October 15, 2007
Solar Energy Seen As Really Plentiful

Engineer-Poet explains we will have huge amounts of energy available once photovoltaics become cheap.

Annual energy consumption of the USA is about 98000 kWh of primary energy per capita.  A square meter in the middle of Kansas receives about 1550 kWh of solar energy per year, so an American's consumption represents about 63 square meters of Kansas.  300 million Americans would need about 7300 square miles out of the 81,815 square miles of the state.  Even if you reduced efficiency to 10%, you wouldn't need the entire state.  We probably have enough area under roofs and roads to do the job already, no further development required.

We have PV made of silicon (27% of Earth's crust) and PV made of organics (representing carbon, possibly reclaimed from the atmosphere) on the way.  Carbon nanowires are already better conductors than copper.  Technology inevitably pushes to the limits of science (just compare the 14-inch Winchester disk drives of 3 decades ago to the one in the iPod).  The science we have today is enough to supply an American level of comfort to billions, albeit using renewables rather than fossil fuels.

E-P thinks he knows a way to extract silicon for photovoltaics at a much lower cost. Not sure he's right about that. But I agree with him that it is a solvable problem.

Our problem is not a general energy shortage. What we are hitting is a liquid energy shortage. The development of technologies to allow electricity to substitute more for liquid fuels will allow us to move past the liquid fossil fuels era and enjoy rising living standards. But we might go through a painful transition before the batteries and other elements of our more electrified society come together.

Share |      Randall Parker, 2007 October 15 09:53 PM  Energy Solar


Comments
Atrox said at October 16, 2007 12:31 AM:

Will airliners fly on batteries? Without sustainable biofuels for jet fuel and long haul trucking, we're screwed.

HellKaiserRyo said at October 16, 2007 1:03 AM:

At least, quotidian cars wouldn't run on gas, but on solar power. That would at least make fossil fuels exclusive to those means of transportation.

russ said at October 16, 2007 4:12 AM:

But how long will it take to roll out this technology? Do we have the decades it will require?

Wolf-Dog said at October 16, 2007 6:12 AM:

If you build 200 additional 1000 megawatt nuclear plants to charge all the 300 million electric cars in the United States, then you will have plenty of surplus oil to use as jet fuel. Or you can even use a small fraction of the nuclear electricity to make liquid hydrogen as jet fuel.

Since a 1000 megawatt costs approximately $1.5 billion to build, building 200 additional 1000 gigawatt reactors would cost less than $300 billion, or 2/3 the cost of the Iraq war (so far.) But if you build only 20 reactors per year, within 10 years you will have built all the necessary 200 reactors to charge 300 million cars. Building only 20 reactors per year is $30 billion per year, which is 10 % of the current $300 billion annual government deficit spending. By 2017 I am sure that cheap 50 mile range plug-in hybrids or slightly more expensive pure electric cars with 300 miles per charge will be reasonably popular.

Wolf-Dog said at October 16, 2007 6:15 AM:

Sorry in the second paragraph I wrote "1000 gigawatt" reactor instead of "megawatt reactor".

David Govett said at October 16, 2007 8:49 AM:

Eventually, we could use dark energy to power everything. If we could only find it...

MG said at October 16, 2007 10:21 AM:

"But how long will it take to roll out this technology? Do we have the decades it will require?"

Your conceptual block is "we". "We" isn't necessary for "you" to roll out the technology for "you".

Once the currently available advanced, mobile electrical storage technologies are put into mass production (within the next five years), you can roll out this technology yourself, whenever you decide to buy a vehicle that uses them.

The solar electrical collectors already exist. As soon as you choose to buy and install them, you can roll out this technology. Ditto for solar thermal collection.

Not everyone will be able to afford the cash required to roll out these technologies when "you" are able to. But that's okay. When they can, their technology options will be superior to the ones you selected.

Brett Bellmore said at October 16, 2007 10:35 AM:

A friend of mine with some experience in solar says E-P's numbers are extremely dubious, in that the available radiance under real world circumstances is far lower than he's basing his calculations on.

Of course, this doesn't go to the merits of using mine wastes as a source of silicon...

Paul Dietz said at October 16, 2007 12:01 PM:

Of course, this doesn't go to the merits of using mine wastes as a source of silicon...

