NF17 


Non-Newtonian Fluid Mechanics & Flow Instabilities


A new mechanism for shear banding of a thixotropic yield stress fluid


October 17, 2018 (Wednesday) 11:05


Track 7 / Plaza II

(Click on name to view author profile)

  1. Geri, Michela (MIT, Mechanical Engineering)
  2. Saint-Michel, Brice (Imperial College, Department of Chemical Engineering)
  3. Divoux, Thibaut (CNRS Bordeaux, CNRS-MIT, MSE2)
  4. Manneville, Sebastien (ENS Lyon, Laboratoire de Physique)
  5. McKinley, Gareth H. (Massachusetts Institute of Technology)

(in printed abstract book)
Michela Geri1, Brice Saint-Michel2, Thibaut Divoux3, Sebastien Manneville4, and Gareth H. McKinley1
1Massachusetts Institute of Technology, Cambridge, MA 02139; 2Department of Chemical Engineering, Imperial College, London SW72AZ, United Kingdom; 3MSE2, CNRS Bordeaux, CNRS-MIT, Cambridge, MA; 4Laboratoire de Physique, ENS Lyon, Lyon, France


Geri, Michela


Shear banding is known to affect the flow of many complex fluids subject to simple viscometric flows. These fluids usually share a common feature in their underlying flow curve: a non-monotonic branch where the stress is a decreasing function of the shear rate. For thixotropic yield stress fluids this unstable branch usually spans between zero and a critical shear rate. Experiments performed on different systems, such as pastes and colloidal gels, show that steady-state banding occurs for any flows where the applied shear rate is smaller than this critical value and is characterized by an arrested band adjacent to a steady sheared region, while the stress measured during flow is constant and equal to the apparent yield stress. Here, we report on a new mechanism for steady-state shear banding observed in a thixotropic yield stress fluid for which the non-monotonic branch can be partially accessed during measurements of the flow curve. The fluid is made of a suspension of microscopic paraffin platelets in mineral oil at different concentrations. We perform rheometric tests in a bespoke Couette cell with roughened walls and a 1mm gap while measuring the local velocity field via ultrasonic velocimetry. The tests consist of decreasing shear rate steps of variable length and the stress response is recorded for the entire duration of each step. Results for different concentrations show the existence of two different critical shear rates corresponding to the local minimum and maximum observed in the flow curve. Below the first critical shear rate slip occurs close to the rotor, an arrested growing band appears close to the stator and the flowing region remains sheared at a constant shear rate. Below the second critical shear rate, the velocity field in the contracting flowing band is characterized by an inhomogeneous shear rate reminiscent of the flow field that develops in the presence of non-local effects. Based on these experimental observations we attempt to describe the velocity field using existing fluidity models.