Cahn-Hilliard Phase Field modelling captures nanoscale contact line dynamics on high-friction surfaces

This study demonstrates that a Cahn-Hilliard Phase Field model, when systematically calibrated using Molecular Dynamics data to capture contact line friction and contact angle dynamics, can quantitatively reproduce nanoscale wetting behaviors on high-friction surfaces, thereby bridging the gap between molecular processes and continuum hydrodynamics.

The Gist

Imagine you are watching a drop of water slide across a windowpane. To the naked eye, it looks smooth. But if you could shrink down to the size of a molecule, you'd see a chaotic, jittery dance where water molecules bump into the glass and each other.

For a long time, scientists have tried to write computer programs to predict exactly how these drops move. They have two main tools:

1. Molecular Dynamics (MD): This is like a high-speed, ultra-microscopic camera. It tracks every single molecule. It's incredibly accurate but requires a supercomputer and takes forever to run.
2. Phase Field Models (CHNS): This is like a smooth, continuous video. It treats the liquid as a fluid blob rather than individual particles. It's fast and easy to run, but it often misses the tiny, messy details happening right where the liquid touches the solid surface (the "contact line").

The Problem: The "Sticky" Edge
When a drop moves, the edge where it touches the surface is the most important part. In the real world (and in the microscopic camera), this edge gets "stuck" or experiences friction. The smooth video models usually struggle here because they assume the liquid slides perfectly or slips in a way that doesn't match reality. They often get the shape of the drop wrong because they can't account for this microscopic "stickiness."

The Solution: A Hybrid Approach
The authors of this paper wanted to fix the smooth video model so it acts exactly like the microscopic camera, but without needing to track every single molecule. They did this by creating a calibration protocol.

Think of it like tuning a musical instrument. The smooth model is the instrument, and the microscopic simulation is the perfect pitch.

1. The Setup: They simulated water and hexane (a type of oil) sliding past each other between two moving walls, like a sandwich being squeezed and slid.
2. The Calibration: They ran the slow, detailed microscopic simulation first. They measured exactly how much the "edge" of the water resisted moving (the contact line friction) and how the surface bent.
3. The Fix: They fed these specific "friction numbers" into the smooth video model. They didn't just guess; they adjusted the model's "friction dial" until the smooth model's edge behaved exactly like the microscopic one.

The Results: A Perfect Match
Once they tuned that one specific "friction dial," the smooth model became incredibly accurate. It could now predict:

• How the drop bends: The curve of the water surface near the wall.
• How far the drop moves: The steady position of the contact line.
• How the water flows: The swirling patterns inside the liquid.

The paper claims that by simply matching the contact line friction (how much the edge resists moving) to the microscopic data, the smooth model can reproduce the complex, messy physics of the real world.

The Catch (The "Slip" Secret)
There is one tiny detail the smooth model still misses. In the microscopic world, the very edge of the contact line actually "slips" a tiny bit more than the rest of the liquid does. The smooth model, even when perfectly tuned, doesn't naturally include this extra slip. The authors suggest that while their method is a huge improvement, future models might need to add a specific rule to account for this extra "slippery edge" to be 100% perfect.

In Summary
This paper is about teaching a simplified, fast computer model to act like a complex, slow one. They found that if you just tell the fast model exactly how "sticky" the edge of the drop is (based on real molecular data), it can accurately predict how the drop moves, bends, and flows, bridging the gap between the microscopic world of atoms and the macroscopic world of fluids.

Technical Summary

1. Problem Statement

Accurate mesoscale modeling of wetting processes is critical for applications ranging from additive manufacturing to energy production. However, describing the motion of moving contact lines (where fluid-fluid interfaces meet a solid surface) remains a significant challenge in computational fluid dynamics (CFD).

• The Singularity Problem: Classical hydrodynamic models (Navier-Stokes) with no-slip boundary conditions predict a non-integrable stress singularity at the contact line.
• The Slip Dilemma: While introducing Navier slip resolves the singularity, experimental and molecular evidence suggests slip lengths on flat, hydrophilic surfaces are often nanometric (or effectively zero). Using unphysically large slip lengths for numerical feasibility compromises physical accuracy.
• The Gap: There is a lack of consensus on how to reconcile macroscopic continuum models with molecular-scale physics, particularly regarding contact line friction and energy dissipation mechanisms on high-friction surfaces.

2. Methodology

The authors employ a multiscale approach, coupling Molecular Dynamics (MD) simulations with a Phase Field (PF) model based on Cahn-Hilliard Navier-Stokes (CHNS) equations.

