First Solar: Quest for the $1 Watt
July 23rd, 2008Photovoltaic cells, once so costly they could be used only to power million-dollar satellites, are today turning up even on humble parking meters. Now a brash Tempe, Ariz., company called First Solar plans to take the technology to the next level by making it cost-effective enough to compete with coal-fired generation.
Achieving grid parity--selling power to the nation's electric grid at a competitive price--has long been a holy grail of the photovoltaic industry and other suppliers of alternative energy. Yet despite the company's soaring price share and its multimilliion-dollar order book, First Solar declines to speak to journalists.
In the August issue of IEEE Spectrum, British writer Richard Stevenson combines a journalist's knack for investigation with the expertise of a solid-state physicist to piece together how First Solar has cracked the problem. He concludes that the secret involves not the photovoltaic cell itself but the way in which it is manufactured. Instead of the familiar silicon, the design uses a compound of cadmium and tellurium.
Not long ago it was little more than a laboratory curiosity, largely because nobody had found a practical way to make the cells much larger than a postage stamp. First Solar has now refined the manufacturing procedure to blow up the cells to poster size.
Already the firm has been able to make a profit selling the panels to utilities in a number of countries--particularly Germany--that subsidize alternative energy sources for environmental reasons. Available figures suggest that the manufacturing cost per watt delivered is still too high to compete with that of power delivered on the grid, but First Solar has told investors that it expects to be able to lower the cost substantially.
It seems likely that such improvements, together with the rising price of fossil fuels generally, will enable the company to reach grid parity within just a few years. Indeed, the technology is so promising that it puts into question whether there will be enough tellurium available to make all the solar panels the world is likely to demand. Stevenson's conclusion is that the answer is yes, because increased demand for the panels will stimulate the search for new supplies of the scarce element.
Meanwhile, other photovoltaic technologies continue to advance, not the least of which is silicon. If, as expected, the current shortage of silicon should abate in the coming years, then First Solar's clear lead on the industry may narrow. In any case, photovoltaic cells seem poised to advance from their current role in niche applications to become one of the more important sources of electricity in the world.
Source: IEEE
Jun 30, 2009 |
not rated yet |
0
It's an intermittent source of power and dealing with the intermittency costs more than just getting non-intermittent power from coal or uranium. Cheap storage is an absolute necessity for solar to take off.
Once solar cells are in orbit they will be a ridiculously expensive proposition.
And we don't have a clue how to transmit that power back to Earth efficiently, cheaply, nor safely; possibly not even continously.
Also what about the efficienty of the panel. Sure if I can get $1/watt but need a football field sized area to achive it, it's a bit impracticle.
This article sounds just like a fish expadition for money by the company.
z
Ditto, Considering it's the most abundant element in the earths crust.
I'm really tired of hearing about the great new solar tech that's going to change everything. This technology seems solid, but you hear about a new whiz-bang solar tech every other day. Let's thin the herd a little bit so that we can get funding to legitimate companies who have shown progress instead of basically giving it to anyone with pretty pictures, an idea and some half-science to back it up.
It hit me as odd to say silicon is rare - i was wondering what planet the author was on.
Talking about cadmium as the great new hope. They have to be kidding - that stuff is dangerous, as well as expensive.
Lets stick to good ideas that will improve our lot overall instead of ideas that are more pie in the sky.
Small scale is fine for this stuff then we have to find a way of making something as good out of cheap non-poisonous products.
The killer for space-based solar is the
launch cost (until carbon nanotubes make a space elevator possible 50 years after everyone stops laughing).
And yes,cadmium teluride is too toxic to be 'green' (cadmium is roughly as toxic as mercury), and some of the ingredients of CIGS are too rare to support significant capacity there, either, so both of the current thin-film favorites are not good answers!
For ground-based solar, cost will come down FAST for photovoltaics for PEAK power, to below fossil fuels, due to advances in multi-junction cells and concentrators.
After the peak is shaved off, the challenge for terrestrial solar for base-load power is indeed storage.
A bit beyond peak we can use hydroelectric dams as 'storage' (hold back the water when the sun is shining, release it in the early evening to shave the tail of the peak). But natural hydroelectric sites are limited, so this is only a partial answer.
Solar thermal will also come down in cost, but not as fast as PV (although it is starting lower). This will then take care of the evening power, leaving the dams for any remaining fluctuations in baseload.
