FEATURE Yale constructs forward osmosis desalination pilot plant

This brief feature article provides details of a novel desalination process that a thermal process, such that the separation is is being investigated and further developed at Yale University in New Haven, achieved using a semi-permeable membrane, but the energy used for that separation is in the form Connecticut. The technology promises lower energy costs, higher recovery and of heat. less brine discharge than conventional desalination systems.

Osmosis is a physical phenomenon that has been It is a well known fact that water will flow Energy costs extensively studied by scientists in various disci- from a dilute to a concentrated solution, (when The heat used by the FO process can be mini- plines of science and engineering. Early research- these solutions are separated by a semi-perme- mized in both quantity and cost, or set to some ers studied the mechanism of osmosis through able membrane), and that a very concentrated balance between the two. In most cases, FO natural materials, and from the 1960s, special solution will draw water from a brackish or sea- will use approximately 25–45% of the ther- attention has been given to osmosis through syn- water saline source. mal energy needed by multi-effect distillation thetic materials. The difficulty of making use of this phenom- (MED), the most efficient existing thermal Following the progress in membrane science enon in practice has been identifying a concen- desalination technology. in the last few decades – especially for reverse trated solution that contains solutes which can FO has the further ability to use heat at much osmosis (RO) applications – the interests in be efficiently and entirely removed. A concen- lower or higher temperatures than MED or engineered applications of osmosis has been trated sugar solution could be used, for instance, multi-stage flash (MSF) systems. FO can use spurred. Osmosis, or as it is currently referred to to effect desalination of brackish water, but this heat as low in temperature as 40°C, which is just as forward osmosis (FO), has new applications would result only in a less concentrated sugar above that typical of steam entering condensers in separation processes for wastewater treatment, solution, not fresh water. food processing, and sea water and brackish The FO process devel- water desalination. oped at Yale uses a unique Other unique areas of FO research include group of removable sol- pressure-retarded osmosis for the generation utes to create a draw solu- of electricity from saline and fresh water and tion for desalination.[1] ‘implantable’ osmotic pumps for controlled drug When ammonia and release. carbon dioxide gases are dissolved in water in the correct proportion, they Pilot-scale plant favour the formation of In the US, researchers at Yale University are a highly concentrated building a pilot-scale plant (Figure 1) to demon- solution of ammonium strate a novel FO desalination process. salts. This solution can The project, led by Professor Menachem have a very high osmotic Elimelech and graduate students Robert pressure, which makes it McGinnis and Jeffery McCutcheon, features a ideal for drawing water process that differs from existing desalination from saline feeds, but technologies in that it uses osmotic pressure, what makes this solution rather than hydraulic pressure or thermal evapo- most advantageous for ration, to separate fresh water from a sea water or use in FO is the ability brackish water source. This approach promises a of the salts to decompose significant reduction in energy consumption and from solution, when costs, as well as high feed-water recoveries and heated, into ammonia greatly reduced brine discharge streams. and carbon dioxide gases again, thus allowing for their efficient and com- Efficient plete removal and reuse The key to the ability of the FO process to (Figure 2). achieve efficient desalination is in the composi- Therefore, this process Figure 1. Yale researchers are building a pilot-scale plant to demonstrate tion of the osmotic ‘draw’ solution used. is both a membrane and a novel forward osmosis desalination process.