A. Molecular Dynamics (MD) Simulations

• System: A water/hexane biphasic system in a two-phase Couette flow configuration confined between silica-like walls.
• Setup: The system uses SPC/E water and OPLS-AA hexane force fields. The solid wall is modeled as a silica-like surface with tunable wettability (hydrophilic and hydrophobic cases).
• Goal: To generate "ground truth" data for interface curvature, dynamic contact angles, and flow profiles under various wall velocities ($u_w \in [1.12, 4.84]$ m/s).
• Observables: The MD simulations allow for the direct measurement of contact line friction coefficients and the extraction of transport properties (viscosity, surface tension) without empirical fitting.

B. Phase Field (CHNS) Modeling

• Governing Equations: The model solves the coupled Cahn-Hilliard equation (for phase evolution) and Navier-Stokes equations (for fluid momentum).
• Key Features:
  • Diffuse Interface: The interface has a finite thickness ($\epsilon$), allowing for a local diffusive slip mechanism that regularizes the contact line singularity without requiring macroscopic Navier slip.
  • Boundary Conditions: Includes a wall-energy relaxation condition to govern contact angle dynamics and a sub-nanometric Navier slip length ($\ell_N$) strictly for numerical stability.
• Calibration Protocol: The authors establish a systematic protocol to parametrize the CHNS model directly from MD data:
  1. Interface Thickness ($\epsilon$): Set to 0.3 nm (matching MD intrinsic width).
  2. Mobility ($M$): Constrained by the Sharp Interface Limit (SIL) condition ($\epsilon \leq 4\sqrt{M\eta^*}$) to minimize unphysical streamline crossing.
  3. Contact Line Friction ($\mu_f$): The only parameter empirically calibrated. It is extracted by matching the MD-derived relationship between contact line speed and dynamic contact angle.
  4. Viscosity: Measured independently from MD.

3. Key Contributions

1. Systematic Calibration Protocol: The paper demonstrates that a Phase Field model can quantitatively reproduce MD results with minimal parameter fitting. Crucially, it identifies contact line friction as the sole physical parameter requiring empirical calibration against MD data. All other parameters (mobility, interface thickness) are determined by numerical constraints (sharp interface limit) or independent MD measurements.
2. Quantitative Reproduction of Dissipation: The study successfully separates and captures the two primary channels of energy dissipation in wetting:
   • Viscous dissipation: Captured by the Navier-Stokes equations (bending of the interface).
   • Contact line friction: Captured by the calibrated friction coefficient (tilting of the contact angle).
3. Resolution of Streamline Crossing: By adhering strictly to the sharp interface limit condition for the mobility parameter $M$, the authors minimize the unphysical crossing of streamlines through the liquid-liquid interface, a common artifact in Phase Field simulations.
4. Identification of Model Limitations: The work highlights a discrepancy between the calibrated CHNS friction and MD-measured friction, attributing it to localized slip at the contact line that is not fully captured by standard Phase Field diffusion alone.

4. Key Results

• Interface Curvature: The CHNS model quantitatively reproduces the interface curvature profiles ($\theta_I(z)$) observed in MD for both advancing and receding contact lines across a wide range of velocities.
• Contact Line Displacement:
  • For hydrophobic surfaces ($\theta_0 \approx 97.3^\circ$), the model shows excellent agreement with MD steady-state displacement.
  • For hydrophilic surfaces ($\theta_0 \approx 80.9^\circ$), the model underestimates displacement at high velocities. This is attributed to the omission of the diffusive flux coupling term in the momentum equation, which becomes significant at high speeds.
• Flow Structure: The velocity fields and streamline patterns (including recirculation zones in the low-viscosity phase) show qualitative and quantitative agreement with MD.
• Parameter Sensitivity:
  • Mobility ($M$): Affects interface curvature away from the wall and steady displacement but does not affect the dynamic contact angle.
  • Friction ($\mu_f$): The sole determinant of the dynamic contact angle behavior.
  • Navier Slip ($\ell_N$): The sub-nanometric slip introduced for stability has negligible impact on the physical observables within the tested range.

5. Significance and Conclusion

This work provides a robust framework for bridging the gap between molecular and macroscopic descriptions of wetting.

• Validation of Mesoscale Models: It proves that Phase Field models, when carefully calibrated using MD data for contact line friction, can accurately capture complex nanoscale hydrodynamics on high-friction surfaces.
• Physical Insight: The study confirms that contact line friction is the dominant molecular input required for accurate mesoscale modeling, distinct from bulk viscosity.
• Future Directions: The authors note that while the current model is robust, it misses a specific "slip component" localized at the contact line. Future work should explore Generalized Navier Boundary Conditions (GNBC) or slip models where slip length is a function of the phase variable to fully reconcile continuum models with molecular physics.

In summary, the paper establishes that contact line friction is the critical empirical parameter for wetting models and demonstrates a rigorous, physics-based protocol to integrate Molecular Dynamics data into continuum simulations, significantly advancing the reliability of multiscale wetting modeling.