SCIENCE & TECHNOLOGY
April 9, 2001
Volume 79, Number15
CENEAR 79 15 pp.32-38
ISSN 0010-2347
WATERWORKS
Research accelerates on advanced water-treatment technologies as their use in purification grows
MAIRIN B. BRENNAN, C&EN WASHINGTONConventional water purification is a tried-and-true process that hasn’t changed much in decades: Coagulation and flocculation, sedimentation, sand or gravel filtration, and chlorine disinfection are the customary steps. Wresting fresh water from seawater is also a long-standing technique, especially in oil-rich, water-starved countries where the cost of the energy-intensive process is not an issue.
The conventional approach will continue to deliver water to most U.S. households in the future. But concerns about water shortages in the U.S. have surfaced, although the plight in the U.S. is not nearly so severe as that endured globally by many countries in dire straits for potable water and water for irrigation (C&EN, Dec. 6, 1999, page 127).
Rivers in Florida are drying up, for example. And severe cutbacks are anticipated in water transported to Southern California from Northern California and from the Colorado River as water demand in locations served by these sources continues to rise. Moreover, groundwater is being depleted in both Florida and Southern California as aquifers are overdrawn.
As a result, advanced water-treatment technologies that can purify saltwater, brackish water, surface water, or wastewater are being adopted to help preserve aquifers while meeting municipal water needs. These technologies include membrane-based filtration systems–reverse osmosis and micro-, ultra-, and nanofiltration–and disinfection with ultraviolet light.
REVERSE OSMOSIS came of age in the 1970s following the development of semipermeable membranes that could efficiently separate salts from water under pressure. The technology is widely used in many areas of the Middle East, North Africa, and the Caribbean, where desalinated seawater is the main source of municipal supplies. But it’s also used in the U.S., Europe, and Japan primarily to produce ultrapure water for industry.
Sometimes called hyperfiltration, reverse osmosis is generally held to be the most complex of the membrane separation processes. In contrast to osmosis, which occurs by water diffusion through a semipermeable membrane, reverse osmosis depends on pressure to selectively filter water through a membrane by overcoming the osmotic pressure from the concentration gradient. Although the process is not completely understood, one theory is that water molecules in the solution form hydrogen bonds with the surface layer of the membrane. As other water molecules approach the membrane surface, a hydraulic pressure gradient is created that drives water across the membrane. Such membranes are usually made from aromatic polyamides or cellulose acetate.
Microfiltration and ultrafiltration are size-exclusion membrane technologies that operate under relatively low pressures. Both technologies are widely used in industry for concentrating and clarifying process streams. In advanced water treatment, they might be used as an initial step in cleaning up the water–depending on the source of the water–or later on in the process as a pretreatment purification step for reverse osmosis. In treating water, the membranes “are really designed to remove particles and pathogens,” notes James S. Taylor, director of the Environmental Systems Engineering Institute and a professor of civil and environmental engineering at the University of Central Florida.
Nanofiltration membranes “are loose reverse-osmosis membranes,” Taylor says, that can remove hardness and natural organic matter from fresh water. These membranes can also remove pesticides with a molecular weight of 190 or above, he adds. “You can see nearly complete removal of pathogens from most sources. Membrane processes typically offer 100 to 1,000 times greater particle and pathogen removal than conventional treatment. There’s just no other process that can compete with membranes–at least at this point in time.” He predicts that all pressure-driven membrane processes will be “boons for improving water quality.”
Reverse osmosis–which also excludes low-molecular-weight compounds–and nanofiltration have some overlapping properties. Both types of membranes exclude many of the small organic molecules that are precursors of trihalomethanes generated during chlorine disinfection. TheEnvironmental Protection Agency strictly limits the allowable amount of these compounds in drinking water based on the possibility that chronic exposure to them could cause cancer.
“THE REGULATIONS for disinfection by-products have been affecting utility operations since the early 1980s,” Taylor points out. “And the regulations are getting more stringent.” Another group of halogenated compounds–haloacetic acids–recently was added to the list of regulated disinfection by-products, he says, while the maximum acceptable contamination for trihalomethanes was cut by 20%.
Reverse osmosis is required for desalination because nanofiltration cannot remove monovalent ions with sufficient selectivity. But nanofiltration membranes efficiently exclude divalent ions, including the calcium and magnesium ions that make water hard and cause scaling in water pipes. A future trend is for nanofiltration to do double duty, especially at facilities not using reverse osmosis. It could be used to replace lime softening for Ca2+ and Mg2+ removal as well as to keep the concentration of trihalomethanes and haloacetic acids within EPA limits.
