03 Aug 2011
Vahan Garboushian on a breakthrough year for concentrated PV, and why the technology promises to be much more than a niche solution.
Cast your mind back to 1989. Aside from the Berlin Wall being torn down, the very first GPS satellite has just been placed in orbit, Voyager II is flying past Neptune and there is talk of cold fusion from a Utah laboratory.
And although there is an awareness of what was then termed the greenhouse effect, man-made climate change is far from most people’s list of major concerns, oil costs less than $20 a barrel and the market for photovoltaic power – well, the PV market simply doesn’t exist in any way that we’d recognize it today.
Neither First Solar nor Q-Cells would exist until another decade had passed, but some of the early solar pioneers were beginning to emerge. Richard Swanson had recently founded SunPower, and Vahan Garboushian was pondering his next move. A self-confessed “semiconductor guy”, the technology entrepreneur had already made a name for himself by forming Power Hybrids, a company he had then sold to M/A-COM for $25 million.
“I wanted to make a dent in the energy world,” says Garboushian of his thinking at that time. For many semiconductor guys, the fast-growing markets of RF and optical communications would have beckoned – ultimately bringing riches for some of them. But during his time at Power Hybrids, Garboushian had worked alongside the space community – and had seen at close hand what high-performance solar cells could deliver in terms of clean electrical power for satellites.
His vision was to turn that technology into something that could deliver reliable, low-cost electricity here on Earth. Two facts were immediately apparent to him: one that conventional crystalline silicon solar cells were too inefficient for the job; and also that to make a truly significant contribution to grid electricity, solar power would need to be delivered on a utility scale. “I had no interest in rooftop solar,” Garboushian says.
What he was much more interested in was higher-efficiency cells, something that the company he founded, Amonix, immediately began researching. As early as 1990, it had demonstrated a 26% efficiency device under concentration – not a new concept, but one that had been held back by the inability of early cells to withstand the power of highly concentrated sunlight. With improved cells and system development programs involving Stanford University and Arizona Public Services, Amonix made progress through the 1990s and by 1998 had six demonstrator systems in the field.
The next step for Amonix was to show that it could make a manufacturable product – and what Garboushian and his colleagues came up with was to set the standard for concentrated PV (CPV) systems. Nobody had seen anything like the “Megamodule” concept before.
Designed at the largest feasible size for transportation via a flatbed truck, these giant arrays of cells and Fresnel lenses would be carefully aligned inside the factory before deployment in the field. Using the extra large arrays means that far less wiring is needed during installation, reducing costs and speeding up the entire process of connecting utility-scale PV to the grid.
The only problem for Amonix was that, more than ten years after its formation, there was still no real PV market to speak of - although the company had deployed a number of systems in Arizona. Another decade on, and not only is PV now a real market – demand for utility-scale CPV has also finally materialized, and Amonix is at the forefront. Garboushian reckons that Amonix is responsible for around three-quarters of all installed CPV systems, including the largest – a 7.8 MW system built in Villafranca in Spain.
In 2011, he expects some 60 MW of CPV systems to go into the ground, in what would amount to a massive expansion of the installed CPV base. Amonix will account for at least 80% of that total and, thanks to a $90.6 million DOE loan guarantee, it has a deal with Goldman Sachs subsidiary Cogentrix for a 30 MW installation in Colorado that should become the most powerful operational CPV system in the world. It isn’t the only CPV company with major deals in the pipeline – rival Soitec has announced even larger projects planned for the San Diego region, although much rests on pending DOE loan guarantees.
Garboushian spoke to optics.org shortly after the InterSolar 2011 show in San Francisco. He said that the widespread acceptance of CPV now contrasts with a previous skepticism that was evident at the same event as recently as 2009. So what has changed? In a word, bankability.
As well as having the largest installed base among CPV system firms, Amonix also now has by far the largest manufacturing capacity. Its Nevada facility, already producing systems in volume, will have an annual output of 100 MW by year-end. And that is thanks to two key pillars of support – financial assistance from the DOE, and the backing of some of the world’s most powerful venture capital groups.
