Landau Critical Velocity

Superfluidity, the ability of a fluid to flow without any viscosity or energy loss, is one of the most striking phenomena in quantum mechanics. It defies our classical intuition about friction. But this remarkable state is not absolute; push a superfluid too fast, and its frictionless nature abruptly vanishes. This raises a fundamental question: what is the speed limit for superfluidity, and what physical principle governs this breakdown? The answer lies in a beautifully simple energetic argument first proposed by the brilliant physicist Lev Landau.

This article delves into the concept of the Landau critical velocity, the theoretical threshold separating frictionless flow from dissipative motion. We will unpack the core ideas behind this criterion, revealing how the secret to superfluidity is encoded in the energy spectrum of a fluid's quantum excitations. The "Principles and Mechanisms" chapter will derive Landau's famous formula and explore how different types of excitations—namely phonons and the more exotic rotons—dictate the critical velocity in systems like Bose-Einstein condensates and liquid helium. Following this, the "Applications and Interdisciplinary Connections" chapter will showcase the principle's remarkable universality, demonstrating its relevance across a vast landscape of quantum systems, from tabletop atomic gases and superfluids in constrained geometries to the dense nuclear matter at the heart of neutron stars.

Imagine trying to stir a cup of honey. It's thick, viscous, and resists your spoon. You have to constantly expend energy to keep the spoon moving. This resistance is friction. Now imagine stirring a cup of superfluid helium. Your spoon would move almost without effort, as if through a perfect vacuum. This is the marvel of superfluidity: flow without friction. But where does friction come from in the first place, and how does a superfluid manage to avoid it?

The great physicist Lev Landau proposed an answer of beautiful simplicity. It all comes down to energy. For a moving object to slow down, it must transfer some of its kinetic energy to the fluid. It does this by "kicking" the fluid, creating a tiny ripple of energy—what physicists call an ​​elementary excitation​​ or a ​​quasiparticle​​. But here's the quantum catch: you can't create just any amount of ripple. The energy of these excitations comes in discrete packets, or quanta. Creating an excitation costs a specific amount of energy, $\epsilon$, and requires transferring a specific amount of momentum, $p$.

Landau's genius was to analyze this process from the fluid's point of view. Consider a superfluid flowing at a velocity $\mathbf{v}$. If an excitation with energy $\epsilon(p)$ and momentum $\mathbf{p}$ is created, the total energy of the flowing fluid changes by $\Delta E = \epsilon(p) + \mathbf{p} \cdot \mathbf{v}$. For the flow to lose energy and slow down (i.e., for friction to occur), this change must be negative or zero: $\Delta E \le 0$.

To see when this is possible, let's consider the worst-case scenario for the superfluid. The term $\mathbf{p} \cdot \mathbf{v}$ is most negative when the excitation's momentum $\mathbf{p}$ points directly opposite to the flow velocity $\mathbf{v}$. In this case, the condition for dissipation becomes $\epsilon(p) - pv \le 0$, which can be rearranged to $v \ge \frac{\epsilon(p)}{p}$.

This is the punchline. Dissipation can only happen if the flow velocity $v$ is fast enough to "afford" the creation of at least one type of excitation. As long as the velocity is below this threshold for all possible excitations, the flow remains perfectly frictionless. The breakdown of superfluidity occurs at the "weakest link"—the specific excitation that is easiest to create. This threshold velocity is the legendary ​​Landau critical velocity​​, $v_c$:

This elegant formula is our master key. It tells us that the entire secret to a fluid's superfluidity is locked within a single function, $\epsilon(p)$, which governs the energy of its excitations.

So what is this mysterious function $\epsilon(p)$? It's known as the ​​dispersion relation​​, and you can think of it as the fundamental rulebook of a quantum fluid. For any given momentum $p$, it tells you the energy cost $\epsilon$ to create an excitation. Every fluid—be it liquid helium, a neutron star, or a cloud of ultracold atoms—has its own unique rulebook, its own $\epsilon(p)$ curve, determined by the intricate quantum dance of its constituent particles.

To find the critical velocity, we don't need to know every microscopic detail of this dance. We just need the rulebook. The problem of finding $v_c$ becomes a beautifully simple geometric one: plot the energy $\epsilon$ versus the momentum $p$. The ratio $\epsilon(p)/p$ is the slope of a line drawn from the origin of this graph to the point $(p, \epsilon(p))$ on the curve. The Landau critical velocity is simply the shallowest possible slope this line can have while still touching the curve.

