Ion exchange is used extensively for the removal of ionised impurities
found in natural waters. The final stage in the production of ultra
pure water is normally a bed of mixed anion and cation exchange resins.
Three areas within the operating cycle of a regenerable mixed bed -
resin separation, resin mixing and anion exchange kinetics - have
Complete separation of the two resins by backwashing, prior to chemical
regeneration, is necessary to prevent the subsequent release of trace
impurities into the purified water. Various published models of
particle segregation by backwashing were examined but none accurately
described the separation of two ion exchange resins with similar bead
size distributions and densities. A new model has been proposed based
on variations in fluidised bed porosity combined with overlapping
bulk circulation cells of particles. A graphical technique has been
developed to predict resin separability and the predictions compared
with practical data. The effects of variations in bead size, bead
density, backwash flow rate and temperature have been calculated. The
variations in bead density with ionic form and polymer/matrix type
of the exchanger have been measured.
Following regeneration the resins are remixed by air agitation of a
resin/water slurry. A mechanism to describe the progressive stages of
air mixing has been proposed, based on bubble transport and bulk
circulation of resin beads. The subsequent sedimentation of the resins
was also considered. Laboratory and full scale studies confirmed the
predicted effects of mixing fault conditions, particularly re-separation
of the mixed resins.
A mass transfer equation has been developed to describe the leakage
of influent ions through a column of exchange resins. In conjunction
with laboratory column tests the equation has been used to investigate
the influence on anion exchange of polymer/matrix type, influent anion
and the presence of foulants on the resin beads. Sulphate and phosphate
ions exchange more slowly than monovalent chloride and nitrate ions.
On a fouled exchanger the rate of sulphate exchange deteriorates more
rapidly and seriously than for chloride exchange. This has been
attributed to steric hindrance of the divalent sulphate ion.
A laboratory method has been developed for the routine assessment of
mixed bed anion exchangers and the prediction of their performance
potential in service, with particular application to condensate
purification for boiler feedwater.
A Doctoral Thesis. Submitted in partial fulfilment of the requirements for the award of Doctor of Philosophy of Loughborough University.