Stability analysis of the flow in a coflowing device

This study demonstrates that in coflowing devices, jet instability precedes meniscus destabilization and that transient breakup is highly sensitive to initial perturbations, thereby challenging the validity of linear stability analysis for predicting polydisperse dripping in such configurations.

The Gist

Imagine you are trying to create a perfect, steady stream of water from a hose to water your garden. You want the water to flow smoothly for a long distance before breaking into a single, uniform spray of droplets. This is what scientists call "jetting." However, sometimes the water starts dripping right at the nozzle, creating a messy, uneven spray. This is called "dripping."

This paper is about a specific type of "hose" setup called a coflowing device. Think of it like a garden hose inside a larger pipe. A fast-moving outer stream of liquid pushes against a slower inner stream, stretching it out into a thin, tapered cone (like a teardrop shape) before it shoots out as a jet.

The researchers wanted to understand exactly when this smooth stream turns into a messy drip. They used two main tools:

1. Experiments: Watching real liquid flow in a lab.
2. Computer Simulations: Using math to predict how the liquid behaves.

Here is the simple breakdown of what they found and why it matters:

1. The "Crystal Ball" That Failed

Scientists often use a method called Global Linear Stability Analysis to predict when a smooth flow will turn into a drip. You can think of this method as a "crystal ball" that looks at the steady stream and asks, "If I poke this stream slightly, will it recover or fall apart?"

Usually, this crystal ball works well. It predicts that if the stream is unstable, the "teardrop" cone at the tip will start wobbling and break apart.

But in this specific setup, the crystal ball was wrong.
The computer model (the crystal ball) said the stream was stable and the cone was perfectly still. However, the real experiment showed that the stream was actually breaking apart and dripping. The model failed to see the problem because it was looking at the wrong thing. It assumed the "teardrop" cone was the weak point, but in reality, the cone was fine; the thin stream coming out of it was the problem.

2. The "Ghost Waves" and the Short-Term Explosion

Why did the model fail? The paper explains that the stream is like a musical instrument with many hidden notes (called eigenmodes).

• The Old Theory: Scientists thought that if the stream was unstable, one specific "loud note" (an unstable eigenmode) would start growing louder and louder until the stream broke.
• The New Discovery: The researchers found that in this device, all the "notes" are actually trying to get quieter (they are decaying). However, for a short moment, these quieting notes can interfere with each other in a way that creates a temporary, massive spike in energy.

The Analogy: Imagine a group of people in a room, all trying to leave quietly. If they all bump into each other at the exact same time, they might create a chaotic, loud pile-up for a split second before they finally exit. The "crystal ball" model only looks at the long-term result (everyone leaving quietly) and misses the short-term chaos (the pile-up).

This short-term chaos is what causes the stream to snap and break into droplets, even though the math says the stream should be stable.

3. The "Push" Matters

The researchers also found that how you disturb the stream matters.

• If you poke the stream right at the tip of the cone, it might not break.
• If you poke it a little further down the stream, it breaks much faster.

This means the length of the stream before it breaks isn't a fixed number written in the laws of physics for that specific setup. It depends entirely on where the initial "nudge" happens. It's like pushing a swing: if you push it at the right moment, it goes high; if you push it at the wrong moment, it barely moves.

4. The Real-World Observation

In their experiments, the researchers watched what happened as they slowed down the inner liquid flow:

• High Flow: A long, steady stream forms and breaks into uniform droplets far away.
• Medium Flow: The stream gets shorter and breaks closer to the tip, but the droplets are still mostly uniform.
• Low Flow: The stream breaks almost immediately, creating a messy spray of different-sized droplets.

The computer model predicted that the transition from "steady stream" to "messy spray" would happen because the cone at the tip started wobbling. But the experiment showed the cone stayed perfectly still the whole time! The instability happened in the stream after it left the cone.

The Bottom Line

This paper tells us that for this specific type of fluid device, the standard mathematical tools used to predict stability are not reliable. They miss the "short-term chaos" caused by the interference of different fluid waves.

Instead of looking for a single "unstable note" that grows forever, we have to understand how a bunch of "quieting notes" can crash into each other to cause a sudden break. This changes how scientists need to think about designing these micro-fluidic devices, as the old rules don't apply here.

