SolarCity to move into former Solyndra site

The building that once housed one of the biggest crash and burn stories in Silicon Valley — Solyndra and its tubular rooftop solar panels — will soon be home to another solar panel manufacturing venture. According to a report in the Silicon Valley Business Journal, SolarCity has quietly leased the former Solyndra building in Fremont, California, just two miles from Tesla’s Fremont factory, with plans to grow its new solar panel division there.

SolarCity bought eight-year-old solar panel startup Silevo last Summer for $200 million worth of SolarCity stock, with plans to become a vertically integrated solar company. SolarCity has long financed and installed solar panels, but soon will make its own solar panels, too. That way the company can control its supply chain, and also lower its costs.

Workers inspecting panels in Solyndra's factory in April

Workers inspecting panels in Solyndra’s factory in April 2011

The company intends to make a gigawatt worth of solar panels a year in a one million square foot facility on 88 acres of land in the development park called RiverBend in South Buffalo. Construction has already started on that factory and it’s supposed to be completed in early 2016. The Fremont building will be Silevo’s headquarters and new R&D center.

Silevo makes low cost and high efficiency solar panels, which SolarCity CEO Lyndon Rive called “generation two” panels, at the recent ARPA-E Summit. Silevo developed a new type of solar cell design using silicon (the main material in traditional solar panels) but also combining different materials for other cell components to make a solar cell that’s more efficient.

That Silevo is now moving into the former Solyndra building shows just how far the solar industry has come since Solyndra’s crash in 2011. Since that time four years ago solar panel prices have dropped dramatically, and there was a shakeout in the solar industry, with dozens of solar panel makers declaring bankruptcy and struggling. Even the leading American solar makers like SunPower and First Solar struggled during this period.

SolarCity panels, image courtesy of SolarCity.

SolarCity panels, image courtesy of SolarCity.

But in the past year or two the large solar panel makers have started to do pretty well. First Solar — the U.S. maker of thin solar panels, which is seen as a bellwether for the industry — has been boosting its solar panel production by as much as 46 percent in 2015.

Due to a combination of federal incentives, private markets, and a decade of lowering technology costs, right now you can build a large utility-scale solar panel farm in the U.S. for the cheapest price in history: as low as $1.68 a watt, according to GTM Research. Both utility-scale and rooftop solar projects (called distributed solar) are booming in the U.S. Almost 4 GW of solar panel projects (both utility scale and rooftops) came online in the U.S. in the first three quarters of 2014 (the latest data from SEIA), up from 2.65 GW for the same period in 2013.

SolarCity’s move to Fremont is also interesting to place it next to its partner electric car maker Tesla. Tesla and SolarCity have spent years working together on combining solar panels and batteries to make an energy storage product. Tesla plans to launch a battery for the home market in a month or so, and has already been bidding on RFPs for energy storage with utilities.

Given the rock bottom prices of solar panels, and potential U.S./China trade wars, many are expecting a return to some difficult times for solar panel makers down the road. That’s another reason why SolarCity wanted to control its manufacturing and bring it stateside.

HelioVolt to finally auction off assets from its solar factory

The obituaries for thin film solar startup HelioVolt were written about a year ago, when the thirteen-year-old company — after over $200 million in funding and no major sales deals — finally said it had halted manufacturing at its factory in Austin, Texas, and was doing layoffs. Now on Tuesday, a company that organizes auctions, Silicon Valley Disposition, said it will hold an auction on Wednesday and Thursday for all of HelioVolt’s assets at its factory.

HelioVolt will be selling off all of its solar panel manufacturing equipment like sputtering tools, laminators, ovens, factory robots, sensing and cleaning systems and office equipment (even the coffee maker). Interested parties can also buy the entire lot, and if that bid is accepted, it will override individual asset sales.

HelioVolt equipment

HelioVolt equipment

HelioVolt developed solar panels that use an ultra-thin layer of semiconductor material from the mixture copper-indium-gallium-selenide (or CIGS) to convert sunlight into electricity. The majority of the solar panels on the market today use a much thicker layer of silicon semiconductor to do the electricity conversion. The founder of HelioVolt, BJ Stanbery — who we first interviewed back in 2007 — got into CIGS back in the 1980’s and invented the CIGS manufacturing process that HelioVolt used in 2001.

