Print 52 comment(s) - last by snyper256.. on Jun 22 at 12:00 AM

Panels are ~50% more efficient than SolarCity's current supplier's, a third better than the market leader

SolarCity Corp. (SCTY), America's largest solar power systems firm, made a bold step forward this week into the realm of panel manufacturing, announcing it has come to terms to acquire top startup Silevo, Inc.  If SolarCity can make good on Silevo's bold claims, which both it and national laboratory scientists claim are as good as they sound, it could be sitting on a veritable gold mine.
According to RE/Code the deal is worth $350M USD.
I. Elon Musk's "Other... Other" Company
Aside from its dominant market position, SolarCity is perhaps best known for its iconic top investor -- Elon Musk.  After cofounding, which became PayPal in 2000, Mr. Musk cashed out in 2002 via the firm's $1.5B USD sale to eBay, Inc. (EBAY).
But Musk -- now an elite Silicon Valley millionaire -- hardly rested on his laurels.  In the time since he's served as a tireless evangelist and technologist for many of mankind’s most ambitious and -- at times -- scoffed at dreams, almost single-handedly driving some of them towards reality.  As CEO of Tesla Motors, Inc. (TSLA) he is revolutionizing the battery industry with his "Gigafactory," which aims to power the first profitable mass-market EV in the next 4-5 years.  Oh, and it just open-sourced its industry-leading patent portfolio to advance the development and proliferation of electric vehicles.
Elon MuskDubbed the "real life Tony Stark", American visionary Elon Musk is now promising to take the solar industry by storm, thanks to a wily acquisition. [Image Source: jurvetson/Flickr]

And as CEO and majority owner of privately held SpaceX, he's offered up the most affordable commercial satellite and space station resupply launch platform.  Not stopping there, Mr. Musk wants to put men on Mars in a decade or so.

By contrast to SpaceX and Tesla, SolarCity is more of a side project for Mr. Musk, but one that he's passionate about.  His holdings are not publicly disclosed, but as Chairman of the board he's thought to be the solar power firm's largest shareholder.  A 2012 report by Bloomberg suggested that Mr. Musk owns a quarter (25 percent) of SolarCity shares, although that figure may have changed.

SolarCity IPO
Elon Musk serves as chairman of SolarCity. [Image Source: AP]

Needless to say, even in sparing quantities, Mr. Musk's tireless optimism, intuition, and vision has been a great asset to the solar firm.
II. Silevo -- Aiming to be the Best Commercial Cell
Silevo was founded in 2007 by a pair of Silicon Valley veterans -- Zheng Xu, Ph.D and Dr. Jianming Fu, Ph.D.  Mr. Xu started his career at Toshiba Corp.'s (TYO:6502) cutting edge thin film solar division in the late 1980s.  He then founded his own firm -- Sierra Solar Power, Inc. -- in the early 90s. However, he was wooed away by coating, display, and semiconductor giant Applied Materials, Inc. (AMAT) -- where he worked from 1992-2007.
It was at Applied Materials that Mr. Fu -- a fellow executive -- formed a strong working partnership with Mr. Xu.  While he had worked at Applied Materials, Mr. Xu had kept Sierra Solar active.  Based out of Fremont, Calif., it bought and installed panels at homes in the region. But Mr. Xu wanted to make panels of his own.

Silevo Triex
Silevo holds key patents on its n-type cell, which it calls "Triex" technology. [Image Source: Silevo]

In 2007 he and Mr. Fu left Applied materials and spun Sierra Solar into a stealth startup.   On the name -- which resembles a certain conservationist organization -- Silevo's business VP Chris Beitel comments, "It was born Sierra Solar Power, and we didn't really have any ties to the Sierras, even though they imply a green environment."
Naming aside, the startup kept a low profile from 2007 to 2011, when it felt its technology sufficiently mature and patent protected to publicize its work.  In 2011 it had $33M USD in seed money.  The next year it raised $55M USD and moved to China, where it built a 30 MW (of annual panel production capacity) plant, located in Hangzhou, China.
III. Silevo's Triumphant Return to American Production has Been a Long Time Coming
But while it jump an ocean away, its roots lingered in the U.S.
VP Chris Beitel in 2012 said an American return might be possible, stating:

We absolutely did an intense assessment of all the various options—we looked at California, Oregon, and Arizona.  [Locating in China is] a commitment that we have for phase one.  When we look to expand to a phase two, we will certainly look into our options.

