In his 1874 science fiction tale The Mysterious Island, Jules Verne predicted, “Water will be the coal of the future.” It is a vision of infinite clean energy available for people to use. More than 30 years ago, Japanese scientists took a seminal step in that direction. With a piece of titanium dioxide and some sunlight, they split water into hydrogen and oxygen. Although researchers have tried to refine the process over the years, nobody has come up with a system that is both efficient and inexpensive enough to produce sufficient hydrogen for use as a clean-burning fuel on the roads, in industry, and at home. Recently, however, researchers have picked up the pace of their pursuit of the ultimate water-splitting system.
With rising oil prices and the specter of climate change that’s due to the burning of fossil fuel, the vision of a hydrogen economy looms ever larger in people’s minds. After all, it’s a fuel for which the only by-product is water. And hydrogen packs more energy per unit mass than any fossil fuel does.
“The conversion to hydrogen has already started. It’s inevitable,” says Nejat Veziroglu, director of the Clean Energy Research Institute at the University of Miami.
But the main source of hydrogen today is natural gas, a non-renewable resource. And the steam-based process for extracting hydrogen from the gas generates carbon dioxide—one of the primary global warming gases. To circumvent these problems, scientists are exploring alternative strategies. Among them are photosynthetic microbes that churn out hydrogen (SN: 10/12/02, p. 235: Hydrogen: The Next Generation) and electromechanical systems that use the electricity from wind turbines to make hydrogen from water (SN: 7/21/01, p. 45: Power Harvests).
However, many scientists contend that catalytic materials that use sunlight to split water on the spot, a process known as direct solar-hydrogen production, could be the most promising strategy. In solar-hydrogen systems, when photons strike the catalytic material, they excite electrons, which then roam about freely until they meet a water molecule at the material’s surface. The extra electrons strip the two hydrogen atoms away from water’s one oxygen atom, producing hydrogen fuel. The oxygen atom simultaneously hooks up with another oxygen atom, forming an oxygen molecule.
Not only is sunlight readily available, “you don’t need a lot of water to make hydrogen fuel,” says John Turner of the Department of Energy’s National Renewable Energy Laboratory in Golden, Colo. If all the 230 million cars and other light-duty vehicles in the United States were running on hydrogen, 100 billion gallons of water per year would be sufficient to supply the fuel, Turner says. The nation’s households collectively consume almost that much water in just a week of drinking, cooking, and washing.
What’s sorely missing, however, is a water-splitting material that’s simultaneously efficient, inexpensive, and stable. Whoever invents a substance that meets all three criteria will add momentum toward a hydrogen economy, revving up progress to highway speeds.
Mix it up
The most widely studied material for solar hydrogen is titanium dioxide, the same stuff used to make white paint. Titanium dioxide is great at splitting water, but it absorbs only ultraviolet (UV) light, which constitutes a scant few percent of the solar energy reaching Earth’s surface. Researchers have tried with some success to increase titanium dioxide’s efficiency by spicing it with different elements. Other chemists are instead layering different light-absorbing materials to combine the best of each into one device that taps the broad band of energy the sun offers.
Several years ago, Turner and his colleagues created such a layered device by placing gallium indium phosphide, which absorbs ultraviolet and visible light, on top of gallium arsenide, which absorbs infrared lights (SN: 4/18/98, p. 246). The resulting device could convert 12.5 percent of sunlight’s energy into the production of hydrogen. This was a feat of efficiency, but “the materials are expensive and they only last about 20 hours” before corroding, says Turner.
To speed up the discovery of suitable materials, chemical engineers at the University of California, Santa Barbara have adapted a robotics-intensive strategy known as combinatorial chemistry, the same approach that pharmaceutical chemists take to synthesize and test new drugs. Led by Eric McFarland, the group designed a system that can rapidly synthesize 120 different materials and test each one’s water-splitting capacity, all in a single day.
