Solar Power

[DonPMitchell]

Sunlight carries 1400 watts per square meter at the Earth's surface, which is really quite a lot of free power!  A problem is to convert that to another form like electricity.

Biofuel depends ultimately on photosynthesis, which is remarkably inefficient.  Photosynthesis only converts 1 percent of sunlight into chemical energy!  Conventional solar cells are about 10 percent efficient at coverting light into electricity.  Solar cells are useful but rather expensive.

The hot new topic (literally) in solar energy is Concentrator Photovoltaics (CPV), where two important ideas are at play.  First, mirrors are cheaper than solar cells, so it is cost effective to focus sunlight from a large area onto a small cell.

Also exciting is the development of new triple junction solar cells that are 40 percent efficient!  These are based on materials like Indium, Gallium and Germanium rather than silicon.  They also work best when they are hot, so they are perfectly suited to use in solar concentrators.  People are saying that this may be the first economically cost-effective technology for generating electricity from sunlight.

Work on CPV is happening all over today.  Here are a few sites:

http://homepage.mac.com/nekins/syracuse/index.html
http://www.parc.xerox.com/research/projects/cleantech/cpv.html
http://www1.eere.energy.gov/solar/pv_sys_concentrator.html
http://www.boeing.com/ids/news/2006/q4/061206b_nr.html

[Johno]

Don, where did you get your numbers?  It's just that I'm covering energy flows in ecosystems with my year 11 biology classes, and I'd love to be able to point the kids in the right direction for efficiency figures.

[jdbenner]

Sunlight carries 1400 watts per square meter at the Earth's surface, which is really quite a lot of free power! 

Don, 1400 watts per m^2 may be a lot of power but the average power output is less than 1/3 (or 1/2, if the panels track the sun) of that due to the rotation of the earth, and weather.  Also the latitude will effect the foot print of a solar power station, although not necessarily the collector aria.

So for a near equilateral power station with 50% conversion efficiency, and average power output of 500Mw.  (1400w/m^2)*0.3 * 0.5 = 210w/m^2
(5*10^8w)/(210w/m^2) = 2.4Km^2

2.4 square kilometers is doable, but imagine the environmental impact, and how many people will be required to clean and repair the collectors?

Another issue is energy storage or long distance transmittion of power.  People will need power at night and usually not in an equilateral dessert. 

Your figures for photosynthesis are correct for the growing season, but plants in some climates don't grow all year long.  What bio fuel does have going for it is the fact that we do not need to manufacture solar collectors and that it automatically provides portable long term energy storage.  A fact useful to both transportation, and to a caveman warming himself by a fire on a cold winters night.

[DonPMitchell]

Another issue with biofuel is the "1 %" efficiency of photosynthesis is compounded with other factors that knock down the efficiency of using plants for energy.

1. The energy costs of cultivating and fertilizing fields.
2. The limited fraction of plant biomass used for fuel.
3. The seasonal and latitude/cosine factors that also apply to solar-electric power generation

The 40% figure for multi-junction solar batteries does not have the first two problems.

Another important issue with biofuel is the social problem of converting the production of food into the production of fuel.  What happens to countries that depend on food exports from America?

Freeman Dyson proposed an interesting but far-out idea -- what he calls "black plants" -- bioengineered to have 10% efficient photovoltaic metabolisms in place of natural photosynthesis.

[jdbenner]

Don, you are right about the limitations of bio fuel. We certainly can't sustain the present world population on bio fuel.  However, bio fuel is the second most common solar energy source used by man at present ( Hydro power being the first, and not counting fossil fuels).

I believe we would be better off if we put all the "alternative energy" research money in to Breeder reactors.  But I don't controle Federal funding

[DonPMitchell]

Breeder reactors seem like a good idea to do.  Certainly it is wasteful to just burn U-235, when there is a way to use the more common U-238 (via conversion to plutonium).

Uranium is still a rare element,  It would be interesting to know what the fuel supply is.  I've heard people claim that there is about a 200 year supply of nuclear fuel in the crust.

[Johno]

That would almost certainly mean Uranium full stop.  U-238 is largely useless of itself, although in a breeder reactor it can be turned into Pu.  There's relatively little U-235 in any sample, but it is (if memory serves) reasonably consistent.

