The effects of electrolytes and eight different cationic polymers on alkyl ketene dimer (AKD) size retention, extent of reaction and sizing efficiency were investigated using a C-14 labelled cationic AKD-dispersion. Sizing experiments were conducted in a standard British hand sheet former, using a Britt Dynamic Drainage Jar (BDDJ) to pre-shear the pulp suspension in order to simulate more practical stock preparation conditions. Some elaboration on shear conditions was also made. The C-14 labelled AKD-dispersion was stabilised using a classical cationic starch/lignosulfonic acid recipe, resulting in a weakly cationic AKD-dispersion (z-potential = + 10.5 mV) at pH = 8.0. In the absence of polyelectrolytes, the cationic AKD-particles are deposited onto negatively charged fibre surfaces through a heterocoagulation mechanism. The deposition is very sensitive to electrolytes and already at a concentration of 2 x 10(-3) M concentration of CaCl2, the AKD retention is essentially zero. As expected more highly charged fibres were found to be less sensitive to electrolytes. In a second series of experiments an anionic AKD-dispersion was used. The C-14 labelled AKD-dispersion was stabilised using a classical lignosulfonic acid, resulting in a strongly negative dispersion (z-potential = - 56 mV) at pH = 8.0. Among the various cationic polyelectrolytes employed for the retention of the anionic AKD-dispersion, the cationic polyacrylamides showed best performance with retention levels up to 90%. The results show clearly that the sizing process is not very sensitive to the state of agglomeration of the AKD-particles as suggested by the Cassie equation. Irrespective of the retention aid the used Cobb(60)-value fell on the same graph if plotted vs. the reacted amount of AKD, irrespective if anionic or cationic AKD-dispersions were used. It was only when the stock was excessively agglomerated, deviations to this behaviour were found.
