Lundquist Number

In the vast expanse of the cosmos and within the fiery hearts of experimental fusion reactors, over 99% of visible matter exists as plasma—a superheated gas of charged particles intertwined with magnetic fields. A core concept in plasma physics is that these magnetic fields are "frozen-in" to the fluid's motion, carried and twisted by the plasma as if they were one. Yet, this simple picture fails to explain some of the most energetic events in the universe, such as solar flares, where magnetic energy is released with explosive speed. This discrepancy highlights a fundamental tension between the plasma's motion that tangles the field and the field's inherent tendency to "slip" and simplify itself.

This article delves into the Lundquist number, the critical dimensionless parameter that quantifies this battle. It serves as the definitive scorekeeper, telling us when the magnetic field is unshakably bound to the plasma and when it can break free. We will first explore the principles and mechanisms behind the Lundquist number, deriving it from the fundamental equations of magnetohydrodynamics and uncovering its profound connection to the timescales that govern plasma behavior. Following this, we will examine its crucial applications and interdisciplinary connections, revealing how this single number explains the paradox of fast magnetic reconnection in solar flares, guides the design of fusion energy devices, and poses one of the greatest challenges in modern computational physics.

Imagine magnetic field lines as immensely flexible, yet powerful, elastic bands threaded through a conducting fluid like a plasma. The plasma, in its motion, drags these elastic bands along with it. This is the essence of being ​​frozen-in​​. If the plasma swirls into a vortex, the field lines are twisted up with it. If it expands, the field lines are stretched and carried outward. This is a wonderfully powerful idea, and it describes a huge range of phenomena in stars, galaxies, and fusion experiments.

But this isn't the whole story. The field lines are not perfectly glued to the plasma. They have a tendency to "slip" or "diffuse" through the fluid, seeking a simpler, lower-energy state. Think of it like a rope slowly slipping through a loose knot. This slipping is a consequence of the plasma having a finite electrical ​​resistivity​​. If the plasma were a perfect conductor (zero resistivity), the field lines would be perfectly frozen-in forever. But in the real universe, every plasma has some resistivity, however small.

This sets up a fundamental tension, a battle that plays out across the cosmos. On one side, the bulk motion of the plasma tries to advect the magnetic field, tying it into complex knots. On the other side, resistivity works to un-tie these knots, allowing the field to diffuse and relax. The entire story of dynamic plasmas—from the gentle heating of the solar corona to the violent eruption of a solar flare—is governed by the ebb and flow of this battle.

To understand this battle, we need to look at its rulebook: the ​​induction equation​​. In its simplest form, it looks like this:

Here, $\mathbf{B}$ is the magnetic field, $\mathbf{V}$ is the plasma velocity, and $\eta$ is the magnetic diffusivity (which is just the resistivity in proper units). The first term on the right describes how the velocity field $\mathbf{V}$ stretches and carries the magnetic field $\mathbf{B}$—this is the frozen-in part. The second term describes how the field diffuses or smooths itself out due to resistivity—this is the slippery part. The question that defines almost all of magnetohydrodynamics is: which of these terms wins?

How do we keep score in this battle? Physicists have a wonderful trick for this: dimensionless numbers. By comparing the characteristic size, or magnitude, of the two competing terms, we can create a single number that tells us the state of play. Let's estimate the size of each term. If our system has a characteristic size $L$ and a characteristic velocity $V$, the advection term scales roughly as $VB/L$, while the diffusion term scales as $\eta B/L^2$.

The ratio of the advection term to the diffusion term gives us our scorekeeper:

This is the famous ​​magnetic Reynolds number​​, $R_m$. It is the general-purpose measure of flux-freezing. If $R_m$ is enormous, say a million, it means advection is a million times stronger than diffusion. The field is, for all practical purposes, perfectly frozen-in. If $R_m$ is small, say $0.01$, diffusion dominates, and the magnetic field slips through the plasma with ease, ignoring the fluid's motion.

