Solar support continues to climb, with more and more becoming aware of the potential of the sun's energy. Australia's clean energy investment package is tipped to provide massive opportunities for large-scale development, and closer to home, initiatives such as the Solar Promise campaign launched today urge communities to embrace solar technology and combat climate change. Meanwhile, over in Silicon Valley, one Kiwi is at the heart of a company making groundbreaking strides toward commercial solar cells which could radically reshape the global economy.
If you happened to be passing Ballroom 6A of the Washington State Convention Centre at 1.30 on the afternoon of June 20, you will have seen a soft-spoken man in his early thirties addressing a crowd a few hundred strong on the subject of photovoltaics. He’d have been referring to wafers and substrates, gallium arsenide and precise control of the dark current, and unless you had a solid knowledge of nanotechnology and semiconductors it might have been possible to listen to the whole speech without being particularly enlightened as to the subject matter or its implications.
The audience, though, were held rapt. They had come to the Photovoltaics Specialists Conference (its 50th anniversary!) precisely for moments like this, when a Aucklander named Brendan Kayes with a PhD from a small college called Caltech would lead off the conference with the announcement that Alta Devices, the Santa Clara–based firm he works for, had broken the world record for single-junction solar cell efficiency, with an independently verified reading of 27.6 percent under 1 Sun intensity.
What does that mean? Right now: very little. But if Alta Devices’ big gamble comes off, its unique approach to solar energy could reach unsubsidised price parity with the filthiest fossil fuels in as little as five years. And while Silicon Valley is better populated than most with startups proclaiming their genius, this particular company is backed both professionally and financially by some of the biggest and most respected venture capital firms in the world.
The origins of Alta and its breakthrough technology lie in the late 70s, in the midst of the last solar boom, when US oil companies were so terrified of the implications of the Arab oil crisis to their businesses that they started pouring money into alternative energy. Then, as now, the dominant material for manufacturing solar cells was silicon. It wasn’t cheap and it wasn’t the best, but it worked pretty well and was a heck of a lot cheaper than gallium arsenide, which while much more efficient was incredibly expensive and thus largely ignored.
Not everyone was disinterested in the material, though. Exxon hired a young Harvard grad named Eli Yablonovitch, who in 1979—around the time Kayes was being born in Wellington—chanced upon a fascinating discovery. He found that you could ‘grow’ a infinitesimally thin layer of gallium arsenide on top of a wafer in the same material, then with immense care separate the two layers and re-use that wafer.
The significance of that discovery was that gallium arsenide (or GaAs) remained an effective solar cell in microscopically thin layers, thus stretching it a long, long way. It was just a starting point, but it seemed to imply that if you could devise a way to get the material somewhere close to its peak efficiency, and figure out a way to manufacture it in vast quantities, then ... But before Yablonovitch could even finish the thought it was 1980 and Morning in America.
Ronald Reagan, with the wind of a disastrous hostage crisis and a tanking economy at his back, trounced incumbent democrat Jimmy Carter in one of the great landslides in US political history. And one of his first acts as president was to tear off those sissy solar cells that Carter had affixed to the White House roof. Reagan was more in the ‘drill, baby, drill’ camp. He slashed petrol taxes, and, when the 80s oil glut saw gas prices slump further still, any wavering oil companies promptly ditched their solar programmes in favour of relapsing to their first love.
The potentially lucrative patents Yablonovitch had filed fell victim to economic circumstances that meant spending hundreds of millions of dollars exploring an obscure idea that only might work looked like pretty strange behaviour for an American oil company in the go-go 80s.
So with the funding tap turned off, Yablonovitch tried a couple of other private sector positions before returning to Harvard to teach, eventually settling into a long-term role as a Professor of Electrical Engineering and Computer Sciences at UCLA.
The solar industry didn’t die just because it became unfashionable. But it undeniably slowed down its evolution and became comfortable as a boutique source of energy. It worked great for satellites, those seeking environmental cachet or houses way off the national grid. And while its cost continued to fall—from $20 per watt in the 1980s to less than $2 in 2010—the mainstream technology and materials stayed broadly the same: silicon panels encased in large rectangular glass structures.
Scientists continued to tinker on the cutting edge, producing all manner of experimental techniques, some with very high efficiencies, and bold claims were made for the likes of Extreme Concentrated Solar Power delivering 5c per kilowatt hour, a price which would make it instantly cost competitive with coal and gas. But the journey from lab to national grid has never been an easy one, and as of now a state-of-the-art solar system will max out at around US$0.22c per kilowatt hour, roughly three times that of its carbon-emitting competition.
