Stability of liquid film coating a horizontal cylinder: interplay of capillary and gravity forces

The Gist

Imagine a horizontal metal pipe sitting in a puddle. Over time, a layer of thick liquid (like honey or oil) coats the outside of this pipe. Gravity wants to pull this liquid down, while surface tension (the "skin" of the liquid) wants to hold it together. This paper is a deep dive into what happens when these two forces fight it out, specifically when the liquid is very sticky (viscous) and moves slowly.

Here is a breakdown of the study's findings using simple analogies:

1. The Setup: The "Hanging Curtain"

When the liquid coats the pipe, it doesn't just slide off immediately. Instead, it gathers at the bottom, forming a shape that looks like a curtain hanging down from the pipe.

• The Tug-of-War: Gravity pulls this curtain down, trying to make it drip. Surface tension acts like a rubber band, trying to keep the curtain attached to the pipe.
• The Breaking Point: The researchers found that if the liquid layer is too thick or gravity is too strong compared to the liquid's "stickiness," the curtain snaps. The liquid accelerates downward and breaks off (drips). However, if the surface tension is strong enough, the curtain stays suspended, hanging there in a "quasi-stationary" state.

2. The Instability: Why the Curtain Wobbles

Even when the curtain is hanging safely, it is never perfectly still. It is constantly wiggling and trying to break up. The paper explains that this wiggling comes from two different "villains" depending on the situation:

• The "Beading" Villain (Surface Tension Dominated):
  • When: This happens when the liquid layer is thin or the surface tension is very strong (like water on a waxed car).
  • The Effect: Surface tension tries to minimize the surface area, just like a soap bubble wants to be a sphere. This causes the liquid to bunch up into pearls that wrap around the pipe. Think of it like a string of beads forming around the cylinder.
• The "Falling" Villain (Gravity Dominated):
  • When: This happens when the liquid layer is thick or gravity is very strong compared to surface tension.
  • The Effect: Gravity pulls the liquid down so hard that the surface tension can't hold it in a smooth shape. Instead of wrapping around, the liquid forms vertical fingers or ripples hanging underneath the pipe, ready to drip.

3. The "Energy" Explanation

The researchers didn't just watch the liquid; they calculated the energy budget to see who was winning the fight.

• The Balance Sheet: They tracked how much energy was being "spent" by gravity pulling the liquid down versus how much energy was being "saved" by surface tension holding it together.
• The Finding: They discovered that while gravity is always trying to break the curtain, surface tension can either help break it (by forming pearls) or help hold it together (by smoothing out the ripples), depending on the thickness of the film.

4. Predicting the Future: The "Regime Map"

The authors created a "cheat sheet" (a regime diagram) to predict what the final pattern will look like based on how thick the liquid is and how strong gravity is:

• The "Pearl" Zone: If the liquid is thin and sticky, it will form a ring of pearls around the pipe.
• The "Drop" Zone: If the liquid is thick or gravity is strong, it will form large drops hanging underneath.
• The "Snap" Zone: If the liquid is very thick, the curtain might break off entirely in a 2D sheet before it even gets a chance to form drops.

5. The "Time Travel" Check

One of the most interesting parts of the study was checking if the liquid changes its mind as it flows.

• The Question: Does the liquid start wobbling in one pattern, then change its mind and switch to a different pattern as it drains?
• The Answer: No. The researchers used a special analysis to look at the very early stages of the flow. They found that the "wavelength" (the distance between the bumps or pearls) is decided almost immediately. The liquid doesn't change its mind later; the pattern you see at the end is the same pattern that started growing right from the beginning. This confirms that their predictions based on the final "hanging curtain" shape are accurate for the whole process.

Summary

In short, this paper explains the physics of a liquid film draining off a horizontal pipe. It tells us that the liquid will either form pearls wrapping around the pipe (if surface tension wins) or dripping fingers underneath (if gravity wins). They mapped out exactly when each happens and proved that the pattern is set early in the process and doesn't change as the liquid drains.

Technical Summary

Technical Summary: Stability of Liquid Film Coating a Horizontal Cylinder

Problem Statement
This study investigates the drainage dynamics and stability of a viscous liquid film coating the exterior of a solid horizontal cylinder under the influence of gravity. The research focuses on the limit of large Ohnesorge numbers ($Oh$), where inertial effects are negligible compared to viscous forces. The primary objective is to understand the transition from an axially invariant draining flow to a quasi-stationary pendant liquid curtain, and subsequently, the linear stability of this curtain. The study aims to link the resulting pattern formation (such as "pearls" wrapping the cylinder or pendant drops underneath) to the interplay between capillary-driven (Rayleigh-Plateau-like) and gravity-driven (Rayleigh-Taylor) instabilities. Unlike previous studies restricted to thin-film limits using lubrication approximations, this work addresses arbitrary film thicknesses, acknowledging that the collected liquid bulk at the cylinder's bottom may violate thin-film assumptions.

