September 14, 2009
Nanosolar Steps Out With Lower Costs Solar Cells

Nanosolar has announced $4.1 billion of orders for their thin film photovoltaic solar cells even though they are just starting up production.

On Wednesday, Nanosolar pulled back the curtain on its thin-film photovoltaic cell technology — which it claims is more efficient and less expensive than that of industry leader First Solar — and announced that it has secured $4.1 billion in orders for its solar panels.

CEO Martin Roscheisen claims lower costs than cost leader First Solar. Great news if true.

Nanosolar has an unusual manufacturing approach using thin aluminum foil and a wet chemistry for depositing CIGS (Cadmium Indium Gallium Selenide) thin film on the foil as the foil moves rapidly. This continuous flow approach holds the prospect of much lower cost of manufacture. Their new plant in Germany can produce 640MW per year when operated 24x7.

Today Nanosolar demonstrated the completion of its European panel-assembly factory as part of an inauguration event attended by Germany's Minister of the Environment, the Governor of the State of Brandenburg, and a host of other leading public officials. Located in Luckenwalde near Berlin, the fully-automated factory processes Nanosolar cells into finished Nanosolar panels using innovative high-throughput manufacturing techniques and tooling developed by Nanosolar and its partners.

The panel factory is automated to sustain a production rate of one panel every ten seconds, or an annual capacity of 640MW when operated 24x7. Nanosolar also today announced that serial production in its San Jose, California, cell production factory commenced earlier this year.

That 640MW is probably in practice equivalent to about a quarter that amount on average - depending heavily on just where the panels are installed. I'm guessing roughly 10 such factories would produce the equivalent of a new nuclear power plant every year. If they can get their costs low enough then raising the capital for a big capacity build will be easy to do. They claim their process makes cost reduction much easier than competing approaches.

Nanosolar claims the electrical capacity of their panels make them especially well suited for large utility installations.

Electrically, it is the industry’s highest-current thin panel, by as much as a factor of six. It is also the industry’s first photovoltaic module certified by TUV for a system voltage of 1500V, or 50% higher than the previously highest certified. Together this enables utility-scale panel array lengths and results in a host of substantial cost savings during the deployment of solar power plants.

Nanosolar's foil PV has produced electricity at 16.4% efficiency.

Our lab and production teams have managed to make more progress on efficiency than we had planned on in any of our business plans. Recall that we print CIGS onto inexpensive metal foil, that is, something that some have been skeptical can work while others have been wondering whether it can deliver efficient cells.

So we are pleased to announce that our low-cost printed-CIGS-on-metal-foil cell stack and process produces quite efficient cells: Earlier this year, NREL independently verified several of our cell foils to be as efficient as 16.4%.

In production efficiency is running at only 11%. If Nanosolar can get their production cell efficiency up to 16.4% that would cut substantially cut their costs.

PV costs are definitely on a downhill slope after several years of stagnant prices. Now we have two main competitors driving costs below $1 per watt. Good news. Will costs of supporting equipment such as grid tie inverters also drop? Or will the development of DC (direct current) appliances avoid the need for grid tie inverters for home PV installations?

Michael Graham Richard has some cool pictures of their manufacturing process.

Update: Be sure to read Chris Nelder's more detailed look at Nanosolar's cost advantages. They cut interconnect costs, reduce need for aluminun frames, and cut costs in other ways relating to installation. They might end up far cheaper than First Solar, the current cost leader. Solyndra has been lauded in the press as the potential big rival to First Solar. But if the claims about Nanosolar are true then Nanosolar might beat both of them. Sure looks like PV costs are headed for a big fall.

Share |      Randall Parker, 2009 September 14 09:27 PM  Energy Solar

Fat Man said at September 14, 2009 10:37 PM:

The cells could be free, but Solar will remain a supplemental and secondary power source, until such time, if ever, as 1) they have solved the problem of the sun's daily disappearances, and 2) the density of insolation at the earth's surface is increased by at least a couple of orders of magnitude.

Michael B. said at September 15, 2009 11:54 AM:

Fat Man,
I don't think anyone expects solar to be a primary source of energy. It's exciting because it could drive down costs and prices because solar energy is generated during the times of peak load. Peak load is one of the main factors that drive up cost. It's also the one of the big reasons that natural gas and petroleum derivatives (like gasoline and diesel) are used for electricity generation.

Randall Parker said at September 15, 2009 5:53 PM:

James Bowery,

A lot of demand is easily shiftable. That's especially true of heating and cooling demand. One can run heat pumps at noon when the sun is shining and make ice or heat up molten salts. One can recharge electric cars at noon.