But while the 2000s won’t be remembered as the decade carbon emissions died, they certainly provided a nexus of science and commerce unseen since the 80s. Between rocketing commodity prices and the renewal of environmental consciousness that lead to governments subsidising solar fields in a big way, particularly in Europe, a second solar boom started rising.
At UCLA Yablonovitch watched these developments with interest. He had never forgotten his patents, and in 2007, nearly 30 years after they were first filed, the world had come to a point where they might prove useful again. Helpfully, they had just expired, allowing him to toy with the idea without worrying about Exxon getting its claws into his work. He set up a meeting with a Harry Atwater, a fellow academic with a deep interest in and knowledge of photovoltaic solar cells, just to mull the opportunity.
Atwater was an acclaimed Professor of Applied Physics and Material Science at Caltech, a college that, while not as large or well known as the likes of MIT or Stanford, has a reputation as one of the most academically rigourous in the world, with 31 Nobel Prize recipients to match. The professor was running a PhD programme at the time, and among its number was Kayes, a University of Auckland graduate who suspected that he had only made it into Caltech at all because he shared Atwater’s deep idealistic streak.
While Kayes is perhaps being modest about his academics, which saw him among New Zealand’s top students, his idealism has been a defining influence from a young age.
“I was never much of practical person,” Kayes says. “I would never get a chemistry set or soldering iron. I was much more interested in arcane knowledge—things like mythology, or strange mathematical theories, or how to make flying cars.”
This is a phrase he repeats frequently and it’s clear flying cars have become a form of shorthand for him, a metaphor for exploring the unknown in pursuit of a grand prize rather than settling for more mundane work within defined parameters.
After graduating the University of Auckland with a double major in physics and philosophy, he struggled with a direction to take. On MIT’s website he stumbled across the Alliance for Global Sustainability, a scholarship to a two-week gathering of undergraduate geeks from around the world that addressed his abiding concerns for the fate of the planet. Deadline for submissions was that day; undaunted, he dashed off the essay that would set the course of his life.
“It was something about, ‘I look at the motorways, and they’re all filled with cars. What about the children?” he jokes. But it impressed the right people, and a few weeks later he was on a plane to Norway. After a week suffering through social scientists’ solutions to global warming, he wandered into a lecture on solar energy.
“I thought, ‘This is what I want to get into.’ I think it was that it was helping to address some of these issues about sustainability but from quite a hardcore science point of view,” he says. “I liked that that I could use arcane knowledge to make this contribution.”
When he landed in Auckland on the morning of September 12, 2001, the awful events (and their more awful response) in the US made his decision more starkly obvious, and by the next summer he was in the system at Caltech.
Kayes saw Caltech as a place he might fine-tune his mind for the good fight, and his classmates frequently had more personal motives, but one of his contemporaries had far more aggressive tactics in mind. William Cottrell was a Caltech PhD candidate who started a series of clandestine molotov cocktail attacks on Hummer dealerships on behalf of the Earth Liberation Front. He would be convicted of arson and jailed in 2005, and despite campaigning from scientists as eminent as Stephen Hawking, he remains in prison, scheduled for release this August.
Luckily, Kayes and Atwater took a less visceral approach to curing the world’s ills. The student and professor bonded over their shared desire to see what solar could do; to play with the potential of approaches as yet untested and see what they might find. Which meant the New Zealander had to get over his disinterest in lab work and learn how to wield a beaker with authority. Somewhat surprisingly, he grew to love the processes.
“It was learning how to build stuff, where the fabrication itself was quite challenging, then using what you’d built to answer a question that no one had answered yet,” he says. “I really loved it in a way I didn’t necessarily expect, because I had very little background in that before, and it became my favourite part of the research experience. It’s something I love at Alta too, the hands-on part.”
As the end of his seven-year graduate-school experience neared, with years of highly technical research under his belt, Kayes performed one last task for Atwater. The professor wanted him to turn his scientist hat backwards and approach his thesis—predominantly concerning the application of nano-technology to solar cells—from an entrepreneurial perspective. He found the experience enlightening, but equally it convinced him that as elevating as his years of research had been, the end idea just wasn’t going to achieve the result he wanted.
About the same time Kayes was deciding his own graduate school research would likely not fly commercially, Atwater and Yablonovitch were reaching a very different conclusion about Yablonovitch’s old patents. The pair had an inkling that this idea, while still very far from proven, deserved further exploration. They set up Alta Devices in Silicon Valley in 2008 but as tenured professors with little desire to give up their academic positions, they needed collaborators who could fund and run this project, one that would need expert stewardship if it were it to be properly explored.
