This proposal combines nanofiltration (NF) and regenerable ion exchange resin (IX) to optimize the removal of PFAS from wastewater. The need that this project addresses is to develop an economically viable method for the treatment of both long and short chain PFAS in wastewater that also contains elevated chloride concentrations. Regenerable ion exchange resin has already proven effective in removing PFAS in large scale treatment systems in the US and Australia, but is sensitive to the presence of chlorides and other anions. A treatment chain that 1) concentrates the PFAS and simultaneously removes anions using NF, followed by 2) regenerable IX to completely remove the PFAS from the water, has the potential to be more economically viable. This is especially true for Flanders (such as the Antwerp region), where several wastewater streams contain elevated chloride concentrations.
The project will focus on the wastewater produced by industrial laundries. This project is divided into three Work Packages (WP). In the first, the wastewater from multiple laundries will be fully characterized. Only laundries that already have a Membrane Bioreactor (MBR) and a Reverse Osmosis (RO) system will be investigated. This combination will allow us to leverage existing treatment technologies to increase our understanding of how the proposed treatment chain will be implemented most effectively. In WP 2, the water selected in WP 1 will be tested in the laboratory with three different NF membranes. Based on these tests, the best regenerable ion exchange resin will be selected to remove the PFAS from the NF-concentrate. Finally, the combination of the NF membrane with the best performing ion exchange resin will be tested on-site at the laundry on a pilot scale.
The added value of the project is to find a technical solution that effectively combines NF with regenerable IX. This is a technology combination that has never been tried before. This treatment chain has the potential to treat short-chain PFAS and even ultra-short-chain PFAS in a more performant and economical way than with the baseline technology of GAC.
Industrial laundry in Herentals that is responsible for cleaning work clothes, catering textiles, personal laundry. This site has its own MBR and RO installation.
Intermediate results:
In the first work package, wastewater streams from three industrial laundries (A, B, and C) were comprehensively characterized. The results revealed substantial variability in organic load, salinity, and PFAS composition between sites, but also for individual sites in time (differences between grab and composite samples). COD concentrations in MBR permeates ranged from 20 to 80 mg/L and increased significantly in RO concentrates. Salts, including chloride and sulfate levels also varied widely. PFAS analyses indicated relatively low concentrations in raw wastewater at laundry A and laundry B (≈0.2 μg/L and 0.3–0.4 μg/L, respectively), but increased concentrations after MBR treatment (9–2.6 μg/L at A, 0.9–1.6 μg/L at B, and 6.6–7.2 μg/L at C), suggesting precursor transformation. At laundry C, PFAS concentrations were already elevated (raw wastewater 6–11 μg/L). Ultra-short-chain PFAS were largely below detection limits, with a few notable exceptions (perfluoropropionic acid (PFPrA) and triflic acid (TFMS)). A non-targeted analysis further broadened the understanding of the chemical complexity of the wastewater. NTA results have shown the presence of unidentified PFAS at the 3 sites, with some PFAS-classes being found at all sites and some other classes were specific for specific sites.
Next, in the second work package, different NF membranes, including polymeric and ceramic membranes, were evaluated at laboratory scale using MBR permeate and RO concentrate from laundry A. Based on performance with respect to suspended solids tolerance, anion rejection and PFAS removal, the hollowfiber dNF40 membrane was selected for pilot testing. The total PFAS (quantitative) removal obtained was > 90% for both NFX and dNF40, with NFX showing a bit higher rejection, with lower anion rejection in the case of dNF40. In parallel, the Sorbix HC5 regenerable IX resin was selected to treat the NF concentrate.
For the pilot scale testing (third work package), containerised pilot units for NF and regenerable IX were installed on-site at laundry A in October 2025. Pilot testing has commenced with MBR permeate, and will continue in 2026 with further testing.
Overall, the project has successfully advanced from concept to pilot scale implementation, laying a strong foundation for assessing the technical and economic viability of the proposed treatment chain.
Status
Duration
15/12/2024
-
14/12/2026
Location
Website
Coordinator
Montrose Environmental Group
Stefan Tassens
sttassens@ect2.com
Technology(ies) considered in project
Environmental compartments
Type of activity
Concern Type