This is a great start, but in plasma physics, we can be more specific. The "characteristic velocity" $V$ is a bit vague. What speed is most characteristic of magnetic phenomena? The answer is the speed at which magnetic information itself propagates. In a plasma, this is the ​​Alfvén speed​​, $v_A$. It's determined by the strength of the magnetic field $B$ and the density of the plasma $\rho$: $v_A = B / \sqrt{\mu_0 \rho}$. It represents the speed of a wave traveling along a magnetic field line, like plucking a guitar string. This is the natural, intrinsic speed for magnetic reconfigurations.

When we take the general-purpose magnetic Reynolds number and use this very special, physically significant velocity, we create a new dimensionless number with profound importance: the ​​Lundquist number​​, $S$.

The Lundquist number is not just another variable; it's a specialized tool. It specifically compares the rate of resistive diffusion to the natural dynamical timescale of the magnetic field itself. It's the perfect parameter for studying phenomena where the magnetic field's own energy and tension are the main drivers, such as magnetic reconnection and instabilities.

There's another, perhaps more beautiful, way to understand the Lundquist number. Let's think about timescales.

How long would it take for a magnetic field structure of size $L$ to simply fade away due to resistivity if the plasma were stationary? From the diffusion term in the induction equation, we can deduce this ​​resistive diffusion time​​, $\tau_R \sim L^2/\eta$. For large objects with low resistivity (like a star), this time can be billions of years.

Now, how long does it take for a magnetic disturbance to cross the system? This is simply the size $L$ divided by the signal speed, $v_A$. This is the ​​Alfvén transit time​​, $\tau_A = L/v_A$. This is the fundamental "heartbeat" of the system, the time it takes for one part of the magnetic structure to communicate with another.

Let's look at the ratio of these two times:

So, the Lundquist number is nothing more than the ratio of the time it takes for the field to diffuse away to the time it takes for the field to rearrange itself dynamically. If $S = 10^{12}$, as it might be in a solar flare, it means the magnetic field can rearrange itself a trillion times in the time it would take for it to decay on its own. The plasma is incredibly "ideal"—that is, its magnetic field is almost perfectly frozen-in.

This immense size of $S$ in most astrophysical and fusion plasmas leads to a profound puzzle. Consider two bundles of magnetic field lines pointing in opposite directions, pushed together. Because the field is so strongly frozen-in ($S \gg 1$), they can't simply pass through each other. They pile up, forming an intensely concentrated ​​current sheet​​ at the boundary.

In this thin layer, the magnetic field gradient is enormous. Even with a tiny resistivity $\eta$, the diffusion term $\eta \nabla^2 \mathbf{B}$ can become large enough to compete with advection. Here, and only here, can the magnetic field lines break their "frozen-in" bonds, "slip" through the plasma, and reconfigure into a new, lower-energy state. This process is ​​magnetic reconnection​​. It is the fundamental mechanism that releases stored magnetic energy, powering everything from solar flares to disruptions in fusion tokamaks.

A simple model, known as the ​​Sweet-Parker model​​, gives a prediction for how fast this happens. By balancing mass flow and magnetic diffusion in the thin layer, one finds that the thickness of the layer, $\delta$, and the speed of the inflow of plasma, $v_{in}$, both depend on the Lundquist number:

For $S = 10^{12}$, this model predicts a reconnection rate of $10^{-6}$ times the Alfvén speed. This is catastrophically slow. A solar flare that we observe to happen in minutes would, according to this theory, take months to unfold. This blatant contradiction between theory and observation was a major crisis in plasma physics, often called the "reconnection problem". The Lundquist number, the very measure of ideality, had led us to a paradox.

What did the simple Sweet-Parker model miss? It assumed that the long, razor-thin current sheet was stable. But is it? A key insight of modern plasma physics is that it is not. A highly elongated current sheet is violently unstable to a tearing-like instability.

To understand why, we must think locally. The stability of the current sheet itself doesn't depend on the global Lundquist number $S$, but on a local Lundquist number based on the sheet's own thickness, $S_\delta = \delta v_A / \eta$. Let's substitute the Sweet-Parker scaling for the thickness, $\delta \sim L S^{-1/2}$:

This is a remarkable result. It means that the current sheet becomes more unstable as the global system becomes more ideal. As $S$ increases, the sheet gets thinner, and its local Lundquist number $S_\delta$ grows, making it more and more prone to tearing apart.