Let's begin with the simplest possible rulebook. What if the excitations are just ordinary sound waves? In the quantum realm, we call these packets of sound energy ​​phonons​​. For a phonon, the energy is directly proportional to its momentum: $\epsilon(p) = c_s p$, where $c_s$ is the constant speed of sound. The graph of $\epsilon(p)$ is just a straight line starting from the origin with a slope of $c_s$.

What is the critical velocity here? The ratio $\epsilon(p)/p$ is simply $\frac{c_s p}{p} = c_s$. Since this value is constant for all $p$, its minimum is just $c_s$ itself. So, for a fluid whose only excitations are phonons, the Landau critical velocity is precisely the speed of sound. An object moving faster than sound can create sonic booms (phonons), dissipating energy; an object moving slower cannot.

This isn't just a hypothetical scenario. A modern marvel of physics, the ​​Bose-Einstein Condensate​​ (BEC)—a state of matter where millions of atoms behave as a single quantum entity—follows this rulebook almost perfectly. The elementary excitations in a weakly-interacting BEC are described by the Bogoliubov dispersion relation. While the full formula, $\epsilon(p) = \sqrt{(\frac{p^2}{2m})^2 + 2gn(\frac{p^2}{2m})}$, might seem intimidating, if you analyze the ratio $\epsilon(p)/p$, you find that its minimum value occurs as the momentum $p$ approaches zero. And what is this minimum value? It's exactly the speed of sound in the condensate, $c_s = \sqrt{gn/m}$. This speed is also beautifully connected to another fundamental property of a BEC, the ​​healing length​​ $\xi$, which represents the characteristic distance over which the condensate can "heal" itself from a local disturbance.

Liquid helium-4, the original and most famous superfluid, plays by a more complicated set of rules. When its dispersion relation was painstakingly mapped out using neutron scattering experiments, it revealed a surprise. After the initial, linear phonon region (where $\epsilon \propto p$), the curve doesn't continue rising indefinitely. Instead, it bends over, dips down to a local minimum at a finite momentum, and then rises again. Landau, with his unparalleled physical intuition, named the excitations that live in this dip ​​rotons​​.

This dip is the crucial feature that makes helium-4 so special. It represents a "soft spot" in the fluid's ability to resist dissipation. We can model the energy of excitations near this minimum with a simple parabolic form:

Here, $p_0$ is the momentum at which the dip occurs, $\Delta$ is the minimum energy required to create an excitation in this region (the "roton gap"), and $\mu$ is a parameter called the effective mass.

What does this dip mean for our critical velocity? Geometrically, it means we can now draw a line from the origin that is tangent to the curve in the roton region, and this line will have a shallower slope than the initial phonon line. A shallower slope means a smaller value of $\epsilon(p)/p$, and thus a lower critical velocity! The rotons provide a "cheaper" way to create an excitation than the phonons do. To find this new, lower critical velocity, one performs a straightforward calculus exercise to find the momentum $p$ that minimizes the ratio $\frac{1}{p}(\Delta + \frac{(p-p_0)^2}{2\mu})$. The result is a precise prediction for the critical velocity limited by roton creation.

For a complete picture of liquid helium, we must consider both possibilities. The fluid can dissipate energy by creating either a phonon or a roton. The actual Landau critical velocity will be determined by whichever process is easier—that is, whichever requires a lower flow velocity. We have two candidates for this "weakest link":

1. The velocity required to create a phonon: $v_{ph} = c_s$ (the speed of sound).
2. The velocity required to create a roton: $v_{rot}$, determined by the minimum of $\epsilon(p)/p$ in the roton region.

For liquid helium-4, it turns out that the roton minimum provides the path of least resistance. A simple but excellent approximation for the roton critical velocity is just the ratio of the gap energy to the roton momentum, $v_c \approx \frac{\Delta}{p_0}$. Using the experimentally measured values for helium ($\Delta \approx 1.2 \times 10^{-22}$ J and $p_0 \approx 2.0 \times 10^{-24}$ kg⋅m/s), this gives a critical velocity of about $59$ m/s. This is significantly lower than the speed of sound in liquid helium (about 240 m/s). The rotons are indeed the Achilles' heel of superfluidity in bulk helium.

