Cadmium-Telluride-on-Silicon, a Novel Approach in Solar

A company in Bolingbrook, Ill., is developing cadmium-telluride cells with a twist.

Using cadmium and tellurium to create solar cells is nothing new. First Solar is the best-known producer of it, and a growing number of startups have followed suit.

But EPIR Technologies, which started in 1998 to develop infrared imaging materials, is coupling cadmium-telluride with silicon. It's an unusual approach, said researchers who are not involved in the project.

EPIR isn't looking at producing cells that would be assembled into solar panels. Instead, the company wants to use its cells for a system that concentrates the sunlight onto the cells to boost their electricity production.

The company's founder, Siva Sivananthan, believes he could produce cells with efficiencies approaching 40 percent (that would include the use of a concentrator). A cell's efficiency refers to the rate at which it could convert sunlight into electricity.

"We are excited about the technology. This is probably the only tech we have a chance for grid parity," Sivananthan said.

The company recently won a grant from the U.S. Department Energy to help it develop the manufacturing process that would make producing the cells commercially possible. The company is still negotiating with the DOE on how much it would get – it's  eligible to receive up to $500,000.

In April this year, it announced it had won a $9 million contract with the U.S. Department of Defense to produce materials for making the solar cells.

EPIR recently signed a five-year, co-development agreement with the National Renewable Energy Laboratory to enlist its researchers' help in improving the solar cell's designs, said David Christensen, senior licensing executive at NREL.

Sivanathan is setting out to mass-produce solar cells based on his research at the University of Illinois at Chicago and the infrared imaging work he has been doing through EPIR. 

Still a professor there, he has figured out a way to grow single-crystal cadmium-telluride compound and a layer of silicon. This is difficult to do. Different semiconductors have different spacing between atoms, so placing them one atop of another could lead to what's called lattice mismatch, said Jud Ready, senior research engineer at Georgia Institute of Technology, where Ready is working on growing carbon nanotubes coated with cadmium-telluride to trap light. That could lead to the degradation of a cell's efficiency. 

Growing cadmium-telluride in a single-crystal form is harder than creating the multicrystal structure, but it promises a higher efficiency, Sivananthan said.

"I pioneered the single crystal cadmium-telluride. Anybody is doing it is either a former employee of mine or works with me," Sivananthan added. 

The only other research group carrying out similar research is at the Shanghai Institute of Technology, which is looking at using the science for infrared applications, not solar, he said.

Other researchers agreed that the breakthrough forms a good starting point to figure out a way to produce the solar cells en masse.

"He's come up with a way grow the materials on top of each other, which is quite unique and novel," said Cyrus Wadia, a researcher at the Lawrence Berkeley National Laboratory. "He probably is doing this under a specific condition. That often has a problem translating into a scalable manufacturable product."

Wadia questioned whether Sivananthan could produce cells with much higher efficiencies and at lower costs than those already available for concentrating PV systems today.

Sivananthan said his formulation includes other materials from the same II-VI family of semiconductors that include cadmium-telluride as well. But he wanted to keep that extra something close to the vest for now. He declined to disclose what kind of cell efficiency he's been able to achieve. 

In general, compounds from the II-VI family have properties that make it possible for them to generate high-voltage current, he said. Silicon, meanwhile, would serve as the substrate partly because its price is falling quickly, Sivanathan added.

Silicon is cheaper than germanium, one of the materials that can be found in solar cells for concentrating PV systems today.

Germanium and those other materials typically come from the III-V semiconductor family. They are more efficient than silicon or cadmium-telluride cells, but they also are more expensive (see a list of III-V cells being developed by different companies).

By using mirrors and lenses to concentrate the sunlight, a solar energy system would only need slivers of those expensive cells. Some of those cells, originally developed for satellites, could have around 30 percent efficiency without the use of a concentrator. With one, the efficiency could advance into the mid to upper 30s, depending on the amount of concentration.

In labs, the efficiency could reach a little more than 40 percent with a concentrator that magnifies the sunlight hundreds of times.

Those cells are typically triple-junction ones, so there are three layers of materials for converting light to power.

Commercial cadmium-telluride solar cells have efficiencies in the mid teens. The best silicon cells on the market today have 22 percent efficiency. EPIR's first solar cell would be dual junction, and it would be designed to withstand a concentration of 200 to 500 suns, Sivananthan said.

The 500-sun goal is ambitious, Ready said.

"As any kid who has burned up ants knows that with a magnifying glass, it burns up a lot of heat," Ready said. "So the same heat that fries your ants can fry your cells as well. Coming up with a way not to do that is a key challenge."

The concentrating photovoltaic approach drew a lot of investors' attention a few years back when silicon, the main material for the most common type of solar panels, was scarce and expensive.

Dozens of companies in the United States and Europe are working on concentrating PV technologies, which are mostly in the pre-commercialization stages (see list 1 and 2 of the companies).

EPIR will be working closely with another company, Sunovia Energy Technologies, to commercialize the solar cells. Sarasota, Fla.-based Sunovia, which also sells LED lighting and infrared detectors, plans to finance and build concentrating PV systems using EPIR's cells and develop solar farms for customers, said Craig Hall, Sunovia's co-founder. EPIR has invested in Sunovia and vice versa, Sivananthan said.

EPIR already has a 5-megawatt pilot production line, Sivananthan said. The plan is to set up a 100-megawatt factory in 2012, he added.

Sunovia, meanwhile, has signed a deal to build a 100-megawatt solar farm for Cybernetic Park, a business park in the Dominican Republic, Hall said. Sunovia plans to build the project over the next several years. The project would cost about $200 million, money that the company still has to raise, Hall said. 

---

Join experts and influencers at Greentech Media's Growth Opportunities in the New PV Market: Projects, Finance and Policy in San Francisco on July 13.