The kinetic nucleation theory of Ruckenstein and Nowakowski (RNNT) computes homogeneous vapor-to-liquid nucleation rates from molecular interactions alone: the rate at which molecules evaporate from a cluster is obtained from an energy-space diffusion (Fokker–Planck) equation and a first-passage-time analysis in the potential field created by the cluster, avoiding the capillarity approximation of classical nucleation theory. As summarized in the source chapter, however, the theory is restricted to rigid spherical molecules with a purely dispersive attraction whose parameters are fitted to macroscopic surface tension and density, and it rests on approximations — the second-order expansion of the kinetic equation in energy space and a Brownian-motion energy diffusion coefficient — whose accuracy for very small clusters and strong interactions has never been assessed. The open problem is to carry this molecular kinetic program to real fluids: account for molecular shape and orientation-dependent (anisotropic, hydrogen-bonding) interactions, establish the validity of the energy-diffusion approximation for evaporation from small clusters, and eliminate parameter fitting to macroscopic quantities, so that nucleation rates and critical cluster sizes follow from intermolecular interactions alone and can be tested against the now-comprehensive experimental dataset for the homologous 1-alcohol series (ethanol to pentanol, 235–265 K).

Background

Homogeneous vapor-to-liquid nucleation is the spontaneous formation of liquid droplets in a supersaturated vapor — a vapor whose pressure exceeds its saturation pressure at the given temperature — without help from foreign surfaces. Classical nucleation theory (CNT) treats a growing cluster as a macroscopic liquid drop and writes its free energy as a volume term plus a surface term proportional to the planar surface tension of the liquid; this is the capillarity approximation. The predicted nucleation rate JJ (droplets formed per unit volume per unit time) is then exponentially sensitive to the surface tension assigned to clusters of only a few tens of molecules, i.e. to the critical cluster size at which growth becomes more likely than decay, and CNT fails badly for small critical clusters — for example underpredicting measured nucleation rates of 1-alcohol vapors by orders of magnitude, with the wrong temperature dependence.

The Ruckenstein–Nowakowski kinetic nucleation theory (RNNT) was built as a molecular alternative. Instead of macroscopic cluster thermodynamics, the rate at which a molecule evaporates from a cluster is obtained kinetically: the probability distribution of a surface molecule moving in the potential field created by the rest of the cluster obeys a kinetic equation, and the mean evaporation time is computed as a first-passage time — the average time to escape the potential well — using a diffusion (Fokker–Planck) equation for the molecule's energy. Combined with the gas-kinetic condensation rate and detailed balance (the equilibrium requirement that condensation and evaporation rates compensate for each cluster size), this yields the nucleation rate without invoking capillarity. Two approximations enter: the kinetic equation is truncated at second order in energy (the diffusion approximation in energy space), and the energy diffusion coefficient is taken from a Brownian-motion analogy. Numerically, RNNT used a pair potential combining a rigid spherical core with a purely dispersive attraction, with the potential parameters — the well depth ϵ\epsilon and the core size η\eta — fitted to the macroscopic liquid density and surface tension; the predictions were compared with measured nucleation rates of the lower 1-alcohols in argon carrier gas (the experiments of Strey, Wagner and Schmeling and others).

The source chapter (Ruckenstein and Berim's 2016 compilation of the RNNT vapor-to-liquid papers) states the outstanding issues explicitly: real molecules such as alcohols have nonspherical shapes and anisotropic, orientation-dependent interactions (dipolar and hydrogen-bonding) that a spherical purely dispersive potential cannot represent, so the observed agreement for the smaller alcohols is only indirect support; the accuracy of the second-order energy-space expansion and of the Brownian energy diffusion coefficient for very small clusters or strong interactions has not been investigated; and because the potential parameters are fitted to surface tension and density, the predictions are not fully ab initio.

