Tunneling Density of States

In the quantum realm of materials, understanding the collective behavior of electrons is key to unlocking the secrets of exotic phenomena like superconductivity and fractionalization. While basic models provide a static picture of electron energy levels, they often fail to capture the dynamic reality of how these interacting particles move and organize. This leaves a critical gap in our ability to probe and characterize complex quantum states of matter. The tunneling density of states (TDOS) emerges as a powerful experimental concept that bridges this gap, offering a direct window into the live, interacting electronic landscape.

This article provides a comprehensive exploration of the tunneling density of states. The journey begins in ​​Principles and Mechanisms​​, where we will delve into the fundamental physics of TDOS, explaining how it is measured using scanning tunneling microscopy and distinguishing it from the simpler single-particle density of states. We will uncover the theoretical underpinnings of TDOS in the strange one-dimensional world of Luttinger liquids, revealing how interactions reshape the very fabric of electronic states.

Subsequently, in ​​Applications and Interdisciplinary Connections​​, we will see the TDOS in action as a diagnostic tool. We will explore how its unique signatures—in the form of power laws and energy gaps—provide crucial evidence for phenomena such as spin-charge separation, the fractionalization of electrons in the quantum Hall effect, and Cooper pairing in superconductors. Through this journey, you will gain a deep appreciation for how measuring the TDOS transforms our understanding of quantum matter from a static map to a vibrant, dynamic world.

Imagine you are trying to understand the inner workings of a bustling city. You could look at a map, which gives you a static picture of the streets and buildings. This is useful, but it doesn't tell you how traffic flows, where people congregate, or how the city truly lives. To understand that, you need to be there, to watch the dynamics, to see how a single car entering a highway affects the flow for miles.

In the quantum world of materials, we face a similar challenge. We want to understand the behavior of electrons, the city's inhabitants. The "map" is the material's basic structure, but the "life" of the material lies in how its electrons move, interact, and organize themselves. The ​​tunneling density of states (TDOS)​​ is one of our most powerful tools for observing this life, for moving beyond the static map to the dynamic reality. It is the quantity we probe when we perform an experiment to measure the availability of electronic states at a given energy.

Our primary tool for this exploration is the ​​Scanning Tunneling Microscope (STM)​​. Think of it as an exquisitely sensitive fingertip that can feel the electronic texture of a surface. The "fingertip" is a metallic tip sharpened to a single atom, and it hovers just above a conductive sample, separated by a vacuum. Classically, this gap is an insurmountable wall. But in the quantum world, electrons have a strange ability to "tunnel" through this barrier, creating a tiny electrical current.

This tunneling current, $I$, is the key. The amount of current that flows depends on two main things: how many electrons in the tip are ready to jump, and how many available "landing spots" or energy states there are for them in the sample. By applying a small voltage, $V$, between the tip and the sample, we can give the electrons the energy they need to make the jump, with the energy gained being $E = eV$, where $e$ is the elementary charge. By precisely controlling this voltage, we can choose which energy levels in the sample we want to probe.

The real magic happens when we measure not just the current itself, but how the current changes as we change the voltage. This quantity, the ​​differential conductance​​, $g(V) = dI/dV$, gives us a direct window into the sample's electronic landscape. For a simple metallic tip where the density of available states is roughly constant, a straightforward calculation shows something remarkable: the differential conductance is directly proportional to the sample's ​​local density of states (LDOS)​​ at the energy corresponding to the applied voltage.

In essence, the STM acts as a spectrometer. By sweeping the voltage, we are scanning through the energy levels of the sample and plotting out its density of states. A peak in the $dI/dV$ curve signifies an energy where there are many available states for electrons, while a valley, or a gap, signifies a scarcity of states. This technique, called ​​Scanning Tunneling Spectroscopy (STS)​​, allows us to create a detailed energy map of the electrons, location by location, across a material's surface.

So far, we have a wonderfully intuitive picture: $dI/dV$ measures the density of available electronic states. But the quantum world, especially when particles interact, is rarely so simple. The electrons in a solid are not isolated individuals; they form a complex, interacting collective. Pushing a new electron into this collective is not like placing a marble into an empty box. It's more like adding a person to a crowded dance floor—the newcomer has to find space, and their arrival makes everyone else shuffle around.

This brings us to a crucial and subtle distinction. There are really two concepts of "density of states":

1. The ​​single-particle density of states ($\rho_{sp}$)​​: This is a theoretical quantity, the one you might calculate in a textbook. It's derived from what physicists call the ​​single-particle Green's function​​, which you can think of as the probability amplitude for a lone electron to propagate from one point to another. Interactions with other electrons modify this propagator, giving the electron a finite lifetime and shifting its energy. These effects are bundled into a term called the ​​self-energy​​.