I boggle a bit at that concept. Silicon is ubiquitous on Earth, second only to oxygen in its abundance (by mass). You'd get it from the source that is overall least expensive to process, which probably means some exceptionally pure silica sand. Not that the raw mineral is itself a significant contributor to the cost of the final product, mind you.

Brett Bellmore said at October 16, 2007 3:54 PM:

Yes, it's true that silicon is common, and the cost of sand to make the silicon in a solar cell is minimal. But reducing and purifying silicon from silica is no easy or inexpensive task, and if producing it from sodium fluorosilicate is cheaper, this would indeed impact the cost of solar cells.

The process described in the link sounds like kitchen sink chemistry. (Yes, you can make metalic sodium in the kitchen.) Refining silicon from silica is definitely in the "Kids, don't try this at home!" category.

Engineer-Poet said at October 16, 2007 10:46 PM:

Quoth Atrox:

Without sustainable biofuels for jet fuel and long haul trucking, we're screwed.
The USA got along just fine before it had jet airliners, or even before the DC-3.  Carbon-neutral fuels will probably price most users out of air transport and move them to high-speed rail.  Fast rail is electric-powered:  TGV, Shinkansen, Acela.

Long-haul trucking is an abomination, not a necessity.  It is exactly the sort of freight transport which is best moved to rail:  heavy loads moving long distances over fixed routes (Interstates, which have medians which could take rails).  I've watched 7 container trains travel the same tracks between Amarillo TX and Clovis NM in an afternoon; there is no question that we can do this.  Freight rail can be electrified also, eliminating the need for liquid fuel.  The final leg from rail terminal to destination can be handled by biofuels or electrics, e.g. ZEBRA sodium nickel chloride batteries.

Quoth Brett Bellmore:

A friend of mine with some experience in solar says E-P's numbers are extremely dubious
Strange, because I got them from NASA's solar-irradiance web app.  1550 kWh/m²/yr is only 177 W/m² average (you have to specify the mid-Kansas location).  If your friend has better data, demand it from him and then tell us all where to find it.

Brett Bellmore said at October 17, 2007 3:57 AM:

I believe it was the phrase, "At the standard 1000 W/m irradiance" he took exception to. It may be the standard for testing solar cells, but he said it's utterly unrealistic as a working irradiance for determining their output through most of the US, and the way it was used was a bid deceptive. And directed me to this map:

http://upload.wikimedia.org/wikipedia/en/thumb/2/2c/Us_pv_annual_may2004.jpg/773px-Us_pv_annual_may2004.jpg

With regards to long haul trucks, would you be interested in dedicated truck lanes, with some provision for supplying the trucks with power? Not quite as energy efficient, but logistically easier.

Engineer-Poet said at October 17, 2007 3:43 PM:

The standard for testing is just that, for testing.  It's what the capacity factor figures use as a reference.  Actual instantaneous production can be higher, such as a panel on a clear sub-freezing day receiving extra light reflected off snow.

We have dedicated truck lanes.  They're called "rails".  They allow total route separation, grade separation, and much lower rolling resistance and total energy consumption than rubber-on-pavement.  They also provide a free current return path for overhead electric supply, cutting the cost of electrification in half (or better).  If we need to package freight in smaller units than entire trains, there is the Bladerunner concept which achieves many of the advantages of rail in an intermodal vehicle.

Randall Parker said at October 17, 2007 6:19 PM:

I agree with E-P that we do not need a lot of liquid fuels for long haul shipping. More shipping can shift to rails as oil prices rise. Also, rails can shift to electric once the price of oil gets high enough.

This will raise shipping costs to places further away from rail lines. But market prices will tell people where to move in order to reduce their energy costs.

What I want to know: At what price of oil does the electrification of various rail lines become cost justified?

Engineer-Poet said at October 17, 2007 9:49 PM:

IIRC, Alan Drake says it's justfied now, if you eliminate the property taxes that railroads have to pay on such improvements (and freeways do not).

But check his words, because my memory might be wrong.

Nick said at October 20, 2007 6:48 PM:

"This will raise shipping costs to places further away from rail lines. But market prices will tell people where to move in order to reduce their energy costs."

Nah. You were right the first time: The use of PHEV/EV trucks for short range transportation will be much cheaper than moving.

The suburbs may be a sterile place to live, but they're not threatened by energy problems.

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

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