Non-Newtonian Fluid Seminar 01 · Universität Heidelberg

This post is the first part of the script of my seminar talk From Maxwell’s Fluid to the Oldroyd-B Model, translated from Typst to Markdown. The series:

  1. Maxwell’s Fluid (this post)
  2. From Maxwell's Fluid to Oldroyd-B (II): Objective Derivatives
  3. From Maxwell's Fluid to Oldroyd-B (III): Oldroyd-B Model and Weissenberg Effect

1. Introduction

You may have seen this picture in the Moodle page. This is one of the most canonical examples of the strange behaviour of non-Newtonian fluids, the Weissenberg effect.

Water forms a whirlpool around a rotating rod; the polymer solution climbs up the rod.
Water forms a whirlpool around a rotating rod; the polymer solution climbs up the rod.

I am going to answer why this phenomenon happens with Oldroyd-B model. Before that, let’s revisit viscosity and elasticity.

2. What are Viscosity and Elasticity?

Every material sits somewhere between two toys: a spring, which remembers, and a dashpot, which forgets. One stores energy, the other burns it.

A dashpot (left) models viscosity; a spring (right) models elasticity. (Morozov, A., & Spagnolie, S. E. (2015))
A dashpot (left) models viscosity; a spring (right) models elasticity. (Morozov, A., & Spagnolie, S. E. (2015))

2.1. Spring: Elasticity

Squeeze a spring and it fights back. Squeeze further, it fights harder. Let go, and it wobbles a little and springs right back: a spring never forgets its original shape.

This memory is elasticity: stress depends on how far you deform it, as Hooke’s law says.

Hooke's Law
(2.1)σ=Gγ.
  • σ : stress
  • G : shear modulus
  • γ : deformation (strain)

2.2. Dashpot: Viscosity

Never met a dashpot? A syringe is close enough. Push slowly and it gives way; push fast and it resists. Let go, and the piston just stays where it is: a dashpot has no memory at all.

Resistance grows with the speed of the piston. (From Wikipedia)
Resistance grows with the speed of the piston. (From Wikipedia)

This is viscosity: stress depends on how fast you deform it:

Newtonian Viscosity
(2.2)σ=ηγ˙.

3. Maxwell Model

How to model viscoelasticity? Namely, how to build a model that puts viscosity and elasticity together?

The simplest marriage: hook a spring and a dashpot in series. Like an (electrical) circuit, with stress ∼ voltage and deformation ∼ current.

Both feel the same stress, but each stretches in its own way, so the strain rates add up:

(3.1)γ˙=γ˙spring+γ˙dashpot=σ˙G+ση.

Multiplying by η gives the Maxwell model:

Maxwell Model
(3.2)σ+λσ˙=ηγ˙=2ηD,λ=ηG,

where D is the rate-of-deformation tensor and λ the relaxation time.

The Maxwell model: a spring and a dashpot in series. (Morozov, A., & Spagnolie, S. E. (2015))
The Maxwell model: a spring and a dashpot in series. (Morozov, A., & Spagnolie, S. E. (2015))
  • poke it fast ( t≪λ ): the spring answers, solid.
  • push it slowly ( t≫λ ): the dashpot wins, fluid.

3.1. When the Maxwell Model Fails

The Maxwell model is only good for simple deformations, such as linear shear. The transition from scalar to tensorial form is not trivial:

(3.3)σ+λσ˙=2ηD.

As σ˙(x+ut,t)=∂tσ+∇σ⋅u , where the convective term ∇σ⋅u is not frame invariant, different observers see different dynamics and obtain different results.

We need a time derivative that every observer agrees on. This is the topic of the next post: From Maxwell's Fluid to Oldroyd-B (II): Objective Derivatives.