Luckily Atwater counted among his friends a man named Andy Rappaport, from venture capital firm August Capital. Rappaport was impressed enough to take it to Kleiner-Perkins, the most acclaimed venture capital firm in the world with early investments in Google, Amazon and Sun on its CV. It had recently picked up a former engineer from Sun Microsystems named Bill Joy, a Silicon Valley legend whose chequebook was aimed squarely at the kind of projects Atwater was envisaging. He seemed a natural partner for a project this complex.
Joy’s been called ‘The Edison of the Internet’ on the cover of Fortune, and when he decides an idea has merit, and Kleiner-Perkins will invest in it, that idea is instantaneously granted a level of credibility most Silicon Valley startups can only dream about. Joy has now committed funds to Alta’s A, B and now C rounds of funding—a rare move for a venture capital firm, which will frequently dip out after a round or two and let their investment ride when private equity starts to come on board for bigger sums. Joy and Kleiner-Perkins presence and continued support, along with that of Rappaport and August, has to be seen as a huge vote of confidence in this company. Both Joy and Rappaport sit on Alta’s board too, and CEO Christopher Norris says they bring very different but complementary strengths to the organisation.
“Bill, as you can imagine from his background, is just extraordinarily technical, a bit of the wild-eyed scientist,” he says. “He’s always, always, always out-of-the-box – I don’t know that he’s ever had an in-the-box idea ever.
“That’s partly how we got to this audacious set of goals, realising that if you’re willing to invent across multiple complicated domains you can transform this overall concept from something which is specialised—there might be a market you can fit into—to something that is foundational and could transform the market. Bill’s the guy who generates those kind of ideas.
“Andy is a true strategist. He spends as much time questioning how we’re thinking about problems as he does the problem themselves. So he has a very different, longer term focus. Together, they provide an extraordinary value, in terms of guidance and visioning for the company, from two very different viewpoints.”
Norris himself came to Alta early, tapped by Rappaport after impressing him with his handling of a previous startup. He’s the son of a nuclear physicist, and grew up in Howe, Idaho, a town of 23 whose main claim to fame was having a post office. Norris studied electrical engineering at the University of Idaho before doing graduate work at the University of Santa Clara. After working for both Intel and Cypress semiconductor for nearly 20 years, he worked for a spell at a VC firm that exposed him to four or five deals a day, and solidified in Norris a desire that his next position would be an area of the tech sector rather different to most of what was coming across his desk.
“I knew I wanted to do something that was a bit more meaningful than the next consumer product, the next memory chip,” he says. “I was looking for something with a bit more relevance. Which was largely cleantech.”
When Rappaport offered him the role of CEO he studied the opportunity for months before committing to this fiendishly complex piece of management. Why so complex? Because it involved sailing into multiple unknowns, hoping more than believing that the immense technical and scientific challenges they represented could be overcome. Oh, and convincing someone to pay for the voyage. Norris sums up the enormity of the challenge.
“The trick in solar technologies, and why new technologies are viewed sceptically here—which they mostly are—is that unless you manufacture in very large scale you can’t achieve the cost-points. You have to build and manufacture at large scale. The factories are very expensive, typically in the hundreds of millions of dollars. So you are going to have to spend at least half a billion, maybe a billion dollars before you’re at scale.
“So there’s a lot of money that has to be deployed to get to the point where you’re delivering product that is at its most economically competitive point. That is the main challenge for solar companies. We can do these,” he waves one of the thin, flexible solar cells laid out on the conference table in front of him, “all day long, and put ‘em together in different versions. The question is, how do you get that to a point from lab samples in small volumes to enormous volumes?”
Thus far, Alta seems to be doing pretty well on all fronts. The initial work fell very heavily on the device team, which Norris characterises as a collective effort, while acknowledging that Kayes has “been the point man for the past nine months”. They have completed a number of well tested ‘devices’ (differently tweaked solar cells), while this month another portion of the company moves into its first pilot factory to begin testing scale models of the machines that will eventually commercially manufacture this stuff.
When we finish our interview, Joe Foster, VP of business development at Alta, takes me on a tour of the labs—strictly no cameras—which drives home both how far they’ve come, and the challenges that remain. After passing through a series of rooms dedicated to inspecting the cells at terrifyingly close quarters, or hammering them to test durability, or testing their efficiency, we come to the last and most intriguing of all. It’s raised up so that “a beautiful laminar flow of air” can pass through, while technicians encased in bunny suits gingerly lift wafers from vats of hydrochloric acid.
