:10.1063/1.1623475

Cited by 94 publications

(107 citation statements)

References 57 publications

Supporting

Mentioning

103

Contrasting

Unclassified

Order By: Relevance

“…Typical statistical uncertainties for ∆ v H and ρ l are 1% and 0.5%, respectively, cf. e.g., [19]. For GROW, this was the case after seven iterations.…”

Section: Application Of Spagrow To Molecular Simulationsmentioning

confidence: 99%

“…A drawback of RBFs is the high computational effort required for the collocation of the matrix, H, in Equation (19). For N = 5, more than one thousand and for N = 6, over 72 million RBF evaluations are necessary.…”

Section: Smoothing Functionsmentioning

confidence: 99%

“…In the last decade, a few approaches were published realizing an automated force field parameterization procedure [16][17][18][19][20]. Thereby, physical target properties, like density, enthalpy of vaporization and vapor pressure, were fitted to their respective experimental reference data at different temperatures and pressures simultaneously.…”

Section: Force Field Parameterizationmentioning

confidence: 99%

See 2 more Smart Citations

SpaGrOW—A Derivative-Free Optimization Scheme for Intermolecular Force Field Parameters Based on Sparse Grid Methods

Hülsmann

Reith

2013

Entropy

View full text Add to dashboard Cite

Molecular modeling is an important subdomain in the field of computational modeling, regarding both scientific and industrial applications. This is because computer simulations on a molecular level are a virtuous instrument to study the impact of microscopic on macroscopic phenomena. Accurate molecular models are indispensable for such simulations in order to predict physical target observables, like density, pressure, diffusion coefficients or energetic properties, quantitatively over a wide range of temperatures. Thereby, molecular interactions are described mathematically by force fields. The mathematical description includes parameters for both intramolecular and intermolecular interactions.While intramolecular force field parameters can be determined by quantum mechanics, the parameterization of the intermolecular part is often tedious. Recently, an empirical procedure, based on the minimization of a loss function between simulated and experimental physical properties, was published by the authors. Thereby, efficient gradient-based numerical optimization algorithms were used. However, empirical force field optimization is inhibited by the two following central issues appearing in molecular simulations: firstly, they are extremely time-consuming, even on modern and high-performance computer clusters, and secondly, simulation data is affected by statistical noise. The latter provokes the fact that an accurate computation of gradients or Hessians is nearly impossible close to a local or global minimum, mainly because the loss function is flat. Therefore, the question arises of whether to apply a derivative-free method approximating Entropy 2013, 15 3641 the loss function by an appropriate model function. In this paper, a new Sparse Grid-based Optimization Workflow (SpaGrOW) is presented, which accomplishes this task robustly and, at the same time, keeps the number of time-consuming simulations relatively small. This is achieved by an efficient sampling procedure for the approximation based on sparse grids, which is described in full detail: in order to counteract the fact that sparse grids are fully occupied on their boundaries, a mathematical transformation is applied to generate homogeneous Dirichlet boundary conditions. As the main drawback of sparse grids methods is the assumption that the function to be modeled exhibits certain smoothness properties, it has to be approximated by smooth functions first. Radial basis functions turned out to be very suitable to solve this task. The smoothing procedure and the subsequent interpolation on sparse grids are performed within sufficiently large compact trust regions of the parameter space. It is shown and explained how the combination of the three ingredients leads to a new efficient derivative-free algorithm, which has the additional advantage that it is capable of reducing the overall number of simulations by a factor of about two in comparison to gradient-based optimization methods. At the same time, the robustness with respect to statistical noise is...

show abstract

“…Typical statistical uncertainties for ∆ v H and ρ l are 1% and 0.5%, respectively, cf. e.g., [19]. For GROW, this was the case after seven iterations.…”

Section: Application Of Spagrow To Molecular Simulationsmentioning

confidence: 99%

Section: Smoothing Functionsmentioning

confidence: 99%

Section: Force Field Parameterizationmentioning

confidence: 99%

See 1 more Smart Citation

SpaGrOW—A Derivative-Free Optimization Scheme for Intermolecular Force Field Parameters Based on Sparse Grid Methods