2. The ​​tunneling density of states ($\rho_t$)​​: This is what is actually measured in the STM experiment. The tunneling event itself—the act of injecting an electron—is a many-body process. The incoming electron can shake the system, creating ripples and other collective excitations (like an extra electron-hole pair). These additional processes, called ​​vertex corrections​​, can modify the probability of tunneling.

In many ordinary metals, described by ​​Fermi liquid theory​​, the distinction is subtle, and the tunneling DOS is a good approximation of the single-particle DOS. However, the possibility of vertex corrections means that tunneling can sometimes reveal features, like a dip in conductance at zero voltage, that are not present in the simple single-particle DOS. The dance is more complicated than it first appears.

Nowhere is this complexity more dramatic and beautiful than in the strange world of one-dimensional (1D) systems, like electrons confined to a nanowire or a carbon nanotube. In 1D, the rules of the game change entirely. The jostling on the crowded dance floor is so intense that the concept of an individual "electron dancer" breaks down completely. The electrons lose their identity and instead move in collective, wave-like motions. This state of matter is not a Fermi liquid; it is a ​​Luttinger liquid​​.

The hallmark of a Luttinger liquid is its stunningly unique tunneling density of states. If you try to tunnel an electron into the middle (the "bulk") of a Luttinger liquid, you find that the TDOS is not constant near the Fermi energy as it would be in a simple metal. Instead, it is crushed to zero in a characteristic power law:

Here, $\omega$ is the energy relative to the Fermi level, and $\alpha$ is a positive exponent. This phenomenon, often called a ​​zero-bias anomaly​​, means it becomes progressively harder to inject an electron as its energy gets closer to zero. It’s as if the collective fluid of electrons actively conspires to repel any newcomers with low energy.

Where does this power law come from? It originates from the very nature of the electron's existence in this 1D world. A theoretical tool called ​​bosonization​​ allows us to switch from the language of individual electrons (fermions) to the language of their collective waves (bosons). Using this tool, we can calculate the electron's Green's function, its probability of surviving over time. In a Luttinger liquid, this probability doesn't decay exponentially (as for a typical particle with a well-defined lifetime), but as a power law in time, $G(\tau) \propto \tau^{-\eta}$. A mathematical procedure known as an analytic continuation (essentially a Fourier transform) connects this time-domain behavior to the energy-domain, turning the temporal power law into the spectral power law we observe, with the exponent being $\alpha = \eta - 1$.

Amazingly, the value of the exponent $\alpha$ is not arbitrary. It is dictated by a single, dimensionless number called the ​​Luttinger parameter, $K$​​. This parameter encodes the strength of the electron-electron interactions: $K=1$ for non-interacting electrons, while repulsive interactions give $K1$. For a simple spinless Luttinger liquid, the exponent for tunneling into the bulk is given by a beautifully symmetric formula [@problem_id:84283, @problem_id:87990]:

For non-interacting electrons ($K=1$), the exponent $\alpha$ is zero, and we recover the constant DOS of a normal 1D metal. But for any repulsive interaction ($K1$), the exponent $\alpha$ becomes positive, and the suppression of the TDOS appears—a direct, measurable consequence of the collective nature of this exotic state of matter.

The elegance of this framework is that it can be extended to describe even richer phenomena, revealing more profound aspects of the quantum world.

​​Spin-Charge Separation:​​ Real electrons have spin. In our 3D world, spin and charge are indivisible properties of the electron. But in the 1D Luttinger liquid, one of the most bizarre phenomena in physics occurs: ​​spin-charge separation​​. When an electron is injected, it shatters into two separate entities: a "holon," which carries the electron's charge but not its spin, and a "spinon," which carries the spin but not the charge. These two quasiparticles can then travel through the wire independently, sometimes even at different speeds!

This schism is written directly into the tunneling exponent. A spinful Luttinger liquid is described by two Luttinger parameters, one for the charge sector ($K_c$) and one for the spin sector ($K_s$). The total tunneling exponent for the bulk is simply the sum of the contributions from each independent sector [@problem_id:1199637, @problem_id:3017403]:

The experimental measurement of a single number, $\alpha$, thus gives us a window into this fantastic separation of fundamental properties.

​​The Edge is Different:​​ The location of the tunneling matters. Tunneling into the infinite "bulk" of a wire is different from tunneling into its end. At an open boundary, the collective electron waves must reflect, altering their behavior. This change in boundary conditions leads to a different set of rules for injecting an electron. Consequently, the tunneling exponent changes! For tunneling into the end of a wire, the exponent becomes [@problem_id:249436, @problem_id:1167996, @problem_id:3017403]:

This is a powerful lesson: in the quantum world, geometry is destiny. The very same material can exhibit different physical laws depending on where you look.

