Geometry Prediction Calculator

VSEPR theory treats the electron groups around a central atom as mutually repelling regions of charge that spread as far apart as geometry allows. The steric number, the sum of bonded atoms and lone pairs, fixes the electron-domain arrangement: two groups give linear at 180 degrees, three trigonal planar at 120, four tetrahedral at 109.5, five trigonal bipyramidal at 90 and 120, and six octahedral at 90. Multiple bonds count as one domain, so carbon dioxide with two double bonds is linear. The molecular shape names only the atom positions, which is why ammonia with a steric number of four is trigonal pyramidal and water is bent even though both have tetrahedral electron geometry. Lone pairs, held by one nucleus rather than shared between two, occupy a wider angular spread and compress the remaining angles, giving the observed sequence of about 109.5 degrees in methane, 107 in ammonia and 104.5 in water. In five-coordinate systems lone pairs preferentially take equatorial sites, which is why SF4 is a seesaw, ClF3 is T-shaped and XeF2 is linear; in six-coordinate systems two lone pairs go trans, making XeF4 square planar. The associated hybridisation labels sp, sp2 and sp3 follow the steric number, but the sp3d and sp3d2 labels traditionally quoted for five- and six-coordinate main-group species are now regarded as poor descriptions, since computational work shows d-orbital participation in period 3 elements is minimal and three-centre four-electron bonding accounts for the geometry better. VSEPR also performs unreliably for transition metal complexes, where crystal field effects dominate, and for some heavy main-group species with stereochemically inert lone pairs. Bent's rule provides the usual correction when substituent electronegativities differ.

The repulsion idea was first stated by Nevil Sidgwick and Herbert Powell of the University of Oxford in a 1940 paper in the Proceedings of the Royal Society, where they surveyed known molecular shapes and noted that the arrangement of valence electron pairs, both bonding and non-bonding, correlated with the observed geometry better than bonding pairs alone. Their conceptual groundwork lay in Gilbert Lewis's 1916 electron pair model of the bond and in the hybridisation scheme Linus Pauling introduced in 1931 to explain the tetrahedral carbon atom. The theory took its predictive modern form in a 1957 Quarterly Reviews article by Ronald Gillespie and Ronald Nyholm, then both at University College London. They added the ordering of repulsion strengths, with lone pair to lone pair exceeding lone pair to bonding pair and both exceeding bonding pair to bonding pair, which explains the systematic compression of angles, and they extended the treatment to multiple bonds and to the axial and equatorial preferences of the trigonal bipyramid. The acronym VSEPR came into use shortly afterwards and the model spread rapidly through undergraduate teaching because it needs no calculation. Gillespie continued refining it for decades at McMaster University, later proposing the ligand close packing model in the 1990s to account for cases where ligand-ligand contact rather than electron pair repulsion sets the bond angles.