Currently electric power is most expensive in the afternoon and early evening. We can easily shift cooling and heating demand several hours. We can easily make washing machines turn on at 11 AM. Given cheaper day time power we've got lots of ways to shift demand to the suddenly very cheap period.

energydude said at September 15, 2009 6:30 PM:

Residential grid intertie inverters cost about 20% of what PV modules cost on a $ / watt basis (today). Grid intertie inverters are about 95% efficient, meaning only 5% energy losses in the DC to AC conversion. Add to that the wonderful job they do automatically balancing with the grid to accomodate things like passing clouds and the "moving sun" (relative to the earth that is).

On the other hand, going to DC loads means you need some way to balance these ups and downs from clouds and sun movement, which probably entails an AC to DC converter running off the grid, or a dual solar and AC charge controller with batteries. Then in either case you'd need a DC voltage regulator unless your loads are running off battery voltage (typically 12, 24 or 48 volts DC) because the voltage output of the PV modules moves with temperature and their current moves with solar irradiance, and the two are moved all around with maximum power point tracking (MPPT) to maximize solar energy. You can probably already figure out that all that DC stuff is more complicated and costly than the AC grid intertie inverter.

AC inverters on the backs of modules are an interesting idea, but right now inverters are the main reliability and service issue with PV. Inverters on the backs of modules need to be absolutely bullet proof and very long life or else the old bell curve of reliability means a hell of a lot of service calls over the life of a system to replace inverter after inverter on the roof. Despite what a sales person may tell you, the technology is not yet reliable enough for 24 inverters on the roof to replace one on the ground for the next 25 years.

Randall Parker said at September 15, 2009 7:13 PM:


Regards grid tie inverter costs: when you say 20% what are your two numbers per watt? I was up on Google Shopping and it looked to me from some examples that grid tie inverter costs ranged as low as 60 cents per watt.

I get that variability of output is a problem. But couldn't some appliances be designed to use varying inputs? I'm thinking in particular cooling and heating pumps.

Though I see one big problem with that: at some parts of the year little heating or cooling is needed. You do end up with extra power that you've got to do something with or it goes to waste.

Wolf-Dog said at September 15, 2009 7:53 PM:

What is the Cadmium content of the solar cells? How well do they seal these cells to avoid any leaks? These questions must be answered before worldwide adoption of this technology.

Engineer-Poet said at September 16, 2009 6:55 PM:

energydude:  any grid-tie inverter worth its salt is going to do MPPT as well.  It's all software anyway.  A buck-type DC-DC converter is going to be pretty simple, though; a whole lot less complicated than the inverter.

I'd love to have the problem of having too much in the way of PV and wondering what to do with the power on sunny days.

energydude said at September 16, 2009 7:40 PM:

Randall, grid intertie inverters cost about 40 to 50 cents per watt from reputable, quality manufacturers - again residential scale and so far almost exclusively through distribution. PV modules on the other hand are now available direct from manufacturers; due to the oversupply the wholesalers are being cut out, at least to volume installers.

Engineer Poet, I don't know of any without MPPT now. Point is, add up all the parts to a DC load PV system that handles clouds, seasons, weather and night-time along with the interconnections and it'll be much more than the cost of a grid intertie inverter. Think $10k vs $3k for a small house. Don't bother, stick with AC loads.

Engineer-Poet said at September 17, 2009 6:30 AM:

Those parts are mostly batteries.  Yes, batteries are expensive, but if you need emergency power you don't have much alternative.

James Bowery said at September 17, 2009 9:01 AM:

Randall, you aren't taking into account the primary driver of utility costs: Meeting peak load.

When people say "baseload" what they mean is the level of continuous electrical capacity that optimizes the total cost of reliably meeting peak load. When you have a variable like insolation in the equation, you need a lot more than than average day-night load leveling working for you.

energydude said at September 17, 2009 4:32 PM:

Engineer Poet, you'd need batteries or a type of grid intertie (AC to DC converter) in a DC load system just to handle the sun going behind the clouds; that is, for more than just emergency power. Most people would not tolerate their loads dimming or shutting down when a cloud passes.

Engineer-Poet said at September 18, 2009 8:11 AM:

Of course you'd need a grid intertie or batteries.  Unless your loads can completely follow the supply (like some water pumps), you have to have some way to buffer or exchange energy.  What about it?  (Unless you are entirely off-grid, you only need batteries for emergency power.)

Energydude said at September 20, 2009 5:25 PM:

True, we're in violent agreement.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©