Non-Fermi Liquids

For decades, our understanding of metals has been anchored by Landau's Fermi liquid theory, a powerful framework that describes electrons as stable, particle-like entities called quasiparticles. However, a growing class of exotic materials, often dubbed "strange metals," defy this conventional picture, exhibiting bizarre properties that suggest the very concept of an electron-like particle is breaking down. This article confronts this profound puzzle in condensed matter physics, exploring the world that lies beyond the standard model of metals. By reading through, you will gain a comprehensive understanding of this fascinating topic.

We will first delve into the fundamental principles, exploring why quasiparticles fail and what mechanisms, such as quantum criticality, give rise to the non-Fermi liquid state. Subsequently, we will examine the experimental fingerprints these strange metals leave behind and uncover their unexpected and deep connections to fields ranging from materials science to the study of black holes, revealing a new, interconnected frontier of quantum matter.

To journey into the world of non-Fermi liquids, we must first understand the state of normalcy from which they so radically depart. Imagine a vast ballroom filled with dancers, each moving in a complex pattern, constantly interacting and avoiding one another. This is the electron sea in a metal—a chaotic, seething collective. In the early 20th century, physicists despaired of ever making sense of this mess. The breakthrough came from the great Soviet physicist Lev Landau, who had a brilliant insight: at low temperatures, this chaos organizes itself in a remarkable way.

Landau realized that an individual electron moving through the crowd creates a disturbance, a wake of other electrons being pushed and pulled. He proposed that we stop thinking about the bare electron and instead focus on a composite object: the electron plus its surrounding cloud of disturbance. He called this entity a ​​quasiparticle​​. In a sense, it is the ghost of a real particle, inheriting its properties but cloaked in the effects of the crowd. This quasiparticle has a well-defined momentum and charge, but its mass is different from a free electron's—it's "renormalized" by the effort of dragging its interaction cloud along.

For this elegant picture to hold, there's one crucial condition: the quasiparticle must be a reasonably stable object. It must live long enough to travel a few wavelengths before decaying into the incoherent slosh of the electron sea. In more formal terms, its energy uncertainty, $\Delta E$, must be much smaller than its own excitation energy, $E$. Since quantum mechanics tells us that the lifetime, $\tau$, is related to energy uncertainty by $\Delta E \approx \hbar/\tau$, this means the quasiparticle must have a long lifetime. This is the golden rule of what we now call a ​​Landau-Fermi liquid​​, the theory that beautifully describes almost all conventional metals, from copper to aluminum.

So, what happens if the quasiparticle dies too quickly? What if the ballroom of orderly dancers turns into a frantic mosh pit? In that scenario, an individual can't maintain a coherent path for even a moment. They are instantly lost in the chaos. The quasiparticle concept breaks down. This, in essence, is a ​​non-Fermi liquid​​.

Let's make this more precise. The "stability" of a quasiparticle with energy $\omega$ (measured relative to the sea's surface, the Fermi energy) can be judged by the ratio of its energy width, $\gamma$, to its energy, $\omega$. The width $\gamma$ represents the fuzziness in its energy, or equivalently, its decay rate. A sharp, well-defined particle requires this ratio to be small. For a true quasiparticle, the defining criterion is that this ratio must vanish as we look at excitations closer and closer to the quiet surface of the Fermi sea:

In a Fermi liquid, the Pauli exclusion principle provides a beautiful protection for quasiparticles. An excited quasiparticle has very few empty states to decay into, as most are already occupied. A careful calculation shows that this restriction leads to a decay rate that scales as $\gamma \propto \omega^2$. This is wonderful! The ratio $\gamma/|\omega|$ scales like $|\omega|$, which goes to zero as $\omega \to 0$. The closer a quasiparticle is to the Fermi energy, the more perfectly stable and long-lived it becomes.

A non-Fermi liquid is any metallic state where this fundamental rule is violated. For instance, in some of the most puzzling materials, experiments suggest that the decay rate scales only as $\gamma \propto |\omega|$. Here, the ratio $\gamma/|\omega|$ is a constant. The quasiparticle is never truly stable; it is born on the verge of death, a "marginal" entity at any energy scale. In other, even stranger cases, the decay rate might scale as $\gamma \propto |\omega|^{\alpha}$ with an exponent $\alpha 1$, meaning the quasiparticle becomes less stable relative to its energy as it gets closer to the Fermi sea—a complete inversion of the normal state of affairs. There is even a more catastrophic way for a quasiparticle to perish: the "quasiparticle residue," denoted $Z$, which measures how much of the original electron's character is left in the composite object, can be driven to zero by violent interactions. The particle-like aspect dissolves completely, leaving only an incoherent, collective excitation.

