Search

Pressure retarded osmosis (PRO) is one feasible technology that can be used to exploit the mixing energy from salt gradients which is commonly referred to as salinity gradient power or osmotic power. This paper focuses on the construction of iso-watt diagrams and their application for detailed evaluation of specific power in PRO on basis of membrane characteristics and site specific design parameters. Iso-watt diagrams are demonstrated to be a useful tool for investigation of the effect of variation in the membrane parameters, i.e. the water permeability, the salt permeability and the structure parameter, and how mutual changes in the characteristic parameters will impact the PRO performance. Thus, iso-watt diagrams can be considered an effective aid for goal-oriented membrane development. In addition, iso-watt diagrams can be considered to be useful also with respect to design related aspects, such as e.g. site selection. Further, membrane performance data are reported in the literature for various experimental conditions, which makes direct comparison difficult. The use of iso-watt diagrams enables an efficient way of presenting and comparing the efficiency of different membranes. This is demonstrated in the current paper for ten states of the art membranes.

Fouling is one of the major challenges to be solved in order to commercialise osmotic power generation by pressure retarded osmosis (PRO). This paper deals with natural organic matter (NOM) fouling. The impact of i.a. NOM concentration and ionic strength in the freshwater feed have been investigated for five different types of PRO membranes by performing fouling experiments in PRO mode at isobaric conditions. An important finding was that fouling propensity in terms of relative flux decline as a function of accumulated NOM load was independent of the NOM concentration at otherwise identical conditions. Further, it was confirmed that increased ionic strength in the support structure, due to internal concentration polarisation and reverse salt diffusion, aggravates membrane fouling in PRO. The observed variation in fouling propensity between different types of PRO membranes was significant. The CA membrane was less susceptible to NOM fouling than the four TFC types of PRO membranes. It is anticipated that the variation in fouling propensity that were observed for different membranes was related to both material and structural properties of the membranes. The results show the importance of applying fouling propensity as an additional main criterion to membrane performance during future research and development of PRO membranes.

The principle of PRO is to utilise parts of the free energy of mixing when mixing freshwater with seawater for production of electrical energy. The mixing process takes place over a semi permeable membrane that will retain salts, and ideally allow only transport of water. The transport of water will occur due to differences in chemical potential across the membrane. The PRO process will be operated with elevated pressure on the seawater side of the membrane, and the net increase in volume due to osmotic transport of water across the membrane, can be utilised to run a turbine and hence produce energy. At present the biggest challenge in PRO is to develop membranes with sufficient specific power. This challenge involves optimisation of the basic membrane characteristics, i.e., high water permeability (A), low salt permeability (B) and low structure parameter (S). In this paper a transport model for water and salt in hollow fibres in PRO has been developed. Further, a structure parameter equivalent to the structure parameter for flat sheet membranes has been defined. The impact of the membrane parameters (A, B and S) on PRO performance in a hollow fibre element has been demonstrated by applying iso-watt diagrams. In general, for a given structure parameter S, increasing the A value and decreasing the B value will increase the specific power. It was also found that if the salt permeability exceeds a certain level, a further increase in the water permeability will not result in a corresponding increase in the specific power. Similarly, if the water permeability is lower than a certain level the effect of increased salt permeability will be low. Further, the PRO performance of different hollow fibre membranes has been estimated using the iso-watt diagrams. Based upon the reported membrane characteristics the PRO performance for the selected membranes ranges from less than 1 W/m2 to above 4.5 W/m2. The membranes having a specific power around 4.5 W/m2 are promising candidates for future PRO plants, given that these membranes can tolerate the typical operation pressure in a PRO plant of 12–14 bar.

A large amount of potential renewable energy can be extracted from the mixing of freshwater and seawater. Two membrane processes can be used to extract this energy, namely pressure retarded osmosis (PRO) and reverse electrodialysis. Both processes need membranes that are similar to the ones used to perform reverse osmosis and electrodialysis. So far neither of the processes has been optimally designed. In the present study PRO has been analysed for potential power production. It appears that PRO exhibits a significant potential for power production and future developments may further improve this. A specific power in the order of 5 W/m2 of membrane seems possible and approximately 40 per cent of the potential mixing energy of freshwater with infinite amount of seawater can be converted to mechanical energy. This article aims to describe the PRO process and to develop guidelines for the development of membranes that are suitable for PRO. Furthermore, some experimental results from laboratory measurements are included both to verify the model and to show present state of membrane development.

In studies of bi-ionic cells containing strontium and potassium chlorides separated by a cation exchange membrane, the water transference was found to have a non linear dependence on ion transference numbers. This finding suggested the existence of local pressure gradients within the membrane during irreversible flow. In order to evaluate this possibility, we developed a method to obtain the concentration profile in a cation exchange membrane from measured fluxes. This permits the calculation of local transport coefficients for the coupled transport of mass and charge by two salts and water within the membrane. The observed variations in transport coefficients across the membrane are used to calculate the hydrostatic pressure profile inside the membrane. The results show that the postulated pressure gradient does exist, but that it has a negligible effect on the chemical potentials. This means that cell potentials can be calculated using the concentration dependent terms of the change in Gibbs free energy.