Direct Computation of Dynamical Response by Molecular Dynamics: The Mobility of a Charged Lennard-Jones Particle

Ciccotti, Giovanni; Jacucci, Gianni

doi:10.1103/physrevlett.35.789

Cited by 125 publications

(89 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They used the D-NEMD method [30] and computed the Soret coefficient when the system had reached a stationary state. The present paper is similar in the sense that the initial phase of the system's response to a thermal perturbation is studied, but we focus on the transient phase and show that it is possible to use such data to determine the Soret coefficient, thus avoiding the need for stationary state.…”

Section: Discussionmentioning

confidence: 99%

Non-equilibrium molecular dynamics simulations of the transient Ludwig-Soret effect in a binary Lennard-Jones/spline mixture

Hafskjold

2017

Eur. Phys. J. E

View full text Add to dashboard Cite

SummaryA binary isotope mixture of Lennard-Jones/spline particles at equilibrium was perturbed by a sudden change in the system's boundary temperatures. The system's response was determined by nonequilibrium molecular dynamics (NEMD). Three transient processes were studied: (1) The propagation of a pressure (shock) wave, (2) heat diffusivity and conduction, and (3) thermal diffusion (the LudwigSoret effect). These three processes occur at different time scales, which makes it possible to separate them in one single NEMD run. The system was studied in liquid, supercritical, and dense gas states with various forms and strengths of the thermal perturbation. The results show that heat was initially transported by two separate mechanisms; (1) heat diffusion as described by the transient heat equation and (2) as a consequence of a pressure wave. The pressure wave travelled faster than the speed of sound, generating a shock wave in the system. Local equilibrium was found in the transient phase, even with very strong perturbations and in the shock front. Although the mass separation due to the Ludwig-Soret effect developed much slower than the pressure and temperature fields in the system at large, it was found that the Soret coefficient could be accurately determined from the initial phase of the transient and close to the heat source. This opens the possibility of a new way to analyse results from transient experiments and thereby minimize effects of gravity and convection due to buoyancy.

show abstract

Section: Discussionmentioning

confidence: 99%

Non-equilibrium molecular dynamics simulations of the transient Ludwig-Soret effect in a binary Lennard-Jones/spline mixture

Hafskjold

2017

Eur. Phys. J. E

View full text Add to dashboard Cite

show abstract

“…To this aim we can exploit the "Onsager-Kubo" relation [11,12,31]. Calling S(t) the time evolution operator of the perturbed dynamics, the following relation holds for the non equilibrium average of the generic microscopic fluxĴ…”

Section: Molecular Dynamicsmentioning

confidence: 99%

“…The fluctuations of the microscopic variables are quite large and dominate the response in the limit of vanishing perturbations. In simple systems, where the response time is comparable to the Lyapunov time of the exponential divergence of nearby starting trajectories, the "subtraction technique" can be used to extract the signal out of the statistical noise [11,12]. If S 0 (t) is the evolution operator representing the unperturbed dynamics with < S 0 (t)Ĵ > 0 = 0, the average values < S(t)Ĵ − S 0 (t)Ĵ > 0 and < S(t)Ĵ > 0 are equal.…”

Section: Molecular Dynamicsmentioning

confidence: 99%

“…We want to emphasize that the theoretical foundation of this class of algorithm is the Linear Response Theory and their use beyond the linear regime is somewhat arbitrary. In this context the "subtraction technique" [11,12] is a very useful tool (at least for simple systems in which the response time is not longer than the typical Lyapunov time) to perform the vanishing perturbation limit.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bulk viscosity of the Lennard-Jones system at the triple point by dynamical nonequilibrium molecular dynamics

2008

View full text Add to dashboard Cite

Non-equilibrium Molecular Dynamics (NEMD) calculations of the bulk viscosity of the triple point Lennard-Jones fluid are performed with the aim of investigating the origin of the observed disagreement between Green-Kubo estimates and previous NEMD data. We show that a careful application of the Doll's perturbation field, the dynamical NEMD method, the instantaneous form of the perturbation and the "subtraction technique" provides a NEMD estimate of the bulk viscosity at zero field in full agreement with the value obtained by the Green-Kubo formula. As previously reported for the shear viscosity, we find that the bulk viscosity exhibits a large linear regime with the field intensity which confirms the Lennard-Jones fluid as a genuine Newtonian fluid even at triple point.