This theoretical distinction between stable and dying quasiparticles would be a mere curiosity if we couldn't see its consequences. Fortunately, the death of the quasiparticle leaves a set of dramatic and unmistakable fingerprints on the measurable properties of a material.

• ​​Anomalous Electrical Resistance:​​ Perhaps the most famous fingerprint is found in electrical resistance. The smooth flow of quasiparticles in a normal metal is primarily limited by two things: scattering off static impurities, which gives a constant residual resistance $\rho_0$, and scattering off each other. The latter process, in a Fermi liquid, gives a contribution that scales precisely as $A_{FL} T^2$. This $\rho(T) = \rho_0 + A_{FL} T^2$ law is a celebrated hallmark of metal physics. However, in many high-temperature superconductors and heavy-fermion compounds tuned to a special point, this law fails spectacularly. Instead, one often finds a resistivity that is perplexingly linear in temperature, $\rho(T) = \rho_0 + A_{qc} T$. This implies a scattering rate that is proportional to temperature itself, a profound departure from the $T^2$ behavior expected from stable quasiparticles. Other strange power laws, like $T^{3/2}$, have also been observed, each signaling a unique brand of non-Fermi liquid physics.

• ​​Anomalous Heat Capacity:​​ How much energy does it take to warm a metal? For a Fermi liquid, the electronic specific heat is perfectly linear in temperature, $C_V = \gamma_{FL} T$. This is a direct count of the number of available quasiparticle states within an energy shell of width $k_B T$ of the Fermi surface. In non-Fermi liquids, this tidy accounting breaks down. The specific heat coefficient, $\gamma = C_V/T$, which is a constant in normal metals, is often seen to diverge as the temperature is lowered. It might grow logarithmically, as $C_V/T \propto \ln(T_0/T)$, or as a power law, $C_V/T \propto T^{\beta-1}$ with $\beta 1$. This implies a massive pile-up of strange, non-quasiparticle excitations at low energies, far more than a Fermi liquid could ever accommodate.

• ​​Direct Spectroscopic Evidence:​​ With modern techniques like Angle-Resolved Photoemission Spectroscopy (ARPES), we can essentially take a direct photograph of the electrons' energy and momentum distribution. For a conventional metal, ARPES reveals a sharp, bright peak corresponding to the well-defined quasiparticle. As predicted, the width of this peak, $\Delta_A(\omega)$, shrinks rapidly as the energy $\omega$ approaches the Fermi energy, following the rule $\Delta_A(\omega) \propto \omega^2$. But when ARPES is turned on a non-Fermi liquid, it often reveals a completely different sight: a broad, hazy continuum of spectral weight with no sharp peak to be found. Or, if a peak exists, its width might shrink much more slowly, perhaps linearly as $\Delta_B(\omega) \propto \omega$, a smoking gun for a marginal quasiparticle that violates the stability criterion.

• ​​Breakdown of a Universal Law:​​ The ​​Wiedemann-Franz law​​ is one of the most beautiful consequences of the quasiparticle picture. It states that because the same entities—quasiparticles—carry both electrical current and heat, the ratio of the thermal conductivity ($\kappa$) to the electrical conductivity ($\sigma$), divided by temperature, is a universal constant of nature: $L = \kappa/(\sigma T) = L_0$. In the strange world of a non-Fermi liquid, where the very notion of a "particle" carrier is ill-defined, this law is often violated. The deviation is not random; it is systematically linked to the other anomalies. For example, in a hypothetical system where the specific heat scales as $C_e \propto T^\alpha$ and resistivity as $\rho \propto T^\beta$, simple kinetic arguments show that the Lorenz number must become temperature-dependent, scaling as $L(T) \propto T^{\alpha-1}$. The shattering of this universal law is a powerful testament to the breakdown of the underlying quasiparticle framework.

What kind of physics is so potent that it can dissolve the very idea of an electron as a particle inside a solid? The answers physicists have found are as strange and beautiful as the phenomena themselves.

