An infinite charge
Fuel cell chemistry is a simple dance.
The trick is to make it cheaper.
The engine of the future is as quiet as a card game. There's no ignition. No exhaust smoke. No knee-trembling rumble or skull-penetrating roar.
Stacks of PEM fuel cells. The large unit produces 5 kilowatts of electricity, enough to power a house. The smaller stacks produce 25 and 30 watts.Photo by Matt Stiveson (DOE/NREL)
In fact, a fuel cell is not an engine at all: It's an electrochemical device. Like a battery, a fuel cell produces power by converting the chemical energy present in a fuel and oxidizer directly into electricity. Unlike a battery, it doesn't store energy, and therefore it doesn't run down.
"It works like a battery with holes in the top and bottom," says Matthew Mench, assistant professor of mechanical engineering at Penn State. "You flow fuel and air through it and it produces a steady current. Keep the flow going and it's always charged."
The Welsh physicist Sir William Grove came up with the idea in 1839. Grove's "gas voltaic battery" reversed the well-known principle of electrolysis, or water-splitting, to create electrical current by combining water's components.
Simply put, a fuel cell produces electricity from hydrogen and oxygen. Its basic structure resembles a sandwich, with two chambers—one negative, the other positive—on either side of a thin electrolyte membrane. Hydrogen atoms flow into the fuel cell at the negative post (or anode). There, a catalyst triggers a chemical reaction that splits the atoms into their two parts, protons and electrons.
The positively charged protons stream through the membrane to the positive post (the cathode). The free electrons can't get through, however. Instead, they move out of the anode into electrical wiring, where they provide the direct current (DC) voltage that lights the lightbulb or powers the drive-train motor.
All electricity works by closing a circuit, so the electrons keep moving through the wiring and re-join the protons on the cathode side. That's where the hydrogen atoms mix and mingle with oxygen, with predictable results: water. The entire process is a simple chemical "dosey-do" that spins electrons away from their proton partners, then reels them back in to the dance.
Today's fuel cells come in several forms, distinguished by the material used for the electrolyte membrane. Alkaline fuel cells have long been used in the U.S. space program, but are expensive and easily contaminated. Direct methanol fuel cells will soon replace batteries in laptop computers and cell phones, but are too inefficient for larger applications. Solid oxide and molten carbonate fuel cells, which operate at temperatures over 600 degrees C, are being developed for large power plants for home and industry.
Inside the box
The sexiest fuel-cell application—and the one that's central to a hydrogen economy—is, the automobile. It's also, by far, the hardest one to achieve. "The system has to be compact," Mench explains. "It has to bear up under the harshest conditions—starting reliably in snow and heat." And the competition is brutal. "The internal combustion engine is a very mature technology."
Gold-plated fuel cell designed for diagnostics by Matthew Mench and colleagues at Penn State's Electrochemical Energy Center.Photo by Matthew Mench
The top contender for meeting this challenge is the polymer electrolyte membrane, or PEM, fuel cell. "It's basically a thin film between two plates," Mench says, hoisting an example. The electrolyte layer is a clear polymer membrane that looks a lot like plastic wrap. The PEM cell, he explains, operates at relatively low temperature—around 80 degrees C. That means it warms up quickly, and doesn't need much housing. It also converts energy efficiently enough that two stacks of about 200 cells—the size of a large suitcase—is enough to power a car.
There is, however, the cost problem. "To compete with a gasoline engine, we have to get to about $60 per kilowatt," Mench says. "Right now industry is at about $3,000." Another serious issue is durability. "We can get good performance at the beginning of operation," Mench says, "but you want it to last for the equivalent of 100,000 miles—that's about 5,000 hours of operation. And that's start-stop, freeze-thaw. Right now we're at about 1,000 to 4,000 hours for this type of operation."
A third major challenge is what Mench calls water management. "Basically a fuel cell is an electrochemical water-generating device. To conduct protons, the electrolyte has to be fully saturated with water. But you can't have too much just sitting in the system—it reduces performance. What if you park the car overnight and it freezes? You have to do something with it."
False-colored neutron image of a PEM fuel cell used to quantify the distribution of liquid water inside the cell. The red areas contain the most water, the dark blue and black areas the least. Imaging by Matthew Mench, in collaboration with Jack Brenizer and Kenan Unlu of Penn State's Radiation Science and Engineering Center.
At Penn State's Fuel Cell Dynamics and Diagnostics Laboratory, which Mench directs, "We're doing research to try to understand where exactly the water is in the system, and how it is causes damage," he says. "Then we can design the fuel cell to put the water where we want it." The Electrochemical Engine Center, directed by professor Chao-Yang Wang, also does computational modeling and experimental studies of fuel cells for automotive and portable applications.
At the Dynamics and Diagnostics lab, Mench uses sensors distributed throughout a fuel cell to measure temperature, moisture, and other factors at various points in the energy producing cycle. He is also working with professors Jack Brenizer and Kenan Unlu to use neutron imaging to visualize the movement of water inside a working cell. "It's analogous to an MRI or X-ray," Mench says. "Our goal is to take the fuel cell from what it has been—a black box—and to observe and measure its performance at as many points as possible."
