The third way
A new method of making electricity from sunlight has just been tested
AT THE moment, there are two reliable ways to make electricity from sunlight. You can use a panel of solar cells to create the current directly, by liberating electrons from a semiconducting material such as silicon. Or you can concentrate the sun’s rays using mirrors, boil water with them, and employ the steam to drive a generator.
Both work. But both are expensive. Gang Chen of the Massachusetts Institute of Technology and Zhifeng Ren of Boston College therefore propose, in a paper in Nature Materials, an alternative. They suggest that a phenomenon called the thermoelectric effect might be used instead—and they have built a prototype to show that the idea is practical.
Thermoelectric devices are not new. They are used, for example, to capture waste heat from car engines. They work because certain materials, such as bismuth telluride, generate an electrical potential difference within themselves if one part is hotter than another. That can be used to drive a current through an external circuit.
The reason thermoelectric materials have not, in the past, been applied successfully to the question of solar power is that to get a worthwhile current you have to have a significant temperature difference. (200??C is considered a good starting point.) In a car engine, that is easy. For sunlight, however, it means concentrating the heat in some way. And if you are going to the trouble of building mirrors to do that, you might as well go down the steam-generation route, which is a much more efficient way of producing electricity. If the heat concentration could be done without all the paraphernalia of mirrors, though, thermoelectricity’s inefficiency would be offset by the cheapness of the kit. And that is the direction in which Dr Chen and Dr Ren hope they are heading.
In their view, three things are needed to create a workable solar-thermoelectric device. The first is to make sure that most of the sunlight which falls on it is absorbed, rather than being reflected. The second is to choose a thermoelectric material which conducts heat badly (so that different parts remain at different temperatures) but electricity well. The third is to be certain that the temperature gradient which that badly conducting material creates is not frittered away by poor design.
The two researchers overcame these challenges through clever engineering. The first they dealt with by coating the top of the device with oxides of hafnium, molybdenum and titanium, in layers about 100 nanometres thick. These layers acted like the anti-reflective coatings on spectacle lenses and caused almost all the sunlight falling on the device to be absorbed.
The second desideratum, of low thermal and high electrical conductivity, was achieved by dividing the bismuth telluride into pellets a few nanometres across. That does not affect their electrical conductivity, but nanoscale particles like this are known to scatter and obstruct the passage of heat through imperfectly understood quantum-mechanical processes.
The third objective, efficient design, involved sandwiching the nanostructured bismuth telluride between two copper plates and then enclosing the upper plate (the one coated with the light-absorbing oxides) and the bismuth telluride in a vacuum. The copper plates conducted heat rapidly to and from the bismuth telluride, thus maintaining the temperature difference. The vacuum stopped the apparatus losing heat by convection.
The upshot was a device that converts 4.6% of incident sunlight into electricity. That is not great compared with the 20% and more achieved by a silicon-based solar cell, the 40% managed by a solar-thermal turbine, or even the 18-20% of one of the new generation of cheap and cheerful thin-film solar cells. But it is enough, Dr Chen reckons, for the process to be worth considering for mass production.
He sees it, in particular, as something that could be built into the solar water-heaters that adorn the roofs of an increasing number of houses. If such heaters were covered with thermoelectric generators the sun’s rays could be put to sequential use. First, electric power would be extracted from them. Then, the exhaust heat from the bottom plate of the thermoelectric device would be used in the traditional way to warm water up. Two-for-one has always been an attractive proposition for the consumer. This kind of combined heat and power might enable more people to declare independence from the grid.
在他看来，尤其热电装置要是被建造成为装饰越来越多的房屋的屋顶太阳能热水器的一部分。 如果这种加热器都安装上能付诸连续使用太阳光热电发电机，那么首先，电能能够从中提取。 然后，从热电装置底板的排气将会用传统的方式加热温水热。 这样合二为一，总是对消费者的主张有吸引力， 这种热电联产可能会使更多的人从电网宣布独立。