Kinetics of binary nucleation: Multiple pathways and the approach to stationarity

A multigrid method for N-component nucleation

A multigrid algorithm has been developed enabling more efficient solution of the cluster size distribution for N-component nucleation from the Becker-Döring equations. The theoretical derivation is valid for an arbitrary number of condensing components, making the simulation of many-component nucleating systems feasible. A steady state ternary nucleation problem is defined to demonstrate its efficiency. The results are used as a validation for existing nucleation theories. The non-steady state ternary problem provides useful insight into the initial stages of the nucleation process. We observe that for the ideal mixture the main nucleation flux bypasses the saddle point.

Nucleation theory in Langevin's approach and lifetime of a Brownian particle in potential wells

The multivariable theory of nucleation suggested by Alekseechkin [J. Chem. Phys. 124, 124512 (2006)] is further developed in the context of Langevin's approach. The use of this approach essentially enhances the capability of the nucleation theory, because it makes possible to consider the cases of small friction which are not taken into account by the classical Zel'dovich-Frenkel theory and its multivariable extensions. The procedure for the phenomenological determination of the nucleation parameters is described. Using the similarity of the Kramers model with that of nucleation, the lifetime of a Brownian particle in potential wells in various dimensionalities is calculated with the help of the expression for the steady state nucleation rate.

Multivariable kinetic theory of the first order phase transitions

The problem of calculation of the steady state homogeneous nucleation rate in the multidimensional space of the variables describing a nucleus is considered. Within the framework of the theory proposed, expressions for the nucleation rate and the steady state distribution function of nuclei are derived. The expression for the nucleation rate is invariant with respect to the space dimensionality and, in particular, involves the result of the one-dimensional theory. The distribution function is obtained in the initial, physical variables. In connection with the analysis of restrictions on the current direction, the question of symmetry of the matrix of diffusivities is considered; on the basis of the detailed balance principle it is shown that this matrix is symmetric. The question of normalizing the equilibrium distribution functions is investigated and the physical picture of the equilibrium state is described. The procedure of reducing the multidimensional theory to the one-dimensional one is described.

Kinetics of binary nucleation of vapors in size and composition space

We reformulate the kinetic description of binary nucleation in the gas phase using two natural independent variables: the total number of molecules g and the molar composition x of the cluster. The resulting kinetic equation can be viewed as a two-dimensional Fokker-Planck equation describing the simultaneous Brownian motion of the clusters in size and composition space. Explicit expressions for the Brownian diffusion coefficients in cluster size and composition space are obtained. For characterization of binary nucleation in gases three criteria are established. These criteria establish the relative importance of the rate processes in cluster size and composition space for different gas phase conditions and types of liquid mixtures. The equilibrium distribution function of the clusters is determined in terms of the variables g and x. We obtain an approximate analytical solution for the steady-state binary nucleation rate that has the correct limit in the transition to unary nucleation. To further illustrate our description, the nonequilibrium steady-state cluster concentrations are found by numerically solving the reformulated kinetic equation. For the reformulated transient problem, the relaxation or induction time for binary nucleation was calculated using Galerkin's method. This relaxation time is affected by processes in both size and composition space, but the contributions from each process can be separated only approximately.

Genuine saddle point and nucleation potential for binary systems

A generalized nucleation potential is constructed for binary systems. The potential consists of the reversible work of cluster formation plus additional terms arising from various kinetic effects. We show that the major nucleation flux passes through the saddle point (termed the genuine saddle point) of this generalized nucleation potential. The generalized nucleation potential reduces to the kinetic potential of a unary system when one component vanishes. The genuine saddle point concept provides a convenient way to identify systems and conditions for which the ridge crossing phenomenon occurs. Our theory agrees approximately with exact numerical results.

Analytical approach to time lag in binary nucleation

We present an analytical formula for the time required to establish steady state in a nucleating binary system. To test our solution, we evaluate the time lag for a range of activities of both components at the vapor-liquid transition, and show that our result is in much better agreement with a purely numerical simulation than other available analytical formulas, which overestimate the time lag by factors of from 2 to 200.