VSEPR Theory: Why Molecules Have the Shape They Do

One rule: electron domains repel

Valence Shell Electron Pair Repulsion (VSEPR) theory starts from one physical fact: every region of electron density around a central atom — whether it's a bonding pair shared with another atom or a lone pair belonging only to the central atom — carries negative charge, and negative charges repel. The molecule's 3D shape is simply whatever arrangement lets those "electron domains" get as far apart from each other as the geometry of 3D space allows.

Steric number = (bonding domains) + (lone pairs) on the central atom

2 domains  →  linear             180°
3 domains  →  trigonal planar    120°
4 domains  →  tetrahedral        109.5°
5 domains  →  trigonal bipyramidal   90° / 120°
6 domains  →  octahedral         90°

Electron geometry vs molecular geometry

These are two different (and often confused) descriptions. Electron geometry counts every domain, lone pairs included, and describes how they're arranged. Molecular geometry describes only where the atoms actually sit — lone pairs are invisible to an X-ray structure or to anyone describing the molecule's shape, even though they're still there, still repelling, and still shoving the visible atoms out of the way.

CH4   4 bonds, 0 lone pairs   →  electron geom: tetrahedral   molecular: tetrahedral
NH3   3 bonds, 1 lone pair    →  electron geom: tetrahedral   molecular: trigonal pyramidal
H2O   2 bonds, 2 lone pairs   →  electron geom: tetrahedral   molecular: bent

All three molecules share the same tetrahedral electron geometry — 4 domains is 4 domains — but they look completely different once you erase the invisible lone pairs and look only at where the atoms are.

Lone pairs push harder than bonding pairs

A refinement matters for bond angles: not all repulsions are equal in strength. A bonding pair is stretched taut between two nuclei, which constrains and slightly weakens its spread; a lone pair belongs to only one nucleus and can spread out more, making it a "fatter," more repulsive domain. The ranking of repulsion strength is:

lone pair - lone pair  >  lone pair - bonding pair  >  bonding pair - bonding pair

This is exactly why NH₃'s H-N-H angle (about 107°) and H₂O's H-O-H angle (about 104.5°) both fall short of the ideal 109.5° from a plain tetrahedral electron geometry — each lone pair present nudges the bonding pairs closer together, and water, with two lone pairs instead of one, gets squeezed a bit harder than ammonia.

The odd one out: trigonal bipyramidal

Five domains is the one case without a single uniform bond angle: three equatorial positions sit in a plane, 120° from each other, while two axial positions sit perpendicular to that plane, 90° from every equatorial position. Because axial positions are more crowded (closer, on average, to more neighbours), lone pairs in a 5-domain molecule strongly prefer the roomier equatorial spots — which is exactly why molecules like SF₄ (one lone pair, 4 bonds) adopt a distinctive "seesaw" molecular shape rather than a simpler alternative.

A genuinely predictive theory

VSEPR's real strength is that it requires nothing more than a Lewis structure — no quantum mechanical calculation — to correctly predict the shape of the vast majority of small molecules, and those shapes have real physical consequences: bent, polar water is an excellent solvent and hydrogen bonder, while linear, nonpolar CO₂ is not, purely because of the geometric consequence of lone-pair count on the central atom.

Frequently asked questions

Try it live

Everything above runs in your browser — open VSEPR Geometry and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.