Technical Summary

Technical Summary: Stability Analysis of the Flow in a Coflowing Device

Problem Statement
The paper investigates the hydrodynamic stability of a coflowing configuration, a common microfluidic device used for tip streaming where an inner liquid is injected through a capillary surrounded by a coaxial outer stream. While global linear stability analysis has been successfully applied to various capillary flows (e.g., gravitational jets, flow focusing) to predict transitions between jetting and dripping modes, previous attempts have failed to accurately predict jet breakup lengths or the specific conditions for the transition to polydisperse dripping in coflowing devices. The central problem addressed is the discrepancy between the predictions of global linear stability analysis and experimental observations regarding the stability of the steady jetting mode and the onset of instability.

Methodology
The authors employ a combined approach of experimental observation, global linear stability analysis, and transient numerical simulations.

• Governing Equations: The flow is modeled using the dimensionless Navier-Stokes equations for axisymmetric, incompressible, immiscible fluids. The system is characterized by the radius ratio $R$, density ratio $\rho$, viscosity ratio $\mu$, Ohnesorge number $Oh_i$, capillary number $Ca$, and flow rate ratio $Q$.
• Global Linear Stability Analysis: The authors assume a perfectly steady base flow and solve for the eigenvalues of the linearized Navier-Stokes operator. They analyze the spectrum to determine if perturbations grow (unstable) or decay (stable) asymptotically.
• Transient Simulations: To study the short-term response of the base flow, the authors perform direct numerical simulations of the linearized equations. These simulations introduce an initial small-amplitude perturbation to the free surface and track the evolution of the interface. Crucially, these simulations account for the superposition of multiple eigenmodes, rather than isolating a single dominant mode.
• Experimental Validation: Experiments were conducted using a coflowing device with fixed geometric and fluid properties ($R=6.67$, $\rho=0.962$, $\mu=36.1$, $Oh_i=0.0264$). The transition from jetting to dripping was observed by varying the flow rate ratio $Q$.

Key Results

1. Failure of Global Stability Analysis for Jetting-to-Dripping Transition: The experiments reveal that as the inner flow rate decreases, the flow does not transition directly from a steady jet to polydisperse dripping via the destabilization of the meniscus. Instead, the flow passes through an intermediate state of monodisperse microdripping. In this state, the tapering meniscus remains stable and steady, while the emitted jet breaks up due to convective instability. Consequently, the global stability analysis, which assumes a steady jetting base flow, fails to predict the transition to polydisperse dripping because the actual marginally stable state (monodisperse microdripping) is fundamentally different from the assumed steady jet.
2. Convective Instability and Non-Modal Growth: In the jetting regime, the flow is convectively unstable. The global stability analysis predicts that the interface remains unaltered near the meniscus and that instability arises from spurious eigenmodes dependent on the domain length. However, transient simulations show that the jet breaks up due to the constructive interference (linear superposition) of decaying eigenmodes triggered by an initial perturbation. This "non-modal" growth allows the perturbation energy to increase significantly in the short term, leading to jet breakup before the modes are damped.
3. Dependence on Initial Perturbation: The jet breakup length is not an intrinsic property of the base flow. The simulations demonstrate that the location of the breakup is highly sensitive to the initial position and nature of the perturbation. Perturbations introduced closer to the meniscus-jet transition region lead to earlier breakup. This explains why experimental breakup lengths are often shorter than those predicted by classical temporal stability analysis, which assumes a specific dominant mode triggered at the jet inception.
4. Nonlinear Effects: Nonlinear effects become significant only near the instant of interface breakup, driving the formation of a neck and subsequent pinch-off. However, the initial growth leading to breakup is adequately described by the linear superposition of modes.

Significance and Conclusions
The paper concludes that the classical global linear stability analysis is insufficient for describing the stability of coflowing configurations and potentially other similar setups. The primary limitation is that the analysis relies on the assumption of a steady base flow, which does not exist at the critical transition point in coflowing devices; the flow actually transitions through a monodisperse microdripping state. Furthermore, the convective instability leading to jet breakup is driven by the non-normal nature of the linearized operator, where the superposition of decaying modes causes transient growth.

The authors assert that their findings question the validity of applying standard linear stability analysis to coflowing devices. Specifically, criteria such as Briggs' pinch condition for absolute instability may be necessary but insufficient for determining stability in these systems. The results suggest that the jet breakup length cannot be predicted solely by the growth rate of a dominant temporal mode, as the process is governed by the interference of multiple decaying modes and is dependent on the initial perturbation. This work highlights the need to consider transient, non-modal growth mechanisms when analyzing stability in microfluidic coflowing configurations.