Sommerfeld expansion

To understand the collective behavior of electrons that governs the properties of a metal, one must often calculate macroscopic quantities like internal energy and heat capacity. These calculations involve complex integrals over all electron energies, weighted by the density of states and the Fermi-Dirac distribution function. While these integrals are straightforward at absolute zero, the introduction of any finite temperature complicates them immensely, seemingly demanding numerical solutions. This presents a significant challenge in bridging the gap between quantum statistics and observable thermodynamics.

This article introduces the Sommerfeld expansion, an elegant and powerful mathematical framework that provides an analytical solution to this problem in the crucial low-temperature regime. It offers a conceptual lens to understand why only electrons near the Fermi surface participate in thermal processes. Across the following chapters, you will gain a comprehensive understanding of this fundamental tool. The "Principles and Mechanisms" chapter will deconstruct the mathematical underpinnings of the expansion, revealing how it leverages the properties of the Fermi-Dirac distribution to simplify calculations. Following that, the "Applications and Interdisciplinary Connections" chapter will showcase its remarkable predictive power, explaining phenomena from the heat capacity and conductivity of simple metals to the magnetic response of electrons and its relevance in modern materials like graphene.

To understand the collective behavior of electrons in a metal, we often need to calculate macroscopic properties like the total energy, particle number, or heat capacity. These quantities are typically found by averaging over all possible electron energies, weighted by two functions: first, a function that tells us how many quantum states are available at a given energy, called the ​​density of states​​, and second, the ​​Fermi-Dirac distribution​​, which tells us the probability that a state at a given energy is actually occupied by an electron. This leads to integrals of the general form:

Here, $\phi(E)$ represents some physical property weighted by the density of states (for example, to get the total energy, $\phi(E)$ would be energy $E$ times the density of states $g(E)$). The function $f(E; \mu, T) = \left[ \exp\left( (E-\mu)/(k_B T) \right) + 1 \right]^{-1}$ is the famous Fermi-Dirac distribution, where $\mu$ is the chemical potential (roughly the "filling level" of the electrons) and $T$ is the temperature.

At absolute zero temperature ($T=0$), this integral is simple. The Fermi-Dirac distribution becomes a perfect step function: it is 1 for all energies below $\mu$ and 0 for all energies above. The integral simply becomes $\int_0^\mu \phi(E) dE$. But what happens when we turn on the heat, even a little bit? The sharp step in the Fermi-Dirac distribution softens into a smooth curve, making the integral fiendishly difficult to solve exactly. Must we resort to brute-force numerical computation every time? Fortunately, the answer is no. A beautiful piece of mathematical physics, the ​​Sommerfeld expansion​​, comes to our rescue. It provides an elegant way to understand the low-temperature world of electrons.

The magic of the Sommerfeld expansion lies not in looking at the Fermi-Dirac function $f(E)$ itself, but at its derivative with respect to energy, $-\frac{\partial f}{\partial E}$. Let’s picture what this function looks like. At low temperatures, $f(E)$ is a very sharp, but smooth, drop from 1 to 0 centered at the chemical potential $\mu$. Its derivative, therefore, must be a sharp, symmetric spike centered precisely at $E=\mu$. This spike has a width on the order of the thermal energy, $k_B T$. Outside this narrow window, the derivative is practically zero.

This function acts like a highly focused spotlight. When we use a clever mathematical trick called integration by parts, we can rewrite our difficult integral $I(T)$ as an integral over a new function, weighted by this "spotlight" kernel $-\frac{\partial f}{\partial E}$. The consequence is profound: the value of the integral is determined almost entirely by the behavior of our physical function $\phi(E)$ within the narrow energy window illuminated by the spotlight around $E=\mu$. Everything happening far below or far above the chemical potential is cast into darkness and contributes very little. This is the physical heart of the matter: at low temperatures, only the electrons near the "surface" of the Fermi sea are involved in thermal phenomena.

