Kohn–Sham Equations

The quantum world of electrons in atoms and molecules is governed by the Schrödinger equation, but its exact solution for more than one electron—the infamous "many-body problem"—is a practical impossibility due to the complex web of electron-electron interactions. This fundamental barrier long stood in the way of predictive quantum chemistry and materials physics. While the Hohenberg-Kohn theorems of Density Functional Theory (DFT) revealed that all ground-state properties are determined by the much simpler electron density, they did not provide a practical map for finding the total energy. This left a crucial knowledge gap: how do we leverage this profound insight to perform actual calculations?

This article explores the brilliant solution to this dilemma: the Kohn-Sham equations. Across the following sections, you will discover the elegant theoretical framework that made DFT the most widely used method in quantum electronic structure calculations. The first chapter, "Principles and Mechanisms," will unpack the ingenious idea of replacing the real system with a fictitious one, deconstruct the effective potential that governs it, and explain the self-consistent procedure used to solve the equations. Following this, the chapter on "Applications and Interdisciplinary Connections" will showcase the immense practical power of this approach, demonstrating how it is used to understand and design materials, predict chemical reactions, and even watch electrons move in real-time.

Imagine trying to predict the precise movement of every planet, moon, and asteroid in the solar system, all at once. The gravitational pull of each body on every other body creates a web of interactions so complex that an exact solution is a fantasy. The world of electrons inside an atom or molecule is a thousand times worse. Each electron repels every other electron, and due to the strange laws of quantum mechanics, their fates are inextricably intertwined. This is all encoded in the infamous many-electron Schrödinger equation, a mathematical monster whose exact solution for anything more complex than a hydrogen atom is, for all practical purposes, impossible.

For decades, this "many-body problem" was the great wall of quantum chemistry and physics. Then, in the 1960s, a breakthrough of sublime elegance occurred. The Hohenberg-Kohn theorems revealed a startling truth: the fantastically complex, multi-dimensional wavefunction of all the electrons—the very thing we thought we needed—is not necessary. Instead, every property of the system's ground state, including its energy, is uniquely determined by a much simpler, three-dimensional quantity: the ​​electron density​​, $n(\mathbf{r})$. This is the probability of finding an electron at a particular point $\mathbf{r}$ in space.

Think of it this way: instead of tracking the individual position of every single person in a crowded city, you could learn almost everything you need to know about the city's life—its traffic, its economy, its energy use—just from a map of population density. This is the promise of Density Functional Theory (DFT). It's an elegant swindle, trading an impossible-to-know wavefunction for a knowable density.

But there's a catch, a rather big one. The Hohenberg-Kohn theorems prove that a magical "universal functional" of the density, $F[n]$, exists, but they don't tell us what it looks like! A major component of this functional is the kinetic energy of the interacting electrons, a term that turns out to be fiendishly difficult to write down as a simple function of density. So, while we know a shortcut exists in principle, we don't have the map. We are left standing at the entrance to a promised land we cannot enter.

This is where Walter Kohn and Lu Jeu Sham entered the scene with a truly brilliant idea, a piece of physical intuition that has been called one of the most ingenious in the history of physics. Their idea was this: If the real system of interacting electrons is too hard, let's invent a fake one that's easy.

Let's imagine a parallel universe populated by fictitious, non-interacting "Kohn-Sham" electrons. Because they don't interact with each other, their Schrödinger equation is trivial to solve—it's just a set of independent, single-particle equations. But here is the crucial stipulation, the masterstroke: we will design this fictitious universe so that the electron density of our fake, non-interacting particles is exactly the same as the density of the real, interacting electrons we actually care about.

If we can pull this off, we have a way out of our dilemma. The largest and most troublesome part of the kinetic energy can be calculated exactly for our simple, non-interacting system. We can calculate it by solving the simple one-electron equations for our fake particles to get their wavefunctions—now called ​​Kohn-Sham orbitals​​, $\phi_i(\mathbf{r})$—and then summing up their individual kinetic energies.

It's important to realize that these "non-interacting" electrons are still fermions. They must obey the Pauli exclusion principle. This is not forgotten; it's enforced by building the total state of the fictitious system as a Slater determinant from the individual Kohn-Sham orbitals. This mathematical construction ensures that no two electrons can occupy the same state, a fundamental feature of the electronic world that gives structure to the periodic table and stability to matter itself.

The problem is now transformed. We are no longer trying to solve the monstrous many-body Schrödinger equation. Instead, we are trying to find the perfect effective potential for our fictitious, non-interacting electrons that dupes them into arranging themselves into the exact density of the real system. The equations these fictitious electrons obey are the celebrated ​​Kohn-Sham equations​​:

Here, $-\frac{1}{2}\nabla^2$ is the kinetic energy operator (in atomic units), $\phi_i(\mathbf{r})$ is the $i$-th Kohn-Sham orbital with its corresponding orbital energy $\epsilon_i$, and $v_{s}(\mathbf{r})$ is the all-important effective Kohn-Sham potential.

