text, no distinction will be made. For all cases the values of equilibrium constants quoted are based on
concentrations and represent approximations to the true values. Equilibrium constants are variously
named depending upon the nature of the equilibria to which they refer, e.g. dissociation constants,
formation constants or solubility products. The most convenient concentration units for analytical
chemistry are mol dm–^3 and the dimensions of the constants will be derived from these and determined
by the stoichiometry of the reaction, e.g.
a + b + c +...
(reactants)x + y + z +...
(products)dimensions of Ke(^33) dimensionless
(^34) mol dm–^3
(^32) mol–^1 dm^3
Care must be exercised in making direct comparisons between Ke values, and due attention should be
given to their dimensions.
Equilibria in Analytical Reactions
In analytical chemistry it is often necessary to shift the position of equilibrium in a reaction so as to
obtain larger concentrations of the desired products. For example, sometimes a complete reaction
between an analyte and a reagent or between an interfering ion and a masking reagent is essential. One
important and widely used method of achieving a shift involves the removal of one of the products from
the system, e.g. by distillation or precipitation. In these circumstances to maintain Ke constant the
reactants will be steadily converted to products until one or all are exhausted. An alternative approach
that is often useful uses an excess of the reagent which in most cases causes a shift towards the products
of the reaction. This is the so-called common ion effect. Too large an excess, however, can partially
reverse the shift or lead to some other undesirable effect (p. 218).
When comparing similar or parallel reactions, consideration of the changes in Gibbs free energy ∆G,
enthalpy ∆H and entropy ∆S can be valuable. The equilibrium constant is related to these quantities by
two fundamental thermodynamic expressions
From these it will be seen that a more negative value of or a more positive value of will in
each case lead to an increase in Ke. Furthermore, there is a hypothesis that if the degree of disorder in a
system increases it will almost certainly be accompanied by an increase in entropy