3.1 Born-Oppenheimer Approximation

Separating Electronic and Nuclear Motions

The Born-Oppenheimer approximation is the starting point for nearly all molecular quantum mechanics. It lets you break an impossibly complex problem (solving the Schrödinger equation for all electrons and nuclei simultaneously) into two smaller, manageable problems. Without it, calculating the properties of even simple molecules would be intractable.

The core idea rests on a physical observation: nuclei are thousands of times heavier than electrons. A proton alone is roughly 1836 times the mass of an electron. Because of this mass difference, electrons move far faster and effectively adjust instantaneously to any change in nuclear positions. You can therefore treat the nuclei as if they're frozen in place while solving for the electronic structure, then use the electronic energy to figure out how the nuclei move.

The Born-Oppenheimer Approximation

The approximation works through a two-step procedure:

1. Fix the nuclei in place. Choose a set of nuclear positions and hold them constant (the "clamped-nuclei" picture).
2. Solve the electronic problem. With the nuclei frozen, solve the electronic Schrödinger equation to get the electronic energy at that configuration.
3. Repeat for many configurations. Solve the electronic problem at many different nuclear arrangements. The collection of electronic energies as a function of nuclear position defines a potential energy surface (PES).
4. Solve the nuclear problem. Use the PES as the potential in a separate Schrödinger equation for nuclear motion. This gives you vibrational and rotational energy levels.

The total molecular wave function is approximated as a product of electronic and nuclear parts:

$\Psi_{\text{total}}(\vec{r},\vec{R}) \approx \Psi_{\text{elec}}(\vec{r};\vec{R})\,\Psi_{\text{nuc}}(\vec{R})$

Here $\vec{r}$ represents all electronic coordinates and $\vec{R}$ represents all nuclear coordinates. Notice the semicolon in the electronic wave function: $\vec{R}$ enters as a parameter, not as a dynamical variable. The electrons "see" the nuclei at fixed positions, and the wave function changes smoothly as those positions change. This smooth adjustment is why the approximation is also called the adiabatic approximation.

Consequences of the Born-Oppenheimer Approximation

Separating electronic and nuclear motion has several far-reaching consequences:

• Potential energy surfaces become well-defined. The PES gives you the equilibrium geometry (the energy minimum), vibrational frequencies (from the curvature near that minimum), and reaction pathways (minimum-energy paths connecting reactants to products). Without the Born-Oppenheimer approximation, the concept of a PES doesn't even exist, because you can't define an electronic energy that depends only on nuclear positions.
• Electronic and vibrational spectra separate. Electronic transitions happen on a much faster timescale than nuclear motion. This is the physical basis of the Franck-Condon principle: during an electronic transition, the nuclei barely move, so the transition is essentially "vertical" on the PES.
• Adiabatic and diabatic representations emerge. Adiabatic states are the eigenstates of the electronic Hamiltonian at each fixed nuclear geometry. They change character smoothly as nuclei move. Diabatic states are constructed differently: they're chosen to minimize the derivative coupling between electronic states during nuclear motion, which makes them useful for describing dynamics near curve crossings.

Born-Oppenheimer Approximation for Molecular Wave Functions

Separating the Molecular Wave Function

The electronic Schrödinger equation at fixed nuclear coordinates is:

$\hat{H}_{\text{elec}}\,\Psi_{\text{elec}}(\vec{r};\vec{R}) = E_{\text{elec}}(\vec{R})\,\Psi_{\text{elec}}(\vec{r};\vec{R})$

$\hat{H}_{\text{elec}}$ contains the electron kinetic energy, electron-electron repulsion, and electron-nucleus attraction. Crucially, it does not contain the nuclear kinetic energy. The eigenvalue $E_{\text{elec}}(\vec{R})$ depends on which nuclear configuration you chose, so it's a function of $\vec{R}$.

Once you have $E_{\text{elec}}(\vec{R})$, it enters the nuclear Schrödinger equation as the potential:

$\left[\hat{T}_{\text{nuc}} + E_{\text{elec}}(\vec{R})\right]\Psi_{\text{nuc}}(\vec{R}) = E_{\text{total}}\,\Psi_{\text{nuc}}(\vec{R})$

$\hat{T}_{\text{nuc}}$ is the nuclear kinetic energy operator, and $E_{\text{total}}$ is the total molecular energy. The solutions $\Psi_{\text{nuc}}(\vec{R})$ describe vibrations and rotations of the molecule on the potential energy surface.