​​Liquid to Liquid:​​ Finally, what if we arrange a meeting between two different 1D worlds by building a tunnel junction between two different Luttinger liquids? The current-voltage characteristic of such a device becomes a conversation between the two systems. The resulting differential conductance also follows a power law, $G(V) \propto V^\beta$, where the exponent $\beta$ is the sum of the end-tunneling exponents for each liquid, $\beta = \alpha_{\text{end},L} + \alpha_{\text{end},R}$. It is as if two different musical instruments are playing together; the resulting harmony depends on the acoustic properties of both.

From a simple measurement of current and voltage, the tunneling density of states leads us on a journey deep into the heart of quantum matter. It reveals the breakdown of our classical intuition, makes manifest the collective dance of interacting electrons, and provides tangible evidence for some of the most profound and exotic concepts in modern physics, such as spin-charge separation. It transforms our picture of matter from a static map into a living, breathing, and often surprising, dynamic world.

If the single-particle density of states we discussed in the previous chapter is an internal census of an electron system—a complete roster of all available energy levels—then the ​​tunneling density of states (TDOS)​​ is what an outside observer actually gets to see. Imagine you are trying to understand a bustling city. The census tells you how many residents there are, but standing at the city gates and watching people come and go gives you a very different, and often more dynamic, picture. This is the role of the TDOS. It is the face a quantum system presents to the world, measured by the ease with which an electron can "tunnel" into it from an external probe, like the tip of a scanning tunneling microscope.

You might think this "external view" is just a blurry version of the "internal truth." But in the quantum world, things are rarely so simple. As it turns out, the act of observation—of adding a single electron—can be profoundly affected by the collective behavior of the electrons already inside. The system can react in startling ways, revealing that the electrons are not just independent residents but members of a complex, interacting society. By measuring the TDOS, we eavesdrop on their collective conversations, and what we hear tells us some of the most profound stories in modern physics: stories of electrons shattering into fractions, of particles forming unbreakable pairs, and of one-dimensional traffic jams that rewrite the rules of electricity.

Nowhere is the drama of electron society more apparent than in one dimension (1D). Imagine a highway with only a single lane. Cars can’t pass each other; the motion of one car inevitably affects all the others. Electrons in a 1D wire, such as a carbon nanotube or the edge of certain exotic materials, face the same constraint. This "no-passing" rule forges them into a remarkable state of matter known as a ​​Tomonaga-Luttinger liquid​​. The electrons lose their individuality and move as a collective, like sound waves propagating through a medium. A single electron is no longer a valid concept; the true excitations are these collective plasmon modes.

So, what happens when we try to tunnel a single, well-behaved electron from a normal metallic lead into this strange 1D world? The answer depends dramatically on where we tunnel.

Think of the 1D wire as a flute. Trying to inject an electron into the very end of the wire is like blowing into the mouthpiece, while tunneling into the middle is like drilling a hole and forcing air in there. The physics is completely different.

When we tunnel into the end of an interacting 1D wire, we are trying to force an individual particle into a perfectly choreographed dance. The collective responds by rearranging itself, making it difficult for the lone electron to join. The result is a dramatic suppression of the TDOS at low energies. The system essentially tells the incoming electron, "You're not welcome here, unless you come with a lot of energy." This aversion is captured by a beautiful power law:

where $\omega$ is the energy of the tunneling electron measured from the Fermi level. The exponent $\alpha$ is not just some number; it's a direct measure of the electron-electron interactions! For a spinless 1D wire, this "end-tunneling" exponent is given by $\alpha = (1/K) - 1$, where $K$ is the Luttinger parameter that encodes the interaction strength ($K=1$ for non-interacting electrons, and $K 1$ for repulsive interactions). For repulsive electrons, $\alpha$ is positive, confirming the suppression of tunneling. By measuring this exponent, experimentalists can directly quantify the strength of the hidden collective dance inside materials like quantum dots or the conducting edges of two-dimensional topological insulators.

Tunneling into the bulk of the wire tells a different, but related, story. Here, the incoming electron has room to maneuver on either side, and the system's response is less severe. The TDOS is still suppressed, but with a different power-law exponent that depends on the same interaction parameter $K$. This physics is exquisitely realized in metallic carbon nanotubes, which act as nearly perfect multi-channel Luttinger liquids, offering a pristine stage to verify these remarkable theoretical predictions.

The story gets even more spectacular when we turn to the two-dimensional world of the quantum Hall effect. When a 2D electron gas is subjected to a low temperature and an immense magnetic field, the electrons abandon their usual metallic behavior and condense into an entirely new, topological state of matter. While the bulk of the material becomes an insulator, its edge comes alive with dissipationless one-dimensional currents. These edges are, in fact, perfect realizations of 1D liquids.

