Me thinks you are applying logic to counter an emotional response. Which to me seems like a fail.
]]>The one good thing that does come out of this panic-driven overengineering is that nuclear power plants last a long time -- most 1970s 1st-gen PWR designs are on schedule for a 60-year lifespan absent economic factors like cheap gas and coal being readily available and no-one caring much about CO2 levels in the atmosphere. The two essential components of such reactors, the reactor vessel and the containment structure are about as strong as they were when they were built forty years ago and all the other parts (pumps, steam generators, turbogenerators etc.) get swapped out and replaced when necessary. The control and management systems have already been upgraded to more precise and adaptable digital hardware, of course.
The bad news is that economic markets are defined by the next quarter's financial figures and long-term proposals like building new nuclear plants are denigrated.
]]>Characterising people with a different opinion to yours as "hysterical" is simply insulting. Please stop doing that.
The one thing that has always marked the comments page of Charlie's blog is that people are polite.
You've now driven me off this blog permanently.
Cheers =:)
]]>Right now the 7 billion people on Earth generate and consume about 15TW of electricity, about 2kW per capita on average. First world countries like the US, the UK, France, Germany etc. consume about double that per capita, the ROTW about half. Assume everyone gets to be First World under today's conditions, that's 30TW demand.
Assume that population doubles over 500 years to 15 billion the demand increases to 60TW of electricity. Assume we've run out of gas by that point and have to heat homes using cheap nuclear electricity (as France does today) then add another 40TW. That's 100TW demand, not the Earth-glowing-dull-red electricity consumption figure you postulated.
It's possible, I suppose that every one of those 15 billion people NEEDS 8MW of electricity for... something. Aircon, perhaps but probably not unless CO2 levels really skyrocket. Let's assume that First World electricity consumption doubles too in that period, to 8kW per capita so we can guess that in 500 years time Planet Earth would be generating and using 200TW of electricity.
Typical fuel burnup figures for today's fleet of PWRs and LWRs means they produce about 40MW of electricity produced per tonne of enriched fuel per year, so five hundred years from now based on my figures above that would require about 5 million tonnes of enriched fuel a year. The enrichment process to 3.5% or so for 0.7%-content U-235 that means 25 million tonnes of raw uranium metal per annum need to be mined, extracted or otherwise sourced.
However, the fuel burnup rate I quoted is quite low -- modern operations in some reactors have improved that to about 60MWe per tonne of fuel annually, but since uranium is currently cheap and plentiful the extra efficiency gains don't provide much of a financial benefit. Other non-LWR reactor designs like the EBR-II have demonstrated burnups of 200MWe per tonne of fuel, including using spent and reprocessed fuel (which doesn't need to be mined). Presuming that's a hard but achievable limit and it can be productionised then the demand for fresh stocks of uranium metal 500 years from now drop to about 4 million tonnes per annum.
That's still a lot of uranium. Then again the Solar System has a lot of uranium and shipping a few million tonnes of it Back Home each year would not be beyond the realms of possibility.
]]>Linear extrapolations like that are very misleadng
]]>I was pointing out that your idea made no sense, not that stripping the crust was a realistic option.
Your revision of my estimates still don't give a scenario where seawater can provide enough uranium. As one commenter said about a previous discussion, "holy moving goal posts batman". Despite containing such gems as "Let's assume that First World electricity consumption doubles too in that period, to 8kW per capita" when the energy consumption in the US is already higher and the highest 1st world energy consumption is nearly 3 times that... Then combine that with a 4 fold efficiency increase in reactors (which takes them to over 100% efficiency, not impossible with fast breeders, but still...), Even with all that white washing, you finally realise that there's not enough uranium in the ocean to cover centuries of use as you previously claimed. (BTW Dirk, obviously the 5 times as much thorium isn't going to last "a few thousand years") Then you get to the "holy moving goal posts", We'll get it from spaaaace.
"...the Solar System has a lot of uranium and shipping a few million tonnes of it Back Home each year would not be beyond the realms of possibility."
Well unless your plan it to drop the uranium into the atmosphere at 20km/s, turn it into dust, let it dissolve in the ocean and extract it later (you don't have to be "hysterical" to think that's a bad idea), you're going to need a re-entry vehicle. The Dragon capsule can return nearly 3 tonnes, but weighs 4 tonnes. So we can say that the re-entry vehicle must weigh at least as much as the payload. That's got to be the lower bound since it neglects the first and second stages needed to get the capsule into orbit and neglects the fuel needed for changing orbital height, inclination and de-orbit. Probably at least 2 orders of magnitude better than reality.
So we'd need to put 4 million tonnes in orbit every year. (assuming we could do that 100% efficiently compared to our current efficiency of far less than 1%). 1/2mv^2... 1/240000000008000*8000 is 1.28x10^17 J. Divide by the number of seconds in a year and you end up with 5.4x10^14 Watts.
