Sharp Convergence to the Half-Space for Mullins-Sekerka in the Plane

This paper revisits the HED Method for Mullins-Sekerka evolution in the plane to establish not only the algebraic rate but also the sharp leading-order constant for convergence to a flat interface by introducing an intrinsic distance and deriving natural assumptions on the flow.

The Gist

Here is an explanation of the paper "Sharp Convergence to the Half-Space for Mullins-Sekerka in the Plane," translated into everyday language with creative analogies.

The Big Picture: Smoothing Out a Bumpy Road

Imagine you have a long, wavy road separating two fields: one filled with blue grass and the other with red grass. This road isn't straight; it has hills, valleys, and maybe even some weird loops or spirals.

In the world of physics and materials science, this road represents the interface between two different phases of a material (like ice and water, or two types of metal mixed together). The Mullins-Sekerka equation is the set of rules that describes how this road moves and changes shape over time.

The main rule is simple: Nature hates sharp curves. Just like a crumpled piece of paper naturally wants to smooth out, this interface wants to become a perfectly straight, flat line. The paper asks: How fast does this happen? And exactly how does it behave as it gets closer to being flat?

The Problem: The "Flat" Goal is Tricky

In a small, closed room (a "compact" space), we know exactly how fast things smooth out. It's like a ball rolling down a hill into a bowl; it stops quickly.

But in the open world (the "plane" or infinite space), things are messier. The road is infinitely long. If you have a tiny bump far away, it might take a very long time to flatten out. Previous studies told us the road does eventually flatten, and they gave us a rough idea of the speed (like saying, "It gets flat in about 10 hours").

However, they didn't know the exact speed limit or the precise shape of the road as it was flattening. It was like knowing a car will arrive in 10 hours, but not knowing if it's cruising at 60 mph or 65 mph.

The New Discovery: The "HED" Method

The authors of this paper, Shi and the Westdickenbergs, revisited a mathematical tool called the HED Method. Think of this as a new, high-tech GPS system for the road.

Instead of just measuring how far the road is from being straight (Distance) and how much energy is stored in the bumps (Energy), they invented a new way to measure the "distance" that is intrinsic to the road itself.

• The Old Way: Measuring the distance from the road to a straight line using a ruler held from the outside.
• The New Way: Measuring the "tension" and "curvature" directly on the road, as if the road were a living thing feeling its own bumps.

By using this new "intrinsic" ruler, combined with the energy and the rate at which the road smooths out (Dissipation), they could derive a much sharper, more precise prediction.

The Key Findings

Here is what they found, broken down simply:

1. The "Sharp" Constant
They didn't just say "it gets flat." They calculated the exact leading number in the formula for how fast it gets flat.

• Analogy: Imagine a runner slowing down. Previous math said, "They will stop in $1/t$seconds." This paper says, "They will stop in exactly$0.178/t$ seconds." That extra precision matters for predicting exactly what the material looks like at any given moment.

2. The "Small Wobble" Rule
They proved that if the road starts out "mostly flat" (mathematically, if the wobbles are small enough), it will stay mostly flat and smooth out perfectly.

• Analogy: If you have a slightly crumpled sheet of paper, it will smooth out nicely. But if you have a sheet that is twisted into a spiral or has a knot, it might get stuck or behave weirdly. The authors showed that as long as the initial "twist" is small enough, the road will inevitably become a straight line.

3. The "Spiral" Warning
The paper also points out a fascinating edge case. If the road is shaped like a giant, infinite spiral (imagine a snail shell that never ends), it technically satisfies the rules to be a "flat-ish" road, but it will never actually become a straight line. It will just keep spiraling forever. This helps explain why the math is so careful about the starting conditions.

Why Does This Matter?

You might ask, "Who cares about a math problem about a wavy line?"

This matters because Mullins-Sekerka describes real-world phenomena:

• Metallurgy: How impurities separate in metal alloys as they cool.
• Geology: How crystals grow in rocks.
• Biology: How cell membranes change shape.