For us, it’s very important to have the right cost ratio.  We do that two ways—one is through the operation strategy, and the second is through our fundamental product. The manufacturing strategy was a difficult decision, but if you look at other peers, like Intel or Apple, all of their manufacturing has been done offshore in China. We thought this was the appropriate stage for us moving forward.

Silevo scaleup
Silevo's temporary exodus to Chinese production was merely one step on its ambitious and well-thought-out roadmap. [Image Source: Silevo]

The pilot plant was a convincing success.  In 2013, Silevo announced a key technical milestone -- its near-commercial 125 x 125 mm (4.9 x 4.9 in.) shingles achieved a whopping 22.1 percent efficiency in Sandia National Labs tests, within 0.5 percent of the record for any silicon cell and in the top three of lab-certified cell efficiencies.  That's a great result for a long-shot R&D firm, but the fact that Silevo was nearing production readiness was eye opening.
The key to Silevo's success, and why it kept under wraps so long is its clever technology, for which it holds over 300 patents.
Silevo's cells use a technique called tunneling oxide junctions (n-type silicon), which is often used in electronics, but was widely considered a dead end in the solar power business.  Most solar power makers use p-type silicon, as it's easy to work with.

Silevo tech
Silevo is on a different road than most other PV panelmakers. [Image Source: Silevo]

Silevo has achieved far better results that past n-type crystalline silicon tunneling oxide coated cells, thanks to a couple tricks.  First it operates at a higher voltage, allowing it to escape some of the format's problems.  Second, it uses a patented, proprietary cell architecture to "improve cell defect or interface trap density (Dit)", according to its company website.
These tricks have Silevo on target to achieve its near term goal of 24 percent panel efficiency.
IV. Already at the Top of the Market in Efficiency
Silevo also enjoys a couple of other advantages.  Thanks to its novel architecture, its cells can ditch the silver paste and use copper electrodes instead.  This not only saves money, but is also easier to scale to larger cell sizes and reduces the silver paste's shading effects, which hurt efficiency in p-type cells.  It also has a relatively simple six step semiconductor fabrication process to make the panels -- an art not that far removed from the CPU-wafer fabrication of Intel Corp. (INTC) and its peers.
SolarCity's current primary supplier -- Kyocera Corp. (TYO:6971) -- was last year offering up efficiencies of 14.74 percent with p-type traditional silicon tile panels.  Costs remained high and product laborious, as the p-type panels typically require silicon paste electrodes on one side.  SunPower Corp. (SPWR) and the Panasonic Corp. (TYO:6752)/Sanyo Denki Comp., Ltd. (TYO:6516) joint venture saw efficiency of around 17.6, the highest among top commercial producers.

200 W panel efficiencies
220 W panels
200 and 220 watt commercial panels available in 2013 topped out at around 17.7 percent efficiency. [Image Source: SRoeCo Solar]

Silevo's panels, if it can scale them up to full production, could be more than a third more efficient than the next closest competitor, and more roughly 50 percent more efficient than the Kyocera panels that SunPower typically buys.  Of course that is assuming that Silevo's claims of better efficiency scaling in p-type cells hold true.  Silevo's current production efficiencies of around 17.6 percent -- on par with top p-type cells, but not necessarily better.

T-Series panels
Silevo already is reaching record commercial efficiencies at impressively low costs.
[Image Source: Silevo]

In case studies in the Netherlands, Silevo showed that its 295 Wp Silevo T-Series (17.6 percent efficient) last year outperformed a 245 Wp standard crystalline cell -- similar to SolarCity's bread and butter -- by near a quarter, producing 23.5 percent more power over the course of a year (9.9 MWh vs. 8.03 MWh).
V. Elon Musk Promises Solar Gigafactory in Two Years
Given Mr. Musk's involvement, you might guess what's coming next -- an American solar gigafactory.
A 2013 slide deck showed Silevo was hoping to build a 230 MW capacity plant in the U.S. or China by the end of 2014.  While it is unclear whether it will make that mark, Silevo and its new owner have an even more ambitious plan -- a 1 Gigawatt "Gigafactory" in the vein of Tesla Motors' upcoming $5B USD Southeast U.S. battery Gigafactory, which will be located in Arizona, Nevada, New Mexico, or Texas (with an announcement of which forthcoming later this month).

