Zeros of Bessel Functions

In the worlds of mathematics and physics, certain concepts emerge repeatedly, forming the very language used to describe the universe. Bessel functions are one such concept, essential for understanding phenomena involving waves and oscillations in circular or cylindrical geometries. While the functions themselves can seem abstract, the true magic lies in their "zeros"—the specific points where their value becomes null. This article addresses a crucial question: why are these discrete points so profoundly important? We will demystify these numbers, revealing them not as mathematical oddities, but as fundamental constants that govern the physical world. The following sections will guide you on a journey, first through the elegant principles and mathematical properties of these zeros, and then into their concrete applications across science and technology. You will learn how these zeros arise, the beautiful order they obey, and how they dictate everything from the sound of a drum to the efficiency of our global communications network.

Now that we’ve been introduced to Bessel functions, you might be thinking they seem a bit abstract, a bit esoteric. Where do they come from? Why should we care about their "zeros"? Well, let's pull back the curtain. My aim here is not to drown you in equations, but to take you on a journey to see how these functions arise naturally from simple physical questions, and how their properties, especially their zeros, reveal a deep and elegant structure governing the world around us, from the sound of a drum to the very fabric of mathematical order.

Let's start with something familiar: sine and cosine. We love them because they are simple. They oscillate up and down in a perfectly regular rhythm. They are the solutions to the simplest equation of vibration. A Bessel function, you might say, is a slightly more worldly cousin of sine and cosine. It also wiggles up and down, but it lives in a world with a bit more... geometry. Specifically, it describes waves that aren't just moving along a line, but are spreading out in a circle. As the waves spread, their amplitude tends to die down.

So, a Bessel function is essentially a ​​damped oscillation​​. Its graph looks like a sine wave that's being squeezed as it moves away from the origin. The "zeros" of the function are simply the points where the graph crosses the horizontal axis—where the value of the function is exactly zero.

Figure 1: Graphs of the Bessel functions $J_0(x)$ (blue) and $J_1(x)$ (orange), showing their oscillating, decaying nature and their zeros (the points where they cross the x-axis).

For most Bessel functions, finding these zeros is a tricky business. But for some special cases, the curtain lifts and we see a familiar face. For the Bessel function of order $-\frac{1}{2}$, it turns out to be nothing more than a disguised cosine function!

Suddenly, the mystery of its zeros vanishes. For positive $x$, the factor $\sqrt{\frac{2}{\pi x}}$ is never zero, so the zeros of $J_{-1/2}(x)$ are simply the zeros of $\cos(x)$. These occur at $x = \frac{\pi}{2}, \frac{3\pi}{2}, \frac{5\pi}{2}, \dots$, or more generally, at $x_n = (n - \frac{1}{2})\pi$ for any positive integer $n$. This simple case gives us a foothold; it tells us that at their core, Bessel functions are deeply connected to the familiar world of oscillation, and their zeros, at least sometimes, follow simple, predictable patterns.

So, what are these zeros good for? Let's get physical. Imagine a circular drumhead. When you strike it, it vibrates, creating sound. But it can't just vibrate in any old way. The edge of the drum is clamped down; it cannot move. This single physical constraint—the ​​boundary condition​​—is the key to everything.

The shape of the standing waves on the drumhead is described by Bessel functions. If the drum has a radius $R$, the condition that the edge doesn't move means that the displacement, described by a Bessel function, must be zero at $r=R$. In other words, the allowed vibrations are determined by the zeros of the Bessel function!

Let's say a particular vibration pattern is described by the Bessel function $J_m(z)$. The allowed wave patterns, or ​​normal modes​​, can only exist if the argument of the function, $kR$ (where $k$ is the wavenumber related to the frequency), is one of the zeros of $J_m(z)$. We denote the $n$-th positive zero of $J_m(z)$ as $j_{m,n}$ (or $\alpha_{m,n}$). This gives us a fundamental equation:

This is a ​​quantization condition​​. It tells us that not just any frequency is possible; only a discrete set of frequencies, determined by the discrete set of zeros, are allowed. Each pair of integers $(m, n)$ defines a unique mode of vibration with a specific frequency. The index $m$ corresponds to the number of nodal diameters (lines running through the center that don't move), and $n$ corresponds to the number of nodal circles (circles that don't move). The zeros of Bessel functions are the fingerprints of musical harmony for a circular instrument. They are, quite literally, the sound of symmetry.

Now that we see the physical importance of these zeros, we can ask a more mathematical question. Is there any pattern to their locations? The answer is a resounding yes, and it is beautiful. It's called the ​​interlacing property​​.

Imagine you have two sets of points on a line, a set of red points and a set of blue points. We say they interlace if, between any two consecutive red points, you find exactly one blue point, and between any two consecutive blue points, you find exactly one red point.

Amazingly, this is precisely what happens with the zeros of Bessel functions of consecutive integer orders, like $J_0(x)$ and $J_1(x)$. If we list their positive zeros in increasing order, we find a perfect, elegant dance:

The first zero of $J_0$ comes first. Then comes the first zero of $J_1$. Then the second zero of $J_0$, followed by the second of $J_1$, and so on, forever. This isn't a coincidence; it's a theorem. It means that the vibrational modes of our drum are ordered in a completely predictable way. This simple rule has a powerful consequence: if you ask how many zeros of $J_1(x)$ lie in the interval between the $m$-th and $n$-th zeros of $J_0(x)$, the answer is astonishingly simple: it's exactly $n-m$.

This interlacing principle extends even further. A similar relationship holds between the zeros of a Bessel function, say $J_0(x)$, and the zeros of its derivative, $J_0'(x)$. Why would we care about the zeros of the derivative? Well, a different physical setup might require it! A "fixed" boundary on our drum means displacement is zero ($J_0(kR)=0$), but a "free" boundary would mean the slope of the displacement is zero, which translates to $J_0'(kR)=0$. The interlacing of these two sets of zeros tells us immediately which type of boundary condition will lead to a higher fundamental frequency, all without having to calculate a single numerical value!

Let's zoom out now. What happens to the zeros way out, for very large values of $x$? Do they continue to spread out, or do they bunch up? The answer lies in looking at the ​​asymptotic approximation​​ of the Bessel function. As we journey far from the origin, the Bessel function "forgets" some of its complexity and starts to look more and more like its simple cousin, the cosine. For large $x$, we have:

The term $\sqrt{2/(\pi x)}$ is the decaying amplitude we talked about. The magic is in the cosine term. It tells us that for large $x$, the wiggles of the Bessel function become incredibly regular. And if the function behaves like a cosine, its zeros must behave like the zeros of a cosine. This leads to a remarkable conclusion: the spacing between consecutive large zeros of $J_\nu(x)$ approaches a constant value—it approaches $\pi$.

Think about this for a moment. The first few zeros might be oddly spaced, but as you go further and further out, a simple, universal rhythm emerges. The spacing between the 1,000,000th zero and the 1,000,001st zero is extremely close to $\pi$. This asymptotic regularity is a profound feature, and with more powerful tools like McMahon's expansion, we can even calculate how the spacing approaches $\pi$ with incredible precision, finding correction terms that depend on the order $\nu$ and the zero's location.

We have seen that zeros dictate physical reality, that they dance in an ordered, interlacing pattern, and that they settle into a majestic rhythm at infinity. But there is one more secret, one final piece of magic I want to share. It answers a seemingly impossible question: can we find a single expression that captures a property of all the zeros at once?

The sum of the zeros themselves, $\sum j_{\nu,k}$, would be infinite. But what about the sum of their reciprocal squares, $\sum \frac{1}{j_{\nu,k}^2}$? This sum does converge, and its value is one of the most beautiful results in the theory.

The trick is to realize that you can describe a function in two ways. One is locally, near the origin, using its Taylor series expansion. The other is globally, using all of its zeros, through something called a Weierstrass product. Both descriptions must be identical. By writing down the first few terms of each and comparing them, you equate a simple coefficient from the Taylor series with an expression involving the sum over all the zeros.

When we do this for the Bessel function $J_\nu(z)$, it leads to an absolutely astonishing result, a formula that connects the infinite, scattered set of zeros to the single number $\nu$ that defines the function in the first place:

This is the kind of thing that gives mathematicians chills. All of that complexity—the infinite list of non-obvious numbers that are the zeros of a Bessel function—is captured by such a simple, compact expression. It reveals a hidden unity, a global rule governing the entire set of zeros. It is a perfect symphony, where every note, every zero, plays its part in a harmony dictated by the number $\nu$. And it is a stunning reminder that in the universe of mathematics, as in the physical world it describes, there is always a deeper, simpler, and more beautiful order waiting to be discovered.