Moreover, in treating waters with very high concentrations of dissolved organic carbon, a situation “not uncommon in Florida,” nanofiltration would be more cost-effective than activated carbon–a highly porous form of graphite–in removing organic and halogenated organic compounds, Taylor says.
A major drawback of membrane-based technologies is cost. “They’re expensive,” emphasizes Jeffrey J. Mosher, director of technical services at the Association of Metropolitan Water Agencies, Washington, D.C. “Utilities employ them only for very specific reasons.” The majority of water-treatment facilities will continue to produce safe water by conventional methods, he says, optimizing their systems as needed to meet any new EPA rules. But some facilities will need to adopt the advanced technologies, he acknowledges.
Two factors contribute to making membrane technology pricey. One is the amount of power consumed by the pressure-driven systems, particularly by high-pressure reverse osmosis. And the other is membrane fouling, a phenomenon that has become a focus of much research.
“We Americans are now getting there in membrane research for wastewater treatment,” comments Menachem Elimelech, director of the Environmental Engineering Program and a professor of chemical engineering at Yale University. “European countries are far ahead of us,” he says, especially France. In the late 1980s, many U.S. researchers went to France to study membrane technology for water treatment, he notes. “Now we all agree that the future of water treatment is membranes.” Membranes would obviate the need to continually have to modify a process to meet new environmental regulations, he points out.
<But ways to prevent or minimize membrane fouling are needed. Fouling occurs when constituents in process water deposit onto a membrane surface or accumulate within membrane pores. Higher feed pressures are required to filter the water, and the quantity or quality of product water can decrease. Membranes currently are cleaned by stopping the operation to wash them. Microfiltration membranes may need physical cleaning every 20 minutes, whereas reverse-osmosis membranes are cleaned chemically after months of operation, Elimelech says. Over the long term, membrane units are replaced when they become too fouled and their performance significantly deteriorates.
MEMBRANE FOULANTS include insoluble inorganic substances, colloidal or particulate matter, dissolved organic substances, and biological matter. Colloids, however–which include clays, iron oxides, colloidal silica, large macromolecules, and calcium carbonate precipitates–are considered to be a major cause of fouling, Elimelech notes. Not many studies have been conducted on the role of physical and chemical interactions in colloidal fouling of reverse-osmosis membranes, he says. Thus he and his colleagues are investigating these interactions using model colloids and a laboratory-sized membrane test unit.
On another front, the Office of Naval Research (ONR) is funding research to improve membrane performance to make it easier to comply with anticipated national and international regulations regarding disposal of wastewater from ships. Currently, ships don’t have to treat “gray water”–the water discarded from showers and kitchens–before discharging it overboard. But regulations on the horizon may require this water to be treated before disposal, notes Benny D. Freeman, a professor of chemical engineering at North Carolina State University. And that’s where membrane technology will help.
“Membranes have a great advantage over other technologies because they are very space-efficient devices, which makes them useful where ‘footprint area’ is at a premium, such as aboard ships,” Freeman notes.
In work funded by ONR and the federal Strategic Environmental Research & Development Program, Freeman and colleagues, including Ingo Pinnau, principal scientist at Membrane Technology & Research, Menlo Park, Calif., “are harnessing the unique properties of block copolymers” to prepare thin, nonporous coatings for porous membranes to provide them with an antifoulant surface. One segment of the copolymer is “water loving,” Freeman says, and this segment is connected to a polymer chain that’s not water soluble.
“The polymer chains self-assemble to form a nanostructured, nonporous coating containing highly hydrophilic channels imbedded in a strong matrix,” Freeman explains. “The coating is highly permeable to water but not to particulate matter.” With proper selection of the polymer coating, separation properties may be tuned, for example, to reject oil droplets in treating oily wastewater such as bilge water.” The technology is designed to be easily integrated into the existing membrane manufacturing process, he says.
OTHER ONR-FUNDED approaches include one being developed by Anne M. Mayes, an associate professor of materials science and engineering at Massachusetts Institute of Technology. In a process known as immersion precipitation, a small amount of a “comb” copolymer–a composite copolymer in which hydrophilic groups dangle from a hydrophobic backbone–is added to the bulk polymer solution from which the membrane is cast. The combs segregate to the surfaces of the membrane, including the interior surface of the pores, and endow them with hydrophilic teeth.
Another “very long term project” being conducted by Douglas L. Gin, an assistant professor of chemistry at the University of California, Berkeley, involves the design of a liquid-crystal-based membrane having uniform pores.
“Usually membranes have a big dispersion in pore size,” notes ONR program officer Paul Armistead, a chemical engineer. “So all the treated water ends up going through the large pores and the areas of the membranes with small pores are wasted space. If all the pores are exactly the same size and there’s a high density of them, you would get optimum water flow through the membrane.”