So, what is this magic potential, $v_s(\mathbf{r})$, that orchestrates the behavior of our fake electrons? It's constructed from three distinct pieces, each with a clear physical meaning:

1. ​​The External Potential, $v_{\text{ext}}(\mathbf{r})$​​: This is the simplest part. It's the same potential from the "real world," primarily the electrostatic attraction from the atomic nuclei. Our fake electrons are tethered to the same atomic framework as the real ones.

2. ​​The Hartree Potential, $v_{\text{H}}(\mathbf{r})$​​: This term accounts for the classical electrostatic repulsion. Each electron feels the repulsion from the average cloud of all other electrons. It’s given by the integral over the total electron density $n(\mathbf{r}')$:

   $$v_H(\mathbf{r}) = \int d\mathbf{r}'\, \frac{n(\mathbf{r}')}{|\mathbf{r} - \mathbf{r}'|}$$

   This is an intuitive, mean-field concept. You can think of it as the electron at position $\mathbf{r}$ interacting not with every other individual electron, but with the smoothed-out charge cloud they create.

3. ​​The Exchange-Correlation Potential, $v_{\text{xc}}(\mathbf{r})$​​: This is the heart of the matter. It is the repository for all the complex quantum mechanical effects that we sidestepped. It's the "magic dust" that corrects our simple picture. It accounts for two main things: (a) quantum ​​exchange​​ effects arising from the Pauli principle (which go beyond simple electrostatic repulsion), and (b) quantum ​​correlation​​ effects, describing how electrons dynamically avoid each other. It also contains the correction for the kinetic energy—the difference between the true kinetic energy of the interacting system and the kinetic energy of our non-interacting fake system. The exact form of $v_{\text{xc}}$ is unknown and unknowable. It is the term we are forced to approximate.

The genius of KS-DFT is that it isolates all our ignorance into this one term, $v_{\text{xc}}$. And, while we don't know it exactly, decades of brilliant work have produced a hierarchy of increasingly accurate approximations for it. This framework can also be readily extended to systems with a net electron spin, like magnets, by defining separate densities and potentials for spin-up and spin-down electrons, leading to spin-polarized KS equations.

At this point, you might notice a circular dependency, a classic chicken-and-egg problem. To find the orbitals ($\phi_i$), we need to know the potential ($v_s$). But to build the potential (specifically the $v_{\text{H}}$ and $v_{\text{xc}}$ parts), we need to know the electron density ($n$), which in turn is built from the orbitals!

How do we break this circle? We don't. We walk around it until we find a stable point. This iterative procedure is called the ​​Self-Consistent Field (SCF) cycle​​, and it is the computational engine of DFT. The process works like this:

1. ​​Guess:​​ We start by making an initial guess for the electron density, $n_{\text{in}}(\mathbf{r})$. A common trick is to just superimpose the densities of the individual, isolated atoms.
2. ​​Construct:​​ Using this guessed density, we construct the Kohn-Sham potential, $v_s(\mathbf{r})$.
3. ​​Solve:​​ We solve the Kohn-Sham equations with this potential to obtain a new set of orbitals, $\phi_i(\mathbf{r})$.
4. ​​Update:​​ We use these new orbitals to calculate a new, output electron density, $n_{\text{out}}(\mathbf{r}) = \sum_i |\phi_i(\mathbf{r})|^2$.
5. ​​Compare and Repeat:​​ We compare the output density, $n_{\text{out}}$, with our initial input density, $n_{\text{in}}$. If they are the same (within some small tolerance), it means our potential has produced a density that generates the very same potential back again. The system is ​​self-consistent​​! We have found the solution. If they are different, we mix the old and new densities to create a better guess for the next iteration and go back to step 2.

The quantity that must converge, that must remain unchanged from one iteration to the next, is the ​​electron density​​. Once the density is stable, all the properties that depend on it, like the total energy, are also determined.

The practical power of the Kohn-Sham method lies in finding good approximations for the exchange-correlation functional. The simplest ones depend only on the local density (Local Density Approximation, LDA), while more sophisticated ones also use its gradient (Generalized Gradient Approximations, GGA).

For even higher accuracy, physicists developed ​​hybrid functionals​​. These functionals mix in a portion of "exact" exchange, calculated directly from the Kohn-Sham orbitals in a manner similar to the older Hartree-Fock theory. This introduces a non-local exchange operator into the equations, meaning the potential at a point $\mathbf{r}$ depends on the orbitals everywhere. These more complex ​​generalized Kohn-Sham equations​​ are computationally more demanding but often cure some of the persistent errors of simpler functionals, yielding much more accurate predictions for things like semiconductor band gaps.

