Mixing in viscoelastic fluids using elastic turbulence

Basic Information

• Paper ID: 2409.06391
• Title: Mixing in viscoelastic fluids using elastic turbulence
• Authors: Reinier van Buel, Holger Stark (Technische Universität Berlin)
• Classification: physics.flu-dyn
• Publication Date: September 2024
• Paper Link: https://arxiv.org/abs/2409.06391

Abstract

This paper investigates the effect of elastic turbulence on the mixing of scalar concentration fields in viscoelastic fluids within a two-dimensional Taylor-Couette geometry, using numerical solutions of the Oldroyd-B model. The flow state is determined through a secondary flow order parameter, indicating the elastic turbulence region. When starting from a turbulent state and subsequently decreasing the Weissenberg number, weak chaotic flow emerges below Wic. Advection in both turbulent and weak chaotic flow states induces mixing, illustrated through the temporal evolution of the standard deviation of solute concentration relative to uniform distribution. Notably, mixing is strong in the elastic turbulence state, quantified through mixing rate, mixing time, and mixing efficiency. All three quantities follow scaling laws. Importantly, the order parameter shows strong correlation with mixing rate, making it a good indicator of mixing within the fluid.

Research Background and Motivation

Problem Background

1. Mixing challenges at low Reynolds numbers: In Newtonian fluids, flow is laminar under low Reynolds number conditions, resulting in slow mixing controlled primarily by diffusion. This represents a significant limitation in microfluidic devices.
2. Discovery of elastic turbulence: Adding high molecular weight polymers to fluids renders them viscoelastic, exhibiting elastic turbulence phenomena even when inertial effects are negligible.
3. Importance of microscale mixing: In lab-on-a-chip devices, efficient fluid mixing is crucial, particularly at micrometer scales.

Research Motivation

• Explore the potential of elastic turbulence in enhancing mixing efficiency
• Understand the physical mechanisms of mixing in viscoelastic fluids
• Provide theoretical guidance for microfluidic applications

Core Contributions

1. Characterization of transition nature: Identified the transition from laminar flow to elastic turbulence as a subcritical transition and discovered the existence of weak chaotic flow states
2. Establishment of mixing characterization methods: Introduced three quantitative indicators—mixing rate, mixing time, and mixing efficiency—all following scaling laws
3. Discovery of strong correlation between order parameter and mixing: Demonstrated strong correlation between the secondary flow order parameter and mixing rate, serving as a good indicator of fluid mixing degree
4. Verification of Batchelor mixing mechanism: Observed exponential decay in the elastic turbulence state, corresponding to the Batchelor mixing region, consistent with experimental results

Methodology Details

Geometric Setup and Governing Equations

The study employs a two-dimensional Taylor-Couette geometry with two concentric cylinders:

• Outer cylinder radius: $r_o = 10\mu m$
• Inner cylinder radius: $r_i = 2.5\mu m$
• Outer cylinder rotates counterclockwise at angular velocity $\Omega = 2\pi s^{-1}$

Fluid motion is described by the generalized Navier-Stokes equation: $\rho\left(\frac{\partial \mathbf{u}}{\partial t} + \mathbf{u} \cdot \nabla \mathbf{u}\right) = -\nabla p + \eta_s \nabla^2 \mathbf{u} + \nabla \cdot \boldsymbol{\tau}$

Oldroyd-B Constitutive Model

The polymer stress tensor $\boldsymbol{\tau}$ employs the Oldroyd-B model: $\boldsymbol{\tau} + \lambda \stackrel{\nabla}{\boldsymbol{\tau}} = \eta_p[\nabla \otimes \mathbf{u} + (\nabla \otimes \mathbf{u})^T]$

where $\stackrel{\nabla}{\boldsymbol{\tau}}$ is the upper convected derivative.

Mixing Modeling

The evolution of scalar concentration field $c(\mathbf{r}, t)$ is governed by the advection-diffusion equation: $\frac{\partial c}{\partial t} + (\mathbf{u} \cdot \nabla)c = D\nabla^2 c$

In this study, $D = 0$, neglecting molecular diffusion.

Key Dimensionless Parameters

• Reynolds number: $Re = \rho\Omega r_o^2/\eta_s = 0.628 \times 10^{-3}$
• Weissenberg number: $Wi = \lambda\Omega$
• Viscosity ratio: $\beta = \eta_p/\eta_s = 1.5$

Experimental Setup

Numerical Methods

• Utilized OpenFOAM® open-source finite volume solver
• Employed RheoTool specialized solver for viscoelastic flow
• Mesh: 100 radial and 120 azimuthal elements in spoke-wheel configuration
• Time step: $\delta t = 10^{-5}s$

Order Parameter Definition

Secondary flow intensity is defined as the order parameter: $\Phi = \sigma = \sqrt{\langle[\mathbf{u}(\mathbf{r},t) - \mathbf{u}_0(\mathbf{r})]^2\rangle}/u_0^{\max}$

Mixing Characterization Indicators

Normalized standard deviation: $\Delta c = \sqrt{\frac{\langle[c(\mathbf{r},t) - c_\infty]^2\rangle}{\langle[c(\mathbf{r},0) - c_\infty]^2\rangle}}$

where $c_\infty$ is the concentration in the completely mixed state.

