Two compounds with the same molecular formula but a different arrangement of atoms are isomers. In coordination chemistry they fall into two families: structural (constitutional) isomerism, where bonds connect different atoms, and stereoisomerism, where the connectivity is the same but the spatial arrangement differs.
Structural isomerism includes:
- Ionisation isomerism — the counter ion and a coordinated ligand swap places, giving different ions in solution, e.g. [Co(NH3)5SO4]Br and [Co(NH3)5Br]SO4.
- Linkage isomerism — an ambidentate ligand binds through a different donor atom, e.g. [Co(NH3)5(NO2)]2+ (nitrito-N, yellow) vs [Co(NH3)5(ONO)]2+ (nitrito-O, red).
- Coordination isomerism — ligands are interchanged between cationic and anionic complexes, e.g. [Co(NH3)6][Cr(CN)6] and [Cr(NH3)6][Co(CN)6].
- Solvate (hydrate) isomerism — water is inside or outside the coordination sphere, e.g. [Cr(H2O)6]Cl3 (violet), [Cr(H2O)5Cl]Cl2·H2O and [Cr(H2O)4Cl2]Cl·2H2O (green).
Stereoisomerism includes geometrical (cis–trans) and optical isomerism. Geometrical isomerism appears in square planar [Ma2b2] (e.g. cis- and trans-[Pt(NH3)2Cl2]) and in octahedral [Ma4b2] and [Ma3b3] (fac–mer) complexes. Tetrahedral complexes do not show cis–trans isomerism. Optical isomerism arises when a complex is non-superimposable on its mirror image (chiral), most commonly in octahedral chelates such as cis-[Co(en)2Cl2]+ and [Co(en)3]3+; the two forms are the dextro (d) and laevo (l) enantiomers.
Valence Bond Theory (VBT) treats bonding as overlap of empty hybrid orbitals of the metal ion with filled ligand orbitals (each ligand donates a lone pair). The hybridisation fixes the geometry:
- $sp^3$ → tetrahedral (CN 4, e.g. [NiCl4]2−).
- $dsp^2$ → square planar (CN 4, e.g. [Ni(CN)4]2−).
- $d^2sp^3$ → octahedral using inner $(n-1)d$ orbitals (inner-orbital / low-spin, e.g. [Co(NH3)6]3+).
- $sp^3d^2$ → octahedral using outer $nd$ orbitals (outer-orbital / high-spin, e.g. [CoF6]3−).
Whether a strong-field ligand pairs the d electrons (inner orbital) decides the number of unpaired electrons and hence the spin-only magnetic moment $\mu=\sqrt{n(n+2)}$ BM. Limitations of VBT: it does not explain colour, gives no quantitative splitting energy, cannot predict the exact magnitude of magnetic moment, and offers no satisfactory reason why some ligands are strong-field and others weak-field.