Coalescence of viscous blisters under an elastic sheet

This paper investigates the coalescence of identical viscous blisters beneath an elastic sheet through experiments, numerical simulations, and a one-dimensional lubrication model, revealing that short-time dynamics are primarily governed by the elastic sheet's bending, which establishes a specific relationship between coalescence speed and the interface's radius of curvature.

The Gist

The Big Picture: Two Bubbles Merging Under a Blanket

Imagine you have a thick, stretchy rubber sheet (like a balloon skin) laid flat over a table. Underneath that sheet, you inject some thick honey (viscous fluid) through two small holes. The honey pushes the rubber sheet up, creating two separate, round "blisters" or domes.

Eventually, these two domes grow big enough that they touch. When they do, they merge into one big puddle. This paper is all about what happens in that split second when they touch and start to merge.

The scientists wanted to know: How fast does the bridge between them form? What shape does it take? And what forces are driving this?

---

The Experiment: Watching the Invisible

To study this, the researchers built a special setup:

1. The Fluid: They used sunflower oil (thick and sticky).
2. The Skin: They used a thin sheet of PDMS (a soft, rubbery material).
3. The Trick: They couldn't just look at the oil because it's hidden under the rubber. So, they placed a checkerboard pattern underneath the whole setup.

As the oil pushed the rubber up, the checkerboard pattern looked distorted, like a funhouse mirror. By taking high-speed photos of these distortions, they could use a computer to "reverse engineer" exactly how thick the oil was at every single point. It's like looking at the ripples on a pond to figure out how deep the water is without touching it.

The Discovery: The "Neck" Grows Fast

When the two blisters touch, a tiny bridge (or "neck") of oil forms between them. The scientists measured how fast this bridge grew upward.

They found a surprising rule: The speed at which the bridge grows depends entirely on how curved the rubber sheet is right at the point of contact.

• The Analogy: Imagine bending a ruler. If you bend it into a tight, sharp curve, it snaps back with a lot of force. If you bend it gently, the force is weak.
• The Finding: The rubber sheet acts like a spring. Where the two blisters meet, the sheet is curved. The tighter the curve (the smaller the radius), the harder the sheet pushes back, and the faster the oil bridge shoots upward.

They discovered a mathematical "recipe" (a scaling law) that predicts the speed of the merge based on this curvature. It's like having a formula that says, "If the bend is this sharp, the bridge will rise at this specific speed."

The Theory: The "Bending" vs. The "Stretching"

The researchers built a computer model to explain why this happens. They had to decide what was more important:

1. Stretching: Like pulling a rubber band tight.
2. Bending: Like bending a stiff piece of cardboard.

They found that for their specific experiment, bending was the boss. The rubber sheet wasn't stretching much; it was just bending like a stiff ruler. Because the sheet is bending, it creates a pressure that squeezes the oil, forcing the two blisters to merge quickly.

The Two Phases of the Merge

The paper describes two distinct stages of this event:

1. The "Snap" Phase (Short Time):
   Immediately after they touch, the bridge grows exponentially fast. It's like a spring being released. The rubber sheet is so curved that it pushes the oil up rapidly. The computer model showed this matches their "bending" theory perfectly.

2. The "Settle" Phase (Long Time):
   After the initial rush, the two blisters have merged into one big blob. Now, the rubber sheet is no longer curved sharply; it's flattening out. The system slows down and relaxes, like a heavy blanket settling onto a mattress. The oil spreads out until everything is flat and calm.

Why Does This Matter?

You might wonder, "Who cares about oil under a rubber sheet?"

Actually, this happens everywhere in nature and industry:

• Geology: Deep underground, magma (molten rock) pushes up the Earth's crust, creating "laccoliths" (dome-shaped mountains). When two magma pockets merge, this physics applies.
• Medicine: Skin blisters (like those from friction or disease) form under the top layer of skin. Understanding how they merge helps doctors understand skin mechanics.
• Printing: Inkjet printers shoot tiny droplets. Understanding how they merge helps make better printers.

The Bottom Line

The scientists took a complex physics problem—how two sticky blobs merge under a flexible skin—and simplified it. They proved that the shape of the skin (its curvature) dictates the speed of the merge.

They used a mix of real-world experiments (watching oil under rubber), clever math (predicting the speed), and computer simulations (watching it happen in a virtual world) to show that nature follows a very specific, predictable pattern when these "elastic blisters" decide to become one.

Technical Summary

1. Problem Statement

The study investigates the coalescence dynamics of two identical viscous fluid blisters situated beneath a thin, flexible elastic sheet. While the spreading of viscous fluids and the coalescence of free droplets are well-studied, the interaction between fluid flow and the mechanics of a covering elastic membrane (specifically in the context of "blisters") presents unique challenges.

• Physical Context: The system involves a viscous fluid (sunflower oil) injected between a rigid substrate and an elastic membrane (PDMS). As the fluid spreads, it forms blisters. When two such blisters merge, the dynamics are governed by the interplay between viscous forces within the fluid and the bending elasticity of the overlying sheet.
• Specific Focus: The authors focus on the onset of coalescence, specifically the formation and growth of the liquid bridge (neck) connecting the two blisters. They aim to understand how the local geometry (radius of curvature) and the elastic properties of the sheet dictate the speed of this merging process.
• Gap in Literature: Previous work (specifically Sæter et al., 2024) addressed this problem but assumed a constant radius of curvature at the coalescence point. This study revisits the problem to test the validity of that assumption and explore the short-time dynamics where the curvature may evolve.

