A structural model that predicts molecular geometry by assuming that electron pairs (bonding and lone pairs) around a central atom repel one another and adopt arrangements that minimize repulsion, determining bond angles and molecular shape.

Principle

Electron pair domains occupy regions around a central atom and arrange in three‑dimensional geometries that minimize electron‑electron repulsion; lone pairs exert greater repulsive influence than bonding pairs, altering idealized bond angles.

Demonstration

Using VSEPR, water (H2O) is predicted to be bent because two bonding pairs and two lone pairs on oxygen adopt a roughly tetrahedral electron‑domain geometry; the greater repulsion of lone pairs reduces the H–O–H angle from the tetrahedral 109.5° to ~104.5°.

Misapplication

Treating multiple bonds as identical to single bonds in repulsion strength, or applying VSEPR to transition‑metal complexes with significant d‑orbital participation and steric/electronic ligand effects, can produce incorrect geometries.

Consequence

Correct application yields quick, qualitative predictions of molecular shapes and approximate bond angles for many main‑group molecules, aiding structure interpretation, nomenclature, and initial model building for further quantum calculations.

Reversal

Deriving geometry from molecular orbital occupancy, ligand field stabilization, or electronic energy minimization (quantum chemical geometry optimization) inverts the approach by predicting shape from electronic structure calculations rather than steric pair repulsion heuristics.

Boundary

VSEPR is most reliable for simple main‑group central atoms with well‑defined valence electron pairs; it is less predictive for hypervalent species, delocalized bonding, transition‑metal coordination complexes, and cases where steric bulk or electronic effects of ligands dominate.

Semantic Tension

Competes with hybridization and MO‑based explanations: VSEPR gives a steric, domain‑based rationale for geometry, while hybridization/MO approaches explain geometry as a consequence of orbital mixing and electronic energy; reconciliation often requires invoking both views.

Synthesis

VSEPR is a pragmatic steric model: by treating bonding and lone pairs as electron domains that repel, it provides an accessible route to predict and rationalize molecular shapes and deviations from ideal angles, while acknowledging its limits in systems with strong electronic or ligand effects.