show abstract

“…5,6 In order to effectively down-select from a large population of potential electrolytes, the parameters of a full-scale model of battery performance must come from more fundamental calculations. On a molecular level, the use of molecular dynamics (MD) to model diffusion, 7-15 ionic mobility, [16][17][18][19][20] and estimate the associated transport coefficients is wide-spread and has been shown to be predictive given representative interatomic potentials. In particular, Refs.…”

mentioning

confidence: 99%

Assessing Electrolyte Transport Properties with Molecular Dynamics

Jones

Ward

Gittleson

et al. 2017

J. Electrochem. Soc.

View full text Add to dashboard Cite

In this work we use estimates of ionic transport properties obtained from molecular dynamics to rank lithium electrolytes of different compositions. We develop linear response methods to obtain the Onsager diffusivity matrix for all chemical species, its Fickian counterpart, and the mobilities of the ionic species. We apply these methods to the well-studied propylene carbonate/ethylene carbonate solvent with dissolved LiBF 4 and O 2 . The results show that, over a range of lithium concentrations and carbonate mixtures, trends in the transport coefficients can be identified and optimal electrolytes can be selected for experimental focus; however, refinement of these estimation techniques is necessary for a reliable ranking of a large set of electrolytes. There is increasing interest in designing and manufacturing highperformance batteries for a wide variety of applications; 1 however, finding the best electrochemical system and configuration is a challenging task given all the possible combinations of electrodes and electrolyte components. In particular, choosing an electrolyte optimal for a given battery configuration and chemistry is daunting. Herein, we propose a computational approach to electrolyte screening and selection. Through a combination of techniques, e.g. ab initio calculations to provide fundamental parameters to molecular dynamics which, in turn, provides transport coefficients to a reaction-diffusion model of the full-scale battery, 2 we can predict battery performance. These predictions can be compared with limited experiments for validation before selecting the best candidates for intensive development. The concept of computational screening is not new, e.g. the work the Ceder group 3 and the battery-focused "Genome" efforts such as the Electrolyte Genome, 4 but here we focus on the selection of ideal electrolytes for Li batteries based on the transport properties which have been identified as critical to performance. 5,6 In order to effectively down-select from a large population of potential electrolytes, the parameters of a full-scale model of battery performance must come from more fundamental calculations. On a molecular level, the use of molecular dynamics (MD) to model diffusion, 7-15 ionic mobility, [16][17][18][19][20] and estimate the associated transport coefficients is wide-spread and has been shown to be predictive given representative interatomic potentials. In particular, Refs. 11,12,21 apply the techniques to measure the diffusivity of nonaqueous Li ion systems and Refs. 17,18 estimate the diffusivity and mobility of Li + in water.To address this big-data computational challenge, in the Theory section we introduce linear response methods to estimate the mutual diffusion and mobility coefficients of the species fluxes, using notation summarized in Table I. Generally speaking, linear response methods have many advantages over alternative methods that use large gradients/unphysical driving forces and large systems to estimate transport coefficients since they are equilibrium method...

show abstract

Direct Computation of Dynamical Response by Molecular Dynamics: The Mobility of a Charged Lennard-Jones Particle

Cited by 125 publications

References 8 publications

Non-equilibrium molecular dynamics simulations of the transient Ludwig-Soret effect in a binary Lennard-Jones/spline mixture

Non-equilibrium molecular dynamics simulations of the transient Ludwig-Soret effect in a binary Lennard-Jones/spline mixture

Bulk viscosity of the Lennard-Jones system at the triple point by dynamical nonequilibrium molecular dynamics

Assessing Electrolyte Transport Properties with Molecular Dynamics

Contact Info

Product

Resources

About