Published by the Natural Resources Defense Council (NRDC) Support Us

Sign Up for Our Newsletter


Heat + New Gizmo = Power

In 1882 Thomas Edison built the nation's first power plant in Lower Manhattan, selling electricity to New York City's rich and powerful. Edison himself flipped the first light switch in the Wall Street offices of J. P. Morgan. But Pearl Street station, equipped with six 27-ton dynamos, also produced tremendous amounts of heat. Edison, always the problem solver, used that heat to make steam, which he then sold to nearby businesses, where it was used to power factories and warm buildings during the winter.

When we convert energy from one form to another, as we do when we burn fuel to make wheels spin and electrons flow, invariably some is lost along the way. Since Edison's day those losses have risen precipitously: today scientists at the Department of Energy's Lawrence Berkeley National Laboratory estimate that nearly two-thirds of the energy generated in the world is lost as heat. If we could reclaim just 3 percent of our waste on a daily basis, they say, we could offset half the nation's electricity needs.

How we go about doing that is no small task, but over the past few years, technology startups, university labs, car companies, and even the Defense Department have begun to take on the challenge in earnest. Researchers are now working furiously to create devices that will skim waste heat on every scale, from the tiniest microprocessors to the largest power plants. Many of these devices will be powered not by boiling water to create steam, as was done in Edison's day, but by taking advantage of the simple fact that electrons will flow from a hot metal surface to a cold one. Connect the device to a wire and a new source of electricity is born.

"The potential is vast," says David Cahill, a professor of materials science and engineering at the University of Illinois at Urbana-Champaign. "What's really exciting about thermoelectrics" -- as the technology is called -- "is that it scales beautifully."

Along with wind turbines and solar panels, the next clean energy frontier could take the form of an array of tiny semiconductor chips wrapped around anything from a smokestack to a computer server, turning heat into usable energy. Imagine a laptop that runs twice as long, powered in part by the heat generated by the microprocessor. Researchers envision cell phones with double the talk time and cars that go farther on a gallon of gas -- all powered in part by heat that would otherwise be wasted.

The basic concept has been around for some time. NASA has been using thermoelectric devices to power deep-space probes for decades; even after 32 years, Voyager 1 is still beaming back radio signals from 10 billion miles away. During the oil embargo of the 1970s, when every kilowatt counted, the Department of Energy started doling out grants to companies and research labs trying to develop heat-scavenging devices for earthbound situations. A 1979 Popular Mechanics story on a related waste-heat capturing technology known as thermionic conversion declared: "Complex engineering problems remain. But there is considerable optimism that thermionic generators may one day play an important part in the country's energy picture."

However, the 1970s were followed by decades of cheap energy and, not surprisingly, a lack of interest in spending money on new ways to collect waste heat. But the cost of fossil fuels has risen sharply in recent years. Today, as the federal government aims to improve the nation's energy efficiency, the Department of Energy is one of the biggest funders of current thermoelectric research.

Before Matthew Scullin became the CEO of a thermoelectric startup called Alphabet Energy, he was a doctoral student at Lawrence Berkeley, where he helped create a device that his company aims to commercialize. Scullin's company is one of a dozen or so that have received government grants or private investments to improve the efficiency of thermoelectrics and bring down their cost by creating new materials, new manufacturing methods, or both.

"We have a core technology -- a chip -- that can be inserted into a wide variety of technologies and applications," Scullin says, adding that his goal is to make Alphabet Energy the "Intel of waste heat." Just as Intel makes silicon computer chips that are used by other companies in a wide variety of devices, Scullin wants to manufacture thermoelectric chips that could be used to recycle waste heat in anything from home appliances to airplane engines.

The concept is simple: a device made of a semiconducting material is sandwiched between two strong conductors. When one side gets hot, electrons begin to flow toward the cooler side. The semiconductor channels them into a wire, creating a flow of electricity that can be directed back into the original gadget, engine, or other heat-generating device, acting as an additional power source.

The first large-scale application of thermoelectric devices could emerge from the auto industry, where the push for higher fuel efficiency is driving car manufacturers to look for anything that can pull another mile or two out of each gallon of gas.

The Michigan-based company Amerigon has been honing its expertise in thermoelectric technology by making seat warmers for cars since the 1990s. Today it receives federal funding for its effort to produce electricity from the temperature differential between the hot gas in a car's exhaust pipe and the chilly fluids in the car's coolant system. That surplus electricity is enough to improve fuel efficiency by 2.5 miles per gallon, according to the Department of Energy's estimates.

Ford and BMW are scheduled to deliver test vehicles equipped with Amerigon's prototype thermoelectric generators to the National Renewable Energy Laboratory in Colorado later this year.

"People have been futzing with this for 180 years," says Dan Coker, Amerigon's CEO, hopeful that his own thermoelectric dreams will finally pay off.

image of Scott Dodd
Scott Dodd is the editor of and an adjunct professor at the Columbia University Graduate School of Journalism. He was a newspaper reporter for 12 years, contributing to coverage of Hurricane Katrina that won a Pulitzer Prize, and has writ... READ MORE >


I really hope this technology does not get hidden from the public. Is there any way in investing on this Comp. ?

Expert, it does not seem like it. The technology has been around for awhile, it just has not been scaled to a recognizable level. Thermocouples are based on this technology and are used in most labs to read temperatures.