Werner’s Theory and Experimental Evidence

Why the empirical formula is not enough

A bottle labelled CoCl₃·6NH₃ tells us how many atoms are present but does not tell us which atoms are directly joined. The question becomes sharper when a related compound, CoCl₃·5NH₃, releases only two of its three chlorides to a rapid silver-ion test. A structural model must explain why chemically identical chloride atoms can behave differently in the same material. Coordination chemistry begins by separating the immediate surroundings of a metal from ions that balance charge outside those surroundings.

In a cold aqueous test, freely available chloride precipitates with silver ions: Ag⁺(aq) + Cl⁻(aq) → AgCl(s). The white solid provides a count of chloride that is immediately accessible under the stated conditions. Chloride already bonded to an inert cobalt(III) centre is not necessarily released on this timescale. Thus the experiment probes both constitution and the timescale of ligand substitution; it does not mean that a coordinated chloride can never react.

From four observations to one structural explanation

Every row has six donor atoms directly bonded to cobalt: six nitrogen atoms, five nitrogen plus one chlorine, or four nitrogen plus two chlorine. The metal therefore keeps an octahedral first coordination sphere in this series while the number of external chlorides changes. The two compounds in the last rows have identical composition and the same count of free ions, yet different colours. A model must therefore describe spatial arrangement as well as charge.

Conductivity supplies supporting evidence because dissolved ions carry electric current. A 1:3 electrolyte here means one complex cation for three monovalent anions, whereas a 1:1 electrolyte produces one of each. Counting ions alone does not give an exact numerical conductivity: charge, mobility, concentration and ion pairing also matter. The evidence supports the formulations when these experimental factors are controlled.

Werner’s two valences

For the cobalt(III) series, primary valence is three and secondary valence is six. Neutral ammonia contributes no formal charge but occupies a coordination site. A coordinated chloride contributes −1 to charge and occupies one site. Chloride outside the bracket balances the charge of the entity but does not increase the coordination number. This is why oxidation state and coordination number answer different questions.

Werner proposed ionisable primary linkages, non-ionisable secondary linkages and characteristic directions for secondary linkages. These ideas explain the cobalt data. In modern language, however, bonding has both electrostatic and covalent aspects, and the coordination number is not permanently fixed for every compound of a metal. Nickel can occur in four- and six-coordinate complexes. The historical postulates are a structural starting point, not a universal rule that forbids ligand exchange or alternative coordination numbers.

Solving the inverse problem

Double salts and persistent complex ions

A double salt is a crystalline combination that supplies its component simple ions on dissolving. Mohr’s salt, (NH₄)₂Fe(SO₄)₂·6H₂O, gives ammonium, sulfate and hydrated Fe²⁺ in water, so suitable Fe²⁺ tests remain possible. Carnallite, KCl·MgCl₂·6H₂O, and potash alum, KAl(SO₄)₂·12H₂O, similarly release component ions. The waters of crystallisation become part of the surrounding solvent and hydration environment.

By contrast, K₄[Fe(CN)₆] dissolves principally into four K⁺ and [Fe(CN)₆]⁴⁻. It does not provide the same free Fe²⁺ and CN⁻ concentrations as a mixture of simple salts. Formation equilibria still exist, but the stable complex is the dominant species under suitable conditions. All aqueous metal ions are solvated; the comparison concerns whether the distinctive coordination entity persists, rather than whether a completely bare metal ion exists in water.

Quick Quiz