• ​​Life on the Edge: The Quantum Critical Point:​​ One of the most prevalent ideas is that non-Fermi liquid behavior emerges in the vicinity of a ​​Quantum Critical Point (QCP)​​. This is not an ordinary phase transition like ice melting, which occurs at a finite temperature. A QCP is a phase transition that has been tuned to occur at absolute zero by some external parameter, like pressure or a magnetic field. At this special point, the material is perfectly poised between two different quantum ground states—for instance, between being a magnet and a non-magnet. It cannot decide which way to fall, so it exists in a state of perpetual, scale-invariant quantum fluctuation. These collective fluctuations act as the "mosh pit," scattering electrons so violently that they can never form stable quasiparticles. This intense scattering by critical modes is believed to be the engine behind many strange metal phenomena, including the famous linear-in-T resistivity. A QCP at $T=0$ doesn't just stay there; its influence extends out to finite temperatures, creating a "quantum critical fan" on the phase diagram—a whole region where strange, non-Fermi liquid physics dominates.

• ​​The Frustration of Screening:​​ Sometimes, non-Fermi liquid behavior can arise from a single, frustrated impurity. The ​​Kondo effect​​ describes how a local magnetic moment in a metal is "screened" by conduction electrons, which form a cloud of opposite spin around it. In the standard case, one channel of electrons is just enough to screen a spin-1/2 impurity, forming a stable, non-magnetic singlet. The low-temperature state is a perfect Fermi liquid. But what if we provide the impurity with two or more channels to screen it? This is the ​​overscreened Kondo model​​. The channels now compete with each other in a state of "screening frustration." The system can't settle into a simple ground state. The result is a quintessential non-Fermi liquid, which for the two-channel case exhibits a bizarre residual entropy of $\frac{1}{2} k_B \ln 2$ at zero temperature—as if half a spin's worth of confusion is left over, a ghostly remnant of the unresolved competition.

• ​​The Ultimate Breakdown: Fractionalization:​​ Perhaps the most exotic source of non-Fermi liquid behavior is the idea that under extreme quantum conditions, the electron itself can ​​fractionalize​​ into more fundamental pieces. In a theorized state called a ​​quantum spin liquid​​, the electron might break apart into a "spinon" (a neutral particle that carries the electron's spin) and a "chargon" (a spinless particle that carries its charge). The spinons can form their own Fermi surface, but they interact with each other via an emergent gauge field—a kind of private electromagnetism that exists only inside the material. The interaction mediated by this emergent force is so singular that it obliterates the spinon quasiparticles, leading to a profound non-Fermi liquid state with bizarre properties, such as a self-energy that scales as $\Sigma(\omega) \propto \omega^{2/3}$ and a specific heat that goes as $C_V \propto T^{2/3}$. This is the ultimate departure from Landau's paradigm: a metal whose elementary excitations are not even electrons anymore.

From the quiet stability of the Fermi liquid to the turbulent chaos of the quantum critical point, the study of non-Fermi liquids forces us to confront the deepest questions about the nature of collective quantum matter. It is a world where our comfortable notions of "particles" dissolve, revealing a richer and far stranger reality underneath.

After our journey through the strange and wonderful principles of non-Fermi liquids, one might be tempted to ask, "This is all very interesting, but what is it good for?" This is a fair and important question. Science is not just a collection of curious puzzles; it is a tool for understanding and eventually manipulating the world around us. The breakdown of Landau's Fermi liquid theory is not a mere academic failure; it is the discovery of a new continent of physical phenomena, teeming with novel properties and profound connections to other branches of science. In this section, we will explore this new world, seeing how the fingerprints of non-Fermi liquids appear in real materials and how the concepts they embody resonate across fields as diverse as materials science, chemistry, and even the study of black holes.

How do we know when we've stumbled upon a non-Fermi liquid? We can't simply look at a piece of metal and see the quasiparticles dissolving. Instead, we must be clever detectives, inferring the breakdown of the established order from the clues left behind in macroscopic measurements. These clues, or experimental signatures, are often "anomalous" power laws in physical properties as a function of temperature or frequency, which deviate sharply from the predictions for conventional metals.

One of the most fundamental clues is found in electrical resistivity. In an ordinary metal, as you cool it down, the resistivity decreases. The residual scattering from impurities eventually sets a floor, but the temperature-dependent part, arising from electrons bumping into each other, falls off as $T^2$. Why $T^2$? Because of the Pauli exclusion principle. An electron near the Fermi surface can only scatter to an empty state, and it needs another electron to scatter off of. The number of available electrons and available empty states are both proportional to the thermal energy, $k_B T$. Two factors of $T$ give us a $T^2$ dependence for the scattering rate. The electrons in a Fermi liquid are, in a sense, very polite; they have a hard time finding partners to scatter with at low temperatures.