Hülsmann

Reith

2013

Entropy

View full text Add to dashboard Cite

show abstract

“…The typical accuracy of reproduction is ±5% for vapor pressure and ±1% for saturated liquid density. [9][10][11][12][13][14][15][16] However, it is usually not known how molecular interaction models perform in homogeneous regions. We performed simulations for some ten molecular models [9][10][11][12][13][14][15][16] representing fluids of very different nature at various state points in the homogeneous regions, and compared A r mn results with the best available FEOS correlations from the literature.…”

Section: Molecular Simulation and Feos Correlationmentioning

confidence: 99%

Communication: Fundamental equation of state correlation with hybrid data sets

Rutkai

Thol

Lustig

et al. 2013

The Journal of Chemical Physics

View full text Add to dashboard Cite

show abstract

“…Extensions of this model consist of 2CLJ models with symmetrical centers and a dipole or a quadrupole located at the center of the molecule [18], and there are even 2CLJ models with asymmetric centers containing both a dipole and a quadruple that had been used to study the composition of planetary atmospheres [19].…”

Section: Introductionmentioning

confidence: 99%

Comparison of molecular models of carbon monoxide for calculation of vapor-liquid equilibrium

Hoyos

Chejne-Janna

2015

Rev. Fac. Ing. Antioquia

View full text Add to dashboard Cite

There are a number of molecular models for carbon monoxide developed from different experimental measurements. This paper aims to compare the results that several of these models produced in the calculation of vapor-liquid equilibrium, in order to recommend which model should be used according to the property and phase to be calculated. The selected models included four non-polar models, with one or two Lennard-Jones sites, and four polar models with dipoles or partial charges to represent the polarity of carbon monoxide. Gibbs-ensemble Monte Carlo simulations in the canonical version (NVT-GEMC) were used to determine the densities of the phases in equilibrium, the vapor pressure and vaporization enthalpy between 80 and 130 K with each of the selected models. It was found that the more complex molecular models, SVH, ANC and PGB, better described the density of the saturated liquid (about 7% average deviation), but these models generated deviations higher than 40% for vapor properties and 20% for vaporization enthalpy. On the other hand, the non-polar BLF model generated the lowest deviations for saturation pressure and vapor density (6.8 and 21.5%, respectively). This model, as the model HCB, produces acceptable deviations for liquid density and vaporization enthalpy (between 10 and 12%). The BLF and HCB models, being non-polar and not requiring the calculation of long-range interactions, can be considered as the molecular models presenting the most satisfactory balance between deviations of the results and calculation complexity.----------Keywords: molecular models, thermodynamic properties, vapor-liquid equilibrium, carbon monoxide ResumenExisten varios modelos moleculares para el monóxido de carbono desarrollados a partir de diferentes mediciones experimentales. El objetivo de este trabajo es comparar los resultados que varios de estos modelos producen en el cálculo del equilibrio líquido-vapor en busca de recomendar qué modelo debe ser usado de acuerdo la propiedad y la fase que se desea calcular. Los modelos seleccionados corresponden a cuatro modelos no polares, con uno o dos sitios Lennard-Jones, y cuatro modelos polares, con dipolos o cargas parciales para representar la polaridad del monóxido de carbono. Simulaciones Monte Carlo en la versión Gibbs canónica (NVT-GEMC) se emplearon para determinar las densidades de las fases en equilibrio, la presión de vapor y la entalpia de vaporización entre 80 y 130 K con cada uno de los modelos seleccionados. Se encontró que los modelos más complejos SVH, ANC y PGB, son los que mejor describen la densidad del líquido saturado (alrededor de 7% de desviación promedio), pero estos modelos generan desviaciones mayores al 40% para las propiedades del vapor y al 20% para la entalpia de vaporización. Por otro lado, el modelo nopolar BLF generó las menores desviaciones para la presión de saturación y la densidad del vapor (6.8 y 21.5%, respectivamente). Este modelo, al igual que el modelo HCB, produce desviaciones aceptables para la densidad del líquido y la entalp...

show abstract

A set of molecular models for carbon monoxide and halogenated hydrocarbons

Cited by 94 publications

References 57 publications

SpaGrOW—A Derivative-Free Optimization Scheme for Intermolecular Force Field Parameters Based on Sparse Grid Methods

SpaGrOW—A Derivative-Free Optimization Scheme for Intermolecular Force Field Parameters Based on Sparse Grid Methods

Communication: Fundamental equation of state correlation with hybrid data sets

Comparison of molecular models of carbon monoxide for calculation of vapor-liquid equilibrium

Contact Info

Product

Resources

About