John Bowers, Efficiency Expert
THE GLOBAL ENERGY CHALLENGE, like economics, is a matter of supply and demand. It means finding new fuels and power sources that cut oil imports and don’t contribute to global warming. That’s the supply side. Just as crucial is the demand side, where the objective is to waste less rather than produce more. The work on the demand side has often been lower-profile, such as designing more power-frugal computers and energy-efficient buildings. But that work is now catching fire because policy-makers are starting to see how the efficiencies discovered by the demand-siders can add up quickly to big economic and societal gains.
John Bowers is a demand-side leader. A longtime professor of electrical and computer engineering at the University of California, Santa Barbara (UCSB), he now holds the university’s new Fred Kavli Chair in Nanotechnology. He also leads UCSB’s Institute for Energy Efficiency (IEE), home of wide-ranging research on energy-saving ideas. Named to head the IEE when it was founded at the beginning of 2008, Bowers has helped get the Institute off to a fast start. In April 2009 it was awarded a $19 million grant in federal stimulus funds to host of one of the federal government’s new Energy Frontier Research Centers.
John Bowers holds one of the first hybrid silicon lasers. The die in the center of the plastic box is an “silicon on insulator” (SOI) wafer with an array of silicon waveguides and a thin layer of indium gallium aluminum arsenide (InGaAlAs) on top. (Credit: Jeffrey Tseng)
Bowers is widely known for his research on optical switching and other uses of light to transmit and process data. Matt Tirrell, the former dean of Engineering at UCSB and the man who picked Bowers to head IEE, calls him “an expert in using light rather than electricity, moving light around rather than electrons.” That makes Bowers an efficiency expert as well, because light can carry far more information than electricity at the same energy level, without generating heat. His research focus, coupled with the efficiency-related work of fellow UCSB researchers in fields such as lighting, building design and thermoelectric materials, is now paying off in public support: “We chose to focus on energy efficiency rather than energy generation, a viewpoint that has become very popular.”
Replacing Electrons with Light
Bowers earned his doctorate at Stanford University and worked for AT&T Bell Laboratories and Honeywell before joining the UCSB faculty in 1987. Along with UCSB colleague Dan Blumenthal, he founded Calient Networks, a maker of photonic circuit switches, in 1999. He is a member of the National Academy of Engineering and a Fellow of the Institute of Electrical and Electronics Engineers (IEEE), the Optical Society of America (OSA) and the American Physical Society. He has received the OSA’s Holonyak Prize, the IEEE Photonics Society William Streifer Scientific Achievement Award, and the South Coast Business and Technology Entrepreneur of the Year Award. In 2007, he and his research team received an EE Times ACE Award for Most Promising Technology for the hybrid silicon laser. He has also earned a reputation as a skilled administrator. Fred Chong, a UCSB professor of computer science and leader of the computing group at the Institute for Energy Efficiency, says Bowers brings a rare combination of qualities to his role. “He has good managerial talents,” says Chong, “but he also has enormous technical credibility.”
That “technical credibility” rests largely on what Tirrell (now chair of the Department of Bioengineering at UC Berkeley) calls Bowers’ “steady, relentless effort to replace electrons with light.” As Bowers explains it, his research aims to take the fiber-optic revolution of telecommunications “to the next level” -- deep into computers. He has helped lead the development of new semiconductor materials that, unlike silicon, emit light and could be the basis for a new generation of photonic devices. His objective is to transform the microcosm of the integrated circuit, he says, “by using light as the interconnect on a chip.” Ultimately, he wants to make it possible for fiber optics to replace all the copper circuitry that carries data in most computers today.
Bowers is working to create new nanomaterials that will boost the power of thermoelectric devices, such as the one shown here, to produce electricity from waste heat in automobiles. (Credit: Gehong Zeng)
The hybrid silicon laser recognized by the 2007 ACE Award marked a key step toward that goal. It combined light-emitting semiconductor compounds with a silicon base in a structure that was compatible with other devices and with standard CMOS (complementary metal-oxide-semiconductor) production technology. In other words, it could be used in computers without a complete revision of their processing. Since then, says Bowers, he and his colleagues have been using this hybrid technology to make more sophisticated devices: “We now are trying to create a whole fleet of devices that you need to create complete photonic circuits.”
What does this mean for the future of energy? Plenty, if you assume that the traffic on the Internet will continue to grow explosively as more users tap into high-bandwidth content such as movies and as more software applications are hosted at huge data centers. Bowers says the Internet is estimated to use about 4% of the world’s electricity now, but that figures are expected eventually to reach 10%. A large part of this consumption – about half, according to a 2007 study of server energy use commissioned by the microprocessor company AMD – goes into keeping the equipment cool. If nothing else, shifting to light from electricity as a carrier of data would sharply reduce cooling costs. It also would cut the direct energy cost of computing by moving data much more efficiently. Bowers points outs that an optical fiber has 10,000 times the bandwidth of a copper wire, and the attenuation of light (signal fading with distance) is 1,000 times less than that of electricity.
Game-Changers in Efficiency
Bowers’ work dovetails with other research at the IEE, such as Fred Chong’s “Greenscale” project focused on cutting energy waste at server centers. Along with tapping into Bowers’ new photonics technology, Chong is also working on software and server designs that allow servers more down time without limiting access to data.
Schematic of thermoelectric device. Core of thermoelectric device shown converts heat to power using a thin layer of indium gallium aluminum arsenide with embedded erbium arsenide particles (ErAs:InGaAlAs) joined to a layer of bismuth telluride (Bi2Te3). Electricity is produced from the temperature difference between the top and bottom of the layered materials. (Credit: Gehong Zeng))
In addition to Computing and Electronics/Photonics, the groups led by Chong and Bowers, the IEE divides its research into four other areas: Lighting, Production and Storage, Buildings and Policy. The Lighting group focuses mainly on developing solid-state lighting such as LEDs (Light Emitting Diodes). Production and Storage has roots in Nobel laureate chemist Alan Heeger’s work on organic photovoltaic materials; it is now branching out into thermoelectric materials, which convert heat to electricity and vice versa. The Buildings group enlists UCSB expertise in mechanical engineering, computer science and control systems to design buildings that are energy-neutral, consuming no more than they use. The Policy group studies the effect of regulations and examines technologies to determine which are truly cost and energy efficient.
Bowers is involved in other IEE research, including its research into thermoelectrics. As with photonics in computing, he sees thermoelectric materials as potential efficiency game-changers in appliances and cars. He has been working with UCSB colleagues and Amerigon Inc., a maker of automotive heating and cooling systems, to create new materials that are highly efficient at the electricity-to-heat conversion. Temperature-controlled car seats, which get warmer or cooler based on the polarity of the current running through their thermoelectric cores, are just a start. The real payoff will come, Bowers says, with the development of materials that are highly efficient at converting the automobile engine’s waste heat to electricity. “The impact of these will be greater than that of the hybrids of today,” he says. Similar materials could also transform the technology of the kitchen. Refrigerators that now rely on the compression and evaporation of gas to cool things down could be replaced by compact, solid-state thermoelectric cooling units.
The key to these advances is nanotechnology, the manipulation of molecules to create artificial materials that improve on nature. So Bowers’ new Kavli Chair in that subject area looks like a perfect fit with his research interests. It also gives him a new freedom to ask questions and follow hunches. “It gives one the opportunity to try out novel ideas,” he says. He can now pursue even “far out ideas” that would not stand much chance of attracting grant money. He notes that the “general idea of nanotechnology is making novel materials that don’t exist in nature” – and that would seem to cover a lot of interesting possibilities.