Treatment of PFAS-contaminated leachate with ion exchange filters has been studied in several pilot studies and showed good results, especially in the case of requirements for the reduction of PFAS with longer carbon chains. The biggest challenge with using ion exchange mass for leachate treatment is that either a very advanced pre-treatment is required or that the filter needs to be backwashed frequently to avoid clogging. Backwashing of filters mixes the ion exchange mass, which causes the column action to be lost.
The aim of the project was to test the design where leachate flows upstream of a filter bed with a sufficient flow to fluidize the bed. The report describes trials carried out on a larger pilot scale at the Sivik waste facility as well as trials on a smaller pilot scale at the Löt waste facility. At Sivik, a standard DynaSand filter has been filled with ion exchange resin and run for 9 months.
At Löt, comparative trials with downstream ion exchange filters and two upstream filters in series have been carried out. Trials at Sivik showed that it was crucial to mix the ion exchange mass in the filter by running the airlift pump to maintain good treatment results. However, it was challenging to find a balance between mixing the resin and the risk of media loss with flushing water. It was also discovered that the flow rate in the trials was too low for homogeneous suspension of the resin, which is believed to have created shortcuts of flow through the filter. This led to varying purification of the shorter PFAS. Despite the challenges, reduction of PFOS averaged 94% during the course of the trial where 14 300 bed volumes (BV) of water was treated.
Effluent PFOS concentrations were approximately 13 times lower than the respective effluent target value for the facility. Complementary lab-scale trials showed that the ion exchange resin had a lot of capacity left when the tests ended and reduced PFOS and other longer PFAS almost as good as new ion exchange resin. In tests at Löt, the 2-stage upstream filter showed similar treatment efficiency as a downstream filter, confirming that capacity utilization of ion exchange resin can be similar with the two process designs. Loss of mass was noted both with downstream ion exchangers (in trials on Löt during backwash) and upstream ion exchangers (due to poor design of a resin mixing system). Measures to reduce the risk of the resin loss as well as the importance of polishing removal of possible smaller amounts of resin are discussed in the report.