Methodology
The authors employ a combination of non-linear transient simulations and linear stability analysis using the finite element software COMSOL Multiphysics.

1. Governing Equations: The flow is modeled using dimensionless mass and momentum conservation equations for an incompressible Newtonian fluid. The system is characterized by the Bond number ($Bo$, comparing gravity to surface tension) and the Ohnesorge number ($Oh$). The study operates in the inertialess regime where $(Bo/Oh)^2 \delta^4 \ll 1$.
2. Base Flow Simulation: The temporal evolution of the film is computed numerically starting from a uniform thickness $\delta$. The simulation tracks the drainage process until a quasi-static pendant curtain is formed or until the curtain ruptures (dripping).
3. Linear Stability Analysis: Once a saturated, quasi-static base state is established, a linear stability analysis is performed. The state vector is decomposed into the steady base flow and an infinitesimal time-dependent perturbation. The perturbation is assumed to be a normal mode with a longitudinal wavenumber $k$. This leads to a generalized eigenvalue problem to determine the growth rate $\sigma_r$ and the most amplified wavelength.
4. Energy Analysis: To quantify the physical mechanisms driving instability, an energy budget is constructed for both the base flow and the perturbations. This analysis partitions the energy exchange into viscous dissipation, surface energy changes (capillarity), and gravitational potential energy changes.
5. Transient Growth: A transient growth analysis is conducted to verify whether the initial flow evolution alters the most amplified wavelength before the system reaches the quasi-static state.

Key Results

• Drainage and Rupture Threshold: Non-linear simulations reveal a critical Bond number ($Bo_c$) for each film thickness $\delta$. Below this threshold, a quasi-stationary pendant curtain forms; above it, the liquid bulk ruptures and drips. The threshold scales inversely with film thickness ($Bo_c \propto \delta^{-1}$), consistent with a scaling argument balancing capillary force against bulk weight.
• Instability Modes: The quasi-static pendant curtain is unconditionally linearly unstable. The instability manifests as a single unstable mode with a peak growth rate at a specific wavenumber $k_{max}$.
  • Low $Bo$ (Capillary Dominance): When surface tension is strong relative to gravity, the instability is dominated by capillary effects. This promotes the formation of "pearls" that envelop the cylinder, driven by a Rayleigh-Plateau-like mechanism.
  • High $Bo$ (Gravity Dominance): As $Bo$ increases, gravity becomes the primary destabilizing force (Rayleigh-Taylor mechanism), while capillarity acts in a stabilizing manner. This leads to vertical finger-like modulations underneath the cylinder.
• Energy Repartition: The energy analysis confirms that while the Rayleigh-Taylor instability (gravity) is always present, the dominant mechanism shifts based on $Bo$. A regime diagram based on the energy ratio ($\Gamma = SUR_1/POT_1$) delineates regions of capillarity-dominated, gravity-dominated, and mixed instabilities.
• Pattern Prediction: By calculating the volume of liquid contained within the most amplified wavelength and comparing it to the maximum static volume a drop can sustain (based on prior non-linear simulations by Weidner et al.), the authors construct a tentative regime diagram. This diagram predicts whether the final state will be an array of saturated pearls, pendant drops, or a three-dimensional pinch-off (dripping).
• Transient Growth: The transient growth analysis demonstrates that the initial flow evolution does not significantly alter the most amplified wavelength. This validates the use of the asymptotic (quasi-static) analysis to predict the final three-dimensional patterns.

Significance and Claims
The paper claims to extend existing knowledge beyond the thin-film limit, providing a comprehensive stability analysis for films of arbitrary thickness.

• Validation: The linear analysis results align with pre-existing experimental data from de Bruyn (1997) and non-linear simulations from Weidner et al. (1997) in the thin-film limit.
• Mechanism Clarification: The study explicitly quantifies the transition between Rayleigh-Plateau and Rayleigh-Taylor instabilities in this geometry, showing that capillarity dominates at small $Bo$ (forming pearls) while gravity dominates at large $Bo$ (forming underside modulations).
• Predictive Capability: The proposed regime diagram offers a method to predict the final pattern (pearls vs. drops vs. dripping) based on the initial film thickness and Bond number, utilizing the volume available for droplet growth derived from the most linearly amplified wavelength.
• Methodological Contribution: The work rationalizes the use of asymptotic analysis for predicting 3D pattern fate by confirming via transient growth analysis that the initial transient phase does not shift the dominant wavelength selection.

The authors maintain a modest tone regarding the regime diagram, noting it is a "tentative" prediction based on linear theory and asymptotic analysis, acknowledging that non-linear effects (such as satellite drop formation) and transient growth could influence the final outcome, though their transient analysis suggests the latter does not alter the wavelength selection.