Taylor dispersion in variable-density, variable-viscosity pulsatile flows

This paper investigates Taylor dispersion in pulsatile pipe flows where the scalar field influences fluid density and viscosity, deriving an effective one-dimensional unsteady mixing model through multiple scale analysis.

The Gist

Imagine you are stirring a cup of coffee. If you just let it sit, the sugar dissolves slowly because of diffusion (molecules wandering randomly). But if you swirl the spoon, the sugar mixes much faster. This is because the swirling motion stretches the sugar out into long, thin ribbons, giving it more surface area to dissolve. In physics, this phenomenon is called Taylor Dispersion.

Now, imagine that the "sugar" you are stirring isn't just passive; it actually changes the properties of the liquid itself. Maybe it makes the coffee thicker (more viscous) or heavier (denser) as it mixes. This is the core problem this paper tackles: What happens to the mixing speed when the thing you are mixing changes the fluid's behavior?

Here is a breakdown of the paper's findings using everyday analogies:

1. The Setup: A Pulsating Pipe

The researchers studied a long pipe where fluid is being pushed back and forth (pulsatile flow), like blood pumping through an artery or air moving in a lung.

• The Old Way: Scientists used to assume the fluid was like water—always the same thickness and weight, no matter what was mixed in.
• The New Reality: In this study, the "scalar" (the thing being mixed, like a chemical or heat) actively changes the fluid. If you add a lot of it, the fluid might get thick like honey or heavy like syrup.

2. The Problem: A Moving Target

When the fluid changes its thickness (viscosity) or weight (density) because of the mixing, the flow pattern changes too.

• Analogy: Imagine a river flowing through a canyon. If the river suddenly turns into thick mud in one spot, the water slows down there and speeds up elsewhere. The "river" (the flow) is constantly reshaping itself based on the "mud" (the scalar) it is carrying.
• Because the flow changes, the way the scalar spreads out (disperses) also changes. It's a two-way street: The flow moves the scalar, but the scalar changes the flow.

3. The Method: The "Slow Motion" Camera

To solve this complex math problem, the authors used a technique called Multiple Scale Analysis.

• The Analogy: Imagine watching a hummingbird. To your eye, it's a blur (fast time). But if you look at the forest behind it, the trees are moving very slowly (slow time).
• The researchers separated the problem into two speeds:
  1. Fast Time: The rapid back-and-forth pulsing of the fluid (like the hummingbird's wings).
  2. Slow Time: The gradual spreading of the scalar down the pipe (like the hummingbird slowly flying across the sky).
• By separating these, they could simplify the messy, fast-moving details to find a clear, simple rule for the slow, overall mixing.

4. The Big Discovery: A New "Mixing Formula"

The paper derives a new equation that predicts how fast things will mix in this variable environment. They found that the "effective diffusion" (the total speed of mixing) is made of three parts:

1. Molecular Diffusion: The natural, slow wandering of molecules (like sugar dissolving in still water).
2. Steady Shear Dispersion: The mixing caused by the steady part of the flow (like the main current of the river).
3. Oscillatory Shear Dispersion: The new mixing caused by the back-and-forth pulsing.

The Twist:
In the old models, the mixing speed depended only on how fast the fluid moved. In this new model, the mixing speed depends on how the scalar changes the fluid's thickness and weight.

• Example: If the scalar makes the fluid heavier (like adding salt to water), the mixing might speed up or slow down depending on the specific conditions. If it makes the fluid thicker, the "ribbons" of scalar stretch differently, changing the mixing rate.

5. The Result: A Simple Rule for a Complex World

After all the complex math involving "Bessel functions" and "perturbation series," the authors boiled it down to a one-dimensional equation.

• The Metaphor: They took a chaotic, 3D, changing, pulsating storm and distilled it into a simple, straight-line highway equation.
• This equation tells engineers and scientists exactly how to calculate the mixing rate for things like:
  • Combustion engines: Where fuel mixes with air, changing the temperature and density instantly.
  • Blood flow: Where blood cells and plasma interact, changing viscosity.
  • Chemical reactors: Where mixing chemicals changes the fluid's properties.

Summary

This paper is like upgrading a weather forecast. The old forecast said, "It's windy, so leaves will blow fast." The new forecast says, "It's windy, but the leaves are wet and heavy, and the wind is gusting back and forth, so here is the exact path and speed the leaves will take."

The authors successfully created a mathematical tool that accounts for the fact that the mixture changes the fluid, and the fluid changes the mixture, providing a more accurate way to predict how things mix in pulsating pipes.

Technical Summary

1. Problem Statement

The paper investigates the phenomenon of Taylor dispersion (shear-induced dispersion) of a non-passive scalar field within a pulsatile pipe flow.

• Non-Passive Scalar: Unlike traditional studies where the scalar (e.g., a solute) is passive and does not affect the fluid properties, this study considers a scalar that actively alters the fluid's density ($\rho$) and viscosity ($\mu$).
• Coupling Mechanism: The scalar concentration changes the fluid properties, which in turn modifies the flow field (hydrodynamics). This altered flow field then feeds back to influence the dispersion of the scalar itself.
• Flow Regime: The flow is driven by a longitudinal pressure gradient consisting of a steady component and an oscillatory (pulsatile) component.
• Objective: To derive an effective one-dimensional, unsteady mixing equation that accounts for these hydrodynamic couplings and variable transport coefficients.

2. Methodology

The authors employ Multiple Scale Analysis (asymptotic analysis) to solve the governing equations under specific scaling limits.

