UCLA

Salinity gradient power (SGP), the controlled mixing of streams with different salinity, is a potential route for clean and renewable base-load power generation. The two most popular forms of SGP production, pressure retarded osmosis (PRO) and reverse electrodialysis (RED), have received renewed research interest in recent years. While previous modeling work has focused on the application of simple mathematical models to these processes, in order to accurately predict process performance on scale-up, robust models must be developed. The objective and focus of this dissertation is to understand through mathematical modeling the practical performance that can be achieved in PRO and RED as the technology transitions from the laboratory scale to full-scale implementation.

First, the thermodynamics of mixing are discussed for both PRO and RED, with the intrinsic differences highlighted for each process. Using simple thermodynamic models, the integration of PRO and RED with reverse osmosis (RO) is discussed as a method for reducing the specific energy consumption (SEC) of seawater desalination. This "osmotic energy recovery" (OER) is evaluated in a seawaterRO plant that includes state-of-the-art RO membranes, plantdesigns, operating conditions, and hydraulic energy recovery (HER) technology. Weassume the use of treated wastewater effluent as the OERdilute feed, which may not be available in suitable quality or quantity to allow operation of the coupled process. A two-stage OER configuration could reduce the SEC of seawater RO plants to well below the theoretical minimum work of separation for state-of- the-art RO-HER configurations with a breakeven OER capital expenditure (CAPEX) equivalent to 42% of typical RO-HER plant cost suggesting significant cost savings may also be realized. At present, there is no commercially viable OER technology; hence, the feasibility of using OER at seawater RO plants remains speculative, however attractive.

In order to probe scale up characteristics of SGP, a process model has been developed for PRO accounting for full-scale system losses such as viscous dissipation, external mass transfer and equipment efficiency. Also, an existing model for RED is adapted to account for analogous full-scale systems losses. We project practical power densities and process efficiencies. The projected power density for PRO (using best available membranes) is much lower than generally predicted by extrapolation of experimental data. For example, a power density of 5 W/m2 extrapolated from laboratory experiments actually yielded negative power at full-scale. Hydraulic energy recovery device (HER) efficiency doubles the maximum power density for HER efficiency increase from 90% to 99%. Furthermore, the operating pressure, load voltage, and crossflow velocities typically applied in laboratory studies appear much too high to be practical in full-scale PRO and RED systems. RED systems should be designed with relatively short lengths compared to PRO. For both processes, energy efficiency does not occur at thermodynamic equilibrium due to hydraulic losses. Finally, maximum power density appears an inadequate parameter for assessing full-scale PRO/RED process feasibility because both processes could produce the same maximum power density, but different power outputs and efficiencies with different system sizes.

While there has been a significant amount of laboratory work focused on improving membrane performance in PRO, there has been little emphasis on developing novel PRO process configurations that might dramatically improve overall process performance. Here, we introduce a novel PRO configuration and apply the developed process model in order to evaluate the performance of the process. This "staged" configuration has the potential to dramatically improve the performance of PRO with up to 33% higher water flux over the standard PRO configuration. In this new arrangement, intermodule hydroturbines are placed strategically in order to modulate the applied pressure over the length of the system. The additional hydroturbines match the applied pressure to the optimum flux condition over each segment and substantially increase the total water recovery. The impact on energy efficiency is significant for some mixing regimes, with seawater-river water and brine-river water mixing corresponding to efficiency increases from 31% to 44% and 34% to 50%, respectively, versus the standard PRO configuration. Of course, performance improvements must be considered in light of the increased capital costs arising from the additional hydroturbines.

Finally, the PRO process model is integrated with cost data in order to determine the change in energy production cost with system size. Since existing cost correlations from the literature are often outdated, new correlations have been developed for each plant component by contacting suppliers and manufacturers. Linking these correlations with power output data, energy cost has been evaluated under different operating conditions and with membrane properties. The results suggest that there is still significant room for improvement in membrane design in order to further reduce the cost of PRO. Furthermore, from an economic perspective, brine-river water mixing appears to be a much more promising mixing regime than seawater-river mixing.