The other interesting choice for storage is pumped hydro, which has been used for decades to absorb excess power at night and release it when demand is higher. When peak PV equals roughly 1/2 the cost of any given fossil fuel, then PV plus pumped-hydro storage will displace that fuel. If we stop subsidizing coal by letting it dump CO2 for free, then coal goes to roughly 8 cents per kWHr for base-load power,
and PV only has to reach 4 cents per kWHr (and unsubsidized) when the sun is shining to repace even coal for base-load power (the hardest challenge).
This is VERY achievable in the U.S. south west, India, southern Europe, Spain, and western China, etc., which are within power-line distance of most of the world's consumption.
So yes, today solar is not cost effective, but then neither was nuclear until the government poured in tens of billion of research dollars (and back when that was big money, too).
But that does not mean that solar won't be cost effective soon. But we will have to sacrifice 10% of Arizona to power the whole U.S....
To start with: If you open with the claim that this technology could be competitive with coal, you ought to say what the cost of coal sourced power is (I believe a new plant is $2.1 or so/watt). Then, explain what you mean: Coal generation if you're going to build a new plant or compete with existing coal plants (which generate at far cheaper once the cost has amortized).
Then, $1Watt is possible... but I assume they mean the production of the product. Not the shipping to a site, building the site, holding this material in a frame... that's going to be more than the cost of original materials for sure. And onward: No solar technology will work for $1/Watt in an Alaskan winter. Under what conditions will this technology work.
Eh, just sloppy stuff. Hope the tech works whatever it is that the article purports to talk about.
One of the advantages of Solar/Wind is its point of use....
From high orbit your beam will not be very collimated. For starters the sun is not a point source, so you'll need parabolic mirrors to avoid having a target area the size of desert; even that won't help very much as you can't get telescope grade reflectors(too heavy, too expensive) and even just heisenberg's uncertainty principle will give you a large spread over long distances unless you have a gigantic minimum aperture in the optical path.
This is still not baseload power. The amount of sunlight your reflector will be able to send back to Earth will vary depending on the angle between the Sun and the Earth from the perspective of the satellite as well as the angle through the atmosphere at which you send your beam and local conditions(smog, clouds, brushfire, sand storms, broken recievers...).
In order to deal with this intermittency you need many arrays of reflectors in orbits such that they are never coincidentally at their minimum power output, that would imply something like a big ring of reflectors in geostationary orbit.
You will need intercontinental HVDC lines with jaw dropping export capacity connecting each person to at least two geographically separate deserts such that sand storms, clouds, Earth quakes or whatever only VERY rarely will interrupt power output to an entire continent. Such giant export lines will be about the juciest terrorist target ever and enormously expensive.
It took 75 years from the production of the first working solar cell to finding a niché use for them in satellites. It is now 125 years since their invention and their use has expanded to powering pocket calculators, parking meters and generating generous subsidies.
You want to build an antenna measuring between several hundred metres and several kilometers in diameter, _in space_?
I see, you're not talking about anything we can do right now. You're talking about a possible solution many decades or centuries into the future.
Modules have come a long way since then.
Space solar is indeed a long term thing. Economics aside, there is an energy problem: current launch tech uses too much energy so it takes too long to pay this back with the space solar station. Clearly, we don't want to use fossil fuels, so electricity in electrolysers for hydrogen would be the best available tech. What will the EROEI be?
The supposed lack of electrical storage capability is a shibboleth, and makes one wonder if it is just being promulgated by oil and coal interests!
The trick is at least for a start, use realtime pricing to get customers to consume when the power is available rather than storing the power. With todays communications and computer intelligence capabilities it would be simple and cheap to connect each customer's meter to a central market database where prices are posted according to availablilty / demand and suppliers can decide what types of generation they want to build to satisfy it. Needs to make sure all generating sources cover all their external costs though. I've seen realistic calculations showing that eg. a gallon of gasoline or diesel made from ME oil should already be priced in the USA at $8.00 to $11.00 just to cover the military expenditures to ensure its reliability. A recent study covered on this site also has shown that children born to mothers who live near coal-fired electricity plants are born with significantly defective neural systems, all other diffenences controlled.
PHEV's charging their batteries whenever the sun shines would make a nice replacement.