Exploiting nature's secret stash

We are, by most accounts, a wasteful bunch. If you are sitting in an office, the chances are that most of the energy produced by the heaters is disappearing outside rather than keeping you warm. If you have your computer switched on, it could be frittering four-fifths of its maximum power consumption simply doing nothing. Just by reading to the end of this article you could expend around 100 kJ of energy – though beware that doing anything else will almost certainly be a bigger drain.

Most people rarely acknowledge the energy that is lost to the environment. Rather than chase it, we have opted to generate more. The past 35 years have seen a tripling of world electricity demand, which has predominantly been met through the burning of additional fossil fuels. With less than a 3% contribution to global energy production, those technologies that exploit what is already around us – solar, wind and tidal power, for example – barely feature.

Recently, however, and partly in response to the drive for lower carbon footprints, scientists have started to rethink the possibilities for salvaging wasted energy – what they term "energy harvesting". They are finding ways to capture the heat lost from buildings, the idle power drain of computers or the motions of the human body, and to convert it into useful, clean electricity. But that is not all: energy harvesting permits devices that are free from a battery or an external power source. With these restrictions removed, the devices can be placed or moved anywhere, scaled to dimensions smaller than ever and designed to exist without maintenance. Bending power Zhong Lin Wang of the Georgia Institute of Technology, GA, focuses on nanoscale devices. He works with ZnO, one of a handful of compound semiconductors that exhibit piezoelectricity – generation of an electric potential when the material is put under stress. In 2000 Wang realized that he could use a vapor-liquid-solid technique to grow ZnO cheaply and simply. His version of the technique involves patterning specks of a catalyst, such as gold, on a substrate. A variety of elaborate ZnO nanowire structures are then created by varying the flow rate and temperature of a ZnO-graphite vapor that impinges on the patterned substrate.

The trick to harvesting energy with ZnO nanowires is to put them in contact with a metal. This creates a Schottky barrier – essentially a diode – which allows electric current to flow in only a single direction. Bending the nanowires in one direction drives charge into the metal; bending the other way means that the charge cannot flow back. Wang calls this current generation process "piezotronics" and he is already developing it for several potential devices via his company, Piezodyne.

Devices that Wang has in mind include nanowire energy harvesters that could be built into the flexible soles of shoes to charge a mobile phone while walking. Or, in a more advanced system, they could be woven into a fabric so that every movement generates electricity. He has been working on this topic for four years: "I believe that in about five years you will see something utilize this technology, in areas like implantable medical devices, mechanical sensors, or powering personal electronics."

The option to use Wang s piezotronics inside the human body separates them from existing piezoelectric energy harvesters. For example, passengers entering the Yaesu end of Tokyo station over the Christmas period might have noticed a display light-up with 1 mW of power each time a traveller walked through a turnstile. But this gadget, like most other piezoelectric energy harvesters, is based on the ceramic lead-zirconate-titinate (PZT) – a toxic substance. In comparison, ZnO, a standard ingredient in sun cream because of its light-absorbing properties, is benign. "If you use PZT in your body," joked Wang, "you re in trouble." Taking the heat Piezoelectrics are not the only energy-harvesting technology to benefit from the nano treatment. Scaling down dimensions can also increase thermoelectricity, the charge flow that results from a heat gradient across a substance, which is most significant in materials that conduct electricity well but heat poorly. In the 1920s researchers found that they could improve thermoelectrics by alloying them with certain metals to introduce defects, which reduce thermal conductivity while maintaining electrical conductivity, but this only took the materials efficacy so far. Last year, however, a group led by Zhifeng Ren of Boston College, MA, showed that it is possible to boost efficacy by powdering thermoelectrics into nanoparticles.

Ren s group works with the compound semiconductor Bi2Te3, which is the most widely used thermoelectric. Traditionally researchers melt Bi2Te3 and then cool it slowly, in such a way that it crystallizes in a preferred direction. Ren s group skips this stage and instead grinds the material into particles 20 nm across before hot-pressing them into ingots. "The whole process is a lot cheaper, a lot faster and uses less energy," said Ren.

These nanocrystalline ingots have a lower thermal conductivity than single, large crystals. This is because phonons – packets of vibrational energy – scatter at each of the numerous crystal grain boundaries, slowing heat propagation. The result is a peak efficacy of 1.4, some 40% better than researchers had previously achieved. What s more, the peak occurs at 120 °C, a higher temperature than other thermoelectrics, which makes this material more valuable for harvesting energy from hot environments.