If we know that only the region near $E=\mu$ matters, we can make a brilliant simplification. If the function $\phi(E)$ is smooth and well-behaved in that small illuminated region—meaning it doesn't have any wild jumps, kinks, or singularities—we can approximate it with a simple polynomial using a Taylor series expansion around $E=\mu$:

Now, we integrate this polynomial term by term, weighted by our symmetric spotlight function, $-\frac{\partial f}{\partial E}$. And here, a second piece of magic occurs, born from symmetry. Because our spotlight kernel is a perfectly even function around $E=\mu$, any integral of it multiplied by an odd function, like $(E-\mu)$, $(E-\mu)^3$, and so on, must be exactly zero.

This beautiful cancellation has a crucial physical consequence: the low-temperature corrections to any quantity calculated this way cannot be proportional to $T$, $T^3$, or any odd power of temperature. The first and most important thermal correction must be proportional to $T^2$. This is not an accident; it is a direct result of the fundamental particle-hole symmetry around the Fermi level in the low-temperature limit.

When we carry out the integrations of the even-powered terms, we arrive at the Sommerfeld expansion. The integrals yield universal numerical constants involving powers of $\pi$. The full expansion to the first few orders is:

Let's dissect this remarkable formula:

• The first term, $\int_{-\infty}^{\mu} \phi(E) dE$, is simply the value of the integral at absolute zero. This is our baseline.
• The second term, $\frac{\pi^2}{6}(k_B T)^2 \phi'(\mu)$, is the leading thermal correction. It's proportional to $T^2$, as predicted by our symmetry argument. Notice that it depends on $\phi'(\mu)$, the slope of our physical function right at the chemical potential. This tells us that the more rapidly the properties of the electronic states are changing at the Fermi surface, the stronger the effect of temperature will be.
• The third term, $\frac{7\pi^4}{360}(k_B T)^4 \phi'''(\mu)$, is the next, much smaller correction. Its accuracy is governed by the bound on the third derivative of the function, and it allows us to systematically improve our calculations when needed.

This expansion is an ​​asymptotic series​​. It's not guaranteed to converge if you add infinite terms, but for small $T$, truncating it after the first few terms provides an exceptionally accurate approximation of reality.

Let's use this powerful tool to solve a real physical problem: why does the electronic contribution to the heat capacity of a metal depend linearly on temperature? We want to calculate the total energy $U(T)$, and then find the specific heat by taking the derivative, $C_V = (\partial U / \partial T)_V$.

This requires a bit of self-consistency. First, we must recognize that as we heat the metal, the chemical potential $\mu$ must shift slightly to ensure the total number of electrons remains constant. We can calculate this shift using the Sommerfeld expansion on the integral for the number of particles $N$. This reveals that $\mu(T)$ decreases slightly from its zero-temperature value, the Fermi energy $\epsilon_F$, with a correction proportional to $T^2$.

Next, we calculate the internal energy $U(T)$ using the expansion, carefully plugging in our expression for the temperature-dependent $\mu(T)$. After a bit of algebra, combining the results gives us the total energy as a series in temperature. Finally, differentiating with respect to $T$ gives the specific heat. For a simple three-dimensional free electron gas, the result is astonishingly elegant:

where $T_F = \epsilon_F/k_B$ is the Fermi temperature. The leading term is linear in $T$, precisely as observed in experiments! The expansion not only gives us the famous $\gamma T$ law but also predicts the next-order correction, a small negative term proportional to $T^3$. It transforms a complex statistical mechanics problem into a clear and predictive physical law. A similar procedure allows us to calculate other thermodynamic quantities like the grand potential.

Like any powerful tool, the Sommerfeld expansion has its limitations. Its validity rests on the assumptions we made. Understanding when it fails is just as important as knowing when it works,.

1. ​​The Low-Temperature Limit is Crucial:​​ The expansion is fundamentally a low-temperature approximation. The condition $k_B T \ll \mu$ must hold. If the temperature gets too high, our "spotlight" $-\frac{\partial f}{\partial E}$ becomes so broad that it no longer samples just the local neighborhood of $\mu$. The Taylor expansion is no longer a valid approximation over the entire illuminated region, and the whole procedure collapses.

2. ​​The Landscape Must Be Smooth:​​ The expansion relies on the function $\phi(E)$ being smooth and differentiable at $E=\mu$. If the chemical potential lies near a point of non-analyticity, the expansion breaks down.

   • ​​Band Edges:​​ If $\mu$ is very close to a band edge (the bottom or top of an energy band), our spotlight effectively shines over a "cliff." The function $\phi(E)$ is not smooth here (it might be zero on one side), derivatives may not exist, and the expansion is invalid.
   • ​​Semiconductors:​​ If $\mu$ lies within a band gap, as in an insulator or semiconductor, the physics is entirely different. There are essentially no states at the chemical potential to excite. Instead, electrons must be thermally activated across the gap, a process governed by exponential factors like $\exp(-\Delta / k_B T)$. This behavior is non-analytic at $T=0$ and cannot be captured by a power series in $T$.
   • ​​Van Hove Singularities:​​ A fascinating case occurs when the density of states itself has a singularity, such as a logarithmic spike, right at the Fermi energy. This happens in certain materials, like graphene or high-temperature superconductors. Here, the derivatives in the standard expansion formula blow up, and the method fails. However, this failure is itself instructive. A more careful analysis shows that transport properties develop unusual, non-analytic temperature dependencies, like corrections proportional to $1/\ln(T)$. Remarkably, even in this case, fundamental relationships like the Wiedemann-Franz law can be recovered in the zero-temperature limit, showcasing a deep robustness in the underlying physics that persists even when our simplest calculational tools need refinement.

The Sommerfeld expansion is more than a calculational trick. It is a conceptual lens that sharpens our focus onto the most important actors in the low-temperature drama of metals: the electrons at the very edge of the Fermi sea. It elegantly demonstrates how the interplay of quantum statistics, symmetry, and the local properties of electronic states gives rise to the simple, universal laws that govern the world of cold matter.

Having acquainted ourselves with the machinery of the Sommerfeld expansion, we are now like explorers equipped with a powerful new microscope. The previous chapter showed us how it works—how it systematically peels back the layers of complexity that temperature adds to the pristine, zero-temperature world of fermions. Now, we turn this microscope towards nature and ask the real question: What can we see with it? What secrets of the material world does it unlock?

You will find that this single mathematical tool is a master key, unlocking doors in seemingly disparate fields of physics and chemistry. From the familiar properties of a copper wire to the exotic behavior of modern materials like graphene, the Sommerfeld expansion provides a unified language to describe how things change when they warm up just a little. It’s a story about the subtle dance between quantum mechanics and thermodynamics, and it begins with the most fundamental properties of metals.

Imagine a metal at absolute zero. Its electrons fill every available energy state up to a sharp cutoff, the Fermi energy $E_F$. This "Fermi sea" is a perfectly still, incompressible quantum fluid. But what happens when we add a little heat? Naively, one might think every electron warms up a bit, just like atoms in a classical gas. If this were true, the electronic heat capacity of metals would be enormous, something flatly contradicted by experiment.

Here, the Sommerfeld expansion provides its first profound insight. It tells us that temperature creates only a gentle "blur" around the Fermi surface. Only the electrons in a narrow energy band of width $\sim k_B T$ around $E_F$ are excited. The vast majority of electrons deep in the Fermi sea are "frozen" by the Pauli exclusion principle; there are no empty states nearby for them to jump into.

This simple picture has dramatic consequences. The first is a subtle adjustment the system must make. To keep the total number of electrons constant as they spread into higher energy states, the chemical potential $\mu$—the effective "sea level" of the electrons—must actually decrease slightly with temperature. The Sommerfeld expansion allows us to calculate this shift precisely, showing it's proportional to $T^2$. This is a crucial first step; it’s like re-calibrating our ruler before making a measurement, ensuring that all our subsequent calculations are consistent.

With this in hand, we can tackle the big questions. The internal energy of the electron gas, for instance, doesn't increase as much as you'd think. The expansion reveals that the increase in energy is not linear with $T$, but proportional to $T^2$. From this, the electronic heat capacity $C_V = (\partial U / \partial T)_V$ immediately follows. We find that it is directly proportional to temperature: $C_V = \gamma T$. This linear dependence is a hallmark of a degenerate Fermi gas and one of the great triumphs of the Sommerfeld model, perfectly explaining decades of experimental data on metals at low temperatures.

This thermodynamic story is completed by looking at the entropy, $S$. Since $C_V = T(\partial S / \partial T)_V$, a heat capacity linear in $T$ implies an entropy that is also linear in $T$. This not only paints a consistent thermodynamic picture but also beautifully satisfies the Third Law of Thermodynamics, which demands that the entropy must go to zero as $T \to 0$. The Sommerfeld expansion shows us how the quantum nature of electrons ensures this happens in a smooth, predictable way.

Understanding the energy and heat capacity of electrons is one thing, but how does this thermal "blurring" affect how they move and transport charge and heat?

Consider thermoelectricity, the fascinating phenomenon behind thermoelectric coolers and generators. If you create a temperature difference across a metal rod, a voltage appears. This is the Seebeck effect. Why? Electrons from the hot end diffuse towards the cold end, and vice versa. At first glance, you might think these flows cancel out. But the Sommerfeld expansion, when applied to the famous Mott formula, tells a more subtle story. The "hot" electrons are slightly more energetic and move differently than the "cold" electrons. The expansion shows that this imbalance results in a net voltage, and that the Seebeck coefficient, which measures the size of this effect, is predicted to be linearly proportional to temperature at low $T$. This prediction is brilliantly confirmed in simple metals.

Another remarkable connection in metals is the Wiedemann-Franz law, which states that the ratio of thermal conductivity ($\kappa$) to electrical conductivity ($\sigma$) is proportional to temperature, with a universal constant of proportionality called the Lorenz number, $L = \kappa/(\sigma T)$. This law suggests that the very same electrons are responsible for carrying both charge and heat. The Sommerfeld expansion allows us to derive this law from first principles. More than that, it allows us to go beyond the ideal law. It can calculate the small, temperature-dependent corrections to the Lorenz number, showing how the "universality" of the law begins to break down as the thermal blurring of the Fermi surface becomes more significant.

Electrons are not just charged particles; they also have an intrinsic magnetic moment (spin) and their motion creates orbital magnetic moments. The Sommerfeld expansion is indispensable for understanding how the collective magnetic response of an electron gas is affected by temperature.

First, consider Pauli paramagnetism, the tendency of electron spins to align with an external magnetic field. At $T=0$, only electrons at the Fermi surface can flip their spins to align with the field. At finite temperature, the thermal blurring allows more electrons to participate, but also creates disorder. The Sommerfeld expansion allows us to calculate the net effect, yielding the temperature-dependent correction to the magnetic susceptibility. Remarkably, the correction depends not just on the density of states at the Fermi energy, but on its curvature (the second derivative). This tells us that the shape of the electronic energy landscape right at the Fermi level dictates how the magnetism changes with temperature.

The same principles apply to Landau diamagnetism, which is the collective orbital response of electrons that opposes an external magnetic field. This is a purely quantum effect, and once again, the Sommerfeld expansion provides the tool to calculate how thermal smearing of the electron states modifies this diamagnetic response, predicting a specific $T^2$ correction.

The reach of the Sommerfeld expansion extends far beyond the simple free electron model. Its conceptual framework is a cornerstone of modern condensed matter physics.

Within solid-state physics, the idea of screening is paramount. A charge placed in an electron gas is "screened" as the mobile electrons rearrange themselves to neutralize its field. The effectiveness of this screening is captured by the Thomas-Fermi screening length. But how does this change with temperature? By applying the expansion to the Lindhard response function, one can calculate the leading temperature correction to the screening length, finding that screening becomes slightly less effective as the electron gas heats up. This has profound implications for understanding effective interactions between particles in a solid.

Perhaps most excitingly, this century-old tool is still at the forefront of research, helping us understand exotic new materials. Consider graphene, a single sheet of carbon atoms where electrons behave as massless, two-dimensional particles described by the Dirac equation. Their energy is proportional to their momentum, not its square. Even in this strange new world, the Sommerfeld expansion is perfectly suited to calculate thermodynamic properties. We can use it, for example, to find the leading temperature correction to the pressure exerted by this gas of "Dirac fermions," providing key insights into the behavior of these revolutionary materials.

From heat to electricity, magnetism to material science, the Sommerfeld expansion is the common thread. It is our quantitative guide to the low-temperature frontier, showing us in exquisite detail how the sharp, perfect quantum world of absolute zero gracefully acquires the fuzziness and complexity of finite temperature. It is a testament to the power of physics to find simple, unifying principles that govern a vast landscape of phenomena.