2. Methodology

The authors employed a multi-faceted approach combining high-resolution experiments, scaling theory, and numerical simulations.

A. Experimental Setup

• Configuration: Sunflower oil ($\mu \approx 5 \times 10^{-2}$ Pa·s) was injected at a constant flow rate between a PMMA plate and a PDMS elastic sheet (Young's modulus $E \approx 1.6$ MPa).
• Geometry: Two blisters were formed separated by a gap ($L_{gap}$) ranging from 1.5 to 6 cm. The elastic sheet thickness ($d$) was varied (0.2, 0.8, 1.6 mm) to control the bending stiffness ($B$).
• Measurement Technique: A time-resolved synthetic schlieren technique was used. A checkerboard pattern placed beneath the substrate was imaged through the fluid and elastic sheet. Using a Fast Checkerboard Demodulation (FCD) algorithm, the apparent deformations of the pattern were converted into a 3D reconstruction of the liquid thickness field $h(x, y, t)$ with a vertical resolution of $\sim 10$ $\mu$m.
• Key Metrics: The team tracked the height of the coalescence point ($h_0$) and the local radius of curvature ($R$) at the neck during the initial merging phase.

B. Theoretical Modeling (Scaling Theory)

• Governing Equation: The fluid dynamics were modeled using the lubrication approximation coupled with the bending equation of the elastic sheet. The thin-film equation is:
  $$\frac{\partial h}{\partial t} = \frac{1}{12\mu} \nabla \cdot \left( h^3 \nabla (\rho g h + B \nabla^4 h) \right)$$
• Simplification: In the coalescence region, the bending term ($B \nabla^4 h$) was found to dominate over gravity and tension terms. The problem was reduced to a 1D approximation.
• Self-Similar Ansatz: The authors sought a self-similar solution where the curvature at the coalescence point is allowed to evolve (unlike previous models that assumed it was constant). They derived a relationship between the vertical velocity of the neck and the local radius of curvature.

C. Numerical Simulation

• The full non-linear 1D equation was solved numerically using a semi-implicit finite difference scheme.
• An idealized polynomial initial condition was used to simulate the merging process, allowing for a quantitative comparison with the scaling laws derived from theory.

3. Key Results

A. Experimental Observations

• Two-Phase Evolution: The height of the coalescence point ($h_0$) exhibits two distinct phases: a rapid initial rise (exponential growth) followed by a stabilization plateau (reaching $\sim 60\%$ of the initial blister height).
• Scaling Law: The experiments revealed a clear power-law relationship between the coalescence speed ($dh_0/dt$) and the radius of curvature ($R$).
• Bending Dominance: When the coalescence speed was normalized by the bending modulus ($B$), data from different sheet thicknesses collapsed onto a single master curve, confirming that bending elasticity is the primary driver of the dynamics.

B. Theoretical and Numerical Findings

• Velocity-Curvature Relationship: The derived scaling law predicts that the coalescence speed is proportional to the bending modulus and inversely proportional to the cube of the radius of curvature:
  $$\frac{dh_0}{dt} \propto \frac{B h_0}{\mu R^3}$$
  The experiments confirmed this with a prefactor $\alpha \approx 1$.
• Short-Time Dynamics: The numerical simulations confirmed that at short times, the system follows a self-similar solution where the curvature remains approximately constant ($R \approx R_0$). This validates the exponential growth phase observed in experiments.
• Long-Time Dynamics: As the system evolves, the curvature changes, and the dynamics transition from exponential growth to an exponential relaxation toward a flat membrane state. The authors successfully "patched" the local self-similar solution with a static outer solution, showing that the dynamics are localized to the coalescence region while the outer membrane shape remains quasi-static.

4. Key Contributions

1. Refinement of Coalescence Theory: The study challenges the assumption of constant curvature in previous models. It demonstrates that while curvature is effectively constant during the initial short-time exponential growth, the theoretical framework must allow for its evolution to describe the full transition to equilibrium.
2. Experimental Validation: Provided the first high-resolution, time-resolved measurements of the thickness field and curvature evolution during the coalescence of elastic blisters, validating the $1/R^3$ scaling law.
3. Unified Model: Developed a composite model that links the local self-similar dynamics of the neck to the global static shape of the membrane, explaining how the system relaxes from the coalescence event to a flat state.
4. Methodological Advancement: Demonstrated the efficacy of synthetic schlieren imaging for quantifying thin-film dynamics under elastic constraints with high spatial and temporal resolution.

5. Significance and Future Directions

• Fundamental Physics: This work deepens the understanding of fluid-structure interaction in soft matter systems, specifically the coupling of lubrication flow and elastic bending.
• Applications: The findings are relevant to various fields:
  • Geophysics: Modeling the coalescence of magma intrusions (laccoliths) and lava flows.
  • Biology/Medicine: Understanding the formation and merging of dermo-epidermal blisters.
  • Industrial: Optimizing inkjet printing and coating processes involving viscoelastic skins.
• Future Work: The authors suggest extending the model to full 2D simulations to capture transverse effects and including membrane tension (which becomes significant for larger deformations). Incorporating tension would allow the study of wrinkling instabilities and orthoradial buckling, which are currently neglected in the pure bending regime. Additionally, investigating the role of adhesion at the blister periphery is proposed as a next step.