Asking a broad spectrum to do a quantum job
 
 

Long wavelength problems

Short wavelength problems
 
 

Using more of the spectrum

Suppose, for example, that a photocell is designed for light of 650 nm wavelength.  When sunlight of  650 nm wavelength (in the red-orange part of the spectrum) strikes the cell, then it causes --- with 100% efficiency --- electrons to be released and to make the cell behave as a battery of 1.91 volts. 

But not all sunlight is of 650 nm wavelength.  None of the light of wavelength longer than 650 nm (redder) can release electrons.  The efficiency for all this too long wavelength is zero.

But there is also the part of the spectrum (orange, yellow, green, blue, violet) for which the wavelength is shorter than 650 nm.  All of this light can release electrons.  The trouble is that it has too much energy.  The excess energy merely heats the photocell, but doesn't produce a higher voltage.

Overall, therefore, the efficiency of the photocell is limited.  The limiting efficiency would depend upon the exact value of the critical wavelength (650 nm in this hypothetical example).  Real photocells available in large quantities are about 10% efficient.  (Furthermore, their efficiency drops as the years go by.)

Wouldn't it be nice to have a photocell that could be tuned to achieve the maximum possible efficiency?  Yes, but the principles remain the same.

It is better to have a multi-band-gap photocell.  The front part would be tuned to the yellow-green part of the spectrum.  That layer would be transparent to the longer wavelengths.  A second layer would be tuned to the red-orange part of the spectrum, and a third layer would react to the red and infrared parts of the spectrum.  Such a design would potentially have an overall efficiency of about 40%.

The promising material is InGaAsN, a crystalline solid made of indium, gallium, arsenic, and a touch of nitrogen.  It is the nitrogen that makes the news, because researchers at Sandia National Laboratories have found that they can make a tunable photocell merely be adjusting the amount of nitrogen in their photocells.

Roger Allan, "InGaAsN Looks Promising as Very Efficient Photovoltaic Power Source," Electronic Design, April 17, 2000, pp. 25-26.