Later work sharpened but did not close the problem. Experimentally, nucleation rates of the 1-alcohol homologous series are now comprehensively mapped between 235 and 265 K (Manka et al., 2012); CNT still fails, the semiphenomenological extended modified liquid drop–dynamical nucleation theory (EMLD–DNT) of Reguera and Reiss predicts 1-propanol to 1-pentanol reasonably but breaks down for ethanol because of vapor-phase molecular association, and the dataset is explicitly offered as the benchmark for nucleation theory. On the theory side, RNNT's own successors kept spherical molecules: the 2005 density-functional refinement by Djikaev and Ruckenstein replaced the uniform-density, sharp-boundary cluster by density-functional-theory density profiles but retained a hard-sphere-plus-dispersion pair potential. Kalikmanov's mean-field kinetic nucleation theory (2006) uses the kinetic approach for cluster kinetics but is explicitly semiphenomenological, containing both a macroscopic and a microscopic surface tension. Molecular dynamics has delivered parameter-free nucleation rates only for simple fluids (Lennard-Jones-type and metallic vapors, argon) and for water with empirical model potentials (Yasuoka and Matsumoto 1998; Matsubara et al. 2007; Toxvaerd 2001), while the microscopic tests of the kinetic framework itself — Ford and coworkers' treatment of cluster decay as escape from a potential of mean force — are likewise confined to argon-like clusters. First-principles quantum-chemical cluster free energies coupled to kinetic cluster dynamics now exist, but for small atmospheric multicomponent clusters rather than for the vapor-to-liquid nucleation of pure molecular fluids. A fully molecular, parameter-free kinetic prediction of homogeneous nucleation rates for real, nonspherical molecules therefore remains open, with the alcohol data as its natural test.

Problem Statement

Extend the molecular kinetic theory of homogeneous vapor-to-liquid nucleation to real molecular fluids and make it fully predictive from intermolecular interactions alone. Concretely, construct and validate a kinetic treatment in which: (1) the evaporation rate of molecules from a cluster (and, by detailed balance, the condensation rate) is derived from a molecular-level kinetic description valid for realistic, orientation-dependent intermolecular interactions of nonspherical molecules — including molecular shape and association such as hydrogen bonding — rather than only a rigid spherical core with a purely dispersive attraction; (2) the accuracy of the energy-space diffusion approximation — the second-order truncation of the kinetic equation in energy and the Brownian-motion energy diffusion coefficient — is established for very small clusters and strong molecule–cluster interactions, where the small-parameter justification of the expansion is weakest; and (3) no parameter of the theory is fitted to macroscopic nucleation-related quantities such as the planar surface tension or the liquid density. These three components are all required, and together they serve one overarching objective: predicting homogeneous nucleation rates J(T,S)J(T,S) and critical cluster sizes from the intermolecular potential alone. A resolution is a treatment that satisfies all three together and yields predictions for the alcohol series as a whole; no single component on its own, and no comparison with a single alcohol on its own, resolves the task. Validation requires that the predicted J(T,S)J(T,S) for ethanol to pentanol (235–265 K) be compared with the measured rates of Manka et al. (2012) and reproduce their magnitude and temperature dependence; a construction that only emits numbers without that comparison does not resolve the task. The predictions can be tested against measured homogeneous nucleation rates of real molecular fluids; the source theory itself was compared with lower 1-alcohol vapors in argon carrier gas, and comprehensive alcohol data (e.g., Manka et al. 2012, ethanol to pentanol, 235–265 K) now exist as a natural test set.

References

  1. Eli Ruckenstein and Gersh Berim, 'Kinetic Theory of Vapor-to-Liquid Nucleation', in Kinetic Theory of Nucleation, CRC Press (2016), pp. 115–228, DOI: 10.1201/b21644-3, https://doi.org/10.1201/b21644-3
  2. B. Nowakowski and E. Ruckenstein, 'A kinetic approach to the theory of nucleation in gases', The Journal of Chemical Physics (1991), DOI: 10.1063/1.459997, https://doi.org/10.1063/1.459997
  3. B. Nowakowski and E. Ruckenstein, 'Homogeneous nucleation in gases: A three-dimensional Fokker–Planck equation for evaporation from clusters', The Journal of Chemical Physics (1991), DOI: 10.1063/1.460719, https://doi.org/10.1063/1.460719
  4. Eli Ruckenstein and B. Nowakowski, 'A unidimensional Fokker-Planck approximation in the treatment of nucleation in gases', Langmuir (1991), DOI: 10.1021/la00055a042, https://doi.org/10.1021/la00055a042
  5. Y. S. Djikaev and E. Ruckenstein, 'Kinetic theory of nucleation based on a first passage time analysis: Improvement by the density-functional theory', The Journal of Chemical Physics (2005), DOI: 10.1063/1.2135777, https://doi.org/10.1063/1.2135777
  6. R. Strey, P. E. Wagner and T. Schmeling, 'Homogeneous nucleation rates for n-alcohol vapors measured in a two-piston expansion chamber', The Journal of Chemical Physics (1986), DOI: 10.1063/1.450396, https://doi.org/10.1063/1.450396
  7. A. Manka, J. Wedekind, D. Ghosh, K. Höhler, J. Wölk and R. Strey, 'Nucleation of ethanol, propanol, butanol, and pentanol: A systematic experimental study along the homologous series', The Journal of Chemical Physics (2012), DOI: 10.1063/1.4739096, https://doi.org/10.1063/1.4739096
  8. V. I. Kalikmanov, 'Mean-field kinetic nucleation theory', The Journal of Chemical Physics (2006), DOI: 10.1063/1.2178812, https://doi.org/10.1063/1.2178812
  9. David Reguera and Howard Reiss, 'Extended Modified Liquid Drop−Dynamical Nucleation Theory (EMLD−DNT) Approach to Nucleation: A New Theory', The Journal of Physical Chemistry B (2004), DOI: 10.1021/jp047168h, https://doi.org/10.1021/jp047168h
  10. David Reguera and Howard Reiss, 'Fusion of the Extended Modified Liquid Drop Model for Nucleation and Dynamical Nucleation Theory', Physical Review Letters (2004), DOI: 10.1103/PhysRevLett.93.165701, https://doi.org/10.1103/PhysRevLett.93.165701
  11. I. Ford and S. Harris, 'Molecular cluster decay viewed as escape from a potential of mean force', The Journal of Chemical Physics (2004), DOI: 10.1063/1.1644533, https://doi.org/10.1063/1.1644533
  12. H. Y. Tang and I. Ford, 'Microscopic simulations of molecular cluster decay: Does the carrier gas affect evaporation?', The Journal of Chemical Physics (2006), DOI: 10.1063/1.2357147, https://doi.org/10.1063/1.2357147
  13. S. Toxvaerd, 'Molecular-dynamics simulation of homogeneous nucleation in the vapor phase', The Journal of Chemical Physics (2001), DOI: 10.1063/1.1412608, https://doi.org/10.1063/1.1412608
  14. K. Yasuoka and M. Matsumoto, 'Molecular dynamics simulation of homogeneous nucleation in supersaturated water vapor', Fluid Phase Equilibria (1998), DOI: 10.1016/S0378-3812(97)00281-1, https://doi.org/10.1016/S0378-3812(97)00281-1
  15. H. Matsubara, T. Koishi, T. Ebisuzaki and K. Yasuoka, 'Extended study of molecular dynamics simulation of homogeneous vapor-liquid nucleation of water', The Journal of Chemical Physics (2007), DOI: 10.1063/1.2803899, https://doi.org/10.1063/1.2803899
  16. D. Reguera, J. M. Rubí and A. Pérez-Madrid, 'Fokker–Planck equations for nucleation processes revisited', Physica A: Statistical Mechanics and its Applications (1998), DOI: 10.1016/S0378-4371(98)00259-3, https://doi.org/10.1016/S0378-4371(98)00259-3
  17. David T. Wu, 'The continuum approximation in nucleation theory', The Journal of Chemical Physics (1992), DOI: 10.1063/1.463129, https://doi.org/10.1063/1.463129
  18. E. Ruckenstein and Y. Djikaev, 'Recent developments in the kinetic theory of nucleation', Advances in Colloid and Interface Science (2005), DOI: 10.1016/j.cis.2005.06.001, https://doi.org/10.1016/j.cis.2005.06.001
  19. Runlong Cai, Xiaoxiao Li, Rujing Yin, Bin Zhao, Chao Yan, Juha Kangasluoma, Xiuhui Zhang, Jun Zheng, Hong-Bin Xie, Veli-Matti Kerminen, Kebin He, James N. Smith, Lin Wang, Markku Kulmala and Jingkun Jiang, 'New Particle Formation and Growth in Urban Atmospheres: From Observations to Molecular-Level Understanding', Chemical Reviews (2026), DOI: 10.1021/acs.chemrev.5c00684, https://doi.org/10.1021/acs.chemrev.5c00684