In a non-Fermi liquid poised at a quantum critical point (QCP), this politeness is thrown out the window. The system is filled with roiling, low-energy collective fluctuations—for instance, the near-critical fluctuations of magnetic order in a metal on the verge of becoming an antiferromagnet. These fluctuations provide a potent and ever-present medium for electrons to scatter off. The scattering is no longer a rare two-body encounter but a constant battle with the collective quantum fog. A detailed calculation shows that for a two-dimensional metal near an antiferromagnetic QCP, the quasiparticle scattering rate becomes proportional to temperature, $\tau^{-1} \propto T$. This linear-in-temperature resistivity is one of the most celebrated and sought-after signatures of non-Fermi liquid behavior.

This electronic strangeness also leaves its mark on the thermodynamic properties of the material, such as its capacity to store heat. The electronic contribution to the specific heat, $C_V$, of a normal metal is proportional to temperature, $C_V = \gamma T$, where the coefficient $\gamma$ is proportional to the quasiparticle effective mass, $m^*$. In a Fermi liquid, $m^*$ is a constant. But in the NFL regime, the very concept of a stable quasiparticle with a fixed mass breaks down. The constant interaction with the critical fluctuations effectively "dresses" the electron in a way that depends on the energy scale. The self-energy, which describes this dressing, acquires a logarithmic dependence on temperature. This translates into an effective mass that diverges logarithmically as $T \to 0$. The result is a specific heat that no longer follows a simple linear law, but instead behaves as $C_V \propto T \ln(\Omega_0/k_B T)$. Measuring this logarithmic enhancement is another powerful piece of evidence that the familiar quasiparticle picture has failed.

The strangeness is not confined to DC properties. When we shine light on these materials and measure their ability to conduct alternating currents—the optical conductivity, $\sigma(\omega)$—we find more anomalies. A standard metal is described by the Drude model, where electrons occasionally scatter and the conductivity at low frequencies is a constant, leading to a peak of finite width at zero frequency. In a quantum critical system, the memory of the electron's path is lost in a scale-invariant way. The "memory function," which encodes the scattering processes, takes on a power-law form in frequency. This can lead to a startling prediction for certain models: the optical conductivity itself may follow a power law, such as $\sigma(\omega) \propto \omega^{-\eta}$, instead of the constant value expected in a simple Drude model. This "Drude-peak-that-isn't" is a direct reflection of the lack of any intrinsic timescale for scattering at a QCP.

These anomalous behaviors are not just theoretical curiosities; they are observed in a growing list of real materials. The quintessential playground for this physics is a class of compounds known as ​​heavy-fermion systems​​. These materials contain elements with localized $f$-electrons, such as cerium, ytterbium, or uranium. At high temperatures, these $f$-electrons act like isolated magnetic moments. But at low temperatures, they try to hybridize with the sea of conduction electrons. This competition—between localized magnetism and itinerant delocalization—can be tuned by pressure, chemical doping, or a magnetic field, driving the system towards a QCP.

Imagine such a material tuned by a magnetic field. At high fields, the local moments are screened by the conduction electrons (the Kondo effect), and the $f$-electrons join the Fermi sea. The result is a "large" Fermi surface that counts both the conduction electrons and the $f$-electrons, and the system behaves as a (very heavy) Fermi liquid. But below a critical field, $B_c$, what if the Kondo screening breaks down? The $f$-electrons might suddenly "localize," withdrawing from the Fermi sea and leaving behind a "small" Fermi surface of only conduction electrons. This is a "Kondo breakdown" QCP. How would we see such a dramatic event? The Hall coefficient, $R_H$, which is sensitive to the volume of the Fermi surface, provides a smoking gun. Experiments on materials like YbRh$_2$Si$_2$ have observed that as temperature is lowered, the crossover in $R_H$ between the small-FS and large-FS regimes becomes sharper and sharper, extrapolating to a dramatic, discontinuous jump precisely at the critical field $B_c$ for $T=0$. This is a beautiful confirmation that the very number of charge carriers in the metal has changed abruptly at the QCP.

The world of non-Fermi liquids extends beyond heavy fermions. In certain multi-orbital materials like the iron-based superconductors, a different mechanism, Hund's coupling, can lead to a "spin-freezing" behavior that also results in non-Fermi liquid properties. In these ​​Hund's metals​​, the electron scattering rate is found to scale not as $\omega$ or $\omega^2$, but as $\Gamma(\omega) \propto |\omega|^{1/2}$, yet another entry in the zoo of anomalous exponents.

