Photonics researchers are harnessing electron excitement for advancements in solar-to-fuel conversion and infrared imaging
The Free Electron Movement
Once elusive, solar-to-fuel conversion is looking like gold in a UCSB lab
by Sonia Fernandez
At UC Santa Barbara, researchers continue to find novel ways of using light — in both the visible and invisible spectra — to address man’s growing need for energy and hunger for information. Through the combination of plasmonics and nanotechnology, researchers have been able to capture a storable form of energy from visible and invisible parts of the spectrum. Manipulating this electromagnetic energy could allow researchers to develop new technology for power generation and imaging.
A new way of harvesting the sun’s energy
In a little water-filled vial in UC Santa Barbara chemistry professor Martin Moskovits ’ laboratory, a tiny disc may hold the key to our pressing present and future fuel needs. When illuminated by the sun, this disc — no bigger than one’s fingertip — is capable of breaking the chemical bonds of water, producing hydrogen and oxygen, thus directly storing sunlight as usable fuel.
“This pursuit has been growing for more than 100 years,” said postdoctoral researcher Syed Mubeen , of the ongoing search for a more robust and efficient way to harvest solar energy and turn it into fuel. Unlike solar-to-electricity applications, where conventional photovoltaics have made great strides in efficiency and affordability in the decades since their inception, developing a technology for sustainable solar-to-fuel conversion processes has been elusive, until now.
“Such devices have been made by many researchers in the past, using conventional semiconductor materials,” said Mubeen. “The problem is, when highly efficient semiconductors, such as silicon or gallium arsenide, are in an aqueous environment, they photocorrode, and stop working after a few minutes.”
There have been some inroads made in the solar-to-fuel quest using semiconductors based on metal oxides, like titanium, for instance. These semiconductors don’t fail as readily the silicon-based types, but the tradeoff is that they absorb only the ultraviolet portion of sunlight — about four percent of the spectrum — so their efficiencies are highly limited. Meanwhile, the search for a viable means of converting the Sun’s energy into fuel intensifies, as concerns over the environmental drawbacks of using fossil fuel mount.
Enter gold, one of the Earth’s most stable and conductive metals. Resistant to corrosion, it can be placed in many aqueous solutions without disintegrating, or otherwise reacting. Enter also an entirely new application for plasmonics.
“We have been working on plasmonic materials for many years in other contexts,” said Moskovits, whose research emphasis is in physical chemistry and materials. For decades, plasmons — the collective oscillation of conduction electrons — have been studied and used in applications such as enhanced spectroscopy, for instance, or to detect molecules adhering to surfaces. However, it was the specific social context, which in this instance is the urgent concern to develop alternative energy resources, that spurred the group into considering plasmonics as a source of non-fossil fuel energy.
Harnessing excited electrons
In conventional photovoltaics, sunlight hits semiconductor material, one side of which is electron-rich, while the other side is not. The photon, or light particle, excites the electrons, causing them to leave their positions, and create positively-charged “holes.” The result is a current of charged particles that can be captured and delivered for various uses, including powering lightbulbs, charging batteries, or facilitating chemical reactions.
In the technology developed by Moskovits and his team, it is not semiconductor materials that provide the electrons and venue for the conversion of solar energy, but the surface of one of the world’s most well known and precious metals.
“When certain metals are exposed to visible light, the conduction electrons of the metal can be caused to oscillate collectively, absorbing a great deal of the light,” said Moskovits. “This excitation is called a surface plasmon.”
However, these excited, “hot” electrons are very short-lived, lasting only about ~ 10 femtoseconds (~ 1014 seconds) before they relax.
To get an idea of just how briefly these electrons stay hot, imagine a stretch of beach that’s 20 feet long by 20 feet wide by five feet deep. That’s one second. Ten grains of sand would be comparable to 10 femtoseconds.
“The question was, can you capture these electrons effectively and put them to useful work?” said Mubeen. To do this, the Moskovits team — which also included chemistry postdoctoral researcher Joun Lee , chemical engineering graduate researcher Nirala Singh , materials engineer Stephen Kraemer, and chemistry professor Galen Stucky  — turned to the very tiny world of nanostructures.
“These hot electrons tend to travel ~10^6 meters per second, which means they could travel at least a few tenths of a nanometer before decaying as heat. The challenge was to come up with an appropriate nanostructured design so that before these electrons decay as heat you use them to do useful chemical reactions,” Mubeen said.
The result is an array of gold nanorods, each rod measuring 80 to 100 nm in diameter and 500 nm in length. Ten billion of these nanoreactors can occupy one square centimeter. Six hundred of them lined up side by side would span the diameter of an average (clean) human hair.
Each nanorod is capped with a layer of crystalline titanium dioxide decorated with platinum nanoparticles. A cobalt-based oxidation catalyst was deposited on the lower portion of the array, and the entire arrangement is submerged in water.
When the negatively charged hot electrons, excited by sunlight, oscillate, they travel up the rod, through the titanium dioxide layer and are captured by the platinum nanoparticles, causing the reaction that splits water molecules. Meanwhile, the positively charged “holes” left behind by the excited electrons head downward to the oxidation catalyst to form oxygen. According to their study, hydrogen production was clearly observable after two hours, and the nanorod array proved to be the durable visible light-harvesting device sought by the researchers.
“The device operated with no hint of failure for many weeks,” Moskovits said. Additionally, according to Mubeen, the use of nanostructures provides the opportunity to scale up for relatively little cost, even with an expensive metal like gold.