By knowing the exact rate at which these interfaces flatten, engineers and scientists can better predict how materials will behave over long periods. They can design stronger alloys or understand how biological tissues heal with much greater precision.

The Takeaway

This paper is like upgrading a weather forecast.

• Old Forecast: "It will rain tomorrow."
• New Forecast: "It will rain at 2:00 PM, with a 95% chance of 0.5 inches of rain, and the wind will be blowing from the North at 12 mph."

The authors took a complex, infinite problem and used a clever new measuring stick (the HED method with intrinsic distance) to give us the most precise "weather forecast" yet for how material interfaces smooth out in the plane. They proved that, under the right conditions, the universe has a very specific, predictable rhythm for becoming flat.

Technical Summary

Here is a detailed technical summary of the paper "Sharp Convergence to the Half-Space for Mullins-Sekerka in the Plane" by Shi, Westdickenberg, and Westdickenberg.

1. Problem Statement

The paper investigates the long-time dynamics of the Mullins-Sekerka evolution in two dimensions ($\mathbb{R}^2$). This is a nonlocal, third-order free boundary problem modeling phase separation in binary alloys, where an interface $\Gamma_t$ evolves with normal velocity $V$ equal to the jump in the normal derivative of a harmonic field $u$ across the interface, with $u$ equal to the mean curvature $\kappa$ on the interface.

Key Challenges:

• Non-compactness: Unlike previous studies focusing on compact interfaces (where convergence is exponential), this paper addresses unbounded interfaces converging to a flat line (the half-space limit).
• Non-convexity: The energy functional (interfacial area) is not strictly convex in the non-compact setting, and the minimizer (the flat line) is not unique (any parallel line is a minimizer).
• Lack of Graph Structure: In 2D, small dimensionless parameters do not automatically guarantee the interface is a graph over the flat line, unlike in 3D.
• Sharp Rates: Previous work established algebraic convergence rates ($E \sim t^{-1}$) but lacked the sharp leading-order constants.

2. Methodology: The HED Method

The authors employ and refine the HED Method (Energy, Dissipation, Distance), originally introduced by Brézis and adapted for non-convex problems by Otto and others. The core strategy involves establishing a system of differential inequalities relating three quantities:

1. Energy ($E$): The excess interfacial area relative to the flat line.
2. Dissipation ($D$): The metric derivative of the flow (related to the $H^{-1/2}$ norm of the curvature).
3. Squared Distance ($H$): An intrinsic distance measure between the evolving interface and the flat equilibrium.

Key Methodological Innovations:

• Intrinsic Distance: Instead of using extrinsic distances (like $L^2$ distance of height functions), the authors define an intrinsic squared distance $H$ based on the normal vector field $\nu$. Specifically, $H = |\nu \cdot \pi^\perp_{e_2}|^2_{-1/2}$, which measures the deviation of the normal from the vertical direction in a negative Sobolev norm.
• Geometric Harmonic Analysis: The analysis relies heavily on the theory of $\delta$-Ahlfors Regular (AR) domains and Chord-Arc Curves. They utilize singular integral operators (Single Layer Potentials) and the Dirichlet-to-Neumann map ($N$) defined on the interface itself, rather than on the bulk domain.
• Regime Control: They define a dimensionless parameter $\epsilon = (E^2 D)^{1/6}$. The analysis assumes $\epsilon$ is small initially, which allows them to control the geometry (ensuring the interface remains a $\delta$-AR domain and eventually a graph) and bound error terms in the differential inequalities.

3. Key Contributions

A. Intrinsic Formulation and Function Spaces

The paper moves away from describing the interface as a graph of a function $h(x)$ (which fails for general admissible interfaces with spirals or overhangs). Instead, they formulate the problem directly on the interface $\Gamma$ using:

• Admissible Interfaces: Defined as boundaries of $\delta$-AR domains with generalized curvature $\kappa \in L^2_{loc}$.
• Function Spaces: They utilize homogeneous Sobolev spaces $\dot{H}^s(\Gamma)$ defined via the Dirichlet-to-Neumann map, allowing for a rigorous treatment of interfaces that are not necessarily graphs.

