Browsing by Author "Van Wyk, Linda Alida"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Solute transport in a submerged forward osmosis membrane system
    (Stellenbosch : Stellenbosch University, 2019-12) Van Wyk, Linda Alida; Burger, A. J.; Stellenbosch University. Faculty of Engineering. Dept. of Process Engineering.
    ENGLISH ABSTRACT: Wastewater treatment with forward osmosis (FO), an osmotically driven membrane process, has been investigated for the osmotic dilution of seawater prior to desalination in an attempt to lower the energy consumption of seawater reverse osmosis (RO). This hybrid FO-RO process provides a dual-barrier for the effective rejection of wastewater contaminants, thereby potentially producing a high quality permeate. Due to the higher rejection capacity of FO membranes compared to ultrafiltration membranes, the FO process can advance wastewater treatment in submerged membrane bioreactors. The aim of this study was to investigate the transport and rejection of selected weakly-rejected solutes in a submerged FO system with a commercially available FO membrane. The benefit of the dual-barrier rejection mechanism of the FO-RO hybrid could then be investigated by simulation of its final permeate quality with the experimentally determined rejections of the selected model solutes. To this end, a bench-scale FO setup was designed and constructed. The baseline performance of the FO membrane was firstly evaluated by considering the effects of the membrane orientation, hydrodynamic conditions and osmotic pressure gradient on the water flux and reverse draw solute flux. Phenol, as an organic water contaminant, and boron and lithium, as inorganic water contaminants, all with different physicochemical properties and potentially weak membrane rejections, were used to study the solute transport and rejection. With a draw solution of seawater quality, water fluxes of 20 L∙m-2∙h-1 and 32 L∙m-2∙h-1 were obtained when the active layer of the membrane was in contact with the feed solution (AL-FS orientation) and draw solution (AL-DS orientation), respectively. The AL-FS orientation exhibited exceptional flux stability at the expense of dilutive internal concentration polarisation. With no hydrodynamic conditions at the submerged membrane surface, concentrative external concentration polarisation (CECP) of the reverse diffused draw solute resulted in a significant water flux decline to below 8 L∙m-2∙h-1 in both membrane orientations. A Reynolds number of 1 100 at the submerged membrane surface was sufficient to mitigate CECP. It was found that the solute rejection improved with an increasing osmotic pressure gradient. The rejection of the neutrally charged solutes, boron and phenol, was independent of their concentration gradients in both membrane orientations. An increase in the ionic strength and decrease in the pH of the feed solution with increasing concentrations of lithium chloride and boric acid increased the rejection of lithium, most likely due to its reduced electrostatic interactions with the negatively charged membrane surface. As opposed to boron and phenol, the lithium rejection in the AL-DS orientation was higher than in the AL-FS orientation as the electrostatic attraction of lithium to the membrane in the AL-DS orientation was perceived to be insignificant. It is postulated that the electrostatic attraction of lithium to the negatively charged membrane surface significantly compromised its rejection, such that it was approximately 16% lower than that of phenol and boron in the AL-FS orientation at neutral pH conditions. The respective experimental phenol, boron and lithium rejections of 91%, 93% and 81% were implemented in the simulation of the FO-RO hybrid process. By its dual-barrier and intermediate dilution effects, the FO-RO hybrid provided an improved permeate phenol concentration of 1.1 μg∙L-1, compared to 9.0 μg∙L-1 provided by a standalone wastewater RO process. The permeate quality of a standalone seawater RO unit could be improved from 315 μg∙L-1 boron and 149 μg∙L-1 lithium to 32 μg∙L-1 boron and 25 μg∙L-1 lithium with typical influent seawater concentrations.