Ever since environmental engineers discovered how membranes could benefit water treatment, surface scientists, polymer physicists, polymer chemists, process chemists, and others have been using their expertise to change the surface of the workhorse polymers such as polypropylene, polysulfone, and polyethersulfone that manufacturers already know how to make–and make well–asserts John Pellegrino, a chemical engineer at the National Institute of Standards & Technology in Boulder, Colo.
Pellegrino likens the design of membranes for water treatment to “making materials that are as close as possible to being like a nonstick frying pan for a broad range of complex solutions.” With regard to water, though, the surface should be such that “to remove water from it would take an awful lot of energy,” he says. Some approaches being investigated aim to mimic biological membranes, such as grafting ligands on the membrane that clear the way for mass transfer of water to occur preferentially, he notes.
At Rensselaer Polytechnic Institute, chemical engineering professor Georges Belfort notes, “We have directly measured the adhesive forces between polysaccharides and proteins and polymeric membrane substrates and have developed some useful ‘rules’ on choosing suitable membranes for water treatment and other applications.” The group has patented an approach to deposit self-assembled monolayers on polymeric substrates to obtain extremely homogeneous surface chemistries.
Belfort and colleagues also have developed and patented a new method for cleaning micro-, ultra-, and nanofiltration membranes during the filtration process. They induce centrifugal vortices under specified flow conditions to create extra shear near the membrane surface, thereby increasing membrane flux by 50 to 300%.
“We named the method after Leonardo da Vinci, who was the quintessential vortex observer,” Belfort notes. Da Vinci-based membrane modules are currently in commercial development, he adds.
VARIOUS TECHNOLOGIES for water purification are being funded by the Defense Advanced Research Projects Agency (DARPA) to make water purification more accessible to military personnel stationed in out-of-the-way places, says William Warren, program manager for DARPA’s Air & Water Purification Program.
At Biosource Inc., Worcester, Mass., a DARPA-funded project aimed at desalinating seawater is under development. It draws on the power of electronics for solid-state water purification, Biosource President Marc Andelman says.
The method is based on a flow-through capacitor that picks up ions when a voltage is applied. The electrically charged capacitor stores the ions until it’s full. At that point, the ions are released to a waste stream by discharging the capacitor. Alternatively, the energy stored in the capacitor can be recovered and used in the next cycle, Andelman notes. Any shortfall in energy can be compensated for by “an inductive energy recovery circuit” that Biosource has developed. It acts like an induction coil, he explains. “The desalination process starts over in harmonic motion. It’s like giving a push to someone on a swing to keep the oscillation going.” As a result, the process can consume very little energy, he says.
Commercially available capacitor technology works well for treating water containing up to 6,000 ppm of salts, Andelman points out. But last month, using a capacitor made from advanced materials, the company partially desalinated water with a salt concentration of 35,000 ppm, the concentration found in ocean water. “Our new material can best be described as a poor man’s nanotube,” Andelman quips. “It’s similar to nanotubes, the major difference being that it’s inexpensive.” The process uses less energy than reverse osmosis, and scale-up efforts to develop it for desalinating seawater are under way, he says.
A mixed-oxidant disinfectant codeveloped by Miox Corp., Albuquerque, and already in commercial use is the basis for a system being developed by Miox in collaboration with DARPA. The system would allow individual soldiers to disinfect water.
A CIGAR-SIZED GADGET that can disinfectant 300 L of water on one set of lithium camera batteries has already been fashioned, notes Rodney Herrington, Miox vice president for engineering. It is intended for use with canteens. But the military now is interested in having soldiers use a water-toting backpack rather than a canteen, so Miox is applying the technology to develop caps operating on AA batteries that would release disinfectant into the backpack when screwed in place. The caps could also be used on standard water bottles. They’re not throwaway items, Herrington says, since they can treat up to 100 L of water. The company also is involved in developing a reverse-osmosis system that would offer soldiers a combination filtration and disinfection package.
Meanwhile, in Florida, membrane-based water-treatment technology has taken hold, its popularity having increased over the past 10 years. Depletion of pristine groundwater has been forcing water-treatment facilities to produce potable water from alternative sources, explains Seungkwan (S. K.) Hong, an assistant professor of environmental and civil engineering at the University of Central Florida. Indeed, Florida has required several water districts to use sources other than groundwater for their drinking supply, Central Florida’s Taylor affirms.
