Online information source for semiconductor professionals

Solar start-up Suniva sees plenty of efficiency, cost innovations left in crystalline silicon PV

21 August 2008 | By Tom Cheyney | Chip Shots

Popular articles

Oberai discusses Magma’s move into solar PV yield management space - 29 August 2008

New Product: Applied Materials new EUV reticle etch system provides nanometer-level accuracy - 19 September 2011

‚??Velocity‚?? the new buzzword in Intel‚??s PQS annual awards - 12 April 2012

Applied Materials adds Jim Rogers to Board of Directors - 29 April 2008

TSMC honors suppliers at annual Supply Chain Management Forum - 03 December 2008

Spun out of the Georgia Institute of Technology's Center of Excellence in Photovoltaics (UCEP) last year, Suniva has emerged as one of the most intriguing start-ups in the solar PV space. Over the past six months, the Atlanta-based company has raised $50 million in Series B funds; started to build, outfit, and ramp its first solar-cell production line; and struck big-ticket, nine-figure, multiyear deals with its supply chain (REC for wafers) and customer base (Solon for modules).

Suniva's goal is not to bring yet another "disruptive" thin-film or concentrator PV technology to market, but to harness the high-efficiency, low-cost potential of solar's workhorse starting material--crystalline silicon.

Company founder and CTO Ajeet Rohatgi has been working on PV since his days investigating the effects of impurities on conversion efficiencies at Westinghouse some 30 years ago, but he is best known for founding and building the Georgia Tech UCEP since its inception in 1985 and naming as a center of excellence by the Department of Energy in 1992--the first of its kind in the country. The center has become one of the preeminent silicon solar research institutions in the world, with expertise all the way up the value chain from materials to systems--and everything in between.

While the research and educational components are essential parts of UCEP's efforts, the most important piece is industry collaboration, Rohatgi told me during an interview at the PV Summit in June, "since every silicon company in the United States has worked with us and is working with us. All of the novelty of the industry was going through Georgia Tech."

After all those years of research, some in the investment community thought it was time for Rohatgi to make a move into the commercial realm. "The thing that we do best is what I call very-high-efficiency cells at low cost, that's what we decided to spin out," he explained. "Our investors came and said, 'Ajeet, you've been helping all these companies, making them rich, why don't you do something for yourself?' I wasn't quite ready to do that but they kind of brainwashed me. I got convinced. It's exciting to do this because it focuses you a lot more than when you are just doing research."

Since the company springs from the substantial scientific prowess and know-how of Georgia Tech's UCEP, Suniva has a built-in advantage over many solar start-ups. "Because of this industry collaboration, we've been able to develop all the right tools over the past twenty-some years. We have so many different technologies in the center.

"If you set up a line, normally you're locked into one kind of technology, you can do diffusion by belt or tube, for example. We can do diffusion by solid source, liquid source, belt, tube, rapid thermal processing, because we've been doing research, we can do so many different things in our lab, which is very rare. We can do plated contacts, screen-printed contacts, evaporated contacts, inkjet contacts. That's what makes our program so powerful, plus the twenty-some years of knowledge base."

"That's what I call our core competitive advantage," he continued, "because most start-ups, they just get a turnkey line and then don't have the luxury of doing high-level R&D because if you stop the line to do R&D, you lose money. This is what is very unique about Suniva. Not only do you have R&D, but it's R&D of the caliber and experience that bigger companies like GE and Dow Chemical have. So we have done this thing for years, so the depth of R&D is so huge for a small start-up. I think that is our biggest advantage.

"You can see how quickly right now we're heading to 20% conversion efficiency. If we had just started the company like any other company, where we are today might have taken three, four years. But because this thing was already attached, and now there is more push and focus, and the results are coming in even much faster, and we are hoping that this will be what differentiates us from any other start-up company."

The company will use what Rohatgi calls "improved versions of our best cell technology...without too much tinkering" in the first manufacturing line, "so we can get the production experience." The first tools started arriving at the factory in July, with more equipment coming in and the ramp-up continuing over the next few months.

Sales and marketing VP Bryan Ashley told me via a recent email that "Suniva has a contract with Centrotherm Photovoltaics, [which] has supplied us with our 32-megawatt line. Deliveries have begun and we're still expecting production to begin in Q4 of this year." The line was originally specified at 25-MW, but the start-up's team "has made improvements that enable us to have an initial capacity of 32 MW." A workforce of about 100 will be running that 32-meg factory (at 4 watts per cell, that's about 8 million cells), with recent hires including a plant manager and quality director, he added.

The company had several phases of capacity additions planned, even before signing the $500 million deal with Solon to supply the module-maker with Suniva cells.

