Avoided Crossing

In the quantum realm, the interactions between energy states define the properties of matter. While it's easy to imagine the discrete energy levels a system can occupy, what happens when these levels, under the influence of changing conditions, appear to be on a collision course? This scenario gives rise to one of quantum chemistry's most fundamental concepts: the ​​avoided crossing​​. This principle, where energy levels of the same symmetry repel each other instead of intersecting, resolves seeming paradoxes and explains a vast array of physical phenomena. This article demystifies the avoided crossing, exploring the rules that govern this quantum 'etiquette.' In the following chapters, we will first unravel the "Principles and Mechanisms" of avoided crossings, examining the critical role of symmetry, the distinction between adiabatic and diabatic viewpoints, and the extension to conical intersections. Subsequently, under "Applications and Interdisciplinary Connections," we will see how this abstract principle has tangible consequences, shaping everything from the geometry of molecules to the dynamics of chemical reactions.

Imagine you are driving down a long, straight road, and far in the distance, you see another car on a different road, headed for the same intersection as you. You watch as the two roads converge. The cars get closer, and closer... and then, instead of colliding, they seem to repel each other, swerving away onto new paths as if guided by an invisible force. In our everyday world, this would imply some last-minute decision-making. In the quantum world, for particles of the same "type," this repulsion is often not a choice, but a fundamental law. This phenomenon, where two energy levels appear on a collision course but then veer away from each other, is known as an ​​avoided crossing​​. It is one of the most subtle yet profoundly important principles in quantum mechanics, shaping everything from the geometry of a water molecule to the flash of a firefly.

First, we must ask: when do energy levels cross, and when do they avoid? The answer lies in the concept of ​​symmetry​​. In quantum mechanics, symmetry is not just about geometric neatness; it's a deep statement about what properties of a system are conserved. Each energy state of a system, like an electron in a molecule, can be labeled with a set of quantum numbers that act like a unique ID badge, specifying its symmetries.

The master rule, known as the ​​Wigner-von Neumann non-crossing rule​​, states that if you are changing a single parameter of a system (say, stretching a bond or applying an electric field), two energy levels belonging to states with the exact same symmetry ID badge cannot cross. They must avoid each other. However, if their ID badges differ in any way—if one state is symmetric and the other is anti-symmetric with respect to some operation, for example—then they are "strangers" to each other. They can pass right through one another, and their energy levels will cross as if nothing happened.

Let's picture a stylized molecule that has a mirror-plane symmetry. We can classify its electronic states as either "symmetric" or "antisymmetric" with respect to reflection in that mirror. Now, let's apply an electric field along an axis that preserves this symmetry. As we dial up the field, we track the energies. We might find that the energy of a certain symmetric state and a certain antisymmetric state are heading towards each other. Because they have different symmetry labels, they are quantum-mechanically invisible to one another. Their energy level curves will intersect at a specific field strength in a ​​true crossing​​.

But what happens if we introduce a small perturbation that breaks the mirror symmetry, like distorting the molecule slightly? Suddenly, the labels "symmetric" and "antisymmetric" are no longer strictly valid. Each state becomes a slight mixture of its former self and the other. They are no longer strangers; they can "see" each other. Now, as we dial up the electric field and the energy levels approach the point where they would have crossed, they instead feel a mutual repulsion. The lower energy state is pushed even lower, and the higher energy state is pushed higher. This is the avoided crossing.

The mechanism behind this repulsion is a ​​coupling​​ between the states, an off-diagonal term in the system's Hamiltonian matrix, let's call it $V$. A true crossing requires two conditions to be met simultaneously: the unperturbed energies must be equal, and the coupling between them must be zero. If the states have different symmetries, the coupling $V$ is forced to be zero for all values of our parameter. So, we only need to meet one condition (equal energies), which is easy to do by tuning our one parameter. But if the states have the same symmetry, the coupling $V$ is generally non-zero. To get a crossing, we would need to tune our single parameter to make both the energies equal and happen to make the coupling term zero at the exact same point—a statistical impossibility. Instead of a crossing, a minimum energy ​​gap​​ of $2|V|$ opens up between the levels.

This picture of swerving energy levels, while correct, can be profoundly misleading. It is a feature of a specific viewpoint called the ​​adiabatic representation​​. The term "adiabatic" here means "happening slowly." In this picture, we solve for the exact energy levels of the system at every single snapshot in time (or for every value of our parameter, like the bond length $R$). These energy levels, the smooth curves you see on a chart, are the adiabatic potential energy surfaces. The problem is, near an avoided crossing, the character of the states themselves can change dramatically.

Let's return to our car analogy. Imagine one car is red and the other is blue. They approach the intersection. In the adiabatic picture, we don't label by color; we label by "lower-energy car" and "higher-energy car." Before the intersection, the red car is the lower-energy car. After the swerve, the blue car is the lower-energy car. From the adiabatic viewpoint, it looks like a single "lower-energy car" approached the intersection and mysteriously changed its color from red to blue! The states appear to have swapped their identities.

This is often not a helpful way to think about what's really happening. Chemists and physicists therefore invented another viewpoint: the ​​diabatic representation​​. In this picture, we follow the properties we care about—the "redness" and "blueness" of the cars. The diabatic states are defined to have a consistent character (e.g., representing a covalent bond versus an ionic bond) that changes as little as possible. In this picture, the red car and the blue car drive straight ahead, and their paths do cross. However, we have to add a new rule: at the intersection point, there is a "coupling" that creates a probability for the red car to suddenly turn blue, and vice versa.