This instability, known as the ​​plasmoid instability​​, has a threshold. When the Lundquist number $S$ exceeds a ​​critical value​​, $S_c$, the growth time of the instability becomes faster than the time it takes for plasma to be flushed out of the sheet. At this point, the sheet shatters into a chaotic chain of magnetic islands, or "plasmoids". Theoretical work and extensive computer simulations have shown that this critical value is surprisingly universal: $S_c \approx 10^4$.

Since virtually all astrophysical plasmas have $S \gg 10^4$, this means that smooth, stable Sweet-Parker current sheets simply do not exist in nature. Instead, reconnection is a violent, multi-scale, chaotic process. This instability provides a pathway for fast reconnection, resolving the paradox. The Lundquist number, which created the puzzle, also held the key to its solution.

Our story so far has been told in the language of fluid dynamics (MHD). But a plasma is made of individual particles—ions and electrons. The fluid approximation is only valid as long as the scales we are considering are much larger than the characteristic scales of these particles' motions.

What happens if the current sheet in our reconnection model gets so thin that its thickness $\delta \sim L S^{-1/2}$ becomes comparable to a fundamental kinetic scale, like the ​​ion inertial length​​, $d_i$? The ion inertial length, $d_i = \sqrt{m_i / (\mu_0 n e^2)}$, is the scale at which ions, due to their inertia, begin to decouple from the motion of the much lighter electrons and the magnetic field.

When $\delta \approx d_i$, the single-fluid MHD model breaks down. We enter a new regime of reconnection governed by two-fluid or fully kinetic physics. We can calculate the critical Lundquist number, $S_c$, at which this transition occurs by setting the Sweet-Parker thickness equal to the ion inertial length:

For a typical solar corona environment, this critical number can be enormous, around $10^{12}$. This gives us a fascinating map of reconnection physics. For $S \lt 10^4$, reconnection is slow (Sweet-Parker). For $10^4 \lt S \lt (L/d_i)^2$, reconnection is fast and dominated by the resistive plasmoid instability. For $S \gt (L/d_i)^2$, reconnection is fast and dominated by collisionless, kinetic effects. The Lundquist number serves as our guide through this complex landscape.

Let's bring this story home, from the distant sun to a laboratory on Earth—a ​​tokamak​​ fusion device. In the complex, donut-shaped magnetic fields of a tokamak, a strange thing happens. Due to the way the magnetic field strength varies, some electrons become "trapped". They can't travel all the way around the torus, but instead bounce back and forth in a limited region, like a ball bouncing between two hills.

These trapped electrons cannot contribute to carrying a steady electrical current along the magnetic field lines. The result? The effective number of charge carriers is reduced. This means the plasma's parallel electrical resistivity, $\eta_\parallel$, is increased compared to what it would be in a simple, uniform magnetic field.

Now think about the Lundquist number, $S \propto 1/\eta_\parallel$. If resistivity goes up, the Lundquist number goes down. In fusion, we often fight against resistive instabilities like ​​tearing modes​​, which can rip apart the magnetic surfaces and let the hot plasma escape. The growth rate, $\gamma$, of these modes often scales inversely with some power of $S$, for example, $\gamma \propto S^{-3/5}$. This means a higher $S$ (lower resistivity) leads to a more stable plasma.

But the trapped particles have lowered our effective $S$. This leads to a beautifully counter-intuitive conclusion: the neoclassical effect of trapped particles, by increasing resistivity, makes the plasma more susceptible to resistive instabilities. Understanding this subtle interplay, all captured by the Lundquist number, is absolutely critical for designing a stable, successful fusion reactor. From the grand scale of the cosmos to the intricate design of a fusion machine, the Lundquist number remains our indispensable guide to the beautiful and complex dance of plasma and magnetic fields.