"As one of the highest quality solar module manufacturers in the world, Solon is a strong industry partner for Suniva," explained Ashley, "and our supply contract with Solon through 2012 is an important milestone in our growth strategy. As we continue to gain more customers, we're planning additional lines to expand capacity to more than 160 MW over the next two years. We had made this expansion decision before the Solon agreement was signed." Deliveries of the cells are "set to begin...in 2009, although smaller shipments will begin late this year."

So what makes Suniva's cells so special, other than benefiting from all those years of Ramblin' Wreck research at Georgia Tech? As Rohatgi explained during his presentation at the PV Summit, it is a combination of their relatively simple technology and architecture with a focus on squeezing more out of low-cost process steps like screen printing to drive efficiencies up to 20% and beyond and manufacturing costs down in the range of a dollar per watt.

He showed a technology roadmap for reaching 20% efficiency screen-printed cSi solar cells. Rohatgi told me he thinks 22% may be possible--NREL has already confirmed 18.5% and 19.9% efficiencies for Suniva cells.

Starting with an industrial cell with 14.4% efficiency, the seven steps to better conversion include improving the fill factor of the front contact to 0.78, increasing the bulk lifetime to 100 microseconds, tweaking the back-surface reflector to get better light trapping and passivation (this step supplies the biggest efficiency pop, at 2%), reducing the cell thickness to 100 microns, making the emitter more selective to improve current collection, enhancing light-trapping through better surface texturing (the second biggest pop, at 1.4%), and reducing resistivity to 0.6 ohms-cm.

One key is developing the right dielectric for the passivation steps on the back contact, which provides a significant voltage enhancement, according to Rohatgi, and is crucial to the architectures of the company's next-generation Star cell technologies. The beta version of Star features a passivated boron capping layer on top of the dielectric layer, accomplished through the use of an oxidation/diffusion process that he said can put the boron and phosphorus down in one step.

When I asked him about the impact of better wafer quality on Suniva cells' conversion efficiencies, his answer was not quite what I expected. "We are constantly looking at quality versus performance [of wafers], and this is another strength that we have. We can do very solid modeling to say what quality we need for what efficiency. What we see is as you go to thinner wafers, the quality relaxes, although it's difficult to give a number, because it's a function of thickness. In other words, as you go to thinner and thinner wafers, you can live with lower and lower quality.

"It's a function of your surfaces, so if you have your dielectric passivation done well, it allows you to go much thinner before you start losing it. If my surface is bad, the moment I go to a thin wafer my performance dies. But once you have done all these things, you have the luxury of thinning the wafer down. You will trap the light in it, so it is like the silicon is still thick, the light is bouncing and hitting the surfaces, and it is not hurting you.

"Once you get the surfaces dielectric and all, you can afford to lower the material quality with very little loss in cell performance. That is where I think we could maybe down the road live with cheaper silicon, thinner wafers, and still get the same performance and without going to exotic cell designs. If you can afford to go thinner, you can afford to go to lower quality and all these will lead to lower [feedstock] costs too."

Rohatgi knows that screen-printed cells will eventually run out of gas in terms of driving further up the conversion efficiency curve toward silicon's theoretical limits of the mid to high 20s, but he doesn't seemed too concerned.

"We are looking at different technologies, ways to make contacts other than screen printing....This is the beauty of having R&D that is so deep. I can do in parallel all kinds of things at Georgia Tech, such as plating, inkjet printing, etc. So we are looking at alternate ways of making contacts, because the screen printing is nice and cheap but it has its own downsides. Because you have to fire the contacts, the dielectric has to be done right, otherwise firing will destroy the dielectric."

Despite his new commitment to the commercial realities of getting Suniva off the ground and running, Rohatgi doesn't stray all that far from the research center he has nurtured all these years. "I keep busy with both UCEP at Georgia Tech and at Suniva. Fortunately, they are very much interrelated--most of my research at UCEP directly benefits Suniva, and vice versa."

If Suniva lives up to its low-cost, silicon-innovating potential, the good professor/CTO's research efforts will also directly benefit the growing global community of solar PV providers and consumers.

Related articles

Applied Materials guides 50% plus increase in sales for FY2010 - 17 February 2010

Applied Materials silicon systems group sees strong growth; Q3 results hit by thin-film PV writedown - 19 August 2010

While First Solar keeps on trucking, others in CdTe thin-film PV pack keep on muddling - 21 August 2008

TSMC creates US$50 solar investment fund - 11 August 2009

Sunovia Energy Technologies and EPIR Technologies Announce Second DOE Contract - 23 July 2009

Reader comments

Thank you for this site which I just found. It is very useful and informative.
By Silk Screennow on 04 October 2010

Post your comment

Name:
Email:
Please enter the word you see in the image below: