December 15, 2006

Osmotic power from the ocean

By Richard L. Stover, PhD and G.G. Pique, Energy Recovery Inc.

Realistic expectations

What level of performance could be expected of a mature osmotic power system? Based on the results of the demonstration tests, it is feasible that osmotic power membranes could provide a constant stream of pressurized water sufficient to drive a turbine—if sufficient crossflow were supplied to achieve turbulent conditions near the membrane surface and to minimize concentration polarization effects. With such an optimized osmotic power membrane, the potential osmotic pressure across such membranes would be the average of the osmotic pressure of the saltwater minus the pressure that was lost to flow friction and other energy sinks.

For the sake of this discussion, it is assumed that permeation and other membrane losses would be less than 25% of the average osmotic pressure. A large-scale osmotic power process would utilize high-capacity, high-efficiency components. Pumps and turbines of up to 90% efficiency are currently available, and the best motors and generators operate at up to 95% efficiency. ERDs that operate at up to 95% efficiency are also currently available. These projected membrane performance characteristics, pumps, turbines, and ERD efficiencies were used to extrapolate the overall efficiency of a hypothetical high-efficiency osmotic power process as a function of dilution rate (Figure 7).

7. Making net power. Overall process efficiency depends on the dilution rate. Source: Energy Recovery Inc.

Figure 6 shows that at low dilution rates, the energy consumed by the pumps exceeds that produced by the turbine, as indicated by a negative efficiency. At higher dilution rates, the osmotic driving force is reduced, as indicated by the flattening of the curve. In addition, membrane performance limits dilution rate.

Although a sustainable dilution rate for osmotic power membranes is unknown, a 45% dilution rate is assumed in the following analysis, based on a typical recovery rate for SWRO. Assuming a turbine flow rate of 367 gpm, water salinity of 37,000 ppm, and the membrane, pump, turbine, and ERD performance characteristics listed above, an osmotic power system will indeed produce positive net power (Table 3).

Table 3. Projected high-efficiency osmotic power system performance (base case: fresh water/seawater). Source: Energy Recovery Inc.

If the membranes perform at the highest specific energy recorded in the demonstration tests, more than 308,000 ft² would be required to generate the net power shown in Table 3. If membrane-specific energy could be increased to 6 x 10^-4 hp/ft², for example, the required membrane surface area could be reduced to 94,000 ft².

The cost to produce osmotic power depends in large part upon the cost of the membranes, which is unknown, and the salinity of the saltwater used. However, for a rough estimate we can assume that a full-scale osmotic power plant would cost about as much as a comparably sized brackish water reverse osmosis plant. Further assuming a membrane power density of 6 x 10^-4 hp/ft² (5 W/m²), high device efficiencies, and a 20-year plant life, osmotic power production costs could be between 20¢ and 50¢/kWh, making it comparable to other renewable power technologies such as solar and wind. Depending on the final osmotic power membrane configuration, topography, and other site-specific features, it is possible that the power requirements of an osmotic power facility could be significantly reduced.

—Richard L. Stover, PhD is the chief technology officer of Energy Recovery Inc. (www.energy-recovery.com). He can be reached at 510-483-7370 or stover@energy-recovery.com). G.G. Pique is president and CEO of Energy Recovery Inc.; he can be reached at pique@energy-recovery.com.