So, an avoided crossing in the "true energy" adiabatic picture is equivalent to a real crossing of "constant character" diabatic states that are coupled together. The adiabatic curves are what you would measure in an infinitely slow experiment, but the diabatic curves often provide a more intuitive picture of chemical processes, where a molecule moving along one diabatic surface can "hop" to another at the crossing point. The strength of the coupling $V$ determines the probability of this hop.

This is not just abstract theory; it has direct, observable consequences. The shapes, colors, and properties of the molecules that make up our world are often dictated by these quantum avoidances.

A classic example is found in the ​​Walsh diagrams​​ that organic chemists use to predict molecular shapes. If we track the energies of the molecular orbitals of, say, an $\mathrm{AH}_2$ molecule as we bend it from linear to a 90-degree angle, we see the energies shift. At some intermediate angle, it may happen that two orbitals of the same symmetry would be driven to cross. They don't. They experience an avoided crossing. One is stabilized (pushed down in energy), and one is destabilized (pushed up). If the molecule has electrons in the orbital that gets stabilized upon bending, the bent geometry will be the molecule's preferred shape. The simple fact that water is bent and not linear—a fact essential for life itself—can be explained by the energetic stabilization gained through an avoided crossing of its molecular orbitals.

Another beautiful example is the bonding in the diatomic molecules of the second period, like nitrogen ($\mathrm{N}_2$) and oxygen ($\mathrm{O}_2$). Naively, one would expect a certain ordering of molecular orbital energies based on simple overlap principles. This holds true for $\mathrm{O}_2$ and $\mathrm{F}_2$. However, for $\mathrm{B}_2$, $\mathrm{C}_2$, and $\mathrm{N}_2$, the experimental ordering is flipped! The reason is an avoided crossing between a molecular orbital derived from atomic $2s$ orbitals and another of the same symmetry derived from atomic $2p$ orbitals. In the lighter elements, the energies of the $2s$ and $2p$ atomic orbitals are quite close, leading to a strong interaction and a large avoided crossing gap. This "s-p mixing" pushes the higher-energy $\sigma$ orbital up so much that it ends up above the $\pi$ orbitals. In oxygen and fluorine, the atomic $2s-2p$ energy gap is much larger, the coupling is weaker, the avoided crossing is less dramatic, and the "normal" orbital ordering is preserved. This subtle quantum effect directly governs the magnetism and reactivity of these fundamental atmospheric gases.

We have seen that for a system depending on a single parameter, levels of the same symmetry do not cross. But what happens in a real polyatomic molecule, which can vibrate and contort in multiple ways at once? What if we have not one, but two or more "knobs" to turn?

In this higher-dimensional world, the non-crossing rule is circumvented. It turns out that to force a genuine degeneracy, you need to satisfy two mathematical conditions. With only one knob (one parameter), you can't guarantee that you'll satisfy both. But with two or more knobs (e.g., two different bond-stretching coordinates), you can! The result is not an avoided crossing, but a ​​conical intersection​​.

Imagine two potential energy surfaces. Instead of swerving to avoid each other, they meet at a single, sharp point, like the vertices of two ice cream cones stacked tip-to-tip. This point of true degeneracy is a ​​conical intersection​​. For a molecule with $N$ internal coordinates, the set of such degeneracy points forms a geometric object of dimension $N-2$.

These are no mere mathematical curiosities; they are the most important gateways for chemistry in the excited state. When a molecule absorbs light, it jumps to a higher-energy adiabatic surface. If its vibrations carry it to a conical intersection, it can "drop" through this funnel down to the lower surface with breathtaking speed—on the timescale of femtoseconds ($10^{-15}$ s). This process is the key to countless photochemical reactions, from the mechanism of vision in your eye to the UV-induced damage of DNA.

-Moreover, conical intersections possess a truly strange topological property. If you trace a path in the nuclear coordinate space that makes a closed loop around the intersection point, the electronic wavefunction comes back to where it started, but with its sign flipped! It acquires a ​​geometric phase​​ (or Berry phase) of $\pi$. This is a profound signature that has no analogue for a simple one-dimensional avoided crossing. It's a testament to the rich and often bizarre geometry of the quantum world, where avoiding a collision on a one-way street is the rule, but in the bustling city of a molecule, there are special points where surfaces can truly meet, acting as the express elevators of chemical change.

Now that we have grappled with the underlying machinery of the avoided crossing, we can step back and admire its handiwork. Where does nature use this elegant little trick? The answer, you may be delighted to find, is everywhere. The same fundamental principle that governs the simplest two-level system orchestrates the grand ballet of chemical reactions, dictates the colors of gemstones, defines the boundary between a metal and an insulator, and may even set the ultimate speed limit for future quantum computers. The avoided crossing is not some esoteric curiosity; it is one of the most versatile and consequential tools in the quantum playbook.

Let's begin our journey in the world of chemistry, where it all comes down to the making and breaking of bonds. Imagine a chemical reaction, say a simple one like a deuterium atom colliding with a hydrogen molecule, $H_2 + D \rightarrow H + HD$, as a journey across a vast, invisible landscape of potential energy. The valleys are stable molecules, and the mountain passes are the transition states. The Born-Oppenheimer approximation gives us these maps by assuming the heavy nuclei move so slowly that the nimble electrons can instantly adjust their configuration for any given arrangement of atoms.

But what happens if there are two different electronic "stories" or configurations that a system could adopt, and their energy landscapes look like they are about to intersect? For instance, one landscape might describe the reactants ($H_2$ bond intact) and another the products ($HD$ bond formed