The fundamental idea of Kohn and Sham is so powerful that it has even been extended to follow electrons in time. ​​Time-dependent DFT (TDDFT)​​ allows us to study how the electron density evolves when a system is perturbed, for example, by a laser pulse. It relies on a time-dependent version of the Kohn-Sham equations, enabling scientists to calculate electronic excitation energies and predict the colors of molecules and materials.

From an impossible problem to a practical, powerful, and predictive tool used by tens of thousands of scientists every day, the Kohn-Sham equations represent a triumph of physical insight. They are a beautiful testament to the idea that sometimes, the best way to solve a difficult problem is to solve a simpler, fictitious one perfectly.

In the previous chapter, we unveiled the elegant deception at the heart of Density Functional Theory: the Kohn–Sham equations. We saw how the intractable problem of many interacting electrons could be magically replaced by a fictitious system of non-interacting "Kohn-Sham" electrons moving in a clever effective potential. This was a triumph of theoretical ingenuity. But is it just a beautiful idea, or is it a useful one?

The answer, it turns out, is that the Kohn-Sham formalism is one of the most powerful and versatile tools in the arsenal of modern science. It is a universal translator, allowing us to ask questions in the language of chemistry, materials science, or physics, and receive answers from the fundamental laws of quantum mechanics. In this chapter, we will take a tour of this vast landscape of applications, seeing how these simple-looking equations form the bedrock for understanding and designing the world around us, from the functioning of a drug molecule to the properties of a solar cell.

The first and most pragmatic application of the Kohn-Sham equations is that we can actually solve them. For a real molecule or solid, this isn't a pencil-and-paper exercise; it's a computational task. The process is a beautiful "dance" of self-consistency. We begin with a guess for the electron density, $n(\mathbf{r})$. From this guess, we construct the effective potential $v_{\mathrm{eff}}(\mathbf{r})$. We then solve the Kohn–Sham equations for a single particle in this potential to get a set of orbitals. From these new orbitals, we build a new, improved density. Then we start over, feeding this new density back into the potential. We repeat this loop—potential from density, orbitals from potential, density from orbitals—until the density stops changing. The system has settled into a stable, self-consistent harmony, where the electrons move in a potential they themselves create. This iterative procedure, known as the Self-Consistent Field (SCF) cycle, is the computational engine of DFT.

Of course, nature presents us with further challenges. For an atom like silicon or gold, most electrons are buried deep in the atomic core, held tightly by the nucleus. These core electrons have wavefunctions that oscillate wildly and form a sharp "cusp" at the nucleus. Representing these sharp features with a smooth mathematical basis, like the plane waves used in solid-state physics, would require an absurdly large and computationally impossible number of functions.

Here again, physical intuition comes to the rescue with a brilliantly pragmatic idea: the pseudopotential. Since the core electrons are chemically inert and don't participate in bonding, we can replace the powerful, singular potential of the nucleus and the complicated core electrons with a much weaker, smoother "pseudopotential." This new potential is carefully crafted to have the same effect on the outer valence electrons as the real thing. The resulting "pseudo-wavefunctions" for the valence electrons are smooth and nodeless in the core region, making them dramatically easier to describe computationally. We essentially "sand down" the sharp, difficult parts of the problem that aren't relevant to chemistry, making calculations on complex materials feasible. It is a masterpiece of focusing our computational effort where it matters most.

Now that we can solve the equations, what do the results actually mean? What is this strange Kohn-Sham world we have constructed? A fascinating insight comes from "inversion" problems. Imagine we have the exact electron density for a simple interacting system. We can then ask: what, precisely, must the Kohn-Sham potential $v_s(x)$ look like to reproduce this density?

When one performs this exercise, even for the simplest case of two electrons in a one-dimensional box, a profound truth is revealed. The Kohn-Sham potential is not the simple, flat-bottomed potential of the box. Instead, it develops bumps and wiggles—features that have no classical analogue. These features are the physical manifestation of the exchange-correlation potential, $v_{\mathrm{xc}}$. This potential is nature's way of encoding all the complex, many-body quantum effects of electron correlation and exchange into a simple, single-particle potential. It contorts the landscape in just the right way to guide the non-interacting Kohn-Sham particles so that their collective density exactly matches that of the real, interacting electrons.

Once we have this precious, self-consistent electron density, $\rho(\mathbf{r})$, we can analyze its structure to extract chemical meaning. One of the most powerful tools for this is the Laplacian of the density, $\nabla^2\rho$. This quantity acts as a local "charge concentration detector." In regions where $\nabla^2\rho < 0$, charge is locally concentrated, as you'd find at the position of a nucleus or in the middle of a covalent bond. Where $\nabla^2\rho > 0$, charge is locally depleted. By mapping this function, we can "see" the atoms and bonds within the computed density. This forms the basis of the Quantum Theory of Atoms in Molecules (QTAIM), which allows us to classify chemical bonds based on the topology of the density at the "bond critical point" between two atoms. A negative Laplacian signifies a shared-shell, covalent bond, while a positive Laplacian indicates a closed-shell interaction, typical of ionic bonds or weaker forces like hydrogen bonds. The Kohn-Sham equations provide the raw numerical density, and topological analysis translates it into the rich language of chemistry.