After our journey through the mathematical landscape of Bessel functions and their zeros, a fair question to ask is: "What's the big deal?" We have discovered a special, infinite list of numbers for each "flavor" $\nu$ of Bessel function. Are these just a mathematical curiosity, a formal game played by mathematicians and physicists? The answer, you will be delighted to find, is a resounding no. These numbers aren't just abstract points on a graph; they are fundamental constants of our physical world. They appear whenever a problem involves waves or fields constrained within a circular or cylindrical geometry. They are, in a very real sense, the secret numbers that nature uses to tune the universe.

Let's explore some of the unexpected places these zeros turn up, from the familiar sounds we hear to the invisible quantum world and the technology that powers our modern lives.

Perhaps the most intuitive and tangible application of Bessel function zeros is in the study of vibrations. Imagine a kettledrum. When you strike its circular membrane, it vibrates to produce a sound. But why does it produce specific, characteristic pitches instead of just a chaotic noise?

The motion of the drumhead is described by the wave equation. Because the membrane is circular and clamped at its edge, there's a crucial boundary condition: the displacement must be zero at the rim at all times. When we solve the wave equation with this constraint, the solutions naturally involve Bessel functions. The condition that the edge remains stationary forces the solution to be zero at the radius of the drum, say $R$. This means that the product of the wave number $k$ and the radius $R$ must be a zero of the appropriate Bessel function, $J_m(kR) = 0$.

This single equation is the key! It tells us that only a discrete set of vibrational modes are allowed. The fundamental tone, the lowest note the drum can play, corresponds to the very first zero of the $J_0$ function, $z_{0,1} \approx 2.405$. Higher frequencies, or overtones, are dictated by the subsequent zeros ($z_{0,2}, z_{0,3}, \dots$) and the zeros of higher-order Bessel functions ($z_{1,1}, z_{1,2}, \dots$).

What's more, for a mode corresponding to a higher zero, say the third zero of $J_0(x)$, something remarkable happens. Not only is the edge of the drum stationary, but there also appear concentric circles on the interior of the membrane that do not move at all. These are the nodal lines. Their radii are not random; they are precisely determined by the preceding zeros of the same Bessel function. If the vibration is governed by the $n$-th zero, the nodal circles will be located at radii proportional to the first, second, ..., $(n-1)$-th zeros. The zeros literally draw the silent lines on the canvas of the vibrating drum.

This principle isn't limited to full circles. If you have a membrane shaped like a slice of pie, the same ideas apply, but the angular boundaries introduce a new constraint that changes the order $\nu$ of the relevant Bessel function. The lowest-frequency mode with a nodal line might be a circular one or a radial one, and we can determine which by simply comparing the numerical values of the corresponding Bessel zeros. The lower zero always wins, corresponding to the lower frequency and thus the more "natural" vibration.

The same mathematics that describes vibrating drums also governs heat dissipation and chemical diffusion. If you have a hot, cylindrical piece of metal cooling in the air, or a chemical diffusing in a circular dish with absorbing walls, the patterns of temperature or concentration decay in modes described by Bessel functions. The slowest-decaying pattern, the one that lingers the longest, corresponds to the first zero of a Bessel function, with its decay rate $\lambda$ being directly proportional to the square of that zero: $\lambda \propto z_{\nu,1}^2$. In all these cases, the zeros are the quantized "answers" that the circular geometry imposes on the laws of physics.

The connection becomes even more profound when we step into the quantum realm. According to quantum mechanics, particles like electrons are also waves—not waves of water, but waves of probability. The Schrödinger equation, which governs these probability waves, is mathematically very similar to the classical wave equation.

Consider a simple, hypothetical "quantum billiard table": an electron trapped inside a two-dimensional, infinitely deep circular well, like a coin in a round, flat-bottomed dish with infinitely high walls. The electron cannot escape, so its probability wave function must be zero at the boundary of the circle. Does this sound familiar? It's the exact same boundary condition as the drumhead!

Consequently, the energy of the electron cannot take on any arbitrary value. Just like the drum can only play certain notes, the electron can only possess certain discrete energy levels. These allowed energies are determined by the zeros of the Bessel functions. The ground state energy, the lowest possible energy the particle can have, is fixed by the first zero of the $J_0$ function. Adding a second fermion, like another electron, forces it into the same spatial state (thanks to its opposite spin), doubling the total ground state energy. The zeros of Bessel functions are, quite literally, dictating the fundamental energy rungs of a quantum system.

