Living Batteries

FEATUREFuel Cells Using sugars, sludge, and the sea floor, can bacteria power the next green-energy alternative?BY JACK LUCENTINIWhen Bruce Rittmann got a $100,000 NASA grant two years ago to find ways for converting human excrement and other organic waste into electricity in spacecraft, he prepared to reach for the stars.When the space agency cut research funding, his ambitions became a bit more grounded. Usin

Jul 1, 2006
Jack Lucentini
FEATURE
Fuel Cells
Living Batteries

Using sugars, sludge, and the sea floor, can bacteria power the next green-energy alternative?

BY JACK LUCENTINI

When Bruce Rittmann got a $100,000 NASA grant two years ago to find ways for converting human excrement and other organic waste into electricity in spacecraft, he prepared to reach for the stars.

When the space agency cut research funding, his ambitions became a bit more grounded. Using private-sector funds, Rittmann, a professor and director of the Center for Environmental Biotechnology at Arizona State University's Biodesign Institute, is now working to convert sewage and other organic matter to electricity here on Earth. He doesn't mind the shift in focus for work that he says could "transform the world."

His experience is emblematic of the halting yet determined progress scientists are making with microbial fuel cell technology. These biological batteries use bacteria to extract electrons from an organic fuel such as carbohydrates, proteins, or raw sewage, and then deliver them to an electrical circuit.

The devices are already in use for certain specialized applications, such as powering scientific instruments. They're also lighting unexplored aspects of microbial life. Scientists hope microbial fuel cells could become a more widespread way to generate relatively clean energy. But before leading the next wave of ecofriendly energy, they must overcome tremendous challenges; their modest power output means that general applications are, by most accounts, far away.

Endeavors like Rittman's - aimed at drawing electricity from waste at sewage plants to assist in treatment - could become economically viable in as few as five years, according to some optimistic predictions. But for most other anticipated uses, such as generating power for remote rural areas, or powering consumer electronics, the picture is murkier. The highest power levels reached to date - at least in experiments generally agreed to approach realistic conditions - are between 2.5% and 15% of what estimates suggest would be needed to make them economically viable (see chart, below). Most of the power improvements have taken place over the past seven years. It could take decades for microbial fuel cells to become practical.

These fuel cells "certainly have strategic small-scale applications," but also face a "credibility gap," says Frank C. Walsh, professor of electrochemical engineering at the University of Southampton, UK. A key problem is that a half-dozen other basic types of fuel cells are under development, some of them 1,000 times more powerful than microbial fuel cells, he says.

Nevertheless, microbial fuel cells have a powerful draw owing to their unique biology. Korneel Rabaey, a postdoc at the Advanced Wastewater Management Center at the University of Queensland in Brisbane, Australia, says, "It's not just a technique to generate electricity; you can also use it as a tool for biotechnology." In addition, he says, the microbes' ability to clean up waste as they generate electricity will make them a viable player for such applications. Microbial fuel cells can run on agricultural and industrial runoffs as well as human waste - "fuels that otherwise cannot be used," he says.

Many of the researchers studying this technology say they believe its limitations can be overcome. Now, groups can tweak several fuel cell parameters to improve performance (See Basic Components of a Fuel Cell), and several parameters remain untested. If harnessed, microbes could unlock the power in vast amounts of waste.

FROM PUGH TO POWER

The founding of the fuel cell is usually attributed to Michael Potter, a professor of botany at the University of Durham, UK, who in 1910 put one electrode into a yeast-filled flask, and another into an organism-free solution.1 He detected a meager current. The concept languished until the 1960s and later, however, when researchers began to understand the principles by which microbes might generate electricity.2

As bacteria break down organic substances anaerobically, they strip away electrons to use them in the generation of ATP, releasing carbon dioxide and protons in the process. Once used, the bacteria must rid themselves of these electrons, generally reducing an environmental electron acceptor such as sulfur, or iron or manganese oxides. The fuel-cell setup gives bacteria an alternative place to dump their electrons. Fuel cells are arranged so that bacteria live on or near an anode, typically made of graphite, and are provided an unlimited supply of organic material. Electrons flow to the positive terminal, or cathode, powering an electrical device in the process. At the cathode, the electrons combine with protons and oxygen to form water.

Early on, says Bruce Logan, professor of environmental engineering at Pennsylvania State University, the currents generated were "negligibly small." Until recently, he adds, everyone thought that fuel cells required added chemical mediators to ferry the electrons from microbe to anode. The mediators were usually toxic and costly, making the devices decidedly impractical.

"The turning point was around 1999," Logan says, when Byung Hong Kim at the Korea Institute of Science and Technology in Seoul and colleagues showed that the mediators were unnecessary.3 They found that an iron-reducing bacterium recently discovered in a rice paddy, Shewanella oneidensis, could pass electrons to an anode on its own, possibly through direct contact. The organism apparently evolved this ability, Kim's group noted, because its natural electron acceptor, Fe (III), is insoluble in its habitat.

There's a gaping lack of knowledge on how to optimize microbial fuel cells, and this is cause for hope.