The system works in the following way. First, the researchers coat a 4-inch-square glass plate with titanium foil to serve as an electrode. They next add a thick layer of Teflon perforated with 120 holes. The researchers then fill the holes with different preformulated solutions containing a dissolved semiconductor material mixed with various metals. For instance, the base semiconductor might have zinc oxide or tungsten oxide, and each sample would be doped with differing proportions of, nickel, copper, or chromium.
In a process called electrodeposition, a robotic instrument dips an electrode into each solution one by one, causing the dissolved materials to form a thin solid film on the titanium-coated glass. Peeling away the Teflon leaves behind 120 thin dots of material, each with a different composition.
To test the water-splitting potential of the newly created films, a second robotic instrument lowers a tiny chamber onto each dot and fills the chamber with a conductive aqueous solution, or electrolyte. The robot then shines light on the chamber and measures the current the film generates. By repeating this quick test on each film, the robot screens the entire array in a matter of hours.
The greater the current produced by a film when illuminated, the more electrons it gives up, and therefore the greater its potential to split water and generate hydrogen.
This technique amounts to an efficient form of trial-and-error. “We can afford to try all sorts of wacky things,” says Thomas Jaramillo, one of the investigators working on the project. “That’s the real power of this technology.”
It also enables the researchers to take inexpensive semiconductors such as zinc oxide and tweak them to improve their properties. Already, the Santa Barbara team has seen some promising results.
The researchers found that when they added cobalt to zinc oxide to create a mixture that was 4.5 percent cobalt by weight, they boosted the zinc oxide’s current-generating capacity fourfold. The extra cobalt enables the material to absorb a larger part of the solar spectrum and thereby free up more electrons, explains Jaramillo.
Changing a material’s chemistry is just one way of devising new candidates for use in solar-hydrogen production. With the advent of nanotechnology, scientists have come to recognize that tweaking a material’s fine structure can have dramatic effects. Take the so-called tandem cell invented by Michael Grätzel of the Swiss Federal Institute of Technology in Lausanne, a leader in the field of photovoltaics.
The tandem cell consists of two separate but electrically connected light-absorbing materials, one of which splits water. The water-splitting material faces the sun and consists of a thin film of either tungsten trioxide or iron oxide in front of a sheet of conducting glass. The back material is a photovoltaic device known as a Grätzel cell (SN: 12/20&27/03, p. 398: Available to subscribers at New materials take the heat).
The nanoscale structure of the metal-oxide film is critical to its water splitting capacity. The film is made of 50 to 100 loosely packed layers of metal-oxide spheres, each about 20 nanometers in diameter. This geometry provides a vast amount of surface—1,000 times as great as its two-dimensional area. The small spheres also make the material more chemically reactive than it would be in bulk form.
When exposed to sunlight, the nanostructured film absorbs UV and blue light. The rest of the spectrum passes through the material to be absorbed by the Grätzel cell. That solar cell provides extra electrons that make the water splitting more efficient.
In September, Hydrogen Solar—a British company that’s working with Grätzel to develop the technology—announced that its tandem cell with the nanostructured film had achieved 8 percent efficiency. This marked a doubling of the performance of earlier devices without the nanostructured film. The firm says it is close to reaching the U. S. Department of Energy’s efficiency goal of 10 percent, says David Auty, chief executive of Hydrogen Solar, which is headquartered in Guilford, England. That’s the benchmark for commercial viability, says Auty.
Auty envisions installing arrays of tandem cells on the rooftops of home garages. The cells would provide drivers with hydrogen for their fuel cell vehicles. These vehicles would consume hydrogen and produce water, essentially reversing the process that generated the hydrogen in the first place.
A rooftop unit working with 10 percent efficiency in a sunny Southern California location could generate enough hydrogen to drive 11,000 miles per year in the small Mercedes-Benz fuel cell car that went on the market in Germany in June, says Auty. To generate larger amounts of fuel, tandem cells could cover the roofs of factories and even central fueling stations, from which trucks would transport hydrogen across the country. This scenario would still require practical solutions for the transportation of this highly explosive gas.
