A pure blue light source makes a test tube glow. Inside is half the reaction needed to turn water into energy.
Sitting amid a tangle of wires and plastic tubing, Morgan Mihok examines a murky orange concoction, her thin frame hunched over to get a better look at what she hopes is the secret to cheaper, cleaner fuel. The vial is capped with an air-tight rubber stopper. Two long needles connected to plastic tubes penetrate the seal, allowing a constant flow of argon gas to purge all traces of atmospheric air from the vial. Mihok turns on a magnetic plate beneath the test tube, and in response a small magnet in the solution causes the orange liquid to swirl like a mini-maelstrom. Flipping another switch, she engages a pure blue light source. The glass of the test tube glows sapphire, and Mihok, a second-year chemistry student, waits.
She has been preparing this experiment for weeks. Hours of studying, meticulous note-taking, and chemical mixing are contained in this one vial. "If everything goes right," she says in a distracted whisper, "we should see something in 30 minutes or so."
Mihok's experiment is based on a simple formula plants have used for eons: water + sunlight = energy. In practical terms, it means using sunlight to break apart the chemical bonds between the hydrogen and oxygen atoms in water—H2O—in order to use the elemental hydrogen for generating energy. Plants accomplish this photosynthesis through a complex series of reactions controlled by carbon-based enzymes. But Mihok's process is more streamlined and entirely inorganic. In a sense, it's artificial photosynthesis. The result—a cheap and plentiful source of hydrogen gas—could revolutionize how machines operate. Instead of internal combustion engines, machines of the future might be powered by fuel cells, a type of battery that combines hydrogen fuel and atmospheric oxygen to create electricity.
After a painfully slow half-hour, Mihok inches closer to her glowing solution. Taking a syringe, she floods its chamber with argon gas to eliminate all traces of air inside, pushes the gas back out, then gently inserts the needle through the rubber seal of the test tube and extracts some air. She rushes three feet down the lab bench to the computer analyzer and injects the syringe into a receptor labeled samples.
Mihok's face, calm an hour ago, looks tense. The monitor flashes on, and a red line begins to inch its way, oh so slowly, across the screen. Her eyes fix on that line. She points to a spot on the screen. "That's about where the hydrogen should show up." Nothing. The red line starts to climb. The screen blinks—and the line is flat again. No hydrogen.
A minute later, the line starts to climb. "That's oxygen," she says. "Some probably crept in when I took the sample, unless there's a leak in the system, which means the whole experiment is ruined . . . Come on, go back down." The line descends, indicating only a small amount of oxygen. "Looks like we're going to be okay." After another minute, the line climbs, showing that some nitrogen has crept in too, but it appears there's no hydrogen at all.
"Chad!" Mihok calls Chad Waraksa, a graduate student who works with her on the project, to take a look at the results. He types a command on the keyboard and the graph changes shape. "You have to format this for what kind of experiment you're running," he says. Another screen appears. Mihok jumps. "Woo hoo! Look!" Pointing to the first line of the analysis, she reads: "Hydrogen."
The lab Mihok and Waraksa share is one in a maze of 12 interconnected laboratories off of an unassuming corridor in Chandlee Laboratory. Tom Mallouk, Penn State professor of chemistry, directs the entire wing. A sign in the hallway announces, "The Center for Miscellaneous Chemistry: Serving Central Pennsylvania since August." The rest of the corridor's wall space is cluttered with diagrams, displays of published papers, and pictures of Mallouk & Co. in strange costumes at Halloween parties. Inside the labs, where students like Mihok and Waraksa work, counters are strewn with centrifuges, agitators, and oscillators. Glassware and chemicals clutter the shelves. The place sounds busy: whirs and hums and clinks muffle the students' voices.
By contrast, Mallouk's office at the far end of the hallway seems quiet and orderly. High-tech micrographs share wall space with crayon drawings done by his children and quilts stitched by his wife. A floor-to-ceiling bin of erector set-type molecule models guards one wall. The shelves are packed with books.
Mallouk himself is not what you'd expect in a chemist. He's a T-shirt and blue-jeans guy, bearded and rugged, with shaggy hair. A smile line or two betray his otherwise youthful face. His quick wit and easy-going manner set students immediately at ease. He's probably never been accused of being too serious.
Mallouk brought his Center for Miscellaneous Chemistry from the University of Texas at Austin seven years ago. "I had some colleagues at Texas who started putting signs on their doors that read, 'The Center for Such and Such,' complete with whizbang logos," he explains. "They were totally serious. The funny thing is that they were only one-man operations." Mallouk chuckles. "When the third one went up I said, that's enough, and decided to create my own 'center.'
