Faber-Krahn inequality

What shape makes for the best drum? If you fix the area of a drumhead, is there a shape that will produce the lowest possible fundamental tone? This simple question from music and physics leads to a deep mathematical principle known as the Faber-Krahn inequality. It provides a definitive answer: the circle is the undisputed champion of low notes. But how can one prove this without testing every imaginable shape? This is the central problem the inequality elegantly solves, revealing a universal preference in nature for the circular form. This article explores this fascinating principle. First, in "Principles and Mechanisms," we will uncover the mathematics of vibrations, eigenvalues, and the clever technique of symmetrization used to prove the inequality. Following this, the "Applications and Interdisciplinary Connections" chapter will showcase how this single idea resonates across physics, engineering, and even abstract mathematics, connecting the sound of a drum to quantum mechanics and the geometry of space itself.

Imagine you are given a piece of stretchable, elastic material, say, for the head of a drum. You have a fixed amount of this material, which means its area is fixed. Your task is to design the shape of the drum's frame. You could make it a square, a long rectangle, a star shape, or a simple circle. Now, if you strike each of these drums, which one will produce the lowest possible fundamental tone? This is not just a musician's idle curiosity; it is a question that cuts to the heart of how shape influences physical reality. The answer is a beautiful piece of mathematics known as the ​​Faber-Krahn inequality​​, and it declares with certainty: the circular drum will always sing the lowest note. Let's embark on a journey to understand why this is so.

In physics, the fundamental tone of a vibrating membrane corresponds to its lowest natural frequency of vibration. In the language of mathematics, this frequency is captured by a number called the ​​first Dirichlet eigenvalue​​ of the Laplacian operator, denoted as $\lambda_1$. Thinking about it can be simpler than its name suggests. Any vibration mode can be described by a function, $u$, that gives the vertical displacement of the drumhead at each point. The total "bending" or "tension" energy of the membrane is proportional to the integral of the square of the gradient, $\int |\nabla u|^2$. The total magnitude of the vibration, a sort of "volume" of displacement, is given by $\int u^2$.

The physicist and engineer in us know that nature is, in a sense, lazy. A physical system tends to settle into its lowest energy state. The first eigenvalue, $\lambda_1$, represents this minimum possible energy for a given amount of displacement. It is found by minimizing the ​​Rayleigh quotient​​:

Here, $\Omega$ represents the shape of our drum. The function $u$ must be zero at the boundary of the drum ($\partial\Omega$), because the membrane is fixed to the frame. The Faber-Krahn inequality answers our initial question with mathematical precision: for any shape $\Omega$ and a disk $B$ with the same area, $\lambda_1(\Omega) \ge \lambda_1(B)$. The disk is the undisputed champion of low energy.

How can we be so sure that the disk is the ultimate minimizer? We don't need to test every conceivable shape—an impossible task! Instead, we can use a clever and deeply intuitive idea called ​​Schwarz symmetrization​​.

Imagine the vibration of a non-circular drum, represented by the function $u$. It looks like a little mountain landscape over the domain $\Omega$. Now, let's do something strange. We'll take this landscape and "rearrange" it into a perfectly symmetrical one. Think of it like a baker taking a lumpy cake and re-forming it into a perfect, round shape. We can do this by slicing our function $u$ horizontally at every height. Each slice is a set of points where the vibration is above a certain level. We take each of these slices, which might be weirdly shaped regions, and replace them with a disk of the same area, centered at the origin. When we stack all these disks up, from the largest at the bottom to the smallest at the top, we get a new, perfectly radially symmetric function, let's call it $u^*$. This new function lives on a disk, $B$, that has the exact same area as our original shape $\Omega$.

Here comes the magic. Two crucial things happen during this rearrangement:

1. ​​The denominator of the Rayleigh quotient stays the same.​​ Because we just shuffled the "stuff" of the function around, the total squared displacement is conserved: $\int_{\Omega} u^2 = \int_{B} (u^*)^2$.