You might think this roton business is just a strange quirk of liquid helium. But one of the most profound pursuits in physics is the search for universal patterns. The roton is no exception. It turns out that similar dips in the dispersion curve can be engineered in other quantum fluids, such as Bose-Einstein condensates with special long-range interactions. The presence of such a "roton-like" minimum is a universal signpost: it indicates that the system's frictionless flow can break down at a velocity below the speed of sound.

But here comes the kind of reality check that makes physics so endlessly fascinating. If you perform an experiment and try to flow superfluid helium through a very narrow tube, you will find that its superfluidity breaks down not at 59 m/s, but at speeds of just a few centimeters per second! Was Landau's celebrated theory wrong?

Not at all. The theory is perfectly correct, but it answers a specific question: what is the critical velocity for creating quasiparticle excitations in the bulk of the fluid? In the real world, especially in a confined geometry like a thin capillary, there is often another, energetically cheaper way to dissipate energy: by creating ​​quantized vortices​​. These are like microscopic whirlpools in the quantum fluid. Richard Feynman showed that the critical velocity for creating these vortices depends on the width of the channel. For a tube just 50 nanometers wide, the predicted Landau velocity is nearly 30 times larger than the velocity needed to spawn a vortex!

This doesn't invalidate Landau's principle; it enriches our understanding. The critical velocity you observe depends on the conditions of your experiment. In the vast, open ocean of bulk helium, creating a roton is the limiting factor. In the tight confines of a narrow channel, the spontaneous formation of a vortex takes over. And all of this is for an idealized system at absolute zero temperature. In any real experiment, thermal energy creates a background gas of excitations, which can further modify and typically lower the observed critical velocity.

Landau's criterion gives us the fundamental energetic speed limit for a pristine quantum fluid. But the real world is a wonderfully messy place, filled with boundaries, imperfections, and thermal noise. These features provide alternative pathways for dissipation, revealing that the beautiful theory of quasiparticles is just the first, essential chapter in the grand, ongoing story of superfluidity.

Having grasped the elegant principle behind Landau's criterion—that superfluidity is a quantum conspiracy to make the creation of excitations energetically unfavorable—we can now embark on a journey to see this principle at work. You might think that such a simple rule, $v \min(\epsilon(p)/p)$, would lead to a monotonous story. But you would be wrong! The true magic, the beauty, and the diversity of the quantum world are hidden in the character of those elementary excitations, $\epsilon(p)$. Each quantum fluid, from a wisp of cold atoms to the crushing density of a neutron star, sings its own unique song, a spectrum of allowed energies. The Landau critical velocity is simply the lowest "pitch" in that symphony. Let us listen to a few of these tunes.

Perhaps the most intuitive and common scenario is found in weakly interacting Bose-Einstein condensates (BECs). In these remarkable systems, millions of atoms lose their individuality and behave as a single quantum entity. If you were to gently poke this fluid, what kind of ripples would you create? At the lowest energies, the excitations are simply sound waves—phonons—ripples of density propagating through the condensate. For these long-wavelength disturbances, the energy is directly proportional to the momentum: $\epsilon(p) = c_s p$, where $c_s$ is the speed of sound.

What, then, is the Landau critical velocity? Applying the criterion, we must find the minimum of $\epsilon(p)/p = c_s p / p = c_s$. The ratio is constant! Thus, the critical velocity is simply the speed of sound itself, $v_c = c_s$. This result is wonderfully intuitive: to create a sound wave by moving through a medium, you must necessarily move faster than the speed of sound. Below this speed, the fluid simply cannot respond by creating a phonon; it is energetically forbidden.

This fundamental connection holds true whether the condensate is in three dimensions, confined to a two-dimensional plane, or even strung out along a one-dimensional line. In each case, the specific value of the speed of sound depends on the details—the particle mass $m$, the density $n$, and the interaction strength $g$—but the principle remains unshakable. For a simple Bose fluid, the barrier to dissipation is set by the speed of its own whispers.

So far, we have imagined our superfluid in a vast, featureless expanse. But what happens if we confine it, forcing it to live within a specific geometry? Consider, for instance, a BEC spread thinly over the surface of a sphere. Now, the game changes. On a finite sphere, you cannot create a wave of any arbitrary wavelength. The curvature and boundary conditions impose a strict quantization; only a discrete set of modes, analogous to the harmonics on a guitar string, are allowed.