Silevo utility scale
Silevo, SolarCity, and its chairman Elon Musk are planning a gigafactory to blanket the U.S. with record-setting solar panels. [Image Source: Silevo]

The SolarCity/Silevo solar gigafactory will be located in New York State and will ramp up to 1 Gigawatt of annual production within two years, according to Mr. Musk.  At the announcement of the Silevo purchase, he cheered:

Our intent is to combine what we believe is fundamentally the best photovoltaic technology with massive economies of scale to achieve a breakthrough in the cost of solar power.

SolarCity was founded to accelerate mass adoption of sustainable energy. The sun, that highly convenient and free fusion reactor in the sky, radiates more energy to the Earth in a few hours than the entire human population consumes from all sources in a year.

This means that solar panels, paired with batteries to enable power at night, can produce several orders of magnitude more electricity than is consumed by the entirety of human civilization. A cogent assessment of sustainable energy potential from various sources is described well in this Sandia paper.

Even if the solar industry were only to generate 40 percent of the world’s electricity with photovoltaics by 2040, that would mean installing more than 400 GW of solar capacity per year for the next 25 years.

We absolutely believe that solar power can and will become the world’s predominant source of energy within our lifetimes, but there are obviously a lot of panels that have to be manufactured and installed in order for that to happen. The plans we are announcing today, while substantial compared to current industry, are small in that context.

SolarCity shareholders are clearly optimistic about the plan.  Shares surged 17.5 percent at close, according to Forbes, following the announcement.

Elon Musk
Maybe Elon Musk will use the stellar cells in his bid to colonize Mars.
[Image Source: Art Streiber/August Image]

Who knows -- maybe we'll see some Triex solar cells powering Mr. Musk's Mars colony in the next few decades.  You know what's in the back of his mind.

Sources: Silevo, [PDF on tech], Forbes, RE/Code

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

By letmepicyou on 6/18/2014 10:14:08 PM , Rating: 2
Lets first consider the fact that we have two solar rovers on Mars that ran for what, 7 years on 1/6th the solar energy earth gets? Solar is obviously feasible once you loosen the oil and steel industry's vice like grip on the auto industry.

Electric cars are the best idea to come out since splinter free toilet paper. Anyone with more than 1st grade knowledge about gas & electric power knows this. Internal combustian is a horribly outdated technology. Why do you think auto makers are putting 6, 7, and 8 speed transmissions in cars? Because internal combustion has a pathetically narrow power band. And in the 100 or so years the best solution thus far is more transmission gears. Solenoid valves seemed to have promise, but nobody seems to be developing the technology, probably once again because of the oil industry.

Electric motors, however, are IDEAL for vehicles. An electric motor doesn't HAVE a torque curve. With an electric motor, 100% current = 100% torque, rpms be da**ed. The Tessla electric vehicles are gen 1 electric vehicles. Given improvements in battery technology and better (non steel) construction materials, gen 2 and 3 purpose built electric vehicles should offer performance that will blow your mind.

Now if we can cover cars with ultra efficient solar cells, we can begin to see all-day drivers we don't need fo gas up OR plug in.

By HoosierEngineer5 on 6/19/2014 11:05:54 AM , Rating: 2
You do realize that the rovers only went a couple of dozen miles, and have not been blasting around Mars at 60 miles per hour?

"On May 16, 2013, NASA announced that Opportunity had driven further than any other NASA vehicle on a world other than Earth.[38] After Opportunity's total odometry went over 35.744 km (22.220 miles) it surpassed the total distance driven by the Apollo 17 Lunar Roving Vehicle"

Solar cells can't provide enough continuous power for propulsion of a practical commuter vehicle without major compromises.

One could argue that infernal combustion engines are IDEAL for vehicles due to the high energy storage density of chemical bonds.

By letmepicyou on 6/19/2014 12:35:09 PM , Rating: 2
Haha. Lets recap. 1/6 the available sunlight. About what, 8 square inches of solar panel area? Surface of a car is what, roughly 40 or 50 square feet? With 6x the energy available?