2. ​​The numerator of the Rayleigh quotient gets smaller (or stays the same).​​ This is the celebrated ​​Pólya-Szegő inequality​​. It tells us that the total bending energy of the rearranged function is less than or equal to that of the original: $\int_{B} |\nabla u^*|^2 \le \int_{\Omega} |\nabla u|^2$. Intuitively, by making the shape smooth and round, we have ironed out all the unnecessary wiggles and sharp corners that contribute to the bending energy. The circle is the most efficient shape for containing the vibration.

Now, look at the Rayleigh quotient. We took the vibrating function from our original drum, rearranged it into a function on a disk, and found that the energy ratio either went down or stayed the same. This means that the lowest possible energy for our original drum, $\lambda_1(\Omega)$, must be greater than or equal to the lowest possible energy for the disk-shaped drum, $\lambda_1(B)$. Equality only happens if our original shape was already a disk to begin with. The argument is simple, elegant, and powerful.

The Faber-Krahn inequality gives us a universal benchmark. For any planar domain $\Omega$ with area $A$, the scale-invariant quantity $A \cdot \lambda_1(\Omega)$ has a universal lower bound, which is achieved by the disk. This minimum value is not some abstract symbol; it's a concrete number, $\pi j_{0,1}^2 \approx 18.168$, where $j_{0,1}$ is the first positive zero of the Bessel function $J_0$, a special function that naturally arises when studying vibrations in circular objects.

This suggests a wonderful new question. If a shape is not a disk, how far is it from this ideal? We can measure its "imperfection" using the ​​Faber-Krahn deficit​​. We can define it, for instance, as a relative quantity:

where $B$ is a disk of the same area as $\Omega$. This deficit is always non-negative, and it is zero if and only if $\Omega$ is a disk. A small deficit means the shape is spectrally "close" to a disk. This leads to profound "stability" results, which state that if the deficit is small, the shape must geometrically look very much like a disk.

The Faber-Krahn inequality is a beautiful and powerful tool, but it is not the only way to peer into the spectral soul of a shape. Another fundamental result, the ​​Cheeger inequality​​, provides a different kind of lower bound for $\lambda_1(\Omega)$. Instead of looking at the total area, Cheeger's inequality looks for "bottlenecks". It defines a quantity $h(\Omega)$, the ​​Cheeger constant​​, which measures the worst-case "perimeter-to-area" ratio for any sub-region within $\Omega$. A domain with a thin neck connecting two larger parts, like a dumbbell, will have a very small Cheeger constant. The inequality states that $\lambda_1(\Omega) \ge \frac{h(\Omega)^2}{4}$.

So we have two different lower bounds, one from Faber-Krahn based on area, and one from Cheeger based on bottlenecks. Which one is better? The answer depends entirely on the shape:

• For a "fat" shape, like a disk or a square, the Faber-Krahn bound is typically stronger. For the disk itself, the Faber-Krahn bound is perfectly sharp (it becomes an equality), while the Cheeger bound is just a decent approximation.

• For a shape with a severe bottleneck, the Cheeger bound tells a more accurate story. Consider a dumbbell made of two large disks connected by a tiny neck, with a total area $A$. The Faber-Krahn bound, depending only on $A$, would suggest a relatively high fundamental frequency. However, we intuitively know that the vibration will mostly confine itself to one of the large disks, barely crossing the neck, which should result in a low frequency. The Cheeger constant $h(\Omega)$ for the dumbbell will be very small due to the thin neck, and the Cheeger bound $\frac{h(\Omega)^2}{4}$ will correctly predict a small $\lambda_1(\Omega)$. In this scenario, the Faber-Krahn bound is misleadingly large, while the Cheeger bound, though not exact, correctly captures the geometry's effect.

• For a long, thin rectangle, the situation is similar. As the rectangle gets thinner while keeping its area constant, its true eigenvalue gets larger and larger. The Faber-Krahn bound remains constant and quickly becomes useless. The Cheeger bound, however, tracks the true eigenvalue beautifully, staying proportional to it.