The galling thing is that we discovered the benefits of nuclear power when (geologically speaking) the available fuel was largely gone.

My suspicion - based more on history than science - is that since we've already put so much time and energy into fusion, sooner or later it's going to pay off.  Historically speaking, though, developments like that usually take a war or other major crisis.  Would global warming qualify as a big enough crisis to stimulate the development of this source?

[Satanic Mechanic]

Been gone most of this week due to work.  Uranium is very plentiful, you just have to process it which is costly.  Most of the time if you want one pound of U-235, you have to process 40,000 lbs of inert Uranium-238. 

SM

[DonPMitchell]

I poked around a little.  Let's say there is a 3.5 million ton supply of Uranium then.  A typical nuclear power plant consumes about 30 tons of fuel per year, and there are currently 442 power plants in the world.  So that's about 13,000 tons per year; however, that fuel is only slightly enriched and not completely consumed before it has to be removed form the reactor.

Fissile isotope U-235 is 0.7 percent of uranium, so there is a total world supply of about 25,000 tons.  One article I found claimed that the estimated world supply will fuel existing reactors for 50 years.

So presumably we must use breeder reactors or we will run out of nuclear fuel quite soon.

[jdbenner]

Does U in this graph means Uranium 308? or does it mean something else. Can either yourself or Joshus verify that?

I don't know what 308 means but since it is 70 AMU larger than U238 it can't be a uranium isotope.

For a breeder reactor Thorium is a more abundant fuel since neutron bombardment can turn it to U233 which is fissionable.  

As far as suplies go:
1 most companies stop investing in exploration when they have found enough for the present and immediate futcher.

2 Ore is understood to mean, minerals from which it is economically reasonable to obtain suplies.  This means that what is defined as ore can change.  Gold is now mined from the tilings of old abandoned Gold mines.  It is now economical due to the fact that most "high grade" ores are gone, and refining technologies have improved.  Ruby is a form of aluminum oxide but it would be crazy to call it aluminum ore.


As fare as fusion goes.  It will be cleaner than fission, and since deuterium is found in water at a nearly constant concentration, rather than in various concentrations, in various minerals with various distributions, AND since we know the size of the oceans we have a good lower bound of the available resource.  If I remember millions of years supply at present levels of energy demand.
But, Fusion will require much more work before it becomes practical.  Breeder reactors provide a proved technology capable of suppling us for thousands of years.  As the proverb goes, A bird in hand is worth more than two in the bush.

[jdbenner]

Science News magazine Vol 171, NO. 5 has an excellent article on bio fuel. Specifically ethanol from corn.  Among outher things, the article warns that if we increase ethanol production to the level proposed in Bush's plan would require half of our current (2008 projections) corn production.  Food prices could rise to record levels.

Here is a link to the on-line version of this article
http://www.sciencenews.org/articles/20070203/note14.asp

[DonPMitchell]

I forgot about Thorium, yes that is very promising.  India is currently making an effort to develop Thorium breeder reactors, and the USA has built some experimental ones.  I have also heard that Thorium is safer in terms of weapons proliferation, since it does not produce plutonium, and I guess U-233 is not easy to weaponize.

Nuclear may be cheaper than solar just in terms of infrastructure costs.  People have built 1000 megawatt nuclear plants, but a 1000 megawatt solar plant would have to be several square kilometers in size.  Not practical unless people figure out ways to make it very cheap.

Ethanol from corn may be mostly a scam by corn farmers to get tax money thrown their way.  Only schemes that can utilize the cellulose are even remotely sensible, and I don't think that techology exists yet.

[jdbenner]

Cellulose in the form of home heating firewood is the most used bio fuel at present.  However cellulose to ethanol is not yet a mature technology. 

[DonPMitchell]

Burning wood also powered the early years of Britain's industrial revolution -- until the nation was virtually denuded of forrest and switched to coal.  Can burning wood be a sustainable energy source on a large scale?

[jdbenner]

Burning wood also powered the early years of Britain's industrial revolution -- until the nation was virtually denuded of forrest and switched to coal.  Can burning wood be a sustainable energy source on a large scale?

No, of course not.  Assuming we have the same definition of "large scale".