Water desalination takes a step forward
A new technology for producing drinking water from seawater has the potential to cost much less than the conventional alternatives.
One-third of the world’s population lives in countries with insufficient freshwater to support the population, according to the UN Environment Programme. For that reason, desalination plants that extract drinking water from seawater are increasingly popular across the world. But many water-poor countries cannot afford the conventional desalination technology—reverse osmosis—because of its relatively high cost. Menachem Elimelech, a professor of chemical and environmental engineering at Yale University, and his graduate researchers Robert McGinnis and Jeffrey McCutcheon are hoping to reduce the cost of desalinating water with a new technology they have developed that they call forward osmosis desalination.
Over the years, better membrane technologies and energy recovery devices have made reverse osmosis more affordable and efficient, says Rhodes Trussell of Trussell Technologies, an environmental engineering firm. But the water produced by desalination still costs at least $2 per 1000 gallons, more than twice the cost of conventional water treatment, he says. A large part of this cost comes from energy use.
And additional costs may be involved. Reverse osmosis typically recovers 35–50% of the volume of seawater as freshwater, with a leftover brine concentrate. Near the coast, the brine waste is simply dumped back into the ocean. However, disposal is not as simple when the technique is used in inland areas, where salt can also be a problem in waters intended for drinking. Desalinating brackish waters inland requires an added step of evaporating the brine, because injecting it underground would affect groundwater.
“The two major resistances to reverse osmosis are cost and brine discharge,” says McGinnis. “That’s one of the things that made us think of doing [forward osmosis] instead.”
The Yale team is the only one to have made the technique work in recent years, and their results are promising so far, says Thomas Mayer, who studies desalination at Sandia National Laboratories, which is run by the U.S. Department of Energy.
In reverse osmosis, high pressure—about 1000 pounds per square inch—is used to push seawater through a semipermeable membrane that holds back the unwanted salts. The pressure is needed to oppose the natural tendency of freshwater to move across such a membrane via osmosis to dilute the seawater.
In the forward osmosis system, the researchers take advantage of this natural tendency. Salt water sits on one side of the membrane, but the freshwater on the opposite side is transformed into a high-concentration solution by adding ammonia (NH3) and CO2. Water naturally flows from the salt water to what is now the “draw solution”, which can have a solute concentration as high as 10 times that of the salt water. “We just let water go in the right tendency... . We don’t apply any pressure,” says Elimelech, who received the National Water Research Institute’s 2005 Clarke Prize for this work and other achievements in water research. The diluted draw solution is then heated to about 58 °C to evaporate off the CO2 and NH3 for reuse, leaving behind freshwater.
At present, some companies are developing forward osmosis systems to treat wastewater and to clean the water leaching off landfills. But Elimelech says that none of them have applied the process to seawater desalination. Only one commercial membrane is available for forward osmosis, he adds. This membrane is used to purify water for drinking, but it is “not optimized for seawater desalination.” Using this commercial system, the researchers removed 95–99% of the salt from the water moving across the membrane, producing 2.1–21.2 gallons of drinking water per square foot per day (gfd).
According to Trussell, these results are good for a prototype system, with a flux that is comparable to the reverse osmosis numbers of 8–15 gfd. However, he says that salt removal should reach 99.9% for the water to be suitable for drinking. Nevertheless, Elimelech’s research is “one of the more important developments in the last decade,” Trussell adds. “He has come up with an idea that has brought [forward osmosis desalination] much closer to feasibility.”
However, the experimentally observed flux numbers are lower than what the researchers calculate from the draw solution’s osmotic pressure. The reason for the disparity, says Elimelech, is that as water crosses the membrane, it dilutes the draw solution inside the pores, gradually decreasing osmotic pressure and flux. He says that a better membrane would give more flux and a consistent 99% salt removal, which is comparable to reverse osmosis.
McGinnis adds that a better membrane could recover 75% of the water from the seawater; that would mean more freshwater from the same amount of seawater and less brine waste. This would make the process useful for landlocked regions like New Mexico, which are rich in saline groundwater.
The researchers have yet to evaluate the performance of an important part of their pilot-scale system—a distillation column that will separate the draw solutes from the drinking water. In the fall, they will test the operation of the column in their prototype system.
The researchers will have to be careful to ensure that most of the NH3 used in the forward osmosis process is ultimately removed, because in order to be suitable for drinking, freshwater should contain less than 2 milligrams of NH3 per liter, Trussell says. However, any additional step to get rid of residual NH3 will add to the system’s cost.
Elimelech calculates that the separation process will consume very little energy. For every 1000 gallons of water produced, the system will require 1 kilowatt-hour of electricity and 1200 megajoules of heat. This could be very low quality heat, he says, “so you can use waste heat, which you can get almost for free.”
The upcoming evaluation of the distillation column will show whether the actual energy used in separating the solute is consistent with the researchers’ calculations. Ultimately, says Mayer, the actual energy cost will determine if the system is cost-effective. “When you get down to it, cost is the key factor that will dominate whether this is going to be a successful technology or not,” he says, “and that depends on the energy consumed by separating the draw agent and how that [energy use] stacks up against reverse osmosis technology.”