B. The HED Inequality with Sharp Constants

The authors derive a refined HED inequality (Lemma 3.16) connecting Energy, Distance, and Dissipation:
$$E \leq C_1 \sqrt{HD}$$
Crucially, they identify the sharp constant $C_1$. For the linearized problem, $C_1 = \frac{1}{4(\sqrt{2}+1)}$. For the nonlinear problem, they show:
$$E \leq \left( \frac{1}{4(\sqrt{2}+1)} + C\epsilon^2 \right) \sqrt{HD}$$
This precise constant is essential for obtaining the sharp convergence rate.

C. Monotonicity of Dissipation

A major technical hurdle is proving that the dissipation $D$ is non-increasing ($\dot{D} \leq 0$) in the non-convex setting. The authors prove that if the interface is sufficiently flat (small $\epsilon$), the Hessian of the energy functional is positive semi-definite along the gradient flow direction. This establishes that $t \mapsto E(t)$ is convex, a prerequisite for the HED method to yield optimal rates.

D. Non-Convexity Proof

They explicitly demonstrate (Proposition 1.7) that the Mullins-Sekerka energy is not convex in any neighborhood of the flat interface in the plane. This distinguishes the 2D case from certain 1D thin-film equations and necessitates the perturbative approach (small $\epsilon$) rather than a global convexity argument.

4. Main Results

Theorem 1.5 (Convergence Rates):
Assuming the initial interface is admissible and the dimensionless parameter $\epsilon_0$ is sufficiently small, the solution $\{ \Gamma_t \}$ satisfies the following sharp algebraic decay rates for all $t > 0$:

1. Energy Decay:
   $$E(t) \leq \min \left\{ E_0, \frac{(C_1 + C\epsilon_0^2)H_0}{t} \right\}$$
   where $C_1 = \frac{1}{4(\sqrt{2}+1)}$. This matches the optimal rate derived from the linearized problem.

2. Dissipation Decay:
   $$D(t) \leq \min \left\{ D_0, \frac{E_0}{t}, \frac{(1/2 + C\epsilon_0^2)H_0}{t^2} \right\}$$

3. Distance Bound:
   $$H(t) \leq H_0 (1 + C\epsilon_0^2)$$
   The distance to the equilibrium remains bounded, preventing the interface from drifting infinitely far from the flat line (unlike the non-unique minimizers in the toy example).

Significance of Constants:
The paper proves that the leading-order constant $C_1$ in the energy decay is sharp. It coincides with the constant obtained by applying the HED method to the linearized Mullins-Sekerka equation. The term $C\epsilon_0^2$ quantifies the deviation from linearity and convexity.

5. Significance and Impact

• Resolution of Sharp Constants: Prior to this work, convergence rates for non-compact Mullins-Sekerka were known only up to generic constants. This paper provides the exact leading-order constants, bridging the gap between linearized analysis and full nonlinear dynamics.
• Intrinsic Geometry: By formulating the problem intrinsically on the interface using chord-arc curves and singular integrals, the authors bypass the limitations of graph-based assumptions. This allows the theory to apply to a broader class of interfaces, including those with mild singularities or non-graph structures, provided they are sufficiently flat.
• Methodological Advancement: The work refines the HED method, demonstrating how to handle non-convex energies by carefully tracking the "flatness" parameter $\epsilon$ and utilizing the monotonicity of dissipation in a perturbative regime.
• Dimensional Distinction: The paper highlights the specific difficulties of the 2D case (critical Sobolev embeddings, logarithmic potentials) compared to 3D, explaining why graph formation is not automatic in 2D and why the intrinsic distance is necessary.

In summary, this paper establishes the optimal algebraic convergence rates with sharp constants for the Mullins-Sekerka flow in the plane, utilizing a sophisticated geometric analysis framework that treats the interface as an evolving geometric object rather than a graph.