“The beauty of this is you invest in this expensive crystal template—which is the wafer—and you can make thousands of films on that one wafer, which is the template,” says Foster. “You treat it more like a fixed asset than a consumable. The only way it falls out of production is if it breaks, essentially. Assuming you treat it with care and you don’t damage it, its lifespan is infinite.”
Behind the technicians is a chamber, less than ten metres long and couple high, that is the company’s first go at a machine that can manufacture these cells. It’s a very early and comparatively rudimentary prototype of what will need to become a machine the size of basketball court that can operate without the clean room. Which is to say that Alta has a long, tortuous road ahead before commercial mass production.
But should it make it to scale, the advantages of Alta’s technology over anything else available could revolutionise the market. Its cells are incredibly light—one micron thick compared with a human hair’s average of 40 microns—and its efficiency means it can generate far more energy per square metre than the current silicon standard. Additionally, the flexibility and thinness could allow the cells to be built into roofing, or cars, or the side of buildings: more surfaces could be solar friendly than you might ever have thought possible.
And for those who still harbour concern about the solar part of this equation—because, ambitious as it is, Alta isn’t attempting to stop the sun from setting each evening—should be somewhat mollified by advances in other complementary areas, particularly combined cycle natural gas plants. These can be powered up or down in as little as three minutes, allowing fossil fuels to potentially play the role once allocated to renewals: providing supplementary off-peak power to ensure the grid stays at capacity.
It’s these elements that make green technology experts think this company has a good shot at being the one that breaks the silicon shackles and sets solar loose on the free market. One man who is mightily impressed by its work is Steve Eglash, executive director of the Energy and Environment Affiliates Programme at nearby Stanford. He’s been following Alta’s progress as closely as it will allow any outsider over the past few years.
With a background in venture capital and private and public sector experience in solar, he’s as well placed as anyone outside the company to judge their decisions thus far, and believes that despite the work still to be completed, the company is performing incredibly well thus far.
“Alta’s real breakthrough is all about finding practical high-volume, low-cost ways to build gallium arsenide solar cells that are still high quality and high performance,” he says on the phone from his Palo Alto office. “And though I don’t have any inside or proprietary insight into what Alta’s doing, I am aware of many of the choices and decisions made along the way, and I think those have been astoundingly good. Where they’ve needed to make bold and courageous decisions to try to differentiate it from the traditional high-cost ways of making gallium arsenide devices, they have done so, and I believe they’re making good progress towards making it work.”
That’s certainly what Norris and his team at Alta believe. And while theirs is far from the only solar technology in the running (cadmium telluride, CZTS, CPV and organic technologies are all at various promising stages of research or manufacture), Alta Devices has completed a formidable amount of fiendishly complex work already. This means that it’s finally emerging from the cocoon of “stealth mode”, as the inquisitive green tech press consistently referred to the way the company initially comported itself, and allowing the fruits of its labour some air.
Despite the optimism within and about the company, Alta is acutely aware that the mission is far from complete, and milestones that might be cause for celebration at other organisations are marked more muted. No one at Alta can remember cracking open a beer when it broke the single-junction efficiency world record, despite it being acknowledged within the company as its finest achievement to date.
Solar has certainly had a strange half-century, filled with peaks and retrenchments, patents squandered and false dawns. Moore’s law, the doubling of transistor capacity every two years, does not apply in this arena. But what Alta is doing holds the distant yet tangible promise of the major commercial breakthrough that solar’s advocates have lusted after for decades. Toward the end of our interview, Norris, drops the caveats for a moment and speaks in definitive terms of his company’s potential impact on the world energy market.
“This is something that when it’s successful, it will transform how solar energy is thought about, and the economics of solar energy.”
If he’s right, then Alta—with a young, gifted New Zealander at its heart—has a shot at being one of the biggest companies on the planet within a few years, a revolutionary force that could de-power the Middle East’s oil cartels and cure global warming in one fell swoop. While Norris, ever the scientist, tends to avoid addressing those prospects head on, he is not above hinting at the scale of the company’s ambitions.
“If you can compete with other forms of energy economically, then the total available market is essentially unlimited. There aren’t many markets like that, where if you can just get to a certain cost point it almost never stops.”
He’s confident they’ve got the science. It’s up to the idealists to figure out the rest.