Molecular-dynamics simulation of homogeneous nucleation in the vapor phase

The classical nucleation theory for homogeneous nucleation is formulated as a theory for a density fluctuation in a supersaturated gas at a given temperature. But Molecular Dynamics simulations reveal that it is small cold clusters which initiates the nucleation. The temperature in the nucleating clusters fluctuate, but the mean temperature remains below the temperature in the supersaturated gas until they reach the critical nucleation size. The critical nuclei have, however, a temperature equal to the supersaturated gas. The kinetics of homogeneous nucleation is not only caused by a grow or shrink by accretion or evaporation of monomers only, but by an exponentially declining change in cluster size per time step equal to the cluster distribution in the supersaturated gas.

“…[10,12]. The size of the critical nucleus depends, however, on the choice of the coordination number.…”

Section: B Determination Of Cluster Propertiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Dynamics of homogeneous nucleation

Toxværd

2015

“…However, the smallest stable, or critical, clusters are nano-sized at typical nucleation conditions, and their properties differ significantly from the CNT assumptions. This results in massive discrepancies between the nucleation rates predicted by CNT and those measured in laboratory experiments [6][7][8][9][10][11][12][13][14][15] and molecular dynamics 13,[16][17][18][19][20][21][22][23][24][25][26] or Monte Carlo simulations. [27][28][29][30][31][32][33][34] More recent nucleation theories have made significant improvements since the introduction of the CNT.…”

Section: Introductionmentioning

confidence: 99%

Large scale molecular dynamics simulations of homogeneous nucleation

Diemand

Angélil

Tanaka

et al. 2013

118

163

We present results from large-scale molecular dynamics (MD) simulations of homogeneous vapor-to-liquid nucleation. The simulations contain between 1 × 109 and 8 × 109 Lennard-Jones (LJ) atoms, covering up to 1.2 s (56 × 106 time-steps). They cover a wide range of supersaturation ratios, S = 1.55-104, and temperatures from kT = 0.3 to 1.0 (where is the depth of the LJ potential, and k is the Boltzmann constant). We have resolved nucleation rates as low as 1017 cm-3 s-1 (in the argon system), and critical cluster sizes as large as 100 atoms. Recent argon nucleation experiments probe nucleation rates in an overlapping range, making the first direct comparison between laboratory experiments and molecular dynamics simulations possible: We find very good agreement within the uncertainties, which are mainly due to the extrapolations of argon and LJ saturation curves to very low temperatures. The self-consistent, modified classical nucleation model of Girshick and Chiu [J. Chem. Phys. 93, 1273 (1990)] underestimates the nucleation rates by up to 9 orders of magnitudes at low temperatures, and at kT = 1.0 it overestimates them by up to 105. The predictions from a semi-phenomenological model by Laaksonen et al. [Phys. Rev. E 49, 5517 (1994)] are much closer to our MD results, but still differ by factors of up to 104 in some cases. At low temperatures, the classical theory predicts critical clusters sizes, which match the simulation results (using the first nucleation theorem) quite well, while the semi-phenomenological model slightly underestimates them. At kT = 1.0 , the critical sizes from both models are clearly too small. In our simulations the growth rates per encounter, which are often taken to be unity in nucleation models, lie in a range from 0.05 to 0.24. We devise a new, empirical nucleation model based on free energy functions derived from subcritical cluster abundances, and find that it performs well in estimating nucleation rates. We present results from large-scale molecular dynamics (MD) simulations of homogeneous vapor-toliquid nucleation. The simulations contain between 1 × 10 9 and 8 × 10 9 Lennard-Jones (LJ) atoms, covering up to 1.2 μs (56 × 10 6 time-steps). They cover a wide range of supersaturation ratios, S 1.55-10 4 , and temperatures from kT = 0.3 to 1.0 (where is the depth of the LJ potential, and k is the Boltzmann constant). We have resolved nucleation rates as low as 10 17 cm −3 s −1 (in the argon system), and critical cluster sizes as large as 100 atoms. Recent argon nucleation experiments probe nucleation rates in an overlapping range, making the first direct comparison between laboratory experiments and molecular dynamics simulations possible: We find very good agreement within the uncertainties, which are mainly due to the extrapolations of argon and LJ saturation curves to very low temperatures. The self-consistent, modified classical nucleation model of Girshick and Chiu [J. Chem. Phys. 93, 1273 (1990)] underestimates the nucleation rates by up to 9 orders of magnitudes at l...

“…As a consequence, the past decade has witnessed a surge of contributions devoted to improving the understanding of the process time scales, with many studies focusing on the large scale calculation of nucleation rates by means of molecular dynamics ͑MD͒, [1][2][3][4] on the benchmark of currently available theories ͑e.g., classical nucleation theory, extended liquid drop modeldynamical nucleation theory, and other semiphenomenological models͒, 1,2,5 and on the calculation of the work necessary to build a cluster from the supersaturated vapor. 6,7 Despite these efforts, the routine application of computational approaches to the calculation of nucleation rates for experimentally relevant systems still appears as a distant goal, the exception being perhaps the prediction of the critical cluster size that can be done by means of thermodynamic integration.…”

Section: Introductionmentioning

confidence: 99%

On possible simplifications in the theoretical description of gas phase atomic cluster dissociation

Mella¹

2009

In this work, we investigate the possibility of describing gas phase atomic cluster dissociation by means of variational transition state theory ͑vTST͒ in the microcanonical ensemble. A particular emphasis is placed on benchmarking the accuracy of vTST in predicting the dissociation rate and kinetic energy release of a fragmentation event as a function of the cluster size and internal energy. The results for three Lennard-Jones clusters ͑LJ n , n =8,14,19͒ indicate that variational transition state theory is capable of providing results of accuracy comparable to molecular dynamics simulations at a reduced computational cost. Possible simplifications of the master equation formalism used to model a dissociation cascade are also suggested starting from molecular dynamics results. In particular, it is found that the dissociation rate is only weakly dependent on the cluster total angular momentum J for the three cluster sizes considered. This would allow one to partially neglect the J-dependency of the kinetic coefficients, leading to a substantial decrease in the computational effort needed for the complete description of the cascade process. The impact of this investigation on the modeling of the nucleation process is discussed.