In fact, almost any hot environment could be suitable for thermoelectrics: chimneys, steam turbines, computer circuits, the human body, even panels that have been placed outside to warm in the sun. Although it will be later this year before Ren s start-up company, GMZ Energy, produces the first prototypes of its energy harvesters, other firms employing the less-efficient single Bi2Te3 crystals have products on the market. Thermo Life, based in the US, manufactures 9 mm diameter thermoelectric devices that can generate up to 100 µW for a 10 °C temperature gradient, while Micropelt, based in Germany, is selling sample devices half the size that can generate 1500 µW with the same temperature gradient. For more extreme environments, the Japanese electronics firm Furukawa is developing 5 cm2 thermoelectric devices based on skutterudite, a cobalt arsenide mineral, which are claimed to generate 33 W when placed beside a vehicle s 720 °C exhaust pipe. "Potential applications are everywhere," said Ren. "And [the technology] is solid-state; there s no moving parts. So the lifetime is long."

The applications of thermoelectrics overlap another heat-exploiting technology – thermophotovoltaics (TPVs). Like conventional photovoltaics, TPVs convert electromagnetic energy into electrical energy, but they differ by focusing on the infrared part of the electromagnetic spectrum. In practice this means that TPVs need another layer in front of the photovoltaic cells, called a thermal emitter, which radiates infrared photons when it gets hot.

Many TPV companies are still in the development stage. Last year, at the end of a three-year project funded by the UK s Technology Strategy Board and the EPSRC (Engineering and Physical Sciences Research Council), a UK consortium led by CIP Technologies announced that it had fabricated a device based on InP – a departure from the usual GaSb – with an efficiency of 12%, breaking the record of 9%. The consortium is now hoping to win more funding to develop a multi-junction version, using InP, which has efficiencies as high as 25%. This could be used for collecting stray heat from domestic boilers, furnaces and glass manufacturing plants.

Meanwhile, MTPV Corporation, a start-up based in Boston, MA, believes that it can boost TPV efficiencies to 50% or more by placing the thermal emitter very close to the cell. The goal is to separate these two by less than the wavelength of the infrared light, because this would prevent any photons from being reflected back. Another TPV specialist, US-based Spire Semiconductor, is developing an InGaAs cell that can create electricity from the lasting warmth of radioactive isotopes. The project has being running since 2006 and is backed by $600,000 of funding from NASA.

For at least one company, energy harvesting is already proving a highly worthwhile venture. Starting in 2001, the Siemens spin-off EnOcean, which is based in Oberhaching, Germany, commercialized a range of self-powered wireless light switches based on PZT piezoelectric technology, and later on dynamo (electromagnetic coil) technology. One of the advantages of such switches is that they can be moved between a building s dividing walls without rewiring, which can cut costs by as much as 80%. In 2005, the famous Semperoper opera house in Dresden commissioned EnOcean to install its lighting because it refused to have cable routes drilled into its 19th-century Baroque ballroom. The company has already sold more than 50,000 energy-harvesting devices, with revenue doubling every year.

Not everyone has a winning story to tell, though. Lewis Fraas founded JX Crystals, a Boeing spin-out, in 1992, which pioneered the idea of generating electricity using TPVs wrapped around flame burners. They netted contracts from NASA, the army and the navy for different GaSb TPV devices. But there were issues integrating the TPVs into effective systems, and by 2001 investment had dried up. Now JX Crystals survives through silicon photovoltaics contracts, mostly from China, and by selling GaSb cells to various research labs across the world.

Fraas thinks that the problem lies with a lack of investment, which is preventing a fall in compound semiconductor manufacturing costs. "It s always an investment, it s not a technology problem," he said. "Investment for scaling up infrared cell supply, or investment in scaling up for lower cost."

It is undeniable that interest in energy harvesting is mounting. An idea that arguably dates back millennia to the invention of the waterwheel, it is only reawakening recently in light of new technology. But it remains to be seen whether any of the nascent ventures will find it too difficult to reap rewards from energy harvesting or whether, like EnOcean, they will find it as easy as switching on a light.   

View pdf of article