Effects of Thermal Boundary Conditions on Natural Convection and Entropy Generation in Non-Newtonian Power-Law Fluids

This study utilizes finite element simulations to demonstrate that in non-Newtonian power-law fluids, shear-thinning behavior enhances heat transfer while uniform thermal boundary conditions promote stronger convection and higher entropy generation compared to non-uniform heating, offering key insights for optimizing thermal system design.

The Gist

Imagine you are trying to understand how heat moves through a thick liquid, like honey or paint, inside a container. This paper is like a detailed recipe and a set of experiments to figure out exactly how that liquid behaves when it gets hot, and how much "wasted energy" (entropy) is created in the process.

Here is the breakdown of what the researchers did and found, using simple analogies:

The Setup: Two Different Containers

The scientists looked at two specific shapes to see how the liquid moves:

1. A Square Box: Think of a square picture frame. The bottom is hot, the sides are cold, and the top is covered (so heat can't escape).
2. A Donut Shape (Annulus): Imagine a large pipe with a smaller pipe inside it. The inner pipe is hot, and the outer pipe is cold.

In both cases, gravity pulls the liquid down. When the liquid near the hot wall warms up, it gets lighter and tries to float up (like a hot air balloon), while the cold, heavy liquid sinks. This creates a natural circulation loop without needing a pump or a fan.

The Special Ingredient: "Smart" Liquids

Most liquids (like water) have a constant thickness, or viscosity. But the liquids in this study are Non-Newtonian. This means their thickness changes depending on how fast they are moving.

• Shear-Thinning (The "Runny" Fluid): Imagine ketchup. The more you shake it or push it, the thinner and runnier it gets. In the paper, these are fluids where the power-law index is less than 1.
• Shear-Thickening (The "Stiff" Fluid): Imagine a mixture of cornstarch and water. If you hit it or push it hard, it instantly turns into a solid block. In the paper, these are fluids where the index is greater than 1.
• Newtonian (The "Normal" Fluid): This is the middle ground, like water or oil, where the thickness stays the same no matter how fast it moves.

The Experiment: Changing the Heat Source

The researchers didn't just heat the containers evenly. They tested two ways of applying heat:

1. Uniform Heating: Imagine turning on a heater that warms the entire bottom wall (or inner pipe) equally.
2. Non-Uniform (Sinusoidal) Heating: Imagine a heater that is hottest in the middle and gets cooler toward the edges, like a gentle wave of heat.

What They Found: The Dance of Heat and Flow

1. How the Liquid Moves (The Flow)

• The "Runny" Fluid (Shear-Thinning): When this fluid gets hot, it becomes thinner and moves much faster. It creates strong, vigorous swirling loops (vortices) that carry heat very efficiently. It's like a high-speed blender.
• The "Stiff" Fluid (Shear-Thickening): When this fluid tries to move, it gets thicker and resists the motion. The swirling loops become weak and sluggish. Heat moves mostly by slowly seeping through the liquid (conduction) rather than flowing. It's like trying to walk through deep mud.
• The Heating Pattern: When the heat was applied evenly (Uniform), the liquid created big, strong loops that filled the whole container. When the heat was applied in a wave (Non-uniform), the liquid only swirled strongly right where the heat was strongest, creating a localized "plume" of rising hot liquid, while the rest of the container stayed relatively calm.

2. How Much Heat Gets Transferred

• The "Runny" fluids transferred heat the best because they moved so fast.
• The "Stiff" fluids transferred heat the worst because they barely moved.
• Interestingly, the "Runny" fluids were even more sensitive to the heating pattern. When the heat was wavy, the difference in performance between the "Runny" and "Stiff" fluids became even more dramatic.

3. The "Wasted Energy" (Entropy Generation)
The researchers also calculated "entropy," which is a measure of how much energy is wasted or lost as disorder during the process. Think of it as the "friction cost" of moving the heat.

• The Big Surprise: For the "Runny" fluids, the biggest waste of energy came from the liquid rubbing against itself (viscous dissipation) as it swirled around fast. It was like the engine of a car revving too high and burning fuel just to spin the wheels.
• The Shift: As the fluid became "stiffer" (moving toward the Newtonian or Shear-thickening side), the friction waste dropped dramatically. Eventually, the main source of waste became the heat itself trying to move from hot to cold areas.
• The Heating Pattern Effect: The "Wavy" (Non-uniform) heating always resulted in less total wasted energy than the "Even" (Uniform) heating. By focusing the heat in one spot, the system didn't have to work as hard to move everything around, making it slightly more "thermodynamically efficient."

The Bottom Line

The study shows that if you want to control how heat moves through special fluids (like paints, polymers, or biological fluids), you have two levers to pull:

1. The Fluid Type: Choosing a fluid that gets thinner when it moves (shear-thinning) will make heat transfer faster but might create more friction waste.
2. The Heating Design: Heating a surface evenly creates strong, widespread currents. Heating it in a specific pattern (like a wave) creates focused currents and generally wastes less total energy.

The researchers built a powerful computer simulation (using a tool called Gridap.jl) to prove these points, and they made their code available so others can check their work. They confirmed that the way you heat a container is just as important as the type of liquid inside it when designing efficient thermal systems.

Technical Summary

Technical Summary: Effects of Thermal Boundary Conditions on Natural Convection and Entropy Generation in Non-Newtonian Power-Law Fluids

Problem Statement
This study investigates the coupled effects of thermal boundary conditions and fluid rheology on natural convection and thermodynamic irreversibility in non-Newtonian power-law fluids. The research focuses on two distinct two-dimensional geometries: a square cavity and a concentric cylindrical annulus. While natural convection in Newtonian fluids is well-documented, the interaction between spatially varying thermal boundary conditions (uniform vs. non-uniform sinusoidal heating) and the shear-dependent viscosity of power-law fluids remains an area requiring systematic analysis. Specifically, the work addresses the gap in literature regarding a unified framework that simultaneously evaluates flow structure, heat transfer performance, and entropy generation across shear-thinning ($n < 1$), Newtonian ($n = 1$), and shear-thickening ($n > 1$) regimes.

Methodology
The authors developed a robust numerical solver using the Gridap.jl finite element framework within the Julia programming language. The governing equations for steady, two-dimensional, incompressible laminar flow were formulated under the Boussinesq approximation, incorporating the power-law (Ostwald–de Waele) constitutive model.

• Numerical Scheme: The study employs the Galerkin finite element method (FEM) with Taylor–Hood elements (quadratic for velocity and temperature, linear for pressure) to ensure inf-sup stability.
• Solution Strategy: Due to the strong nonlinearity introduced by the power-law viscosity term, particularly at extreme power-law indices, the discrete nonlinear residual equations are solved using both the Newton–Raphson method and a trust-region method (via the NLsolve.jl package) to ensure convergence in ill-conditioned cases.
• Validation: The solver was rigorously validated against established benchmark solutions for both Newtonian and non-Newtonian natural convection in square and cylindrical geometries. Comparisons included isotherm fields, streamlines, local Nusselt number distributions, and entropy generation contours, demonstrating good agreement with literature data (e.g., Basak et al., Turan et al., Matin and Khan).
• Post-Processing: Key physical quantities were evaluated, including the stream function, heat function, local and average Nusselt numbers, and entropy generation rates. Entropy generation was decomposed into contributions from heat transfer ($S_\theta$) and fluid friction ($S_\psi$), with the average Bejan number ($Be_{av}$) used to quantify the dominance of irreversibility mechanisms.

Key Results
The simulations were conducted at a Prandtl number of $Pr = 100$ and a Rayleigh number of $Ra = 10^4$ for power-law indices $n = 0.6, 1.0,$ and $1.4$.

1. Rheological Influence:

   • Shear-thinning ($n=0.6$): Reduced apparent viscosity enhances buoyancy-driven circulation, leading to stronger vortices, steeper thermal gradients, and higher heat transfer rates. However, this also results in significantly elevated viscous dissipation.
   • Shear-thickening ($n=1.4$): Increased apparent viscosity suppresses fluid motion, shifting the transport mechanism toward conduction dominance, with weaker circulation and lower heat transfer rates.
   • Newtonian ($n=1.0$): Exhibits intermediate behavior, serving as a baseline for comparison.

2. Thermal Boundary Condition Effects:

   • Uniform Heating: Produces stronger, more spatially distributed convective structures. In the square cavity, this leads to symmetric circulation cells; in the annulus, it generates intense, asymmetric counter-rotating vortices.
   • Non-Uniform (Sinusoidal) Heating: Localizes thermal forcing to specific regions (e.g., the center of the cavity or the upper sector of the annulus). This localization consistently reduces the total entropy generation compared to uniform heating by confining high-gradient regions and reducing the overall intensity of fluid friction.

3. Entropy Generation and Irreversibility:

   • Dominance Mechanisms: In shear-thinning fluids, viscous dissipation is the primary source of irreversibility, particularly in the annular geometry where circulation is most vigorous. As the power-law index increases, the contribution of fluid-friction irreversibility drops sharply, and heat-transfer irreversibility becomes dominant.
   • Bejan Number Trends: The average Bejan number ($Be_{av}$) increases monotonically with $n$. For shear-thinning fluids under uniform heating, $Be_{av}$ is very low (e.g., $\approx 0.034$ in the annulus), indicating viscous dominance. As $n$ increases to 1.6, $Be_{av}$ approaches unity, confirming a transition to heat-transfer-dominated irreversibility.
   • Boundary Condition Impact: While thermal boundary conditions significantly alter the magnitude and spatial distribution of entropy generation (non-uniform heating reduces total entropy), the nature of the dominant irreversibility mechanism is primarily governed by the power-law index.

Significance and Claims
The paper claims that appropriate thermal boundary design, when considered alongside fluid rheology, provides an effective route for controlling heat transfer performance and minimizing thermodynamic losses in non-Newtonian convection systems. The study demonstrates that non-uniform sinusoidal heating can be utilized to localize thermal forcing and reduce total entropy generation without fundamentally altering the flow regime's rheological characteristics. Furthermore, the work establishes that the power-law index is the critical parameter determining whether a system is limited by viscous dissipation or heat transfer irreversibility. The authors conclude that their findings offer reliable guidelines for the optimization of energy-efficient thermal systems involving complex fluids, such as polymer processing and industrial cooling applications, by providing a comprehensive understanding of the interplay between boundary conditions and non-Newtonian rheology. The source code and metadata are made publicly available to ensure reproducibility.