With funding from the Department of Energy, Hydrogen Solar is collaborating with the University of Nevada at Las Vegas to develop its technology. The company plans to have a pilot fueling station up and running near the campus in 3 years.
Auty concedes that hydrogen from tandem cells, at least in the near term, will cost at least twice of much as hydrogen produced from natural gas. But the price of natural gas fluctuates widely, he says. What’s more, unlike the method used to extract hydrogen from natural gas, the tandem-cell technique doesn’t generate carbon dioxide.
Like many people in the industry, Auty anticipates that regulations eventually will dictate that oil and gas companies capture and sequester the carbon dioxide they generate. “I don’t think anyone knows just how much extra it’s going to cost, but it’s certainly going to add to the price of the fuel,” he says.
And that, Auty adds, should help make solar hydrogen more economical.
While Hydrogen Solar continues to refine its materials, other groups are pursuing a different approach. These investigators are engineering complex molecular machines that can split water using solar energy.
Consider the work of Karen Brewer of Virginia Polytechnic Institute and State University in Blacksburg. She says the reason it’s been so difficult to make efficient solar hydrogen materials is that each water molecule needs two additional electrons to strip off its hydrogen atoms. Her lab is developing molecular structures that can deliver multiple electrons simultaneously to a central reaction center, which then catalyzes the splitting of water.
Many of the other solar-hydrogen materials are inefficient in gathering up the two electrons needed to split water. Because the reaction is thus energy intensive, materials such as titanium dioxide work only under high-energy UV light. By designing materials at the molecular scale, Brewer says she can build greater efficiency into the system. For instance, she has designed molecular complexes that absorb visible light and thereby tap into energy carried in a larger part of the solar spectrum.
The Virginia Tech team tested different combinations of components over many years before it created successful supramolecular complexes. “We figured it out through an awful lot of work and a lot of wrong choices along the way,” says Brewer. Her group presented its results in August at the national meeting of the American Chemical Society in Philadelphia.
Brewer’s molecular complexes mimic natural photosynthesis. The machine, a combination of organic and metal-containing components, comprises three main units. A chemical bridge connects each of the two light-absorbing units to a catalytic central unit.
The light-absorbing units contain ruthenium atoms. As in a chlorophyll molecule, a photon hitting a ruthenium atom excites one of its electrons. The electron is then shuttled to the central unit, which contains a rhodium atom. The rhodium collects electrons, two at a time, to perform the reaction.
To ensure that the excited, mobilized electrons would gather in the central unit, the researchers designed the complex’s chemical bridges to attract the electrons from the light-absorbing segments and shuttle them in the right direction. Once the team got the bridges in order, the next challenge was finding a metal for the central catalytic unit that would be strong enough to pull the electrons. “The problem initially was that the electrons would just sit there on the bridges,” says Brewer.
Eventually, the team found the answer in rhodium. “This has been a major breakthrough for us,” Brewer says. Not only is rhodium a strong electron acceptor, but it’s also reactive enough to split water and it’s stable. In short, it could be just right for making solar-hydrogen systems.
The researchers have done preliminary experiments in which they mixed the molecular complexes with water in a glass vial and exposed the vial to visible light. Soon thereafter, hydrogen began to bubble from the system. These initial studies indicate that the efficiency of the system is already “reasonable,” Brewer says.
As the world begins to shift toward a hydrogen economy, other hydrogen-generating technologies initially might win out over solar-generated hydrogen. “Countries will use whatever source is cheapest at the time,” says Veziroglu. “It may be natural gas, coal, wind, or hydropower, but eventually, it will be solar energy.”
And if water and sunlight are all it takes, then Jules Verne’s fantasy of burning water like coal will have been realized.