Being dead serious. Do you work for the oil industry, and are here to sway public opinion or what? Internal combustion engines in modern cars are PATHETICALLY inefficient. A modern I.C. engine is roughly 12% efficient. Oh yeah gas has awesome energy density, which a modern engine wastes 82% of as heat, converting just over 10% to workable mechanical energy.

SUPER great technology. Diesel hits like, 15%, and gas turbines hit maybe 25% or 30%. Electric motors are roughly 85% efficient, depending on the motor. If you think internal combustion is ideal timeto either give up your job with BP or go back to physics class.

By M'n'M on 6/19/2014 1:25:41 PM , Rating: 2
Haha. Lets recap. 1/6 the available sunlight. About what, 8 square inches of solar panel area? Surface of a car is what, roughly 40 or 50 square feet? With 6x the energy available?

Yes, let's recap. I have no idea where you think you have 40-50 sq ft on a car that could be covered w/PV cells (I just measured 6.5' x <4' on the roof of a Ford Escape) but let's say it's a big SUV and it's 50 sq ft. The solar power available at high noon at the Earth's surface is about 1kW/m^2. On your 50 sq ft that's 1000 W x 4.64 sq meters or 4645 W. The solar cells mentioned in the article can convert that solar power into DC electrical power at 18% efficiency. Lab PV cells have hit about 50%. But let's even assume you get to 99% efficiency someday. Now you have 4600 W to move your car. That's just below 6.2 HP.

I hope you're not planning to go very fast as even a Prius (or Tesla S) needs 14 HP to go 70 MPH.

As for electric motors not having a torque curve ... I can only hope your EE degree isn't from an accredited university.

By HoosierEngineer5 on 6/19/2014 1:57:52 PM , Rating: 2
Let's assume you live in Detroit and want to drive your solar-powered car in December (many folks do). The AVERAGE solar irradiance in Detroit is 2.4 kWh/m2/day, assuming your car has a 2-axis tracking flat-plate collector.

Your 50 ft^2 roof (about the size of 1.5 sheets of plywood by the way), 4.65 m^2, would collect about 11 kWh of energy per day, average (no snow on the roof).

Now let's assume you have a commute of 1/2 hour each way, and don't go out to do anything else. In order to complete your hour drive, you would need to produce less than 15 horsepower on the average.

I have assumed 100% efficient solar cells which is a theoretical impossibility

You would need to drive a very small car for limited distances or very low speeds in order to achieve reasonable utility (i.e. year-round in most of the U.S.). Driving continuously, you would get around 0.5 horsepower, enough to move a bicycle at a pretty good clip.

Ok, I showed you mine, let's see yours.

BTW, Do you work for Disney?

By letmepicyou on 6/20/2014 3:06:20 AM , Rating: 2
In 2013 the Nuon team won the challenger class of the Solar Challenge with an average (average, meaning they actually went faster) speed of 90 kmh.

According to your data, such a thing is a theoretical impossibility.


By atechfan on 6/20/2014 6:18:59 AM , Rating: 2
And how much did the Nuon team's vehicle weigh? We are talking actual vehicles here, ones that have to carry passengers and cargo, not a sheet of solar panels with a bubble sticking up for the driver's head and some bicycle wheels.

I'm all for electric cars, when they make sense. But battery tech isn't there for us to go all electric.

By HoosierEngineer5 on 6/20/2014 9:52:00 AM , Rating: 2
And you don't think that vehicle embodied any major compromises?

OK, I'm going to stop feeding the troll.

By letmepicyou on 6/21/2014 10:50:01 PM , Rating: 1
So, knowing the technology exists (or is being purposely held back by the big oil companies, which is KNOWN to happen) and demanding it as a consumer makes me a troll?

If I'm a troll for this, then you're DEFFINITELY a big oil employee paid to makes posts like this on popular message boards. How much does manipulating public oppinion pay these days, anyway?

By snyper256 on 6/21/2014 11:59:11 PM , Rating: 1
You're acting like someone in the 1970's who thought a computer would never make it into the home, sir.

By Nexos on 6/19/2014 2:37:13 PM , Rating: 2
its not 1/6 of the sunlight, its closer to 1/2.

"It's okay. The scenarios aren't that clear. But it's good looking. [Steve Jobs] does good design, and [the iPad] is absolutely a good example of that." -- Bill Gates on the Apple iPad

Copyright 2015 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki