Rayleigh-Bénard thermal convection in emulsions: a short review

This paper reviews recent progress on Rayleigh-Bénard thermal convection in emulsions, highlighting how their complex, concentration-dependent rheology couples with buoyancy-driven flows to produce unique stability, transient, and morphological phenomena in soft materials.

The Gist

Imagine a pot of soup on the stove. Usually, when you heat the bottom, the hot liquid rises, cools at the top, and sinks back down, creating a swirling dance called convection. This is how heat moves through fluids like water or air.

Now, imagine that soup isn't just water; it's a thick, creamy emulsion—like mayonnaise or a very rich salad dressing. It's a mixture of tiny droplets of oil floating in water. This paper explores what happens when you heat this "thick soup" from the bottom.

Here is the story of their findings, broken down into simple concepts:

1. The Two Types of "Soup"

The researchers looked at two very different states of this mixture:

• The Thin Soup (Dilute): When there are only a few oil droplets, the mixture acts like normal water, just a bit thicker. It flows easily.
• The Thick Sludge (Concentrated): When the pot is packed with oil droplets, they bump into each other constantly. The mixture becomes a yield-stress fluid. Think of it like toothpaste or a crowd of people packed so tightly in a subway car that no one can move unless you push really hard. It acts like a solid until a certain force breaks it loose.

2. The "Sticky" Problem (Stabilization)

In real life, oil and water hate each other. If you mix them without a "glue" (called a surfactant), the oil droplets will eventually merge into one giant blob.

• Without the glue: The mixture changes its identity. If you start with mostly water and a little oil, it stays that way. But if you start with mostly oil and a little water, the water gets squeezed out, and the whole thing flips inside out (becoming water-in-oil).
• With the glue: The droplets stay separate, like individual bubbles in a foam. This allows the mixture to keep its complex, "jammed" structure even when heated.

3. The Surprise: The "Sleeping Giant"

When they heated the thick, jammed emulsion (the toothpaste-like one), something weird happened that doesn't happen with normal water.

Instead of a steady, smooth flow, the heat transfer went into a stop-and-go mode:

• The Nap: For a long time, the mixture sits still. The heat can't get through because the "toothpaste" is too stiff to move. It's like a sleeping giant.
• The Burst: Suddenly, a tiny bit of the structure breaks loose. This creates a "heat burst"—a sudden, violent swirl of hot liquid shooting upward.
• The Cycle: After the burst, the mixture settles back down, but the structure is slightly weaker now. Sooner or later, it bursts again.

The researchers call this intermittency. It's like a heartbeat that skips beats, then thumps hard, then skips again. This happens because the "jammed" droplets are constantly rearranging themselves, storing up energy until they snap free.

4. The Great Flip (Phase Inversion)

If you keep heating the thick mixture hard enough, something dramatic happens: The mixture flips inside out.
Imagine a room full of people (water) holding hands, with a few balloons (oil) floating around. If you push hard enough, the balloons might push the people apart so much that the balloons end up holding hands, and the people become the floating bits.
In the experiment, the "oil-in-water" mixture suddenly became a "water-in-oil" mixture. Once this happens, the mixture behaves like a normal, thin fluid again, and the heat flows smoothly. You can't easily reverse this; it's a one-way trip.

5. Why Computers Were Needed

You might wonder, "Why didn't they just do this in a lab?"

• The Problem: Thick emulsions are opaque (you can't see through them). Trying to watch tiny droplets move inside a thick, cloudy mixture is like trying to watch individual ants in a jar of mud.
• The Solution: The team used a super-powerful computer simulation. They built a "digital pot" where they could track every single oil droplet, watching how they bumped, broke apart, and merged. This allowed them to see the invisible "dance" of the droplets that causes the heat bursts.

The Big Picture

This research tells us that soft, complex materials (like emulsions, foams, or even magma in the Earth's crust) don't just flow when heated; they have a personality. They can get stuck, store energy, and then release it in sudden, unpredictable bursts.

Understanding this helps us predict:

• How lava flows (which is often a mix of rock and gas).
• How to design better industrial mixers.
• How heat moves through the Earth's mantle.

In short: Heating a thick emulsion isn't just about boiling water; it's about waking up a sleeping giant that moves in fits and starts before finally flipping its entire identity.

Technical Summary

1. Problem Statement and Motivation

The paper addresses the fundamental physical understanding of thermally driven emulsions, a system relevant to both industrial applications and geophysical phenomena (e.g., lava flows, mantle convection). While emulsion rheology under isothermal conditions is well-studied, their behavior under thermal forcing (specifically Rayleigh-Bénard convection) remains largely unexplored.

Key challenges identified include:

• Complex Rheology: Emulsions exhibit concentration-dependent rheology, transitioning from Newtonian (dilute) to shear-thinning and yield-stress (concentrated/jammed) behaviors as the volume fraction ($\phi$) of the dispersed phase increases.
• Multiscale Coupling: The problem involves a complex interplay between large-scale buoyancy-driven flows and microscopic interfacial dynamics (droplet breakup, coalescence, and rearrangement).
• Limitations of Continuum Models: Traditional hydrodynamic equations with prescribed constitutive laws often fail to capture the discrete nature of droplets, finite-size effects, and the resulting plastic rearrangements that drive intermittent dynamics.
• Experimental Difficulties: Direct experimental observation is hindered by the optical opacity of dense dispersions and the difficulty of tracking breakup/coalescence dynamics.