This reveals a deeper truth: there is no single magic lens to understand the world. The beauty lies in having different tools that reveal different aspects of reality. Faber-Krahn sees the bulk, while Cheeger sees the constrictions.

The journey does not end on the flat plane of a drumhead. We can ask the same question in a more exotic setting: what if our membrane lives on a curved surface, like a sphere or a saddle-shaped hyperbolic plane? Does a "geodesic ball" (the analogue of a disk in a curved space) still have the lowest fundamental frequency for a given area?

The answer is a breathtaking generalization of the Faber-Krahn inequality, and it connects the vibration to the very curvature of space itself. The ​​generalized Faber-Krahn inequality​​ states that on a manifold with positive Ricci curvature (a space that curves like a sphere, on average), any domain has a first eigenvalue $\lambda_1$ that is greater than or equal to the eigenvalue of a ball of the same volume in a standard sphere of that curvature. In essence, positive curvature makes things "stiffer," raising all the fundamental frequencies. Conversely, negative curvature makes things "floppier," lowering the frequencies compared to the flat case.

This is a profound statement. It tells us that by listening to the "sound" of a domain—its fundamental frequency—we can learn something about the intrinsic geometry of the space it inhabits. The simple question about the best shape for a drum has led us all the way to a principle that unifies analysis (eigenvalues), geometry (shape and curvature), and physics (vibrations), painting a unified and beautiful picture of the world.

After a journey through the principles and mechanisms of the Faber-Krahn inequality, we might be tempted to see it as a beautiful but esoteric piece of mathematics. Nothing could be further from the truth. Like a master key, this single, elegant idea unlocks doors in a surprising variety of fields, revealing a deep and often hidden unity in the workings of the world. It shows us that nature, in many different contexts, seems to have a clear preference for a particular shape: the circle. Let us now explore some of these remarkable applications and connections.

Perhaps the most intuitive place to start is with music. Imagine you have two drumheads, one shaped like a square and the other a circle, but both are made from the same material and have the exact same area. If you strike them both, which one will produce a lower-pitched fundamental tone? Our intuition might suggest the circle, as it feels more "perfect" or "efficient." The Faber-Krahn inequality turns this intuition into a certainty. The fundamental frequency of a vibrating membrane is proportional to the square root of the first Dirichlet eigenvalue, $\lambda_1$. Since the inequality states that $\lambda_1(\text{square}) > \lambda_1(\text{circle})$ for equal areas, the square drum will always have a higher fundamental frequency. The circle, of all possible shapes, sings the deepest note.

This is more than a musical curiosity. The very same mathematics that governs the vibration of a drum—the Helmholtz equation, $-\Delta u = \lambda u$—appears again and again throughout physics. Consider the cooling of a hot metal plate whose edges are kept at a constant cold temperature. The temperature distribution over time is governed by the heat equation, whose solutions are built from the same eigenfunctions of the Laplacian. The long-term cooling rate is dictated by the slowest-decaying thermal mode, and its time constant $\tau$ is inversely proportional to the first eigenvalue, $\tau = 1/(\alpha \lambda_1)$, where $\alpha$ is the thermal diffusivity. A circular plate of a given area, having the smallest possible $\lambda_1$, will therefore hold onto its heat the longest; it has the largest time constant and equilibrates with its surroundings more slowly than a square plate or any other shape of the same area. The shape that produces the lowest tone is also the one that cools the slowest.

The story continues into the strange and wonderful world of quantum mechanics. Imagine a particle trapped in a two-dimensional "box." Its behavior is described by the Schrödinger equation, which, for a particle in an infinite potential well, is mathematically identical to the Helmholtz equation. The allowed energy levels of the particle are determined by the eigenvalues of the Laplacian on the shape of the box. The lowest possible energy the particle can have—its "ground state" energy—is directly proportional to $\lambda_1$. Consequently, for a given confinement area, a particle in a circular quantum well will have a lower ground state energy than a particle in a square well. The circle is, in a sense, the most stable or "relaxed" shape to be confined within. From the audible vibrations of a drum to the thermodynamics of cooling and the quantum energy of a particle, the Faber-Krahn inequality reveals a universal principle: among all shapes of a given size, the circle is the ground state.

