Material soaks up the sun

By Kimberly Patch, Technology Research News

Capturing solar energy efficiently means finding materials that absorb as many wavelengths of light as possible that the sun sends our way.

Researchers from Lawrence Berkeley National Laboratory, the University of California at Berkeley and Cornell University have discovered that measurements of the semiconductor indium nitride taken two decades ago were wrong. The measurements led researchers to erroneously classify the material as a mediocre photovoltaic.

Instead, the material's band gap falls squarely in the solar spectrum, making it potentially more efficient than currently used photovoltaics. A material's band gap determines which wavelengths of photons can excite electrons in the material to create a flow of electricity.

The reclassified material promises to boost the efficiency of solar cells, allowing for smaller cells or cells of the same size that collect more electricity.

The best photovoltaic material currently available, a combination of gallium arsenide and gallium indium phosphide, has a theoretical efficiency of 32 percent; indium nitride has the potential to increase that to 50 percent, said Wladek Walukiewicz, a senior staff scientist at Lawrence Berkeley National Laboratory.

The researchers' initial testing also shows that the material withstands high energy particle irradiation without breaking down. This bodes well for satellite and other solar collectors that work in space.

When they made their surprising discovery, the researchers were investigating the mysterious lack of emissions from the material at the conventionally understood band gap of two electron volts. Instead of finding unusual reasons for the lack, they discovered that the material was simply misclassified, and instead has a band gap of 0.7 electron volts. "Once we looked at a lower energy range we could easily see all the features characteristic of [a] direct band-gap semiconductor," said Walukiewicz.

The misclassification took place at a time when samples were prepared by sputtering the material onto a surface. "Such samples contain large amounts of oxygen, as much as 30 percent. So the samples were indium-oxygen-nitride alloys rather than indium nitride," said Walukiewicz.

The more modern samples, in contrast, were grown by molecular beam epitaxy, a process that takes place in a vacuum. "Our material... contains undetectable levels of oxygen," he said.

A solar cell works by separating the positive and negative charges -- holes and electrons -- produced in a semiconductor unit in the when photons of sunlight hit it. The atomic structure of the semiconductors used in solar cells determines how many holes and electrons they can generate. The efficiency of a material has to do with microscopic measurements: the band gap, or spacing between the levels electrons occupy in a material, and the wavelengths of solar radiation.

Materials whose electron level spacing matches up with the wavelengths of the photons hitting it can absorb the photons. When a material absorbs a photon, some of the photon's energy causes an electron, which holds a negative charge, to change to a higher-energy position, leaving behind a positively-charged hole.

To separate these charges, the substance must be doped, or mixed with another substance to create separate paths for positive and negative charges. The separate types of semiconductor are n-type, which guides electrons, and p-type, which guides holes. "When an electron-hole pair is produced by a photon... the electron is pulled to the n-type side, and the hole is pulled to the p-type side," said Walukiewicz.

Because the charges naturally attract each other, the charge separation creates a change in electric potential, much like rolling a rock up a hill stores energy that can be released by simply starting it on the path downhill. In a solar circuit the potential stored by the separated charges can be released in a current of electricity.

The efficiency of a solar cell depends on how much of the total solar photon flux, or flow, can be converted into charge carriers, said Walukiewicz. "For a solar cell made of one semiconductor with a specific energy-gap, only the photons close to the absorption [range] contribute to the electric current," he said.

To increase the efficiency, however, solar cells can be made by stacking several cells of semiconductors with different band gaps to catch the different light waves. "Although each of the cells will convert the solar energy only from a limited range of photon energies, all of them together can make use of more photons," said Walukiewicz.

Indium nitride is important because its band gap sits squarely in the middle of the light spectrum; this makes it possible to produce gallium-indium-nitride semiconductors that have any gap within the solar spectrum range, and makes it possible to put more semiconductors in tandem, said Walukiewicz. "This is a solar-cell-designer paradise [because] one can maximize... performance by optimizing the number of cells and their band gaps," he said.

There's a lot of work to be done before practical solar cells can be made from indium nitride, however. The researchers have not yet made the p-type form of the material. "One of the biggest challenges is to make p-type doped indium nitride," said Walukiewicz. The indications are good, however. It is theoretically easier to make p-type doped indium nitride than to do the same with gallium nitride, which has already been done, he said. Gallium nitride is also a direct band-gap material.

The researchers' next step is to make p-type indium nitride. They are also working to make p-type gallium indium nitride, he said. And they are more thoroughly testing the properties of the two materials under high-energy particle irradiation, he said.

The researchers have only tested a few samples, said Cheng Hsiao Wu, a professor of electrical and computer engineering at the University of Missouri at Rolla. The reasons for the measurements are not yet clear; there could be a mechanism involved other than a different band gap, he said.

Given the benefit of the doubt, however, a 0.7 reading could be useful, but finding such a material is a long way from making a solar cell, said Wu. "For actual solar applications, either the current or the voltage of each solar cell using a particular [part of the] solar spectrum has to be matched," and this is a particularly tricky proposition for a full-spectrum solar cell, he said. If the material works out, however, it "may add another variety of the solar cell we already have for the full visible spectrum range," he said.

It will take three to four years to develop indium-nitride-based solar cell technology, said Walukiewicz.

Walukiewicz's research colleagues were Junqiao Wu and Eugene E. Haller from the University of California at Berkeley and Lawrence Berkeley National Laboratory, W. Shan, Kin Man Yu and Joel W. Ager III from Lawrence Berkeley National Laboratory, and Hai Lu and William J. Schaff from Cornell University. They published the research in the November 15, 2002 issue of Physical Review B. The research was funded by the Department of Energy.

Timeline:   3-4 years
Funding:   Government
TRN Categories:  Semiconductors; Materials Science and Engineering
Story Type:   News
Related Elements:  Technical paper, "Effects of the Narrow Band Gap on the Properties of InN," Physical Review B, November 15, 2002.




Advertisements:



December 11-25, 2002

Page One

DNA prefers diamond

Material soaks up the sun

Design links quantum bits

Microscopic mix strengthens magnet

Laser pulses could speed memory

News:

Research News Roundup
Research Watch blog

Features:
View from the High Ground Q&A
How It Works

RSS Feeds:
News  | Blog  | Books 



Ad links:
Buy an ad link

Advertisements:







Ad links: Clear History

Buy an ad link

 
Home     Archive     Resources    Feeds     Offline Publications     Glossary
TRN Finder     Research Dir.    Events Dir.      Researchers     Bookshelf
   Contribute      Under Development     T-shirts etc.     Classifieds
Forum    Comments    Feedback     About TRN


© Copyright Technology Research News, LLC 2000-2006. All rights reserved.