Now that we have acquainted ourselves with the principles of the Lundquist number, let us embark on a journey to see where this simple ratio of timescales becomes more than a theoretical curiosity. Where does it become a central character in the story of our universe? The answer, it turns out, is almost everywhere a plasma is found—which is to say, almost everywhere. The Lundquist number, $S$, acts as a master dial, setting the character of the plasma from the fiery heart of our Sun to the intensely confined fuel in our fusion experiments on Earth. A large value of $S$ means the magnetic field is "frozen-in," shackled to the plasma fluid, and for the most part, this is the rule in the cosmos.

The "frozen-in" picture is beautiful, but it presents a puzzle. If magnetic field lines are forever tied to the plasma they thread, how can they ever break, rearrange, and change their connections? How can the immense energy stored in the twisted and sheared magnetic fields of a star ever be released? Nature, in its ingenuity, has an answer: magnetic reconnection. This process finds the tiny cracks in the seemingly impenetrable armor of ideal magnetohydrodynamics (MHD), allowing the magnetic field to slip through the plasma, release its energy, and form new configurations.

The simplest model for this process, known as Sweet-Parker reconnection, imagines a thin layer where oppositely directed magnetic fields are forced together. Here, resistivity, however small, becomes important and allows the field lines to diffuse and annihilate. Through simple scaling arguments, we can discover that the rate at which plasma is drawn into this layer, $v_{in}$, is agonizingly slow in a highly conducting plasma. It is governed by the Lundquist number, scaling as $v_{in} \sim v_A / \sqrt{S}$, where $v_A$ is the characteristic Alfvén speed.

Now, let us turn our gaze to a solar flare, one of the most spectacular explosions in our solar system. In the Sun's tenuous outer atmosphere, the corona, a typical plasma might have a temperature of a million Kelvin and a density of $10^{15}$ particles per cubic meter, threaded by a magnetic field spanning tens of thousands of kilometers. When we use these representative parameters to calculate the plasma's resistivity and Alfvén speed, we are confronted with a colossal Lundquist number, easily reaching $10^{12}$ or more.

What does our simple Sweet-Parker model predict for this situation? With $S \sim 10^{12}$, its square root is a still-enormous $10^6$. The predicted reconnection speed is thus a millionth of the Alfvén speed. If we calculate the time it would take to reconnect a large magnetic structure and power a flare, the answer is not the minutes or hours we observe, but weeks or even months! This is a beautiful catastrophe. Our simple, elegant model has made a prediction that is wildly, spectacularly wrong.

This is where physics gets truly exciting. A wrong prediction is not a failure but a signpost, pointing toward new, richer phenomena. The flaw in our reasoning was the assumption that the long, thin reconnection layer was stable. What happens when you stretch any sheet—of paper, of rubber, of plasma current—incredibly thin? It tears.

At the immense Lundquist numbers found in the corona, the Sweet-Parker current sheet becomes violently unstable to what is known as the "plasmoid instability." The sheet shatters into a dynamic, chaotic chain of magnetic islands, or "plasmoids." The growth rate of this instability is itself driven by the Lundquist number, scaling roughly as $\gamma \propto S^{1/4}$ or $\gamma \propto S^{1/5}$, depending on the specific model assumptions. The higher the value of $S$, the more explosively the sheet tears itself apart. This turbulent process is vastly more efficient at reconnecting magnetic fields and provides a viable path to explain the rapid energy release we witness in solar flares and other cosmic explosions. The Lundquist number, which at first seemed to forbid fast reconnection, turns out to be the very ingredient that triggers the instability making it possible.

Let us now journey from the heavens to the laboratory, where humankind is striving to build a star on Earth. In a tokamak, a donut-shaped magnetic bottle designed to confine a plasma hotter than the sun's core, we are also deep in the realm of high Lundquist numbers. For a typical large fusion device, with its powerful multi-Tesla magnetic fields and dense, hot plasma, the Lundquist number is enormous, routinely reaching values from $10^7$ to $10^{11}$.

