very inspiring.... evergreen? kick ass!

There is one SoG-Si manufacturer in the USA, Solar Grade Silicon LLc in Washinton State. They say that the dearth will last till 2007 when MANY SoG-Si manufacturers open their shipping departments.... (around the same time nanosolar is supposed to revolutionize everything anyway..)

and glad the venture cpaitalists are getting in on it! that's why we keep seeing new and inspiring co.s like this one and this one and this one popping up all overthe place

One other problem facing potential US buyers of PVs is that what little supply we do have in the States is worth more in Europe. So many panels sold here end up overseas, reducing availability for US projects.
===== author's response follows ====
Amazing. You could have filled a warehouse with panels two years ago and now re-sell all of them overseas at a profit after paying back the loan ; meanwhile, our government's "strategy" here has been focused on ANWR drilling and the like. Once the panels make some progress toward commodity status, the next struggle will be over green electricity. The US' most electricity-intensive industrial operations, such as those which produce aluminum or chlorine, are able to take advantage of much green hydro-power, often subsidized by taxpayer investments. As demand for silicon chip making rises, the "new" electricity is looking like it will come from coal, which means a very high carbon debt per Kg sold. The high-level point is that energy intensive industries are going to compete for green power in order to justify their carbon neutrality from a climate change perspective. My prediction is that astute silicon foundary owners are going to locate in countries with plenty of hydroelectric power. Hello Canada.

Yes, there is http://www.evergreensolar.com, publically traded, and their primary product is solar photovoltaic panels.

But there others based here in the US not necessarily publically traded, nor exclusively manufacturing solar photovoltaics.
http://www.uni-solar.com, Unisolar a division of http://www.ovonic.com, Energy Conversion Devices manufactures their solar panels in Michigan. (Publically traded)

Also http://www.firstsolar.com, First Solar (privately held) manufactures in California and Ohio.

And there are numerous other privately held and venture capital funded firms with various specialized solar photovoltaic technologies - high energy density, low cost, flexible, transparent, and so on.

To imply that we would be or are dependent on foreign solar photovoltaic manufacturers is completely wrong. In fact most of these modules made here in the US are exported to other countries where the demand has traditionally be higher...
=== author's response follows ====
Thanks for providing the links. I wrote the post thinking that the term "pure play" was self explanatory. In using it, I was referring to publicly traded companies with exclusive focus on SPV manufacture, which is what is in short supply. In that sense my inference, although based on a single example, is emplematic of the larger issue. If you take a look at cumulative global manufacturing capacity you'll see that the US is and will remain in a weak position for some time. Japan and Germany dominate. Even Evergreen is pursuing a JV of some sort in Germany. I haven't seen the numbers but I'll bet that it's that way even if you include all factories in the states, regardless of which country that parent company is based in. I think its a cloudy day carbon dioxide emitting day in Phoenix when folks have to import their solar panels from half way around the world and a bunch of the net profits are flowing out. I realize that this is a provacative analogy. But getting some discussion going sometimes takes a ripple in the pond, Thanks again.

I remain amazed that we have a shortage of the second most common element on the planet.

Yeah, I know it's all about manufacturing capacity, but still...
===== author's response =====
It was so much more fun investing in the internet bubble stuff when the only equity was desks and PC's. 'This silicon stuff takes all this expensive hardware and talent, can't we just have more search engine thingies?'

evergreen's JV in germany also has a lot to do with the manufacturing of solar grade silicon

The shortage is because the abundantly available silicon is all locked up in silicon dioxide (quartz/sand) and even then full of various impurities. Pure silicon is not found in nature. To be useful for computer chips or silicon-based photovoltaic materials, the silicon must be extracted and purified through expensive and energy-intensive processes to very high levels of purity. Because of the nature of semiconductor devices, very small amounts of impurities can make silicon useless, particularly for some elements that can typically contaminate silicon (Gold, of all things, is one of the worst). So when silicon feedstock suppliers are ramping up capacity, it's a matter of building some pretty significant infrastructure. The bright side is that they are ramping up, so prices for crystalline silicon photovoltaics will eventually fall, depending on the complex interplay of silicon supply and demand for solar power.

Anonymous: based on evergreensolar.com, it looks like the Evergreen JV with Q-Cells is for manufacturing solar 'wafers, cells, and modules', not the silicon itself. Or in other words, I think it is meant to mean that they take silicon (from somewhere else, presumably they buy it from a supplier), and manufacture into wafers, which they process into cells, and then package into modules.

These guys in Russia have been working with the NREL and the good folks at Sandia and have developed a pilot project to turn Metal Grade silicon into SoG-Si. it's awesome, uses about 80% less chemicals and no chlorene, and 75% less electricity than traditional methods. ( PDF available from NREL )

...but the thing is-- while everyone is racing to produce more SoG-Si (esp SOGSI LLC in WA, Solarworld/Astropower in germany/usa)... it does take 2.5 years to build a facility... and guess what? In two years Google's Nanosolar will be puoring down the pipe, as will many other nano solar manufacturers, and if the cost is that much lower than traditional solar (I heard $0.50/Watt) then all the SogSi in the world wont make siliconPV anyone's first choice
=========== author's reponse follows ==========
Astropower's claim to fame formerly was that they re-used computer chips to make solar cells, bypassing the raw ingot purchase and saving a lot of energy and material. Anyone know if that is still happening?

Regarding metallurgical grade silicon, it would indeed be awesome if people can find out how to make efficient wafers and cells from it; all existing silicon-based solar cell manufacturers would probably be able to take advantage of that.


Regarding Astropower, I have no idea what they do currently, but in terms of 're-us[ing] computer chips to make solar cells', that was probably not the case - a computer chip would require lots of processing to strip off the ceramic packaging and all the transistors and interconnects before you could get at the bare silicon. What would make sense would be to take silicon wafers originally destined for computer chips but rejected for quality issues, and make solar cells out of those, and I think this is probably what Astropower is/was doing.

From photon magazine: AstroPower saves even more production costs, though, since they receive the 200 mm wafers directly from the semiconductor industry’s waste material. No ingots have to be cut into slices with wire-saws; instead the integrated circuits have to be etched off from the wafer surface. »Modules with round cells have the lowest cost per watt,« explains AstroPower’s director of communications Michael Right.

The reason, however, for the domination of square crystalline cells lies in the miserable module efficiencies – in AstroPower’s case 10.3 percent – of modules with round cells, since the round shape allows no tight filling of the available module area. ms For technical details see page 30.

Thanks John! So, I was wrong twice in my guesses about what Astropower does and doesn't do (or did and didn't). At any rate, the other big issue with recycling rejected wafers from integrated circuit manufacturing is that there aren't enough rejected wafers; especially as the silicon consumption of the solar companies of the world begins to surpass and then dwarf the amount of silicon used in computer chips. It is certainly an excellent idea to recycle what existing wafers there are, though.