"I came to Penn State for family-related reasons," he adds, "but also because of this university's strength in materials research across the board. I was looking forward to collaborating with different departments, and we're doing just that—with electrical engineering, materials science, and physics. That's an opportunity I probably wouldn't have had at Texas."
Mallouk has been on the trail of artificial photosynthesis since his undergraduate days at Brown. He was in high school when the oil embargo hit in the early 1970s and was taken with the push for alternative fuels. "People were seriously concerned that we were going to run out of oil and needed some other source of fuel," he says. "In college I got my start with a really inspiring professor and I got to do my first real experiment—no one had ever done it before, no directions. We were developing new kinds of semiconductor crystals for photosynthesis reactions. That avenue of research eventually led to a dead end, but I was hooked."
Artificial photosynthesis could provide a clean, renewable alternative to fossil fuels: renewable because it would require only water, clean because its only by-product would be oxygen. What's more, the right system could produce hydrogen cheaper than the price of gasoline—once it's running, the only energy required is light. "There's enough energy in sunlight to vaporize the oceans," says Mallouk. "All you need is the right catalyst."
Almost 30 years ago, a team of scientists from the University of Tokyo took the first step toward making artificial photosynthesis a reality. They announced in Nature that they'd successfully split water using ultraviolet light and a catalyst of titanium dioxide. While a system using ultraviolet light doesn't have many practical uses (not enough ultraviolet light penetrates Earth's atmosphere), the experiment proved that artificial photosynthesis wasn't just a dream. Mallouk calls it proof of concept. "That paper really got the science world in a buzz about artificial photosynthesis," says Mallouk. "It showed us that it could be done."
About the same time, American chemists discovered a class of compounds called ruthenium trisbipyridines that are sensitive to blue light. These orange-colored molecules, which give Mihok's test-tube solution its mashed Creamsicle appearance, fire off an electron when struck by a photon of light. "Ru-bipy"—as it's known in chemistry circles—"is a fantastic reducing agent when stimulated by light," says Mallouk. "Water-splitting requires 1.23 volts of energy. The excited state of Ru-bipy gives off 2.1 volts—almost twice what we need."
In 1989, Japanese scientists were studying potassium hexaniobate, a semi-conducting material made of potassium, niobium, and oxygen. Like all semi-conductors this material can receive or donate electrons. But unlike most, the researchers found, its many layers can be chemically split apart into individual sheets. "The proportions of one sheet are about the same as this," says Mallouk, holding up a piece of notebook paper, "only on a very, very microscopic scale. This feature is important because we've learned how to split apart a stack of these sheets and embed pockets of catalyst between them." More recently, British scientists discovered that another material, iridium dioxide, is especially good at catalyzing the production of oxygen from water, an important part of the water-splitting process.
Artificial photosynthesis, Mallouk-style, would work like a molecule-sized machine. When light energizes a Ru-bipy molecule, an adjacent stack of potassium hexaniobate gladly accepts the ejected electron and ushers it to its core, where microscopic pockets of platinum have been implanted. "Think of the platinum as a singles bar," explains Mihok. "It serves as a meeting place for electrons and hydrogen ions to hook up." Hydrogen ions (protons) are naturally present in water. And since acids, by definition, have a higher concentration of hydrogen ions than plain water, Mallouk's recipe calls for an acidic solution to increase the rate of the reaction. So when an electron arrives at the platinum, it invariably finds a waiting proton. But like many couples out on a first date, this newly formed union doesn't like to be alone, chemically speaking, so it finds another recently formed hydrogen atom. They bond together and voilà: hydrogen gas.
Meanwhile, the Ru-bipy molecule is in dire need of an electron to replace the one it ejected. If it doesn't get another one soon, it will fall apart. Fortunately, iridium dioxide, suspended in the mix, is an electron giver, and the ravenous Ru-bipy snatches up an electron from the iridium dioxide molecules (only to surrender one again when struck by another photon of light). Soon, the iridium dioxide molecules, in turn, badly need replacement electrons.
At this point, the only possible source of electrons is water, and without hesitation iridium dioxide rips two electrons from each of two adjacent water molecules. The former water molecules' two oxygen atoms float off together as oxygen gas, and the four remaining protons find themselves strangely drawn to a platinum pub not far away.
In theory, this cycle will continue until all of the water has been photosynthetically converted to hydrogen and oxygen. "But we're not there yet," says Mallouk. In order to develop a fully functional water-splitting system, Mallouk's team has taken one step back, isolating the two parts of the reaction to more closely study and fine-tune them.