We can explore more exotic confines as well. Imagine a particle constrained to move on the surface of a cone. Once again, solving the Schrödinger equation for this geometry leads us to Bessel's equation. But this time, the half-apex angle $\alpha$ of the cone determines the order $\nu$ of the Bessel function. A sharper cone leads to a different order than a wider cone. The allowed energy levels are then determined by the zeros of that specific Bessel function. The geometry of the space itself selects which set of magic numbers will govern the quantum states. Finding the energy of the first excited state becomes a game of checking which zero, from which order of Bessel function, comes next after the ground state's zero. Often, states with opposite angular momentum ($\ell$ and $-\ell$) will have the same energy, leading to a degeneracy dictated by the structure of the Bessel function solutions.

These ideas are not confined to the laboratory or to thought experiments. They are at the very heart of technologies that shape our world, most notably in telecommunications. An optical fiber is essentially a "waveguide for light," consisting of a central core with a high refractive index surrounded by a "cladding" with a lower refractive index. This difference allows light to be trapped in the core via total internal reflection and guided over vast distances.

A light wave propagating in the fiber can be described by different modes, which are the stable patterns the electromagnetic field can form. For a wave to be successfully guided, its field must be evanescent—it must decay to zero—within the cladding. There is a critical condition, a tipping point, where a mode is no longer tightly bound to the core and is on the verge of leaking out into the cladding. This is called the "cutoff" condition.

For the important family of Transverse Electric (TE) modes in a step-index fiber, the characteristic equation that governs the modes involves both Bessel functions (for the field in the core) and their cousins, the modified Bessel functions (for the field in the cladding). At the exact moment of cutoff, this equation simplifies dramatically, and we find that the condition for a mode to exist is determined by a zero of a Bessel function. For example, the cutoff for the $\text{TE}_{01}$ mode occurs precisely when the fiber's "$V$-number" (a dimensionless quantity that combines the core radius, wavelength, and refractive indices) is equal to $j_{0,1}$, the first zero of $J_0(x)$.

This is a profoundly important practical result. Engineers designing optical fibers use this exact principle to create "single-mode" fibers. By carefully manufacturing the fiber's properties so that its $V$-number is less than $j_{0,1}$ (approximately 2.405), they ensure that all higher-order modes are "cut off," and only the fundamental mode can propagate. This prevents modal dispersion—where different modes travel at different speeds, blurring the signal—and allows for the high-speed, long-distance data transmission that underpins the internet. Every time you stream a video or make a call over a fiber-optic network, you are relying on the fact that engineers have designed the system to operate below a threshold defined by a zero of a Bessel function.

Finally, the zeros of Bessel functions are not just passive results of physical constraints; they are active participants in the very definition and structure of the functions themselves. In the field of complex analysis, the Hadamard and Weierstrass factorization theorems tell us something astonishing: an entire function (a function that is "well-behaved" everywhere in the complex plane) can be completely reconstructed from its set of zeros, much like a polynomial is defined by its roots.

The Bessel function $J_\nu(z)$ is an entire function, and it can be written as an infinite product over all its zeros, $j_{\nu,k}$. The function is, in a deep sense, "made of" its zeros. This allows us to construct and manipulate new functions defined entirely by these zeros. For instance, we can build a function whose zeros are the zeros of $J_\nu(\sqrt{z})$ and whose poles (infinities) are the zeros of $J_{\nu+1}(\sqrt{z})$. By using the infinite product representations for both, we can discover that this complicated product simplifies down to a neat ratio of the original Bessel functions themselves.

Even more elegantly, the zeros act as singularities whose effects can be perfectly cancelled. Consider two different functions, one built from a sum involving the squares of the Bessel zeros ($F(z) = \sum_{n=1}^\infty (z^2 - j_{\nu,n}^2)^{-1}$), and another built from a ratio of Bessel functions ($G(z) = J_{\nu+1}(z) / (z J_{\nu}(z))$). Both of these functions have poles—they blow up to infinity—at every single zero of $J_\nu(z)$. Yet, a remarkable thing happens. If you combine them in just the right way—by taking $2F(z) + G(z)$, for example—the residues at every single pole cancel out perfectly. The resulting function is entire, smooth, and well-behaved everywhere. It's a beautiful act of mathematical harmony, where an infinite number of infinities are simultaneously tamed by choosing a specific ratio, a ratio that is a simple integer, 2.

From the tone of a drum to the energy of an electron, from the capacity of the internet to the deep structure of analysis, the zeros of Bessel functions are an inescapable feature of our world. They are not merely answers to a boundary value problem; they are a fundamental part of the mathematical language that nature speaks, revealing a beautiful and unexpected unity across seemingly disparate fields of science and engineering.