Quest for efficiency
Currently, efficiencies for this plasmonic technology are at about .25 percent, which is comparable to silicon semiconductor-based photoprocesses almost a century ago. And, plasmonic technology is still more costly than that for conventional semiconductors.
“We still have a lot of work to do,” said Mubeen, ticking off a list of ideal qualities that would make nanostructured plasmonic materials competitive with conventional semiconductors. “We need to test cost-effective plasmonic metals, so we can make fuels cheap enough. We need to re-engineer the system design to be more efficient.”
Copper and silver are being eyed as alternatives to gold, and an efficiency of 5 percent or more is one of the early targets for the research.
“If the last century of photovoltaic technology has shown anything, it is that continued research will improve on the cost and efficiency of this new method - and likely in far less time than it took for the semiconductor-based technology,” said Moskovits.
“In view of the recentness of the discovery, we consider .25 percent to be a ‘respectable’ efficiency,” he said. “More importantly, we can imagine achievable strategies for improving the efficiencies radically.”
Catching the (invisible) wave
Meanwhile, in another lab on the UCSB campus, researchers Hong Lu , Art Gossard  and Mark Sherwin  have performed a feat that may provide a wide array of applications, from more efficient solar cells to higher-performance telecommunications to enhanced imaging and sensing technologies.
It comes in the form of a compound semiconductor of nearly perfect quality with embedded semimetallic nanostructures, and it capitalizes on the manipulation of the infrared (IR) and terahertz (THz) range of the electromagnetic spectrum. These invisible areas of the spectrum — with longer wavelengths and lower frequencies than the naked eye can sense — offer much in the way of information they can provide. However, the development of instruments that can take advantage of their range of frequencies is still an emerging field.
Bridging optics and electronics
To cope with the demands of today’s information technology — more data, faster transmission, better energy efficiency — researchers have been turning to optics, using IR light to transmit information.
However the transition between optics and electronics is a difficult one because they operate at vastly different scales, with electron confinement possible in spaces far smaller than light waves. The size gap between the technologies have been a hurdle for scientists and engineers trying to integrate the two with a circuit that can take advantage of the speed, capacity and energy efficiency of optics with the compactness of electronics for information processing.
Here plasmonics plays a vital role, by providing the highly sought bridge between the two technologies. Key to this technology is the use of erbium (Er), a rare earth metal that has the ability to absorb light in the visible as well as infrared wavelength, and has been used for years to enhance the performance of silicon in the production of fiber optics. Pairing erbium with the element antimony (Sb), the researchers embedded the resulting compound — erbium antimonide (ErSb) — as semimetallic nanostructures within a semiconducting matrix of gallium antimonide (GaSb).
When IR light hits the surface of this semiconductor, electrons in the semimetallic nanostructures begin to resonate — that is, move away from their equilibrium positions and oscillate at the same frequency as the infrared light — preserving the optical information, but shrinking it to a scale that would be compatible with electronic devices.
“This is a new and exciting field,” said Hong Lu, project scientist in materials and in electrical and computer engineering. But the ability to translate optical information into electronic data is only one benefit of this unique semiconductor.
‘A new kind of heterostructure’
In the world of semiconductors, structural quality is of utmost importance: the more regularly repeating and aligned — “flawless” — the arrangement of atoms in the semiconductor’s crystal lattice is, the more reliable and better performing the device in which it will be used will be.
Generating these perfect structures is no minor feat. Any mismatch in size or alignment becomes magnified and could result in cracking. The difficulty becomes even greater when incorporating different atoms, which may be desired for their properties, but not so for their potential to result in defects. While semiconductors incorporating different materials have been studied for years — a technology UCSB professor and Nobel laureate Herbert Kroemer pioneered — a single crystal heterostructured semiconductor/metal is in a class of its own.
ErSb, according to Lu, is an ideal material to match with GaSb because of its structural compatibility with its surrounding material, allowing the researchers to embed the nanostructures without interrupting the atomic lattice structure of the semiconducting matrix, each atom aligned with the matrix around it.
“The nanostructures are coherently embedded, without introducing noticeable defects, through the growth process by molecular beam epitaxy,” said Lu. “We can control the size, the shape and the orientation of the nanostructures.” The term “epitaxy” refers to a process by which layers of material are deposited atom by atom, or molecule by molecule, one on top of the other with a specific orientation.
“It’s really a new kind of heterostructure,” said Arthur Gossard, professor of materials and electrical and computer engineering.
Seeing things in a new light
The researchers have already applied for a patent for these embedded nanowires as a broadband light polarizer.
“For infrared imaging, if you can do it with controllable polarizations, there’s a lot of information there,” said Gossard.
The researchers credit the collaborative nature between departments on the UCSB campus for this multidimensional breakthrough.
“One of the most exciting things about this for me is that this was a ‘grassroots’ collaboration,” said Mark Sherwin, professor of physics, director of the Institute for Terahertz Science and Technology at UCSB. The idea for the direction of the research actually came from the junior researchers in the group, he said, grad students and undergrads from different laboratories and research groups working on different aspects of the project, all of whom decided to combine their efforts and their expertise into one study. “I think what’s really special about UCSB is that we can have an environment like that.”
Researchers on campus are also exploring the possibilities of this technology in the field of thermoelectrics, which studies how temperature differences of a material can create electric voltage or how differences in electric voltages in a material can create temperature differences. Renowned UCSB professors John Bowers  (solid state photonics) and Christopher Palmstrom  (heteroepitaxial growth of novel materials) are also investigating the potential of this new semiconductor.