Furthermore, in the constrained world of one dimension—as found in carbon nanotubes or quantum wires—the Fermi liquid is exceptionally fragile. Any interaction, no matter how weak, is enough to destroy the quasiparticle. The electron effectively splinters into two separate collective excitations: one carrying its charge (the "holon") and one carrying its spin (the "spinon"). This state of matter is called a ​​Luttinger liquid​​. The tell-tale sign of this separation is that the single-particle Green's function, which measures the probability of finding an electron at a certain distance from where it was created, no longer decays as $1/|x|$ but as a faster power law, $1/|x|^\alpha$, where the exponent $\alpha$ depends on the interaction strength. The electron as a whole ceases to exist as a long-lived excitation.

The study of non-Fermi liquids has pushed physicists to develop new theoretical tools and has revealed astonishing connections between previously disparate fields.

One of the most theoretically clean examples of a non-Fermi liquid occurs in the ​​two-channel Kondo model​​, where a single magnetic impurity is coupled to two independent channels of conduction electrons. The impurity spin becomes "overscreened," a situation that cannot be resolved into a simple, non-magnetic ground state. The system flows to a non-Fermi liquid fixed point. The properties of this state can be solved exactly using the powerful machinery of boundary conformal field theory (BCFT). These exact calculations reveal a clear non-Fermi liquid signature: the local electron spectral density at the impurity vanishes at the Fermi energy as a power law, $A(\omega) \propto |\omega|^{1/2}$, in stark contrast to the constant density of states in a normal metal. Even more bizarre is the prediction for the ground-state entropy. While a normal system would have zero entropy at $T=0$ (the Third Law of Thermodynamics), the overscreened impurity retains a peculiar, non-zero residual entropy of $S_{imp} = \frac{1}{2} k_B \ln 2$. This "fractional" entropy arises because the residual degree of freedom is not a simple spin-up/spin-down qubit, but something more exotic: a single, free Majorana fermion. This connects the physics of magnetic alloys to deep ideas in topological quantum matter and quantum information.

Perhaps the most breathtaking interdisciplinary connection is the ​​holographic correspondence​​, or AdS/CFT, which emerged from string theory. This duality proposes a stunning equivalence: a strongly interacting quantum field theory (like the one describing a non-Fermi liquid) in $d$ dimensions is mathematically equivalent to a weakly interacting theory of gravity (like Einstein's general relativity) in a curved $(d+1)$-dimensional spacetime known as Anti-de Sitter (AdS) space. A fiendishly difficult calculation in a strange metal can be mapped onto a much more tractable problem: calculating the trajectory of a particle or field near the event horizon of a black hole in this higher-dimensional universe. The thermal properties of the strange metal are related to the Hawking temperature of the black hole. Using this holographic dictionary, one can calculate properties like the fermion spectral function in a strongly interacting system. For certain types of charged black holes, the calculation reveals fermionic excitations that do not have a quasiparticle interpretation, exhibiting spectral functions with non-Fermi liquid characteristics, a result that is very difficult to obtain by conventional means. The fact that the esoteric physics of black holes can describe the behavior of electrons in a piece of metal is a profound testament to the unity of physics.

Finally, the consequences of this physics can even spill over into ​​physical chemistry​​. The rate and equilibrium of a chemical reaction are governed by the change in the Gibbs free energy, $\Delta_r G$. The Gibbs energy is related to the entropy and heat capacity of the reactants and products. If a reaction takes place between solid-state species, and one of them is a metal tuned to a QCP, its singular contribution to the heat capacity ($C_p^{sing} \propto T^{1-\alpha}$) will propagate through the thermodynamic relations. This leads to an anomalous, singular contribution to the reaction Gibbs energy itself, scaling as $\Delta_r G_{sing} \propto -T^{2-\alpha}$. This implies that quantum criticality could, in principle, alter the thermodynamic driving force of chemical reactions at low temperatures, a fascinating and unexpected connection.

From the resistance of a strange metal to the entropy of a single magnetic atom and the properties of a distant black hole, the study of non-Fermi liquids forces us to abandon old certainties and embrace a richer, more interconnected view of the quantum world. It is a field that is not just about breaking the rules of the old "standard model" of metals, but about discovering the new rules that govern some of the most exotic and potentially useful forms of quantum matter.