This means that momentum $p$ is no longer continuous. There is a smallest possible non-zero momentum, $p_1$, corresponding to the lowest-energy "sloshing" mode of the condensate on the sphere. You cannot create an excitation with less momentum than this. The minimum of $\epsilon(p)/p$ is therefore not found by taking the limit $p \to 0$, but by evaluating the ratio at this first allowed momentum, $p_1$. The critical velocity now depends not only on the intrinsic properties of the fluid (like density and interaction strength) but also on the radius of the sphere, $R$. The superfluid, in a very real sense, feels the size and shape of its universe. This is a beautiful marriage of quantum mechanics and geometry.

The story gets even more intricate and fascinating when we turn from bosons to fermions. In fermionic systems like electrons in a superconductor or liquid Helium-3, superfluidity arises from the formation of Cooper pairs. The crucial new feature is the appearance of a superfluid energy gap, $\Delta$. This is the minimum energy required to break a Cooper pair and create a quasiparticle excitation.

This gap fundamentally changes the low-energy landscape. In a simple s-wave fermionic superfluid, like the one found in the crust of a neutron star or in a unitary Fermi gas of cold atoms, the excitation energy is given by a relation of the form $E(p) = \sqrt{(\epsilon_p - \mu)^2 + \Delta^2}$, where $\epsilon_p$ is the single-particle energy and $\mu$ is the chemical potential. Unlike the bosonic phonon spectrum, this energy does not go to zero as $p \to 0$. It has a minimum value of $\Delta$ right at the Fermi momentum.

The resulting critical velocity, often called the pair-breaking velocity, is no longer the speed of sound. Instead, it is determined by a delicate balance between the energy gap $\Delta$ and the Fermi momentum $p_F$, scaling roughly as $v_c \sim \Delta/p_F$. Think about the implications: deep within a spinning neutron star, a city-sized ball of nuclear matter, quantum mechanics dictates that its neutron superfluid can flow without friction, helping to explain why these stellar corpses spin for so long. The Landau criterion is at work in the heavens!.

The plot thickens further in exotic superfluids like the A-phase of Helium-3. Here, the Cooper pairs have internal orbital angular momentum, leading to an anisotropic energy gap. The gap vanishes at two "poles" on the Fermi surface. These nodes act as weak points. It is far easier to create an excitation near these nodes than elsewhere. Consequently, the critical velocity becomes dependent on the direction of flow relative to the orientation of the Cooper pairs. Superfluidity is more fragile in certain directions, a macroscopic manifestation of the intricate, microscopic dance of the paired fermions.

Is the story always about phonons or breaking pairs? Not at all. Nature is more imaginative. In his groundbreaking work on liquid Helium-4, Feynman realized that the excitation spectrum $\epsilon(p)$ was more complex than a simple linear curve. After an initial linear (phonon) region, the curve dips to form a local minimum, called a "roton," before rising again. This roton minimum represents a different kind of short-wavelength excitation, almost like a tiny quantum vortex ring.

This feature can have a dramatic effect on superfluidity. If this dip is pronounced enough, the line from the origin with the shallowest slope might not touch the curve at $p=0$, but rather be tangent to the roton minimum. In such a case, it is the rotons, not the phonons, that are the "cheapest" excitations to create, and they set the critical velocity. A modern analogue can be found in condensates of exciton-polaritons, quasiparticles of light and matter, whose complex interactions can be described by a phenomenological dispersion relation featuring a roton-like minimum.

Finally, what if a system has internal degrees of freedom, like spin? Consider a ferromagnetic BEC, where all atomic spins are aligned. This fluid supports not one, but two types of elementary excitations. The first are the familiar phonons (density waves). The second are "magnons"—spin-flip waves, where one atom's spin is locally tilted against the ferromagnetic background. An object moving through this fluid can lose energy by creating either a phonon or a magnon. Nature, ever economical, will choose the path of least resistance. The true Landau critical velocity is therefore the minimum of the critical velocities for each channel: $v_c = \min(v_{c, \text{phonon}}, v_{c, \text{magnon}})$. Superfluidity can be destroyed not by a ripple in density, but by a ripple in magnetism.

From the simple hum of sound in a BEC to the complex interplay of geometry, pairing anisotropy, and multiple excitation channels, the Landau criterion serves as a unifying lens. It reveals that the robust phenomenon of superfluidity is ultimately governed by the subtle and varied energetic landscape of the quantum world. The same principle applies to a vial of atoms in a lab and to the heart of a distant star, a testament to the profound power and unity of physics.