Hot stuff
For materials chemist Tom Mallouk, "the fuel cell problem is simple. We have good fuel cells, but they're made of expensive materials." A PEM cell membrane, for instance, includes a few milligrams of platinum to help catalyze the electrochemical reaction. Add that up for 200 cells and it comes to about 150 grams. When it comes to breaking into the automobile market, that little bit of precious metal is cost-prohibitive. "Cars are sold by the pound," Mallouk says. "The price of a car is less, per pound, than the price of hamburger. You can't add tens of thousands of dollars to that."
There are several ways to cut cost, he adds. One alternative is to build a fuel cell that runs hotter. "Any reactive catalyst works better when you jack up the temperature." At higher temperatures, the polymer membrane would require far less platinum to do its job. Take it high enough, Mallouk says, and you could substitute nickel, a less efficient—but much cheaper—catalyst.
Unfortunately, raising the temperature also runs the risk of material failure. The current polymer membranes, Mallouk explains, "are flexible, compliant, good proton conductors, but they only work when swelled with liquid water. If the water is boiled away, they won't work. So right now the big push is for fuel cells that run hotter but still below the temperature where polymer membranes lose all their water, about 200 degrees C."
An alternate approach, he says, would be to find a completely new material, something that can do what membranes do—conduct protons, hold up mechanically—at well above the boiling point. "If you could make a membrane that works at 300 degrees C," Mallouk says, "the catalysis problems would go away. That's one of the things we're working on."
Mallouk's idea is to move from polymers to inorganic materials, like ceramics. "They can retain water up to 300-400 C," he says. "They contain water within their crystalline structure." Ceramics can also be good proton conductors. The problem, though, is that ceramics tend to be brittle. They lack the mechanical strength needed for a flexible membrane.
"A few years ago," Mallouk remembers, "a colleague of mine, Eugene Smotkin at the University of Puerto Rico, had the idea of making a membrane that was a composite of a metal foil and a proton-conducting polymer. Metal foil is all wrong electronically—metal is a conductor not an insulator, and it won't allow protons through—but mechanically it's strong. And actually, certain metals, like palladium alloys, would let protons through." On came the lightbulb: Why not a composite? "The foil supports this patchy layer of ceramic deposited on top of it, which in turn acts as an insulator," Mallouk explains.
"We tried this in 1998 with a real simple system, and showed that it works," he says. Since then, he and his students have been fine-tuning. "The challenge is to make the perfect inorganic material for that thin film on top of the foil."
Help for hybrids
Serguei Lvov is also interested in making membranes that work at higher temperatures. "High temperature should improve performance," notes the professor of energy and geo-environmental engineering. "At higher temperatures the membrane can better survive impurities like carbon monoxide," which choke current fuel cells. Equally important, he says, hotter running cells are compatible with existing automotive technology.
"I think nobody now believes that a straight fuel-cell car is going to run in the near future," Lvov explains. Instead, "we're probably going to have a hybrid—consisting of a battery, a fuel cell, and a traditional heat engine." To make this combined system economical, "you need to run the fuel cell at high temperature—the automakers want 120 or even 130 degrees C." That bump in temperature, he says, makes humidity inside the cell a concern. "Below 100 degrees C, where PEM fuel cells run now, relative humidity is not a serious issue," he explains. "At 120, just a little bit makes performance drop off."
Composite membrane
Lvov thinks he and his team may have come up with a membrane material that can stand both high temperature and low humidity. It's based on combining the proton-conductive polymer Nafion with titanium oxide, or titania—the same white powder Tom Mallouk is using to make cheaper solar cells. "Nafion is a very well-known material, very well tested, and titania is very inexpensive, and also a proton-conductive inorganic," he says. "If you put the two together you end up with a cheap composite material that is mechanically stable, and can operate at a higher temperature. And we have shown that it works in an atmosphere of as low as 25 percent relative humidity."
The discovery, he notes, was somewhat serendipitous. Lvov was directing parallel projects, one with Department of Energy support aimed at developing high-temperature fuel cells, and the other, with support from Oak Ridge National Laboratory, whose goal is understanding the basic electrochemical surface properties of titania. "I saw that people were starting to use composite materials for fuel cell membranes," he says, "and one of those composites was Nafion and silica. At the same time, I had learned enough about titania's surface properties, and the differences between it and silica, that we decided to try titania for this purpose. And it looks like we were successful." Penn State has submitted a patent on Lvov's Nafion-titania composite membrane, and Dupont has signed on to sponsor further studies.
For Lvov, one lesson is clear. "Fuel cell research is very complicated. If you want to end up with something new, you have to bring together expertise from many different areas—polymer science, inorganic materials, electrochemistry." He has assembled a team consisting of 17 Penn State researchers, along with colleagues from other universities and industry, with the hope of starting a center for developing composite membranes.
Lvov isn't leaving future success up to happy accident. "In order to advance further," he emphasizes, "we must be very deliberate."
Matthew Mench, Ph.D., is assistant professor of mechanical engineering and director of the Fuel Cell Dynamics and Diagnostics Laboratory, mmm124@psu.edu. Thomas Mallouk, Ph.D., is Dupont professor of materials chemistry and physics, and director of Penn State's Center for Nanoscale Science, tom@chem.psu.edu. Serguei Lvov, Ph.D., is professor of energy and geo-environmental engineering, sxl29@psu.edu.
