Something to reflect on
Nanotechnology largely eliminates reflection waste from solar cells
The average solar cell that you might see on the roof of a house has an efficiency of about 10 to 15%. That is, only about one in ten of the photons of sunlight striking it are converted into electrons of usable electricity. To a large extent this limitation is set by the inherent properties of the silicon that such cells are made of. In recent years scientists have been looking at ways of improving this situation by using other semiconductors to create cells and even sandwiches of several materials each able to absorb a part of the solar spectrum that the ones above can’t.
One class of semiconductor frequently used in efficient solar cells is the III-Vs. Compounds like gallium arsenide GaAs, that have one atom from group three of the periodic table and one from group five. Using III-V cells, efficiencies of over 40% are possible when the cells are coupled with external optical concentrators.
However even III-V cells are limited by two fundamental physical processes. One is reflection from the surface. Semiconductors have very high refractive indices which means that incident light is reflected far more strongly than it would be from glass or plastic. As much as 30% of the sunlight can be lost in this way.
The second problem relates to the junction. Solar cells are made from junctions between an n-type semiconductor in which electrons are the predominant carrier of electricity and a different version of the same material in which holes carry the charge – so called p-type material. The physics of the p-n junction dictate that there are essentially no charge carriers at all in the junction region which can range from nanometres to microns in size. When a photon of sunlight hits the junction an electron and a hole are created which rapidly migrate to the n and p type material respectively thus creating a current in the external circuit.
The problem for a cell designer is that there are competing requirements in the size of this active junction. The wider the junction the more photons will be absorbed in it. But a wide junction also means a long journey for the electrons holes and a greatly increased chance that they will recombine with each other inside the junction yielding no external current.
However recent work at the Australia National University may be set to change the rules on how cells are made by making clever use of nanotechnology. Associate Professor Hoe Tan leads a group specialising in the growth of exotic structures in III-V semiconductors. “We’ve been adapting nanowire growth technology to produce solar cells.” Professor Tan says, “Because the physical properties of nanowires should enable us to solve the twin problems of reflection and junction absorption.”
Essentially the nanowire cell consists of countless ultra thin projections from the surface of a conventional semiconductor wafer almost like fur on animal skin. Each wire is several µm long but only a few 10’s of nanometres wide. The core of the wire can be grown p type GaAs whilst the cladding n type so in effect each wire is a coaxial p-n junction.
The light trapping properties of the nanowire arrays reduces reflection to a tiny fraction of that from a solid chunk of GaAs with light reflected from one wire being absorbed by one of its neighbours. Likewise the microscopic coaxial junction leaves very little room for recombination loss since electrons or holes have to travel only a few 10’s of nm to the contacts.
“Making nanowire junctions isn’t the same as making conventional solar cells though,” Professor Tan explains, “once we have the nano structure there are quite a few steps involved in creating the final cell.”
The fine “fur” of nanowires is mechanically delicate so it has to be planarised – that is turned into a single solid mass by the addition of a polymer that fills the caps between the wires. The polymer is plasma etched back to expose the ends of the wires so that a transparent electrode can be applied to allow transmission of sunlight to the nanowires and draw off the electric current.
“We’ve been experimenting with several polymers some of which even enable us to peel off the nanowire cell layer right off the underlying wafer.” Professor Tan says, “So in effect what we are creating is a flexible nanowire solar cell that you can wear.”
In their flexible or rigid forms these new cells are attracting lots of interest from those to whom efficiency really matters such as space engineers and designers of large scale solar concentrator power farms. “If you are building a concentrator system chances are you’re investing a lot of money, “Professor Tan says, “so generally you’re going to want to use the most efficient cells at the focal point of the concentrator even if they’re a bit more expensive than silicon cells. Additionally, III-V semiconductor materials are able to withstand much higher solar concentration ratio than silicon.”