It turns out that an array of metal-reducing bacteria can accomplish similar feats. A flood of work on "mediator-less" microbial fuel cells followed. In 2003, researchers at the University of Massachusetts, Amherst, reported that Geobacter sulfurreducens, a bacterium discovered in a contaminated ditch nine years earlier (pictured on the opposite page), could convert its fuel to electricity with a dramatically greater efficiency than organisms in previous studies, by oxidizing its acetate fuel completely and forming a film on electrode surfaces.4

Researchers today typically harness species including Rhodoferax ferrireducens, Desulfuromonas acetoxidans, Pseudomonas aeruginosa, and Clostridium butyricum. Often a mix of microbes inhabits the fuel cell. They use a variety of pathways to oxidize fuels ranging from glucose to starch, acetate, cysteine, and wastewater, and use cytochrome proteins expressed on their surfaces or electron-shuttling compounds for reduction reactions. If denied their natural electron acceptors, they will use electrode materials instead.

Different organisms and configurations have produced a plethora of fuel cell designs. The most successful to date may be the Benthic Unattended Generator, or BUG. While not the most powerful, it's the only one already practical for certain applications, according to some researchers. Clare Reimers of Oregon State University in Corvallis, Ore., and Leonard M. Tender, a research chemist at the Naval Research Laboratory in Washington, DC, created fuel cells in which an anode is planted directly in rich underwater sediments. Here, naturally occurring bacteria associate with the anode and produce electricity. A cathode operates in the oxygenated water above. BUGs generate up to about one watt, says Tender, which is enough to power small scientific instruments such as temperature or pressure gauges.

"They have immediate applications in powering instruments in remote locations," says Peter Girguis, assistant professor of biology at Harvard University, because they keep working where conventional batteries die and solar cells become fouled. They've been tested in the Potomac River, and the laboratory plans to deploy BUGs to power an instrument measuring water current velocities this summer in the Gulf of Mexico, says Tender.

But these remain specialized applications. Researchers say the next most likely microbial fuel cell application, and the first relatively widespread one, will be in sewage treatment plants. The idea is to use the sewage as fuel for the microbes in the fuel cells. The bacteria would offset some of the huge costs of running the plants by simultaneously producing electricity and helping to break down wastes.

This is the application that "most engineers are focusing on," says Derek Lovley, a professor of microbiology at the University of Massachusetts, Amherst. It would have huge benefits, Logan adds. Some 1.5% of the electricity produced in the United States goes toward maintaining wastewater infrastructure, he says. "Many countries can't afford to develop it" fully, so any cost offset would be valuable. Pressed for an estimate of when sewage fuel cells will see real-world use, Girguis and Logan both venture five years. "We're talking to a few companies" about commercializing the technology already, Logan says.

The highest power output his group has reported for an experimental, sewage-fueled fuel cell is 464 milliWatts per square meter of anode surface (mW/m2).5 That's almost halfway to the benchmark number of one watt per square meter, which Logan estimates would constitute economic viability.

Rittmann's forecast is less sanguine. "I am very enthusiastic" about the fuel cells, he writes in an E-mail, but adds that he doesn't want to give people "an unrealistic idea that diverts R&D funding to the wrong directions." He estimates that economic viability would require power levels fully 100 times higher than those commonly attained now. He allows that Logan's halfway estimate might be accurate if one overlooks the one-time costs of building new plants to incorporate the novel technologies, but that's unrealistic, he says. "Treatment of sewage is not like dunking a few electrodes in existing treatment reactors. This will be a totally new design," he writes.

Gloomy as this may sound, microbial fuel cells have an advantage when it comes to the wastewater application: There is limited competition from other sorts of fuel cells, and an irresistible scientific draw. "We have microbial ecology, biofilms, electrochemistry, materials, and kinetics all wrapped together," Rittmann writes. "So, it is scientifically exciting, to say the least."

OUT OF THE SEWER

Other proposed fuel cell uses include generating electricity in rural areas far from power plants, and powering an array of consumer devices. For such applications, estimates of what constitutes commercial viability range from 10 to 60 watts per square meter. "For a lot of applications we need a really substantial breakthrough," Lovley says. To make those ideas realistic, researchers propose two strategies: Increase power, or reduce costs.

Researchers are examining various ways to increase power, but benchmarking is difficult because different laboratories measure fuel cell performance differently. The measures used most often show power per unit area of anode, or mW/m2. Most researchers say a more useful measure is ultimately power per unit volume of the cell, such as mW/m3, because the space it occupies is more relevant economically than electrode area. Rough translation between the two measures is possible.

CREATED WITH INPUT FROM BRUCE LOGAN
A CHALLENGE: The bar chart represents progress to date in microbial fuel cell power output in milliWatts per square meter of anode surface, compared to a range of scientists' estimates as to the power-output goals required for the devices to become economically viable. The bars represent the highest power output, for each year, in tests generally agreed to approach realistic conditions. The inset scatterplot shows some individual results (on a logarithmic scale) as well as the goal estimates for economic viability.