This immediately tells us something profound: for the most part, the plasma should behave ideally. The magnetic field lines form beautiful, nested surfaces that confine the hot fuel, and resistivity is a minor, slow-acting nuisance. However, these machines can be afflicted by sudden, violent instabilities called Edge Localized Modes, or ELMs, which can dump a crippling amount of energy onto the reactor wall in the blink of an eye. To understand and control them, we must know their nature. Are they a resistive process, a slow tearing of the magnetic structure, or are they an ideal process, a rapid buckling of the entire field?

The Lundquist number provides the answer. An ELM crash occurs on the timescale of microseconds. The resistive diffusion time in the tokamak's edge pedestal, given its Lundquist number of $S \sim 10^{11}$, is on the order of hours. On the rapid timescale of an ELM, resistivity simply does not have time to act. The crash must be an ideal MHD instability. This fundamental insight, guided by the magnitude of $S$, directs the entire theoretical and computational effort to predict and mitigate these dangerous events.

The story continues in other magnetic confinement concepts, such as Field-Reversed Configurations (FRCs). When scientists merge two FRCs to create a hotter, more stable plasma, a reconnection layer is formed at the interface. For typical experimental parameters, the Lundquist number in this layer might be around $5 \times 10^4$. This number is fascinating because it sits near the critical threshold (theoretically predicted to be around $S_c \sim 10^4$) where the plasmoid instability is expected to take hold. By calculating $S$, experimentalists can predict whether their reconnection process will be a slow, smooth, collisional affair or a dynamic, turbulent event filled with plasmoids. The Lundquist number thus becomes a direct, predictive tool for designing and interpreting experiments.

We have seen that nature delights in creating plasmas with enormous Lundquist numbers. But what of our attempts to imitate nature in a computer? Here, we run into a profound difficulty, a tyranny imposed by the very largeness of $S$.

Think back to the Sweet-Parker current sheet. Its thickness, $\delta$, shrinks with the Lundquist number, as $\delta \sim L/\sqrt{S}$. To accurately simulate this layer, our computational grid must have cells that are smaller than $\delta$. If we want to simulate a solar corona region with $S=10^{12}$, we would need to resolve a physical layer that is a millionth of the size of the overall system! A simulation attempting to resolve this directly would require an astronomical number of grid points, far beyond the capacity of any conceivable computer.

But the problem is even deeper. The time step of a simulation is limited by the fastest processes, which are typically Alfvén waves. The physical process of reconnection, however, evolves over a much longer timescale, which itself scales with $S$. This means that to simulate one full reconnection event, we must take an enormous number of tiny time steps. A careful analysis shows that, due to the dual constraints on grid resolution and the total number of time steps we can afford, the maximum Lundquist number we can truly simulate is often only a few hundred or perhaps a few thousand.

Compare this achievable $S_{sim} \sim 10^3$ with the Sun's realistic $S_{real} \sim 10^{12}$. This vast, unbridgeable chasm is famously known as the "Lundquist number problem" in computational plasma physics. It means we cannot simply simulate a solar flare or a tokamak disruption from first principles. Instead, physicists must be clever, developing "anomalous resistivity" or other "sub-grid" models that are designed to mimic the effects of the unresolved physics. The Lundquist number is therefore not just a descriptor of nature, but also a formidable barrier that shapes the very strategy of modern computational science.

The Lundquist number does not act in isolation. It is one member of a whole family of dimensionless numbers that orchestrate the physics of fluids and plasmas. Just as $S$ compares ideal motion to resistive diffusion, the Reynolds number compares inertia to viscosity, and the Hartmann number pits viscous forces against magnetic forces. The complete behavior of a plasma in a star or a fusion device is a complex symphony conducted by these competing effects. Is the plasma resistive? Is it viscous? Is gravity important? The answers are not simple "yes" or "no"s, but are found in the relative magnitudes of $S$, the Hartmann number $Ha$, and their brethren. To understand the physics is, in a profound sense, to understand these ratios. The Lundquist number is a particularly brilliant character in this cast—a single number that tells a rich story of stability, instability, cosmic explosions, and our quest for a star on Earth.