Many venture-backed startups, as well as big companies, have worked on making CIGS panels over the years, betting that if the thinner semiconductor material could be used (so it’s less expensive) and if/when silicon prices rise, CIGS would be cheaper. Well, instead of rising, the price of silicon solar panels has actually dropped dramatically over the past several years, in essence halting most of the potential cost benefits of CIGS panels for the time being.

HelioVolt solar assets up for auction.

HelioVolt solar assets up for auction.

High profile CIGS flame-outs include Solyndra, MiaSole, and Nanosolar. Billions of dollars of venture capital funding went into CIGS companies. However, down the road, CIGS panels could still play an important role for more flexible, thinner solar panels. Japanese company Solar Frontier has been one of the only ones producing CIGS panels at any kind of scale.

Even as HelioVolt struggled along with the other CIGS makers, it held on for a long, long time. In late 2011, Korean giant SK Group showed an interest in the company and made a $50 million investment. And in 2012, HelioVolt was still plodding along with small pilot installations in Austin.

But in early 2014, according to local reports, SK Group pulled any more support for the company, and HelioVolt started looking for an acquirer or other investment. Clearly Stanbery didn’t find the deal he wanted and now the company’s technology might have to live on piecemeal at other companies.

Quantum dots, a Ghostbuster backpack & the future of solar technology

Sometimes, it just takes a Google search to overcome a big technical hurdle in a lab. OK, so it didn’t quite happen that way, but for Illan Kramer, a University of Toronto researcher and IBM employee, a simple search turned up the right tool that helped him make a working solar cell out of quantum dots, with are ultra tiny semiconductor bits.

Kramer’s efforts have now demonstrated how to spray a coating of quantum dots using a nozzle that Japanese company, Ikeuchi, developed for steel makers to cool steel with a fine mist of water. Previously, quantum dot solar cell research was mostly focused on using spin-coating, where the semiconductor compound is applied on a substrate that spins to spread the materials. But that process is slow, which means it’d be an expensive way to mass produce solar cells because the equipment cannot coat a large area quickly.

By contrast, silicon solar cells, which dominate the market today, are made by melting and shaping the semiconductor material into rods that are then sliced into wafers. Those wafers then undergo chemical treatment before being assembled into panels.

I caught up with Kramer, a post-doc fellow, because I wanted to learn more about what his research could mean for the future of solar technology. The breakthrough create the ability to quickly print solar materials on a variety of surfaces, including those that are currently unsuitable for placing the thicker and more brittle silicon solar cells. IBM doesn’t make solar cells, but it’s run a solar research lab for years and works on finding manufacturing partners that can license its technology.

U. of Toronto IBM solar quantum dots Illam Kramer

To motivate himself to achieve the next research milestone, Kramer said he conjured up a futuristic image of men with Ghostbusters-style backpacks spraying solar materials onto home roofs. “If only you could one day buy a (solar) spray can at Home Depot the way you can do with a can of paint. But that’s a really long way off,” said Kramer, laughing over the phone during an interview. More likely, a first-generation product would involve a thin layer of quantum dots sprayed onto glass or a flexible material on an assemble line that would then be installed on a building.

I love that Ghostbuster backpack idea, too. After writing about solar technology for six years and witnessing the death of many startups attempting to bring to market ultra-thin and flexible solar cells with all sorts of non-silicon materials, I wonder how solar technology will evolve in the next few decades. Will we mostly see increasingly efficient and thinner silicon cells, which would increase energy production while reducing the material cost? Will there be more color choices for cells beyond blue and black? And do solar panels always have to be rectangular?

Not that there’s been a dearth of research in academic and company labs around making cells that look quite different than what’s available to consumers today. But efforts to commercialize these types of cells haven’t been very successful, mostly because the technology hasn’t been nearly as cheap or efficient as the silicon version.

Konarka Technologies tried to make money from printing an organic compound onto plastic but filed for bankruptcy in 2012. And, yes, there was the spectacular demise of Solyndra, whose novel design of lining copper-indium-gallium-selenide cells inside tubes drew huge crowds at trade shows.

There’s still a lot of cool solar technology research underway that may one day make a market debut, such as using silicon or other materials to build nanowires to improve a cell’s efficiency. Research into producing perovskites, a class of solar compounds with a particular crystalline structure, appears very promising at creating efficient and cheap solar cells that can be tuned into different colors.

U. of Toronto IBM solar quantum dots spray coating

Quantum dots make for an intriguing area of solar research because the tiny size and shape of these nanocrystals make it possible to engineer them to be able to absorb different parts of the light spectrum. The way silicon is created today doesn’t allow it to offer that flexibility.

Kramer is using lead sulfite, which has been shown to work well for creating quantum dots, to make solar cells measuring 1 inch by 1 inch. The best cell that his team has made with the spay-painting method can covert just over 7 percent of the sunlight. That is lower than the 9.24 percent efficiency, a world record for quantum dots, that his fellow researchers at the same lab recently created using spin coating, he noted.

The goal, of course, is to not only to exceed the spin-coating method but to chase after the efficiencies of silicon solar cells, which are mostly in the mid- to high teens. Kramer also wants to develop equipment that could produce cells more cheaply by printing quantum dots, which are suspended in a liquid, on rolls of substrates that could range from glass to plastic. Once the solvent in the liquid evaporates, the nanoparticles are left and then will undergo further chemical treatment to activate their ability to produce electricity by absorbing light.

The path to designing equipment and a process that can mass produce quantum dots is still a long way off. It will likely take 5 to 10 years, Kramer noted.

But there is still cause for optimism. “We are really excited about this process of making nanoparticle ink and using a printing process. You don’t need a billion dollar foundry,” to do it, he said.

Images courtesy of University of Toronto and IBM

Remember that cleantech is creating jobs

Remember when everyone was complaining about the Department of Energy’s loan guarantee program, particularly amidst the collapse of Solyndra. For those who don’t recall the immensity of the political fallout, Solyndra got a $535 million loan guarantee, went bankrupt, laid off 1100 employees and couldn’t repay its loan. Solyndra was part of a larger portfolio of 28 companies that the DOE maintained under the program.
I recall that the DOE wound up having to defend its program with a jobs analysis report, that said that more than 60,000 jobs had been created as a result of the program. This led to a reasonable debate about whether these were permanent jobs or temporary construction jobs for short term capital projects. Things got so politicized that Obama’s first political ad for his 2012 re-election campaign was a defense of his energy policies.
Well, here we are almost three years later and states are fighting over Tesla’s coming gigafactory which will bring over 6,000 jobs and $4 billion in investment to the winning state. And then SolarCity’s acquisition of Silevo should mean a factory will be built in New York that is an “order of magnitude” bigger than plants currently churning out solar panels.
So this is all to point out that renewable energy is a small but real part of the current renaissance of the U.S. manufacturing industry. And the irony here may well be that as natural gas prices creep north and electricity prices go higher, renewables will become even more important in keeping the U.S. manufacturing industry stable. The growth of the manufacturing industry has been largely dependant on a competitive advantage related to cheap domestic natural gas, an advantage that is unlikely to last too much longer.
So the next time critics go after renewable energy for all of its supposed failures, let’s remember how that early investment created lots of initial jobs and laid the groundwork for the larger industry creating long term manufacturing jobs.

Solar job growth solid at 20 percent

Remember the arguments over the impact of the loan guarantee program on job creation? Amid the political fallout of Solyndra’s bankruptcy the DOE issued reports on the impact of that program on the economy. The idea was to justify a program based on an outcome–economic benefits.

I’ve never felt that it was a strong position for the DOE to have to defend its programs on the basis of job creation when the programs are there for national security and environmental reasons. But with all that said, there’s promising data showing that the solar industry added almost 24,000 jobs last year, bringing the total number of those employed by solar to 142,000. That’s a 20 percents solar jobs growth rate, year over year.

Interestingly, two thirds of the new jobs were in installation as opposed to manufacturing, which is the segment of the industry Solyndra occupied. Solar installation continues to chug along at a good pace as financing schemes have lowered barriers to installations, retail electricity rates creep up, and solar panel prices creep down.

SolarCity CEO Lyndon Rive is quoted as saying, “Cost has come down so much that we are starting to work without state-based incentives.” As we near a time when solar can thrive without subsidies, that’s when we should see an even further hockey sticking of job creation in solar.