Governing Equations

The problem is modeled using the low Mach-number Navier-Stokes equations coupled with a scalar transport equation:

• Continuity & Momentum: Low Mach-number approximation including variable density and viscosity.
• Scalar Transport: Advection-diffusion equation with source/sink terms $Q(c)$ and wall leakage $S(c)$.
• Constitutive Relations: Density $\rho$, viscosity $\mu$, and diffusivity $\lambda$ are functions of the scalar concentration $c$.

Scaling and Parameters

The analysis assumes the Taylor dispersion limit, characterized by:

• $\varepsilon = a/l_m \ll 1$: The ratio of pipe radius ($a$) to the mixing length scale ($l_m$) is small.
• Peclet Numbers: $Pe \sim 1$ and $\tilde{Pe} \sim 1$ (steady and oscillatory components), indicating advection and diffusion are balanced.
• Dimensionless Groups:
  • Schmidt number ($Sc \sim 1$).
  • Frequency parameter ($\beta \sim 1$), linking oscillation frequency to radial diffusion time.

Perturbation Strategy

1. Coordinate Transformation: A moving frame of reference is adopted with velocity $U$ (mean steady flow) to eliminate net mass flux at infinity.
2. Time Scales: Two time scales are introduced:
   • Slow time ($t$): Associated with longitudinal mixing ($l_m^2/D_\infty$).
   • Fast time ($\tau$): Associated with radial diffusion and oscillation ($a^2/D_\infty$).
3. Series Expansion: Variables (velocity, pressure, concentration) are expanded in powers of the small parameter $\varepsilon$.
   • $c = c_0 + \varepsilon c_1 + \dots$
   • $u = u_0/\varepsilon + u_1 + \dots$
4. Averaging: The solution is averaged over the fast spatial (radial) and temporal (oscillatory) scales to derive the macroscopic 1D equation.

3. Key Contributions

The paper makes several significant theoretical advancements:

1. Generalization of Taylor Dispersion: Extends the classical Taylor-Aris dispersion theory to variable-density and variable-viscosity flows in pulsatile regimes.
2. Hydrodynamic Coupling: Explicitly derives how the scalar-induced changes in density and viscosity modify the velocity profile and, consequently, the dispersion coefficient.
3. Effective 1D Model: Derives a closed-form, one-dimensional unsteady advection-diffusion equation that captures the complex 3D physics through an effective diffusion coefficient ($D_{eff}$).
4. Analytical Solution for Oscillatory Flow: Provides explicit formulas for the velocity and concentration profiles in the oscillatory regime with variable properties, utilizing modified Bessel functions ($I_0, I_1, I_2$).

4. Key Results

The Effective Diffusion Coefficient ($D_{eff}$)

The most critical result is the derivation of the effective diffusion coefficient, which comprises three distinct terms:
$$\rho D_{eff} = \underbrace{\lambda}_{\text{Molecular}} + \underbrace{\frac{Pe^2 \rho^2}{48\lambda}}_{\text{Steady Taylor}} + \underbrace{\text{Oscillatory Term}}_{\text{Shear-induced}}$$

• Term 1 (Molecular): The base molecular diffusivity $\lambda(c)$.
• Term 2 (Steady Taylor): Represents dispersion due to the steady Poiseuille flow.
  • Novelty: It depends on $\rho^2/\lambda$. The authors note that if $d\rho/dc > 0$ (e.g., saline water), the effective Peclet number increases, enhancing dispersion. If $d\rho/dc < 0$ (e.g., thermal mixing), dispersion decreases.
  • Note: Unlike the oscillatory term, the steady Taylor term is independent of viscosity $\mu$.
• Term 3 (Oscillatory): Represents shear-induced dispersion due to the pulsatile component.
  • This term is complex and depends on both viscosity ($\mu$) and diffusivity ($\lambda$).
  • It involves integrals of modified Bessel functions and the variable Schmidt number $\sigma = \mu Sc / \lambda$.
  • The formula accounts for the phase lag and amplitude modification of the oscillatory velocity profile caused by variable fluid properties.

Velocity and Concentration Profiles

• The leading-order velocity $u_0$ consists of a standard Poiseuille profile (steady part) and a complex oscillatory part that evolves with large-scale coordinates $(x, t)$ due to property variations.
• The first-order concentration correction $c_1$ captures the radial distribution of the scalar, which drives the axial dispersion.

Governing Equation

The final macroscopic equation for the scalar concentration $c$ in the Lagrangian coordinate system ($\psi$) is:
$$\frac{\partial c}{\partial t} = \frac{\partial}{\partial \psi} \left( \rho^2 D_{eff} \frac{\partial c}{\partial \psi} \right) + \frac{1}{\rho}(Q - S)$$
This equation allows for the simulation of scalar transport in complex pulsatile flows without resolving the full 3D Navier-Stokes equations.

5. Significance and Applications

• Combustion and Reacting Flows: The model is highly relevant for combustion systems where temperature (acting as a scalar) drastically changes gas density and viscosity, and flows are often unsteady/pulsatile.
• Biomedical Engineering: Applicable to drug delivery or blood flow analysis where solute concentration affects blood viscosity (non-Newtonian effects approximated here) and density in pulsatile vessels.
• Environmental Flows: Useful for modeling pollutant dispersion in estuaries or atmospheric flows where salinity or temperature gradients alter fluid properties.
• Theoretical Foundation: The paper bridges the gap between classical Taylor dispersion (steady, constant properties) and modern complex flow scenarios, providing a rigorous mathematical framework for "active" scalars in unsteady flows.

In summary, Rajamanickam and Weiss successfully decouple the complex 3D interaction between scalar transport and hydrodynamics in pulsatile flows, reducing it to a tractable 1D problem with a rigorously derived, property-dependent effective diffusion coefficient.