Extended Wertheim theory predicts the anomalous chain length distributions of divalent patchy particles under extreme confinement

TL;DR

This paper extends Wertheim's theory to predict the chain length distribution of divalent patchy particles under extreme confinement, explaining experimental anomalies and matching simulation results without adjustable parameters.

Contribution

The authors develop a modified Wertheim theory incorporating chain length-dependent reaction constants to accurately model confined patchy particle polymerization.

Findings

The extended theory predicts chain length distributions in excellent agreement with simulations.

It explains the high monomer concentration observed in experiments under confinement.

The approach is applicable to various interaction potentials and confinement conditions.

Abstract

Colloidal patchy particles with divalent attractive interaction can self-assemble into linear polymer chains. Their equilibrium properties in 2D and 3D are well described by Wertheim's thermodynamic perturbation theory which predicts a well-defined exponentially decaying equilibrium chain length distribution. In experimental realizations, due to gravity, particles sediment to the bottom of the suspension forming a monolayer of particles with a gravitational height smaller than the particle diameter. In accordance with experiments, an anomalously high monomer concentration is observed in simulations which is not well understood. To account for this observation, we interpret the polymerization as taking place in a highly confined quasi-2D plane and extend the Wertheim thermodynamic perturbation theory by defining addition reactions constants as functions of the chain length. We derive the theory, test it on simple square well potentials, and apply it to the experimental case of synthetic colloidal patchy particles immersed in a binary liquid mixture that are described by an accurate effective critical Casimir patchy particle potential. The important interaction parameters entering the theory are explicitly computed using the integral method in combination with Monte Carlo sampling. Without any adjustable parameter, the predictions of the chain length distribution are in excellent agreement with explicit simulations of self-assembling particles. We discuss generality of the approach, and its application range.