Between 150 and 175, or about a third of the nation’s membrane facilities, are located in Florida, Taylor notes. They variously treat surface, brackish, groundwater, and seawater. Surface waters can be highly turbid and easily contaminated by wild or domesticated animals, he points out. In Florida, surface water from many river systems “looks like tea,” he notes, because it contains a lot of humic materials, the natural decay products of leaves and trees.
The coagulation step during treatment doesn’t remove enough of these organic materials to meet the disinfection by-product regulations when chlorine is used as the disinfectant. And that’s a problem, Taylor says. But there are ways to treat the water without having to resort to membrane technology. For brackish water, though, which contains high concentrations of dissolved solids, and seawater “reverse osmosis is the only practical choice,” he insists.
Reverse osmosis coupled to microfiltration was the technology of choice for the Dunedin, Fla., water-treatment facility that came on-line in 1992. The city had been plagued with severe water-quality problems because of the high content of iron and sulfur in its various wells. Moreover, the groundwater source for the city–the Floridian aquifer–was providing less water and the threat of saltwater intrusion into the aquifer was becoming a reality. In addition, EPA’s drinking water standards had tightened.
THE CENTRALIZED PLANT is tailored to produce potable water from a blend of well water and brackish water, and it’s designed to be easily upgraded should the need arise. It’s one of the more modern reverse-osmosis plants, designed to be easily operated and cost-effective, says Robert H. Brotherton, director of Public Works & Utilities for Dunedin. Seen as a model operation, the plant has been visited by dignitaries and scientists from Singapore and Middle Eastern countries, among others, he adds.
Tampa Bay Water, a regional agency that provides drinking water to three counties on the West coast of Florida as well as to three cities–Tampa, St. Petersburg, and New Port Richey–is constructing a seawater desalination plant due to be completed by December 2002, agency engineer Donald E. Lindeman says. The plant will be the largest desalination plant in North America when it comes on-line. It’s being built on the site of a Tampa Electric Co. power station, which will provide power at a reduced cost.
A second cost-cutting asset for the facility is that the water in Tampa Bay is not as salty as ocean water, so more moderate pressures will drive membrane filtration. Another advantage is that the plant will be able to dilute the briny concentrate from the process before discharge by mixing it with cooling water being discharged from the power plant back into the ocean.
THE WATER SITUATION in the Tampa Bay Area currently is “very, very tenuous,” Brotherton says. “Tampa is in a state of emergency. Water is being pumped out of sinkholes and all kinds of sources to keep the city in water.”
In Yuma, Ariz., a large desalination plant is on standby, ready for action when needed. Constructed by the U.S. Bureau of Reclamation, the plant is intended to desalinate irrigation water diverted near Yuma as it drains back into the Colorado River south of the city. A treaty signed in 1944 entitles Mexico to an annual allotment of Colorado River water, but drainage water was raising the salt content of the river going into Mexico. The product water from the plant is put into the river to help meet the treaty requirements.
“We started up the plant–the world’s largest reverse-osmosis plant–in 1992,” notes plant manager Paul McAleese. Then the rains came. First the Gila River, which empties into the Colorado River, overflowed. Then El Niño came along, and heavy snowfalls followed. Precipitation in general has been more than adequate since 1993, averting the need to use the plant, he says.
Drainage water, which contains about 3,000 ppm of salts, is currently being diverted to a wetlands site in Mexico, where it has expanded the wetlands there from about 400 to 10,000 acres. “It has become a very important wildlife habitat down there,” McAleese says. Meanwhile, he demonstrates the plant’s operations to many international visitors who are interested in establishing or modernizing reverse-osmosis plants in their home countries.
Like Florida, Southern California’s Orange County is feeling the brunt of a water squeeze, which promises to get worse as the population increases and the supply of imported water shrinks. Since 1976, Orange County’s Water District (OCWD)Water Factory 21, in Fountain Valley, has been purifying a portion of treated wastewater effluent and blending it with well water to maintain the underground water dams–injection wells held at high water pressure that prevent seawater from intruding into the region’s sea-level aquifer.
Under a project jointly sponsored by OCWD and the Orange County Sanitation District, the factory is getting set to step up the quota of wastewater effluent that is being purified. Called the Groundwater Replenishment System (GWR System), the project was approved on March 28. It entails construction of a water-purification plant on the same site as Water Factory 21, notes OCWD civil engineer Shivagi Deshmukh. Deshmukh was a member of the team that developed the full-scale advanced technology demonstration plant for the project.
