Novel semiconductor alloy can capture near-infrared light

In what could be a major step forward for the development of a new generation of solar cells called ‘concentrator photovoltaics’, researchers at the University of Michigan (U-M) have created a new semiconductor alloy that can capture near-infrared light.

Easier to manufacture and at least 25% less costly than previous formulations, the alloy is believed to be the world's most cost-effective material for capturing near-infrared light – and is also compatible with the gallium arsenide semiconductors often used in concentrator photovoltaics.

Concentrator photovoltaics gather and focus sunlight onto small, high-efficiency solar cells made of gallium arsenide or germanium semiconductors. They're on track to achieve efficiency rates of over 50%, while conventional flat-panel silicon solar cells top out at around 25%.

"Flat-panel silicon is basically maxed out in terms of efficiency," said Rachel Goldman, U-M professor of materials science and engineering, and physics, whose lab developed the alloy. "The cost of silicon isn't going down and efficiency isn't going up. Concentrator photovoltaics could power the next generation."

Varieties of concentrator photovoltaics exist today; they are made of three different semiconductor alloys layered together. Sprayed onto a semiconductor wafer in a process called molecular-beam epitaxy – a bit like spray painting with individual elements – each layer is only a few micrometers thick. The layers capture different parts of the solar spectrum; light that gets through one layer is captured by the next.

But near-infrared light can slip through the layers unharnessed. For years, researchers have been working toward an elusive ‘fourth layer’ alloy that could be sandwiched into these solar cells to capture near-infrared light. It's a tall order, though, because the alloy must be inexpensive, stable, durable and sensitive to infrared light, with an atomic structure that matches the other three layers in the solar cell. Getting all those variables right isn't easy, and until now the only options have been prohibitively expensive formulas that use five elements or more, including arsenic and bismuth.

To find a simpler mix, Goldman's team devised a novel approach for keeping tabs on the many variables in the process. They combined on-the-ground measurement methods, including X-ray diffraction at U-M and ion beam analysis at Los Alamos National Laboratory, with custom-built computer modeling.

Using this method, they discovered that a slightly different type of arsenic molecule would pair more effectively with the bismuth. They were also able to tweak the amount of nitrogen and bismuth in the mix, allowing them to eliminate an additional manufacturing step that previous formulas required. And they found precisely the right temperature that would enable the elements to mix smoothly and stick to the substrate securely.

"'Magic' is not a word we use often as materials scientists," Goldman said. "But that's what it felt like when we finally got it right."

This latest advance, which is reported in a paper in Applied Physics Letters, comes on the heels of another innovation from Goldman's lab that simplifies the ‘doping’ process used to tweak the electrical properties of the chemical layers in gallium arsenide semiconductors. This was also reported in a paper in Applied Physics Letters.

During doping, manufacturers apply a mix of chemicals called ‘designer impurities’ to change how semiconductors conduct electricity, and to give them a positive and negative polarity similar to the electrodes of a battery. The doping agents usually used in gallium arsenide semiconductors are silicon on the negative side and beryllium on the positive side.

The beryllium is a problem, though: it's toxic, costs about 10 times more than silicon dopants and is sensitive to heat, which limits flexibility during the manufacturing process. The U-M team discovered that by reducing the amount of arsenic below levels that were previously considered acceptable, they could ‘flip’ the polarity of the silicon dopants, allowing them to use the cheaper, safer element for both the positive and negative sides.

"Being able to change the polarity of the carrier is kind of like atomic 'ambidexterity,'" explained Richard Field, a former U-M doctoral student who worked on the project. "Just like people with naturally born ambidexterity, it's fairly uncommon to find atomic impurities with this ability."

Together, the improved doping process and the new alloy could make the semiconductors used in concentrator photovoltaics as much as 30% cheaper to produce. That would be a big step toward making these high-efficiency cells practical for large-scale electricity generation.

"Essentially, this enables us to make these semiconductors with fewer atomic spray cans, and each can is significantly less expensive," Goldman said. "In the manufacturing world, that kind of simplification is very significant. These new alloys and dopants are also more stable, which gives makers more flexibility as the semiconductors move through the manufacturing process."

This story is adapted from material from the University of Michigan, with editorial changes made by Materials Today. The views expressed in this article do not necessarily represent those of Elsevier. Link to original source.