7 Membrane Technology January 2007 FEATURE

Main advantages of using forward osmosis The main advantages of using forward osmosis (FO) are that it operates at low or no hydraulic pressures, it has high rejection of a wide range of contaminants, and it may have a lower membrane fouling propensity than pressure-driven membrane processes.[2] Because the only pressure involved in the FO process is caused by the flow resis- tance in the membrane module (a few bars), the equipment used is very simple and membrane support is less of a problem. Furthermore, for food and pharmaceutical processing, FO has the benefit of concen- trating the feed stream without requiring high pressures or temperatures that may be detrimental to the feed solution. For medi- cal applications, FO can assist in the slow and accurate release of drugs that have low Figure 2. Schematic of Yale’s novel forward osmosis desalination process. oral bio-availability because of their limited solubility or permeability.[3] approximately 21% of what is required by MED, information for detailed estimates of total cost. Versatile technology and 9% of what is required by RO. The combina- Yale researchers have begun the construction of The increased attention to FO from various tion of low-cost heat and minimal electrical con- the FO desalination pilot, with funding provided disciplines arises from the fact that FO can sumption promises to make FO the best desalina- by the Office of Naval Research, with completion be employed in many fields of science and tion process with respect to energy cost. expected by spring of 2007. engineering, including water and wastewater treatment, sea water or brackish water desal- ination, food processing, drug delivery and High recovery References the production of electrical power. The high osmotic pressures characteristic of the 1. J.R. McCutcheon, R.L. McGinnis and M.A. Despite the lack of robust membranes and ammonium salt draw solution allow for fresh water Elimelech: Novel ammonia–carbon dioxide membrane modules for FO, basic research recovery from highly concentrated saline feeds. forward (direct) osmosis desalination process, on FO and the development of new applica- Laboratory tests show effective desalination of Desalination 174 (2005) 1–11. tions of FO are steadily growing. Currently, 3.4 M NaCl solutions, a salinity corresponding 2. R.W. Holloway, T.Y. Cath, K.E. Dennett and the most important measure to be taken in to around 85% recovery from a typical sea-water A.E. Childress: Forward osmosis for concentration order to advance the field of FO is the devel- source. Realization of this potential will require of anaerobic digester centrate. Proceedings of the opment of new membranes in both flat- investigation of appropriate pretreatment strat- AWWA Membrane Technology Conference and sheet and hollow-fibre configurations. egies, but the capability of FO to produce very Exposition, Phoenix, Arizona, USA, 2005. The membranes need to provide high high recoveries is unmatched by existing desalina- 3. P.P. Bhatt: Osmotic drug delivery systems water permeability, high rejection of solutes, tion methods. for poorly soluble drugs. PharmaVentures substantially reduced internal concentration One significant impact of this increased capa- Ltd, Oxford, UK, 2004. Available online at: polarization (CP), high chemical stability bility will be the reduction in the volume of brine www.drugdeliveryreport.com/articles/ and high mechanical strength. discharge streams from desalination plants. In ddcr_w2004_article3.pdf the case of brackish water desalination, the very 4. R.L. McGinnis and M. Elimelech: Energy high recoveries made possible by FO may allow requirements of ammonia–carbon dioxide for- in an electrical power plant. At this temperature, for zero liquid discharge (ZLD) operation, a ward osmosis desalination, Desalination 207 the cost of thermal energy is very low. capability critical to the adoption of desalination (2007) 370–382. FO can also use higher temperature heat in inland environments. ZLD may one day also sources, which gives the benefit of greatly be possible in sea-water desalination, provided reducing the total amount of heat required. At that a use can be found for the large quantities of Further reading 200–250°C, for instance, FO may achieve a salt that would be produced. T.Y. Cath, A.E. Childress and M. Elimelech: gained output ratio (GOR) – commonly used Forward osmosis: Principles, applications, and to compare thermal desalination efficiencies, recent developments, J. Membrane Science 281 with higher numbers indicating less heat used Useful desalination process (2006) 70–87. – of up to 26.5. MED desalination typically The combination of low energy costs, high has a GOR of approximately 8–15, depending feed-water recoveries and minimized brine Contact: on its configuration. discharge promise to make FO a highly useful Professor Menachem Elimelech, Mason Laboratory, The electrical consumption of FO is also desalination process. Room 313A, 9 Hillhouse Avenue, Yale University, New much lower than that of existing desalination It is expected that the total water cost of FO Haven, CT 06520, USA. Tel: +1 203 432 2789, Email: technologies. In most cases, FO requires less will also be much lower than that of RO or MED, [email protected], Web: www.yale.edu/ than 0.25 kWh/m3 of water produced. This is but only pilot testing will provide the necessary env/elimelech/research_group.htm

8 Membrane Technology January 2007