Liquid-like growth of colloidal nanocrystals of coalescence

TL;DR

This paper reveals that colloidal nanocrystals grow via coalescence in a manner similar to liquids, with high reaction rates and predictable behavior, challenging traditional views of crystal growth.

Contribution

It introduces a new understanding of colloidal nanocrystal growth through coalescence, supported by experiments, modeling, and simulations, showing high reaction rates and control mechanisms.

Findings

High coalescence rates comparable to bimolecular reactions

A simple model predicts coalescence behavior based on ligands

Crowding effects explain large coalescence rates and low activation energy

Abstract

Our understanding of the growth of crystals is dominated by the classical description according to which individual atoms or molecules, driven by supersaturation, add to crystal facets. As a result, the growth of hard matter is still mostly considered to be fundamentally incomparable to the growth of soft matter, like polymers or liquids. By a combination of experiment and modeling we here show how amine-capped PbS colloidal nanoparticles grow in the absence of supersaturation by coalescence, like droplets in an emulsion. Specifically, we (i) determine that the rates of crystal-crystal coalescence are remarkably high (10^-2 to 10^1 M^-1*s^-1) in spite of the steric stabilization of the particles, and are comparable to those of bimolecular reactions, (thereby providing a new avenue for the development of a form of chemistry where the reactants are colloids rather than molecules), (ii) elucidate the rate limiting steps of crystal-crystal coalescence leading us to propose design rules to control it, and (iii) demonstrate a simple, two-parameter model that predicts quantitatively this process and its dependence on the ligands. Lastly, we use Brownian dynamics simulations to show how crowding effects and the relatively large size of the particles compared to their mean free path explain these remarkably large rates of coalescence and, at the same time, the puzzlingly low values of activation energy previously reported for oriented attachment processes.