Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Brackish water treatment using pressure retarded osmosis (PRO) as a driving force for reverse osmosis (RO)

Tshuma, Ivonne (2021) Brackish water treatment using pressure retarded osmosis (PRO) as a driving force for reverse osmosis (RO). PhD thesis, Murdoch University.

PDF - Whole Thesis
Download (2MB) | Preview


The desalination process removes salts and contaminants from water to make it suitable for drinking and other beneficial purposes. Although Reverse Osmosis (RO) is currently the most energy-efficient, widely used desalination technology, it still requires a great deal of energy to create the high pressure necessary to desalinate seawater. The largest operating cost is the energy consumed in overcoming osmotic pressure and membrane resistance. Pressure Retarded Osmosis (PRO), on the other hand, utilises the salinity gradient between two solutions of different salt content to produce pressure, which can subsequently be used to generate electrical energy.

This thesis describes how PRO can be used directly as the sole energy source for RO in an autonomous PRO-RO system without additional energy input without converting osmotic energy into electrical energy. The PRO-RO proof of concept was experimentally verified in a simple combined cell without cross-flow resulting in 12.5 bars of hydraulic pressure, and flux of 3.5 L m-2 hr-1 in the RO permeate. The PRO-RO system used PRO and RO brackish water feed solution and concentrated brine PRO draw solution (200 g/L NaCl).

PRO is typically driven by freshwater to seawater gradient, but far greater energy can be predicted using hypersaline draw solutions. This study investigates the power density and maximum flux obtainable from two such saltwater solutions. The experimental data was verified by a transient model that predicts well (within 10%) PRO flux for all draw solution concentrations coupled with deionised water as feed. However, lower agreement with laboratory results was found for draw concentrations above 100 g/L when coupled with salty feed. Draw and feed cross-flow velocity were optimised at 0.1 m/s and 0.17 m/s, respectively.

A cellulose tri-acetate forward osmosis membrane was used for experimental evaluation of power production by a PRO apparatus with pressure generation up to 40 bars. A numerical model was produced from the first principles and established theory on osmotic systems to aid understanding and project beyond the practical experimental results.

The concept of combining the PRO driven by hypersaline and brackish water with a brackish water RO was investigated by modelling and simple proof of concept experiments. Theory suggests that the energy recovered from a PRO system keeps increasing with the draw solution's salt concentration. However, draw concentrations above 150 g/L did not result in a further increase in observed flux and observed energy (with a maximum pressure of 40 bars). Despite no gain in flux, modelled optimal power densities of 1.29, 12.19, and 62.4 W/m2 were obtained with increasing draw solution concentrations of 70, 150, and 300 g/L, respectively, with seawater feed solutions.

A novel, autonomous pressure retarded osmosis (PRO) driven reverse osmosis (RO), with an energy recovery device (ERD), was proposed to replace RO high-pressure pumps. The experimental PRO power density outputs from seawater/brackish water feed with 300 g/L draw and RO power density requirements were analysed. Coupling PRO with RO at PRO maximum power densities was found not to be economically viable due to high pressures of 160 bars.

The power density of 19.1 W/m2, achieved with brackish (10 g/L) feed and 300g/L draw, was sufficient to desalinate brackish water by RO at 30 bar pressure. Overall, brackish water feed and 300 g/L draw solution were used to model direct PRO to RO coupling efficacy. Using these feed and draw concentrations and the projected fluxes using currently available membranes, direct PRO to RO coupling of 2 m2 of PRO membrane to 1 m2 of RO membrane would produce 12.4 L m2 hr-1 of RO permeate at 30 bar applied pressure in an autonomous PRO-RO system.

Item Type: Thesis (PhD)
Murdoch Affiliation(s): Engineering and Energy
Supervisor(s): Ela, Wendell
Item Control Page Item Control Page


Downloads per month over past year