These examples are all instances of a broader class of problems in a field known as ​​shape optimization​​. The fundamental question is simple: if you have a fixed amount of resources (e.g., a fixed area or volume), what is the best shape to use for a particular purpose? The Faber-Krahn inequality provides one of the most elegant and definitive answers in this field. If the goal is to minimize the fundamental frequency, the slowest rate of decay, or the ground-state energy, the inequality guarantees that the disk (in 2D) or the ball (in 3D) is the unique optimal shape. This principle has profound implications for design and engineering, from the construction of resonant cavities and waveguides to the design of cooling fins and quantum dots.

But the story of optimality does not end there. What if we need to tile a large surface with many small cells, and our goal is to minimize the sum of the fundamental frequencies of all the cells? The Faber-Krahn inequality tells us that each individual cell wants to be a circle to achieve the lowest possible $\lambda_1$. However, it is impossible to tile the plane with circles without leaving gaps. This creates a beautiful mathematical tension between the optimality of a single shape and the geometric constraint of tiling. The problem then becomes finding the best compromise: a shape that is "circle-like" enough to have a low $\lambda_1$, but can also tile the plane perfectly. The leading candidate, supported by extensive numerical evidence and deep mathematical reasoning, is the regular hexagon. This "hexagonal conjecture" for spectral partitions suggests that the honeycomb pattern, long admired in nature, may also be the optimal design for minimizing spectral energy. This question pushes the simple idea of Faber-Krahn into the complex and fascinating world of geometric tiling problems, a vibrant area of current research.

The true power of a great mathematical idea is often measured by the unexpected bridges it builds between different fields. The Faber-Krahn inequality is a prime example, connecting the modern field of spectral theory with classical geometry and deep questions in analysis.

One of the most ancient problems in geometry is the ​​isoperimetric problem​​: of all shapes with a given perimeter, which one encloses the largest area? The ancient Greeks knew the answer was the circle, but a rigorous proof was elusive for centuries. Astonishingly, one can prove this purely geometric fact using a theorem about eigenvalues. By considering a related spectral problem with different boundary conditions, mathematicians have shown that a spectral isoperimetric inequality (a cousin of Faber-Krahn) logically implies the classical one. This reveals a hidden, profound relationship: information about the geometry of a shape, like its perimeter-to-area ratio, is encoded in its vibrational spectrum.

The connections run even deeper, into the heart of modern analysis. A central object in the study of diffusion is the ​​heat kernel​​, $p(t, x, y)$, which describes how an initial point-source of heat at location $y$ spreads throughout a domain over time $t$. This is a fundamentally local description of a dynamic process. The first eigenvalue, $\lambda_1$, is a global property of the domain as a whole. It seems remarkable that the two should be related, yet they are. The Faber-Krahn inequality is a key ingredient in proving what are known as Gaussian upper bounds for the heat kernel. It provides a strict "speed limit" on how fast heat can diffuse, showing that the global geometry of a space exerts powerful control over local phenomena unfolding within it.

Finally, the principle of the circle's optimality is not confined to the smooth, Euclidean world we are used to. Mathematicians have asked whether similar ideas hold true for more exotic spaces, such as fractals. On the famous Sierpinski gasket, a shape of infinite complexity and fractional dimension, an analogue of the Faber-Krahn inequality exists. It tells us that among all connected subsets of the gasket with a fixed "fractal measure," the minimum first eigenvalue is achieved by one of the gasket's basic self-similar triangular components. This demonstrates the robustness and universality of the underlying principle, extending it far beyond the realm of classical geometry.

From the simple and audible to the abstract and profound, the Faber-Krahn inequality stands as a testament to the interconnectedness of scientific thought. It shows us how a single, simple statement—that among all shapes, the circle holds a special place—can resonate through physics, engineering, and mathematics, revealing the hidden music that unifies our understanding of the world.