Logan has explored boosting power by modifying the spacing between anode and cathode. A smaller spacing can increase power by reducing electrical resistance in the circuit, although too little space raises the risk that oxygen near the cathode might diffuse to the anode. A potent electron acceptor, oxygen short-circuits the system by stealing electrons, and it kills strictly anaerobic bacteria. In one study, Logan and colleagues found that a 2-cm space between anode and cathode was optimal.5 They also found that pumping fluids from the anode toward the cathode increased power, by pushing oxygen away from the anode.

Other ways of boosting power output involve adjusting the combinations of fuels or the anode and cathode materials. Foamy or fibrous materials work well, Walsh says, because they increase the microbe's working surface area per unit volume. "If you use a three-dimensional porous electrode such as carbon foams, felts, and fibrous materials ... you can get much more current," he explains.

Lowering costs is also critical, Walsh says, adding that part of this will happen automatically. The current expenses for electrode materials, for instance, reflect "high development costs. They will come down appreciably as large-scale use goes ahead." New materials typically drop tenfold in price within five years, he observes.

However, the most expensive component, the proton-exchange membrane, has tended not to fall in cost, Walsh adds. This membrane prevents oxygen from flowing to the anode, while allowing protons generated from microbial metabolism to migrate in the other direction. These are needed at the cathode to balance out the negative charge buildup there.

An ion exchange membranes costs "$500 to $1,000 per square meter, which makes it the most expensive component," says Walsh. Researchers need to figure out ways to reduce the need for this, but it's not easy, he adds. Logan says features of his setup, including the pumping, made the membrane unnecessary. Walsh acknowledges that Logan showed this could work for short periods, but "long-term proof of pudding needs to be shown." Many problems could arise with regular use, including anode corrosion, he adds.

LESSONS FROM NATURE

Less explored fuel cell innovations involve working with microbial physiology. "We may have to engineer a superbug to really drive the power output," Girguis says. Lovley says that genome-based modeling of Geobacter has suggested ways that the organisms could be engineered to drive up respiration rates and provide more electrons.6

But nature may outperform any engineered design (see Taming Electricigens). In natural settings, evolution changes the mix of organisms populating the anode and tends to improve the system over time, researchers have found. In 2004, Rabaey, then at Ghent University in Belgium, and colleagues reported that the fuel cell environments in their tests had selected for bacteria that could deliver electrons to the anode more efficiently.7 This occurred, they explained, because the anode was the strongest electron acceptor in the system. Thus organisms using it, rather than other available acceptors, could take advantage of a greater drop in redox potential between acceptor and donor, and could therefore generate more ATP.

The selective pressure spurred an evolution that over 10 weeks, resulted in a sevenfold increase in power per unit area, Rabaey and colleagues reported. The higher output was associated with growing production of compounds that shuttle electrons to the anode, suggesting that the system was selecting for bacteria that produced more of these, Rabaey argues.

The microbes "truly turn toward living with an anode as their final acceptor," says Willy Verstraete, head of the Laboratory of Microbial Ecology and Technology at Ghent University and a Rabaey collaborator. Microbial communities probably evolve both in their individual physiologies, and in the way different species in the mix work together, he adds. For instance, Rabaey says, certain thick-walled, Gram-positive bacteria in the system probably can't reduce electrodes on their own, and live by exploiting electron shuttles produced by other organisms such as P. aeruginosa. "We don't know the return service" they provide, if any, he adds.

Better understanding of the microbial communities may provide a plan for directing the evolution of more powerful fuel cells, and unknowns still outweigh the knowns. Lovley says there's a gaping lack of knowledge on how to optimize the fuel cells, and this is actually cause for hope. "We've seen a dramatic increase" in power outputs even with that limited understanding, he points out. As to whether they will ever be commercially viable, he says, "I would err on the optimistic side."

References
1. D.R. Lovley, "Bug Juice: harvesting electricity with microorganisms," Nat Rev Microbiol, in press.
2. A.K. Shukla et al., "Biological fuel cells and their applications," Curr Sci, 87:455-68, 2004.
3. B.H. Kim et al., "Direct electrode reaction of Fe(III)-reducing bacterium, Shewanella putrefaciens," J Microbiol Biotechnol, 9:127-31, 1999.
4. D.R. Bond, D.R. Lovley, "Electricity production by Geobacter sulfurreducens attached to electrodes," Appl Environ Microbiol, 69:1548-55, 2003.
5. S. Cheng et al., "Increased power generation in a continuous-flow MFC with advective flow through the porous anode and reduced electrode spacing," Environ Sci Technol, 40:2426-32, April 1, 2006.
6. R. Mahadevan et al., "Characterization of metabolism in the Fe(III)-reducing organism Geobacter sulfurreducens by constraint-based modeling," Appl Environ Microbiol, 72:1558-68, February 2006.
7. K. Rabaey et al., "Biofuel cells select for microbial consortia that self-mediate electron transfer," Appl Environ Microbiol, 70:5373-82, 2004.