AT THE NEW PLANT, wastewater effluent will skip conventional water-treatment steps. Instead, it will be processed in sequence by microfiltration, reverse osmosis, and UV light, which will serve as the disinfectant. Results from the demonstration plant indicate that the near-distilled-quality water produced meets California’s drinking water standards, Deshmukh notes. However, it will not be used to supply drinking water directly. Instead, some of it will be injected underground to expand the number of water dams, and the rest will be transported to nearby Anaheim, where it will be percolated into the aquifer to help lower the mineral content of the groundwater.
“What’s really interesting is the technology allows us to produce this water with half the energy it takes to pump water here from Northern California,” observes GWR System program manager Thomas M. Dawes. “And right now, saving energy is a big deal in California.”
<UV light is a highly potent disinfectant for water. It destroysCryptosporidium, a pathogen resistant to chlorine that’s found in animal wastes. Since microfiltration also removes the offending parasite, the GWR System plant deals it a double whammy.
In 1993, more than 50 people died and an estimated 400,000 became ill in Milwaukee from water contaminated withCryptosporidium. The tragedy sent shockwaves around the world, says Lon A. Couillard, water-quality manager for the City of Milwaukee Waterworks. Since then, new plants using microfiltration have been constructed while others have been retrofitted to use ozone, which also kills the parasite, instead of chlorine.
UV DISINFECTION of water is new and not widely used. This year, EPA intends to propose regulations that would requireCryptosporidium levels to be monitored in source water, notes EPA environmental engineer Stig Regli. If they exceed allowable limits, utilities would be required to provide additional treatment, and they could opt to use UV disinfection, he adds.
The new plant at Water Factory 21 will be the largest membrane-based wastewater reclamation plant in the world. But it won’t be the first to practice wastewater reuse. In Northen Virginia, for example, the Upper Occoquan Sewage Authority, Centreville, has been doing this since 1978, notes Executive Director James L. Bannwart. Treated wastewater effluent is discharged into an “effluent” reservoir, a 55-acre surface lake, before flowing over a spillway into Bull Run, a stream that joins up with the Occoquan Reservoir at its northen edge. This reservoir, a sort of slow-moving river, meanders south over 20 miles before water is drawn from it to be processed into drinking water.
>The Centreville plant replaces 11 small plants in the Northern Virginia watershed. In the 1960s, effluent from these plants was found to be polluting the reservoir, which led to construction of the Centreville facility. It attracts visitors from water- and wastewater treatment facilities worldwide who are interested in planned indirect water reuse to augment water supplies.
The practice of infusing treated wastewater into groundwater for indirect potable use is of great concern to Dunedin’s Brotherton. Although purified wastewater may meet drinking water standards even before it’s percolated, that doesn’t mean it won’t contain various contaminants–including drug metabolites, endocrine disruptors, and pesticides–because it’s not being tested for many such compounds, he says. “It’s what we do not know about the compounds in wastewater that gives me the greatest pause. The few tests for drinking water standards are not adequate when starting with wastewater rather than a pristine water supply.”
Even reverse osmosis, “the best technology, in my opinion, for making pure water, won’t provide a 100% barrier to everything that may be in the water,” Brotherton continues. The reason is that seals in the membranes can leak, allowing molecules normally excluded to escape. “What happens in the laboratory does not necessarily replicate itself in normal plant operation,” he notes.
With regard to the Tampa Bay Area, Brotherton believes the use of reverse-osmosis filtration needs to be expanded to ensure high-quality water. “But the local agencies are hesitant to do this, because they don’t understand the main problem–the quality of the water people are drinking, not the fact that we’re running out of it.”
At Water Factory 21 in California, ongoing research is directed at fingerprinting bacteria in the highly treated clear-looking wastewater effluent received from the sanitation plant. The idea is to “characterize the range and diversity” of bacteria in the water that can cause membrane biofouling, explains Harry F. Ridgway, director of water resources and technology. Bacteria are characterized and cataloged according to their 16S ribosomal RNA, a classic, well-established approach. The knowledge gained is expected to be useful in designing membranes that resist biofouling.
In Florida, other concerns have surfaced. Switching from groundwater to surface water as the source of potable water may be expected to adversely affect the linings of water-distribution pipes, Cental Florida’s Taylor indicates. “If you change the water quality, you destroy the film and get the release of undesirable substances.” Taylor currently is collaborating with Tampa Bay Water on ways to pretreat surface water so this situation doesn’t arise.
The ultimate goal is to get the last bit of economic efficiency out of all water-treatment technologies, NIST’s Pellegrino asserts. “Water is a commodity, and it has to stay a commodity–it’s like gasoline in this country now. And there’s a finite amount of water in the world. When we increase industrial activity and we increase populations, we have to turn water over at a faster rate than nature allows.” Treatment processes that run faster and more efficiently and create high-quality water are needed, he insists.
