Hybrid Membrane Processes for Micropollutant Removal from Wastewater
Hans David Wendt is a PhD student in the department Soft matter, Fluidics and Interfaces. (Co)promotors are prof.dr.ir. R.G.H. Lammertink; prof.dr.ir. W.G.J. van der Meer and dr.ir. A.J.B. Kemperman from the faculty Science and Technology.
The quantity and quality of available water is at risk. An aspect related to quality is the growing issue of organic micropollutants (OMPs) present in surface waters. These OMPs can be pharmaceuticals, personal care products, per- and polyfluoroalkayl substances (PFAS), industrial chemicals, or pesticides. One pathway for OMPs to enter surface water is through wastewater, since typical wastewater treatment plants (WWTPs) are not designed to remove such compounds. They are mainly based on biological treatment, which can only partially remove part of the OMPs.
A possible solution to remove OMPs from wastewater is to add additional treatment steps to conventional WWTPs. When nanofiltration (NF) or Reverse Osmosis (RO) membranes are used as such, the OMPs can be retained by the membrane, while the permeated water has lower OMP concentrations. However, the concentrate with higher OMP concentrations, as well as higher salt concentrations, requires further treatment. This work focuses on treating the concentrate by recirculating it to conventional biological treatment. This can lead to higher removal of OMPs due to a longer residence time of OMPs as well as to the adaptation of biological treatment to specific OMPs.
The goal of Chapter 2 is to investigate which type of dense membrane has the highest potential for practical application. This is done with a study based on literature data and modeling. Recirculation of concentrate has been investigated before. However, the focus was never on OMPs. Four model were analyzed with a mass balance based model with five different commercially available membranes. The model shows that the combination of conventional treatment, membrane filtration and concentrate recirculation can substantially improve the overall removal of OMPs with low bioremoval and high membrane retention. However, attention should be given to the potential accumulation of salts. Therefore, a balance between salt and OMP retention by the membrane is required. From the investigated membranes, the dNF40 and NF270 membrane strike the balance the best. Membrane development is recommended to further optimize this balance.
The dNF40 hollow fiber nanofiltration (HFNF) membrane is investigated in more detail in Chapter 3. Real wastewater effluent is used on benchtop (0.5 L/hr) and continuous pilot scale at an actual WWTP (1 m3/hr). The effects of scale, recovery, flux, crossflow velocity and staging are investigated in detail, with a focus on their effect on retention of ions and OMPs and calculated energy consumption. The results indicate that scale notably affects retention, with smaller scales achieving higher retention due to shorter module lengths and lower recovery. Decreasing recovery and increasing flux compared to the base case are most effective in increasing retention on the pilot scale. The calculated energy consumption of the Christmas Tree configuration is lower than for the Feed&Bleed configuration.
In Chapter 4, the dNF40 membrane is used on pilot scale to investigate the effect of concentrate recirculation on the total treatment system. To do so, the pilot system also includes a dedicated biological treatment system. The results with and without recirculation are compared throughout this chapter. The system could operate in a stable manner for 45 days. Recirculation of concentrate leads to substantial accumulation of ions with high retention such as sulfate. The combination of biological treatment, the HFNF membrane and concentrate recirculation can increase overall removal compared to only biological treatment for several compounds. A mixture of 23 OMPs shows that overall removals beyond 80% can be obtained for 12 OMPs.
An alternative to removing OMPs in one step with a dense HFNF membrane, is to use the membrane as pre-treatment for a subsequent oxidation step. The primary focus of Chapter 5 is to achieve a balance between enhancing the transmittance of the membrane permeate, thereby reducing the energy consumption of the oxidation step, and minimizing the energy required for membrane filtration. Six HFNF membranes with a wide range of permeability, MgSO4 retention, molecular weight cut-off, and UV retention are investigated. This data is used as input for the calculations for the energy consumption for both the membrane and the UV/H2O2 oxidation step. The one-to-most dense membrane yields the lowest total specific energy consumption at 0.17-0.18 kWh/m3 permeate for 70 or 80% removal, respectively.
The findings of Chapters 2-5 are summarized in Chapter 6 together with an outlook on future research and a broader perspective on the work.