In early 2010, Amonix announced $129 million in funding from a consortium including Kleiner Perkins Caulfield & Byers, while Goldman Sachs is also an investor. That financial credibility, coupled with US utilities starting to take CPV seriously and Amonix bringing down costs through the use of higher-efficiency multi-junction cells all added up to “bankability” – the perception that CPV is now seen as a serious and reliable way to generate electricity.
Headroom to improve
PV systems are typically rated in terms of their installation cost, in terms of $/W. Indeed, the DOE’s target under its SunShot initiative is to reduce that cost from the $3/W level that is typical today to around $1/W by 2020.
But this metric does not suit CPV – even though up-front capital costs are lower than for conventional PV. A much fairer comparison for CPV is instead the levelized cost of electricity (LCOE), a simple-sounding calculation of the total energy produced by a system over a given time (usually 25 years) and its total cost over that time. On this metric, CPV usually looks a lot more competitive.
A central message from Garboushian’s plenary talk at SPIE Optics + Photonics will be that CPV has more headroom to improve LCOE than any other PV technology, and could get very close to the DOE’s solar electricity target cost of $0.06-0.07 per kWh.
Much of the headroom relates to using multi-junction cells based on III-V semiconductors, which Amonix switched to from silicon in 2007. Production cells are now available at close to 40% efficiency, and Garboushian believes when cells reach 48-50%, CPV will be closing in on the SunShot target.
“That means we need to learn how to use and manufacture cells with four or five junctions,” says the CTO, the key benefits being the ability to match the absorption spectrum of the cell to the terrestrial spectrum of sunlight. That’s a tough engineering problem to solve, but one that is set to benefit from the attention that volume optoelectronic component manufacturers such as JDSU are now paying CPV. In essence, says the CTO, this kind of cell is like a high-brightness LED in reverse – and the equivalent progress in LED efficiency over recent years has been nothing short of phenomenal.
Aside from the cells, Garboushian anticipates improved system designs to dramatically reduce balance-of-system costs, and – perhaps recalling his power semiconductor days – the potential for integrating chip-scale inverters to convert DC generation to an AC output directly within cells.
Focus on execution
But in a phase of rapid expansion, the priority for Amonix right now is not technological. “Our major challenge is execution, as a business that is trying to grow fast,” Garboushian says. At such a sensitive stage in the evolution of this technology, it is crucial that no mistakes are made.
2012 will likely mean significant global expansion for Amonix, with the CTO seeing potential for the technology across the sun-belt of Africa, the Middle East and Asia. Last month the company signed a deal with Thermax to bring CPV technology to India and, as costs decline, even the less-sunny regions of these regions could benefit. “Although CPV excels in areas of high DNI, that does not mean that it can’t be competitive in lower-DNI areas,” he says. If he’s proved right, that could add up to an awful lot of megawatts.
“We are very ambitious, and we intend to grow quickly to the gigawatt level,” Garboushian told optics.org. All along, he and Amonix have chosen to operate at the utility level, and ultimately that means competing in a commodity business. “The plan is to play big, with the big guys.”
The development of Amonix has been more of a slow-burner than a roller-coaster ride, and there’s no doubt that this technology entrepreneur could have made a faster buck in other areas. But Garboushian has no regrets: “The important thing is that you don’t give up,“ he says. “If you believe in something, then you make it happen.” It might have taken 22 years, but CPV is now starting to happen - the dent that Garboushian wanted to make in the energy industry is just beginning to appear.
Vahan Garboushian will present a talk entitled “CPV: competitive now; plenty of headroom” during the Solar Energy Plenary Session at SPIE’s Optics+Photonics conference in San Diego, California, on August 22.
Time-lapse video showing a 500 kW Amonix CPV installation in one day:
About the Author
Mike Hatcher is the Editor in Chief of optics.org