The flexibility of the Kohn-Sham framework allows it to be adapted to an incredible variety of physical systems. For a materials scientist studying a crystalline surface, the system is periodic in two directions (the plane of the surface) but finite in the third (perpendicular to the surface). The Kohn-Sham equations gracefully handle this mixed geometry. By combining the 2D Bloch theorem for the in-plane periodicity with open boundary conditions for the finite direction, we can build accurate models of surfaces, interfaces, and thin films—the fundamental components of modern electronics and catalysis.

What about heavier elements, like gold or platinum, where electrons move so fast that relativistic effects become important? Once again, the Kohn-Sham Hamiltonian can be extended. By incorporating the leading-order relativistic corrections from the Pauli Hamiltonian—the "mass-velocity" term that accounts for the increase of mass with speed, and the "Darwin" term that arises from the electron's jittery quantum motion (Zitterbewegung)—we create scalar-relativistic Kohn–Sham equations. This allows DFT to provide reliable predictions for the properties of materials across the entire periodic table, where relativity is not a subtle correction but a dominant effect that determines color, chemical reactivity, and stability. In atomic units, this leads to a modified Hamiltonian of the form:

This power, however, comes with a responsibility to be a skilled craftsperson. The beautiful, systematic convergence seen when improving the basis sets for traditional wavefunction methods is not always guaranteed in DFT. The reason is profound: DFT's main source of error is often the approximation in the exchange-correlation functional itself, a flaw that isn't cured by simply using a bigger basis set. This means that unlike in wavefunction theory, the path to the "right" answer in DFT is not always a smooth, monotonic descent, and requires careful choice of both the functional and the basis set.

So far, we have focused on static, ground-state properties. But the world is dynamic; it is full of motion, reactions, and responses to external stimuli. The Kohn–Sham framework can be extended to describe this too, through Time-Dependent DFT (TD-DFT). By making the external potential a function of time, we can solve the time-dependent Kohn-Sham equation, $i\partial_t \psi(\mathbf{r}, t) = \hat{H}(t) \psi(\mathbf{r}, t)$, and watch the system evolve in real time.

For example, we can simulate what happens when a molecule is hit by a laser pulse. By adding a time-dependent electric field term to the Hamiltonian, we can watch the electron density slosh back and forth, driving charge transfer from one part of a system to another. This is precisely the kind of simulation used to understand the initial steps of photocatalysis, where light absorption on a catalyst surface leads to the charge separation needed to split water. TD-DFT provides a quantum-mechanical movie camera, giving us direct insight into the ultrafast electronic processes that drive photochemistry and photovoltaics.

We can also use the theory not just to observe, but to control. What if we want to study a charge-transfer excited state, where an electron has moved from a donor molecule to an acceptor? Such a state is not the ground state. Using the technique of Constrained DFT (cDFT), we can force the system into this configuration. By adding a constraint potential via a Lagrange multiplier to the Hamiltonian, we can enforce a specific condition—for instance, that a certain number of electrons must reside on a given fragment of the system. The SCF cycle then finds the lowest-energy state that satisfies this constraint. This is an incredibly powerful tool for calculating the energies of excited states, understanding electron transfer reactions, and designing molecules with specific electronic properties.

The genius of the Kohn-Sham mapping is its generality. It provides a blueprint for replacing a complex, interacting problem with a simpler, non-interacting one. This idea is so powerful that it continues to be extended to new frontiers of physics. One of the most exciting recent developments is Quantum Electrodynamical DFT (QED-DFT).

In this emerging field, we treat not only the electrons but also the photons of the electromagnetic field as quantum-mechanical entities. This is essential in environments like optical cavities, where light and matter can become so strongly coupled that they form new, hybrid quasi-particles called "polaritons." The Kohn-Sham idea can be generalized to this coupled electron-photon system. The result is a set of coupled Kohn-Sham-like equations: one for the electrons moving in an effective potential, and another for the quantum photon modes driven by the motion of the electrons. Solving these equations self-consistently gives us the properties of the polaritonic states.

This marks a full circle. From a clever trick to solve the electronic structure problem, the Kohn–Sham idea has evolved into a paradigm for tackling coupled many-body systems of all kinds. It is a testament to the fact that a truly deep physical insight has reverberations far beyond its original application, continuing to provide a framework for new discoveries at the very edge of our understanding. The journey that began with a fictitious system of electrons is now leading us into a world where light and matter dance together as one.