Experimental Results

Flow State Transitions

1. Subcritical transition: The transition from laminar to elastic turbulent flow exhibits subcritical characteristics with hysteresis phenomena
2. Critical Weissenberg numbers:
   • Upper branch: $Wi_c = 9.95$
   • Lower branch: $Wi_c^I = 9.26$
3. Weak chaotic state: Weak chaotic flow is observed below $Wi_c^I$

Mixing Characteristics Analysis

Weak Chaotic Flow State ($5.03 < Wi \leq 8.17$)

• Standard deviation exhibits power-law decay with exponent close to $-1/6$
• Power spectrum shows power-law scaling with exponent $-5$ or larger

Elastic Turbulence State ($Wi \geq 9.42$)

• Initial power-law decay (exponent ~$1/5$), followed by exponential decay
• Exponential decay form: $\Delta c \sim \exp(-\alpha_{mix}t)$
• Power spectrum exhibits Batchelor spectrum characteristics ($k^{-1}$ scaling)

Scaling Law Relationships

1. Mixing time: $t_{mix} \sim (Wi - Wi_c^I)^{-0.52}$
2. Region II concentration: $c_{II}/c_\infty \sim (Wi - Wi_c^I)^{0.28}$
3. Scaling relationship: $c_{II} \sim t_{mix}^{-1/2}$

Correlation Between Order Parameter and Mixing Rate

Experimental results demonstrate strong correlation between order parameter $\Phi_d$ and mixing rate $\alpha_{mix}$, indicating that enhanced velocity fluctuations directly improve fluid mixing efficiency.

Related Work

Elastic Turbulence Research

• First discovery of elastic turbulence in polymer solutions by Groisman & Steinberg
• Linear and nonlinear stability analysis in Taylor-Couette geometry
• Observation of spiral and banded flow patterns in three-dimensional numerical simulations

Mixing Enhancement Research

• Extensional flow mixing in four-roll mill geometry
• Elastic instability in cross-slot geometry
• Convective heat transfer enhancement in serpentine channels

Conclusions and Discussion

Main Conclusions

1. Transition nature: Confirmed the subcritical nature of elastic turbulence transition in Taylor-Couette flow
2. Mixing mechanism: Verified that elastic turbulence enables efficient mixing at extremely low Reynolds numbers
3. Batchelor region: Observed Batchelor mixing characteristics in the elastic turbulence state
4. Predictive indicator: Order parameter serves as a reliable predictor of mixing efficiency

Limitations

1. Two-dimensional constraint: Study is limited to two-dimensional geometry; three-dimensional effects may produce different results
2. Grid dependence: Hysteresis behavior disappears on finer grids, indicating effects of numerical accuracy
3. Initial condition sensitivity: Mixing characteristics depend on the choice of initial concentration distribution

Future Directions

1. Elastic wave effects: Investigate the influence of elastic waves on mixing efficiency, mixing rate, and mixing time
2. Three-dimensional extension: Verify discovered scaling laws in three-dimensional geometry
3. Practical applications: Explore potential applications in microfluidic devices

In-depth Evaluation

Strengths

1. Systematic investigation: Comprehensive analysis of mixing characteristics during the transition from laminar to elastic turbulent flow
2. Quantitative characterization: Introduction of multiple quantitative indicators to characterize mixing degree and efficiency
3. Theoretical verification: Numerical results are consistent with Batchelor mixing theory and experimental observations
4. Practical value: Provides important theoretical guidance for microfluidic applications

Weaknesses

1. Numerical accuracy issues: Grid dependence of hysteresis phenomena suggests limitations in numerical methods
2. Mechanism understanding: Physical mechanism explanation for power-law decay in weak chaotic state is insufficient
3. Parameter range: Relatively limited parameter space explored, lacking broader parametric investigation

Impact

1. Theoretical contribution: Provides important insights for understanding mixing mechanisms in viscoelastic fluids
2. Application prospects: Offers significant guidance for microfluidic device design
3. Methodological value: Established characterization methods can be generalized to other flow geometries

Applicable Scenarios

• Mixing optimization in microfluidic devices
• Mixing control in polymer processing
• Microscale mixing in biomedical applications

References

The paper cites 57 related references covering elastic turbulence, viscoelastic flow stability, mixing theory, and other aspects, providing a solid theoretical foundation for the research.