Structuring of Fluid Adlayers upon Ongoing Unimolecular Adsorption

Abstract

Fluids with spatial density variations of single or mixed molecules play a key role in biophysics, soft matter and materials science. The fluid structures usually form via spinodal decomposition or nucleation following an instantaneous destabilisation of the initially disordered fluid. However, in practice an instantaneous quench is often not viable, and the rate of destabilisation may be gradual rather than instantaneous. In this work we show that the commonly used phenomenological descriptions of fluid structuring are inadequate under these conditions. We come to that conclusion in the context of surface catalysis, where we employ kinetic Monte Carlo simulations to describe the unimolecular adsorption of gaseous molecules onto a metal surface. The adsorbates diffuse at the surface and, as a consequence of lateral interactions and due to an ongoing increase of the surface coverage, phase separate into coexisting low-and high-density regions. The typical size of these regions turns out to depend much more strongly on the rate of adsorption than predicted from recently reported phenomenological models. We discuss how this finding contributes to the fundamental understanding of the crossover from liquid-liquid to liquid-solid demixing of solutioncast polymer blends.

Morphology formation in binary mixtures upon gradual destabilisation

“…However, there is limited understanding of how the early-stage structure development is affected by the physical properties of the molecules and by the processing conditions [13,14,15,16,17,18,19,20,21]. Indeed, in case that the mixture is gradually destabilised, e.g., in case of slow cooling or of an ongoing concentration increase due to solvent evaporation, mean-field theory fails [22] and is unable to predict the rather strong dependence of the coarseness of the morphology on the quench rate [18]. In the present work, we rationalise this observation in terms of universal critical dynamics [23].…”

“…Recently, kinetic Monte Carlo (kMC) simulations of a 2D lattice fluid seemed to indicate that liquid-liquid demixing upon a gradual concentration quench leads to a one-fourth power, rather than one sixth [18]. This power-law exponent was interpreted to be originated by the diffusion of material towards nuclei: Phase separation takes place at the point in time where diffusion becomes faster than the rate at which the mean-free path becomes shorter, due to an increasing concentration.…”

“…In that case, the size of the clusters is not coupled to the mean-free path, but to the stability of spherical critical nuclei as described by classical nucleation theory [29]. Although both theories seem to agree with the observations, the diffusion-limited nucleation argument seems to be specific to the case studied [18], and a simple classical nucleation theory is inaccurate near the critical point [30].…”

“…Phenomenological phase-field models are highly successful in explaining how coarsening and arresting are affected by viscoelastic effects [6], hydrodynamics [7,8], chemical reactions [9,7,10,8], turbulent flow [11], and the presence of a surfactant [12]. However, there is limited understanding of how the early-stage structure development is affected by the physical properties of the molecules and by the processing conditions [13,14,15,16,17,18,19,20,21]. Indeed, in case that the mixture is gradually destabilised, e.g., in case of slow cooling or of an ongoing concentration increase due to solvent evaporation, mean-field theory fails [22] and is unable to predict the rather strong dependence of the coarseness of the morphology on the quench rate [18].…”

“…This power-law exponent was interpreted to be originated by the diffusion of material towards nuclei: Phase separation takes place at the point in time where diffusion becomes faster than the rate at which the mean-free path becomes shorter, due to an increasing concentration. While this argument compares the time scale of diffusion-limited nucleation to the quench rate [18], the power one fourth may also be obtained by comparing the time scale of reaction-limited nucleation to the quench rate [28]. In that case, the size of the clusters is not coupled to the mean-free path, but to the stability of spherical critical nuclei as described by classical nucleation theory [29].…”