"Morgan and Chad are working on the hydrogen side," Mallouk explains. "They're using hydrogen iodide, which is a pretty strong acid, as a source of protons instead of water. On the oxygen side, we're using a chemical that gets used up to induce the reaction. Through these methods, we've been able to experimentally demonstrate that both sides of the reaction do what they're supposed to do." All that remains is to put them together. Unfortunately, that's not as easy as it sounds.
Deep in the basement of Chandlee Lab, in a dark box of a room, Chad Waraksa blasts away at Ru-bipy molecules. "Chad's got destructive tendencies," jokes Mallouk. "He really likes playing with lasers." The room is uncomfortably warm, the lighting low. Waraksa wears green-tinted safety goggles that look like 1970s sunglasses. Beads of perspiration dot his forehead. The room hums with electricity. Waraksa hits a button, and a beige box the size of a small desk fires a blinding seven-nanosecond-long pulse of emerald laser light. The bright green blast careens through one prism, gets redirected by another, and finally smashes into a sample of Ru-bipy. The impact of laser photons energizes the light-sensitive molecules, setting off a torrent of electron expulsions. A scanner that detects ultraviolet and visible light analyzes the electron flow and sends the results to the monitor where Waraksa sits.
"What this process does is determine the efficiency of the light absorption of the Ru-bipy molecule," he says. Explains Mallouk, "Efficiency has really become the focus of our research. If every electron energized by a photon of light met up with a hydrogen ion, then the reaction would be functioning at 100 percent efficiency.
"Right now, we're at about 3 percent. That means 3 percent of the electrons are ending up where they're supposed to. The remaining 97 percent are probably going back where they came from."
In order to decrease the fraction of electrons slipping into this "back reaction," Mallouk's team has found a way to more precisely place the platinum within the potassium hexaniobate. "Previously, the layered potassium hexaniobate was stacked like a deck of cards, with the platinum randomly interspersed like jokers," says Waraksa. "But we've developed a way to slip a layer of platinum between only two layers of potassium hexaniobate—sort of like a platinum sandwich." The experiment that Mihok has been working on uses this technique. The idea is that if the electron has a shorter distance to travel to reach the platinum, it will be more attracted to a night at the pub than to going back home.
Waraksa is also formulating another approach to increase the reaction's effectiveness, based on an "energy gradient." "Instead of one layer of potassium hexaniobate, I'm trying to create three layers of different semiconductors that have an increasing affinity for electrons," he says. "That way, each successive layer pulls the electron further from its starting point and closer to the platinum, making it even harder for electrons to go back."
Another side to the efficiency problem is a phenomenon in which the rates of all the reactions are determined by the rate of the slowest one. "They bottleneck," explains Mallouk. "Think of a tractor-trailer going up a steep mountain road. It doesn't matter how fast the cars behind it are, they'll all end up going the speed of the semi. In our water-splitting system, the hydrogen reaction takes about one millisecond, but the oxygen reaction takes one second. That still means it's a thousand times slower than the hydrogen reaction. But if we can shrink that difference to just one power of ten instead of three, then we'll have a total system efficiency of ten percent, and we'll be in business."
Despite the hurdles, Mallouk seems confident that a working system is just a matter of time. "Well," says Mallouk with a laugh, "Department of Energy grants run in three-year cycles, so I hope we get our system running within the next year or so." They won't be waiting for fireworks, either. "We're not looking for perfection by any means. We're shooting for a few percent—maybe ten percent would be interesting. If we demonstrate that we've designed a reproducible method of sunlight-powered artificial photosynthesis that works, at that moment we publish a paper—proof of concept." That paper will ignite the field much like the first successful artificial photosynthesis experiment did 25 years ago.
And Mallouk will be just as happy if it comes from someone else's lab. "I don't know who will ultimately find the breakthrough artificial photosynthesis system," he says. "It's my job to take the science as far as it can go. When someone says to me, 'Can you split water for me,' I see a very complex series of kinetic reactions. But beyond cutting-edge materials and breakthrough formulas, I'm developing good scientists. And that's the real goal. Ultimately, even if someone else comes out with artificial photosynthesis first, we still win."
Morgan Mihok is a chemistry major in the Eberly College of Science and the Schreyer Honors College. Her research adviser is Thomas Mallouk, Ph.D., the DuPont Professor of Materials Chemistry in the Eberly College of Science, 215 Davey Lab, University Park, PA 16802; 814-863-9673; tom@chem.psu.edu. Chad Waraksa is a doctoral candidate in chemistry. This project is being funded by a grant from the U. S. Department of Energy. Andrew Gathman graduated from Penn State in May 2000 with a degree in English. He is currently writing for Genesis Publishing in Baltimore.