Let's first look at the simplest case: the ​​integer quantum Hall effect (IQHE)​​ at filling factor $\nu=1$. Here, you might expect the strong interactions and magnetic field to create a complicated mess. But when we measure the TDOS by tunneling into this edge, we find that the exponent $\alpha = 0$. This means the TDOS is constant, just like in a boring, non-interacting metal! This is a profound result. The underlying topology of the quantum Hall state protects the edge channel, forcing it to behave like a perfect, one-way "chiral" Fermi liquid. The interactions are still there, but they are rendered harmless; they can't scatter electrons or change the fundamental nature of the state.

Now for the main event: the ​​fractional quantum Hall effect (FQHE)​​. At a filling factor like $\nu = 1/3$, things change completely. Tunneling experiments reveal a TDOS that follows a power law $\rho(\omega) \propto |\omega|^{\alpha}$ with an exponent $\alpha = 1/\nu - 1$. For $\nu = 1/3$, this gives $\alpha = 2$. This is a massive suppression! Why is it so incredibly difficult to add an electron to this system at low energy?

The answer is one of the most beautiful in all of physics: the electron you are trying to inject cannot exist inside the $\nu=1/3$ FQHE state. Instead, it must ​​fractionalize​​. The collective fluid of electrons is organized in such a way that its fundamental excitations are not electrons, but mysterious quasiparticles carrying exactly one-third of an electron's charge. To add one full electron, you must create three of these fractional charges. This is a highly non-trivial, many-body process, and its difficulty is reflected in the severe power-law suppression of the TDOS. Measuring the tunneling exponent is, therefore, a direct observation of charge fractionalization—seeing an electron shatter into pieces.

Let us now turn to a completely different type of collective quantum behavior: superconductivity. Here, electrons with opposite spin and momentum form "Cooper pairs," which then condense into a single, massive quantum state. This collective harmony allows them to flow without any resistance.

What face does this system present to our tunneling probe? A very different one from the Luttinger liquid. Instead of a power-law "soft gap" that fades away to zero at the Fermi energy, the TDOS of a superconductor exhibits a hard, unforgiving ​​gap​​. Below a certain energy threshold $\Delta_0$, the TDOS is strictly zero. It costs a finite amount of energy, the "binding energy" of the Cooper pair, to break a pair apart and create two particle-like excitations. There are simply no states available for a low-energy electron to tunnel into.

As the tunneling electron's energy $\omega$ approaches the gap edge $\Delta_0$, the TDOS doesn't just turn on; it diverges, forming sharp "coherence peaks" before settling down to the normal metallic value for $\omega \gg \Delta_0$. The full shape of the TDOS, including the effects of scattering which can smear these sharp features, is a powerful diagnostic tool:

Here, $N_s(\omega)$ is the superconducting TDOS, $N_0$ is the DOS of the normal metal, and $\Delta(\omega)$ is the complex energy-dependent gap. Scanning tunneling microscopy can map this TDOS with exquisite precision, allowing us to literally "see" the superconducting gap, measure its size, and watch how it changes with temperature or magnetic field. It is our primary window into the health and character of the superconducting state.

These power-law exponents and gap structures are beautiful theoretical constructs, but how do we connect them to what is measured in a laboratory? An experimentalist may not have a perfect energy-resolving spectrometer, but they do have control over two other knobs: voltage ($V$) and temperature ($T$).

Amazingly, the physics of the energy-dependent TDOS is directly mirrored in the transport properties as a function of temperature. A key insight is that both voltage and temperature provide the energy needed for tunneling. Through a simple and elegant argument, one can show that if the TDOS has a power-law form $\rho(\omega) \propto |\omega|^\alpha$, then the zero-bias electrical conductance will have a corresponding power-law dependence on temperature: $G(T) \propto T^\alpha$. This is a fantastic gift to experimentalists! To measure the critical exponent $\alpha$, one simply has to measure the conductance of a device while cooling it down.

This leads to an even more powerful idea: ​​scaling collapse​​. For a Luttinger liquid, the conductance $G(V,T)$ isn't a messy function of two independent variables. The two energy scales, $eV$ and $k_B T$, combine into a single, dimensionless variable $x=eV/(k_B T)$. This implies that the behavior of the system is governed by a single universal scaling function. If one measures the conductance at many different voltages and temperatures and plots the combination $G/T^\beta$ versus $V/T$ (where $\beta$ is the expected exponent), all the data points collapse onto a single, universal curve. The emergence of such a scaling collapse is often considered the "smoking gun" evidence for Luttinger liquid physics, cleanly distinguishing it from the behavior of conventional, non-interacting systems that lack such elegant simplicity.

In the end, the tunneling density of states is far more than a technicality. It is our portal into the rich social lives of electrons in quantum materials. It allows us to witness them fractionalizing, pairing up, and dancing in choreographed unison. It is the crucial bridge connecting the abstract beauty of many-body theory to the concrete, measurable realities of the laboratory, revealing time and again the deep and often astonishing unity of the quantum world.