What's being neglected here? The exact molecular Hamiltonian contains terms where the nuclear kinetic energy operator acts on the electronic wave function (these are the nonadiabatic coupling terms). The Born-Oppenheimer approximation drops these terms entirely, which is justified when the electronic state is well-separated in energy from other states.

Constructing Potential Energy Surfaces

Building a PES in practice means solving the electronic Schrödinger equation at a grid of nuclear geometries and mapping out $E_{\text{elec}}(\vec{R})$. For a diatomic molecule, the PES is just a curve: energy as a function of internuclear distance $R$. For a triatomic molecule, it becomes a surface over multiple internal coordinates.

The PES encodes key molecular properties:

• Equilibrium geometry: the nuclear configuration at the global energy minimum. For $\text{H}_2$, this is the bond length of about 0.74 Å.
• Vibrational frequencies: determined by the curvature (second derivative) of the PES at the minimum. A steeper well means a stiffer bond and higher vibrational frequency.
• Dissociation energy: the depth of the potential well relative to the separated-atom limit.
• Reaction pathways: the minimum-energy path on a multidimensional PES connecting reactant and product valleys, passing through a transition state (saddle point).

For polyatomic molecules, the PES is a high-dimensional object ($3N - 6$ internal coordinates for a nonlinear molecule with $N$ atoms), so it can't be fully visualized. Computational chemists typically map out slices or paths along specific coordinates of interest.

Validity and Limitations of the Born-Oppenheimer Approximation

Conditions for Validity

The approximation holds well when the electronic state of interest is energetically well-separated from all other electronic states across the relevant nuclear configurations. Quantitatively, the energy gap between electronic states should be much larger than the energy scale of nuclear motion (vibrational quanta are typically on the order of $10^{-1}$ to $10^{-2}$ eV, while electronic gaps are often several eV).

Ground-state molecules near their equilibrium geometry usually satisfy this condition comfortably. The ground electronic state of $\text{H}_2$, for example, is separated from the first excited state by several eV at the equilibrium bond length, so the Born-Oppenheimer approximation works extremely well there.

The approximation becomes less reliable for:

• Excited electronic states, where energy gaps to neighboring states are often smaller
• Stretched or distorted geometries, where electronic states can approach each other in energy
• Heavy atoms with strong spin-orbit coupling, where relativistic effects mix states of different spin

Limitations and Breakdown

The Born-Oppenheimer approximation fails when electronic and nuclear motions become strongly coupled. The most important scenarios are:

• Conical intersections: These are points (or seams) in nuclear coordinate space where two electronic PESs become exactly degenerate. Near a conical intersection, the nonadiabatic coupling diverges, and the system can efficiently transfer between electronic states. Conical intersections are central to photochemistry; they provide the "funnels" through which excited molecules return to the ground state on ultrafast timescales.
• Avoided crossings: When two PESs of the same symmetry approach each other but don't actually cross (per the noncrossing rule in one dimension), the gap can become small enough that nonadiabatic transitions occur with significant probability.
• Jahn-Teller distortions: For molecules with degenerate electronic states (due to symmetry), the nuclei spontaneously distort to break the symmetry and lift the degeneracy. The coupling between electronic degeneracy and nuclear displacement is inherently non-Born-Oppenheimer.
• Spin-orbit and vibronic coupling: Spin-orbit coupling mixes electronic states of different spin multiplicity, while vibronic coupling mixes electronic and vibrational degrees of freedom. Both effects are neglected in the standard Born-Oppenheimer framework.

These breakdowns matter in real chemistry. Photochemical reactions, internal conversion, intersystem crossing, and electron transfer processes all involve nonadiabatic dynamics where the Born-Oppenheimer picture is insufficient.

To handle these situations, several post-Born-Oppenheimer methods exist:

• Surface hopping and other nonadiabatic dynamics methods propagate nuclear trajectories that can switch between electronic surfaces.
• Diabatic representations recast the problem so that the coupling appears as off-diagonal potential terms rather than derivative couplings, which are numerically easier to handle.
• Multireference electronic structure methods (CASSCF, MRCI) properly describe near-degenerate electronic states that single-reference methods cannot.

Despite these limitations, the Born-Oppenheimer approximation remains the foundation of molecular quantum mechanics. Nearly every electronic structure calculation you'll encounter assumes it, and understanding where it works and where it breaks down is essential for interpreting molecular spectra and reaction dynamics.