2. Methodology

The authors rely on numerical simulations using TLBfind, an open-source 2D multicomponent Lattice Boltzmann Method (LBM) code.

• Physical Model:
  • Two fluid components ("oil" and "water") are represented by kinetic distribution functions.
  • Phase Separation: Intra-component interactions create diffuse interfaces.
  • Stabilization: An additional interaction generates a positive disjoining pressure, mimicking surfactants to prevent full phase separation and stabilize droplets against coalescence.
  • Geometry: A 2D Rayleigh-Bénard cell of height $H$ and length $L$, heated from below ($T_{hot}$) and cooled from above ($T_{cold}$) under gravity $g$.
• Key Parameters:
  • Volume Fraction ($\phi$): Ranging from dilute ($\phi \approx 0.16$) to jammed ($\phi \approx 0.84$).
  • Rayleigh Number ($Ra$): Defined based on the continuous phase properties to measure buoyancy forcing.
  • Stabilization: Comparing stabilized (proper emulsions) vs. non-stabilized (dispersions prone to coalescence/inversion) systems.
• Analysis Metrics:
  • Nusselt Number ($Nu$): Global heat transfer efficiency.
  • Droplet-scale Nusselt Number ($Nu_{drop}$): Local heat flux fluctuations associated with individual droplets.
  • Displacement Fluctuations ($\tilde{\delta}d$): Spatio-temporal correlations of droplet motion.

3. Key Contributions and Results

A. Impact of Interfacial Stabilization

• Dilute Regime ($\phi \lesssim 0.5$): Both stabilized and non-stabilized systems exhibit similar heat transfer behavior ($Nu$ vs. $Ra$). The emulsion behaves as a Newtonian fluid with effective viscosity.
• Concentrated Regime ($\phi > 0.5$):
  • Non-stabilized: The system undergoes phase inversion. The continuous water phase becomes the dispersed phase (water-in-oil), effectively behaving as a complementary oil-in-water emulsion with volume fraction $1-\phi$.
  • Stabilized: The system maintains its oil-in-water structure but undergoes a sharp transition from a conductive state ($Nu=1$) to a convective state ($Nu>1$) as $Ra$ increases. The stabilization mechanism prevents the catastrophic phase inversion seen in non-stabilized dispersions.

B. Role of Finite-Sized Droplets

• Heat Flux Fluctuations: In stabilized emulsions, the suppression of coalescence leads to frequent droplet collisions. This generates non-Gaussian tails in the probability density functions (PDFs) of local heat flux fluctuations ($Nu_{drop}$).
• Mechanism: These fluctuations are "bursts" of heat transport caused by local fluidization and droplet-droplet interactions.
• Spatial Correlations: As $\phi$ increases, spatial correlations among droplets strengthen. In jammed systems, displacement fluctuations become coherent over extended regions near boundaries, whereas they are uncorrelated in dilute systems.

C. Intermittency and Phase Inversion in Jammed Emulsions

• Intermittent Convection: For highly concentrated, yield-stress emulsions ($\phi \approx 0.79$), the system does not transition smoothly to turbulence. Instead, it exhibits intermittent dynamics:
  • Resting Periods: The emulsion is mostly conductive ($Nu \approx 1$), but local plastic rearrangements occur, relaxing stored elastic energy.
  • Convective Bursts: These rearrangements trigger sudden, intense convective bursts ($Nu \gg 1$) where the material temporarily fluidizes, forming convective rolls.
• Irreversible Phase Inversion: During convective bursts, strong velocity gradients promote droplet coalescence. This reduces the interfacial area and can lead to partial phase inversion (islands of the original emulsion embedded in a dilute, inverted matrix). Once this occurs, the system cannot return to its original state upon reducing $Ra$, leading to hysteresis in the $Nu$ vs. $Ra$ curve.
• Model Limitation: Continuum models with local constitutive laws fail to capture this intermittency because they lack the discrete "granularity" required to model plastic rearrangements that spontaneously restart convection.

4. Significance and Perspectives

• Novel Phenomenology: The paper establishes that the coupling between soft matter rheology (yield stress, plasticity) and thermal convection creates a rich dynamical landscape distinct from standard Newtonian or simple multiphase flows.
• Methodological Validation: It demonstrates that mesoscale methods (LBM with resolved droplets) are essential for capturing the interplay between interface dynamics and large-scale convection, which continuum models miss.
• Experimental Guidance: The authors propose specific experimental parameters (e.g., silicone oil emulsions with low yield stress, specific $Ra$ and $\Delta T$) to observe these intermittent regimes, noting that such controlled experiments are currently missing in literature.
• Future Work: The authors acknowledge current limitations to 2D simulations and identify 3D simulations as a necessary next step to fully capture the complexity of these flows, despite the computational cost.

In summary, this review highlights how finite-size effects and interfacial stabilization fundamentally alter thermal transport in emulsions, leading to unique phenomena like intermittent convection and dynamically induced phase inversion, which are critical for understanding complex soft materials in geophysical and industrial contexts.