Nonlinear mechanics of colloidal gels: creep, fatigue and shear-induced yielding

TL;DR

This paper investigates the nonlinear mechanical behavior of colloidal gels, focusing on their yielding, fatigue, and shear-induced failure, and distinguishes between reversible and irreversible yielding mechanisms using rheological and mesoscopic experiments.

Contribution

It introduces a classification of colloidal gels based on their yielding behavior and discusses modeling approaches for creep and fatigue experiments to predict gel failure.

Findings

Reversible gels reform after load removal.

Irreversible gels undergo fracture and phase separation.

Modeling creep and fatigue helps predict yielding.

Abstract

Colloidal gels are formed through the aggregation of attractive particles, whose size ranges from 10~nm to a few micrometers, suspended in a liquid. Such gels are ubiquitous in everyday life applications, from food products to paints or construction materials, in particular thanks to their ability to easily "yield", i.e., to turn from a solid to a liquid under the application of a weak external load. Understanding and controlling the mechanical response of colloidal gels is therefore of prime importance. Depending on the details of the system, however, the resulting gel networks present different microstructural organisations that may lead to widely different mechanical responses. This raises important challenges in fully characterizing yielding and in uncovering the mechanisms of nonlinear response in colloidal gels. In this paper, we distinguish between two classes of colloidal gels showing respectively reversible yielding, where the gel network reforms upon load release, and irreversible yielding, where the network is fully destroyed through fractures and phase separation. This broad, empirical distinction is achieved through rheology and local experiments at a mesoscopic scale, intermediate between the network characteristic size and the sample size. We further discuss how the observables derived from creep and fatigue experiments may be modelled to predict yielding and highlight open questions and future research directions in the domain.