the Born-Oppenheimer approximation
/ born OH-pen-hy-mer /
A molecule is a swarm of light electrons around a few heavy nuclei, and solving for all of them at once is hopeless. But there is a saving grace: nuclei are thousands of times more massive than electrons, so they lumber while the electrons zip around almost instantly. The Born-Oppenheimer approximation exploits exactly this separation of timescales — it lets the electrons follow the nuclei adiabatically, so you can pretend the nuclei are momentarily frozen while you solve for the electrons.
Concretely, the total wavefunction is approximated as a product, Psi(r, R) ≈ psi_elec(r; R) chi_nuc(R), where r are the electron coordinates and R the nuclear ones. First, you solve the electronic Schrödinger equation with the nuclei clamped at fixed positions R; this gives an electronic energy E_elec(R) for every nuclear geometry. That function E_elec(R), plus the nuclear-nuclear repulsion, acts as a potential-energy surface on which the nuclei then move, described by their own Schrödinger equation. The whole scheme is controlled by the small parameter (m_e/M)^(1/4), the fourth root of the electron-to-nuclear mass ratio. Out of this fall the concepts of equilibrium bond length (the minimum of the surface), molecular vibrations (oscillations about it), and rotations.
The Born-Oppenheimer approximation is the conceptual foundation of essentially all of molecular physics and quantum chemistry — it is what makes a 'molecular structure' a meaningful idea at all. But be honest about where it fails: it breaks down wherever two electronic potential-energy surfaces approach or cross, at avoided crossings and conical intersections. There the nuclear motion can no longer be pinned to a single electronic surface, non-adiabatic couplings become large, and electrons and nuclei exchange energy — precisely the regime that governs photochemistry, radiationless transitions, and vision. It is an approximation, not an exact separation.
For H2+, clamping the two protons a distance R apart and solving for the single electron gives an electronic energy curve E(R) with a minimum near R = 0.106 nm — the predicted bond length. Small oscillations of the protons about that minimum are the molecule's vibrational levels, spaced by hbar omega.
The electronic energy versus nuclear separation IS the potential well the nuclei live in.
The approximation fails at conical intersections and avoided crossings, where electronic surfaces meet and non-adiabatic couplings drive electrons and nuclei to exchange energy — the very processes behind photochemistry and vision.