So even assuming orbital launch is well over 100 times more efficient than anything we could do today, that's still over 500 TW, or 2.5 times the amount of energy you were planning on generating by bringing that 4 million tonnes of uranium to Earth. I haven't even bothered to add in the energy needed to move the uranium from the asteroidal refinery or decelerate it from escape velocity down to orbital.
Uranium from Spaaaace isn't the answer.
]]>I made a basic maths error.
Bringing uranium back from space is somewhat energy positive (but only just)
]]>Both you and Nojay are confusing total primary energy use with electricity use. Divide the numbers in this table by 8760 (24 * 365) to get continuous electrical power per capita: http://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC
According to the latest data in that table Norway is the champion of electricity use at 2.6 kilowatts per capita. The USA is at 1.5.
When it comes to primary energy consumption per capita, the USA has been stagnant to slightly down since the 1970s. The USA has also been at below-replacement total fertility rates since 2007, joining other developed countries who got there some years or decades ago. Contrary to the hopes of economists, human demand for economic activity is not infinite and exponential growth does not go on forever.
]]>As for low-energy concentrating, I'd find an organism (probably a bacterium or possibly a plant) that concentrated uranium, grind up the asteroid, and feed it into a bioreactor. That's a really slow way to do it, but on the other hand, it might conceivably give you some useful uranium in a few years, a decade or two at the most...
]]>I'm a huge space fanboy, but I don't think I can make a case for colonising planets based on Uranium mining. The concentration of U into the crust seems to depend on plate tectonics. There's none on Mars. Venus seems to have (or had) a form of plate tectonics and may have concentrated U in the crust but has no weathering that would lead to ore deposits. There's also the matter of delta V for return of uranium from the surface of Venus to Earth. Despite my incredible mistake in maths in the previous post, it's still pretty marginal returning U from space if it's actually in the form of ingots floating in low earth orbit. Even if U was in the form of ingots on the surface of Venus... You'd need to launch a vehicle from here, that could survive re-entry into Venus then have enough oomph to punch through 90 atmospheres of gas, then get to orbit then get to Earth, then be de-orbited. Rockets work slightly better in a vacuum, but much worse under 90 bar of 750 K CO2. The atmospheric pressure on Venus is so high that gas would flow into the combustion chamber of a Merlin engine rather than out.
So even assuming we can engineer a magic bacteria that lives at 750 K and makes U into ingots for us to collect and carry back, that's still pretty hard.
11 km/s to get out of Earth's gravity (payload under 1%). 7 km/s to get to Venus (payload about 5%). Re-entry and associated shielding (payload about 10%) at Venus unless you do a powered decent, then that's 10 km/s (payload 1%). 10 km/s to get out of Venus's gravity (Payload 1%, assuming you can figure out how to get out of the atmosphere first which I can't). Another 7 km/s to get back to here (payload 5%) followed by a re-entry at 11-12 km/s (think Apollo rather than Dragon, payload maybe 10% of the re-entry vehicle mass, rather than 40%). If you multiply all those together you get 0.00000025%
Even assuming some sort of magic drive and being wildly optimistic, the returned payload couldn't make up more than 0.001% of the launch mass. The only even roughly equivalent sample return mission to date brought back one speck of dust that was visible to the naked eye despite launching over 150 tonnes for a return ratio of 0.000000003% and that didn't include matching orbital speed, landing or take off. So for 4 million tonnes returned from the magic Venus bacteria you need to launch 400 000 million tonnes of rocket every year. Even if Mars had plate tectonics, the figures are hardly any better. There's a 5 km/s improvement for escape velocity and no atmosphere problem on launch. The other figures are about the same.
Ceres (basically all the asteroids) works out slightly more delta V than Mars. 10 km/s each way from LEO. That's about the same energy requirement as launch from surface to LEO both going there and coming back. Each change in Delta V of 10 km/s seems to end up with about 1% ending up at the final velocity. So that's 1% of 1% of 1%. Then 10% of that for the re-entry. About one 10 millionth of the launch mass is returned as payload. You could do better with solar sails, but then missions are multiple decades long and there's no uranium there anyway. Solar sails might help between planets, but they don't solve the launch from Venus's surface problem which seems insurmountable to me.
]]>At this point only nuclear electricity is a "primary" energy source that has to fill in for everything we do now, including synthesising liquid fuels for aircraft, making fertiliser etc. Yes it's not cheap but it will work and it won't double the CO2 levels in the atmosphere.
At that point the electricity demand worldwide could well be 8kW per person. It's a ballpark figure -- 20kW is possible if we go for intensive greenhouse agriculture or direct synthesis of foodstuffs or indulge in other energy-intensive processes like direct decarbonisation of the atmosphere. I don't see the consumption as being 8MW per capita though which a steady 2% annual growth figure would suggest.
]]>First off, the arguments that population will keep expanding and energy use per capita will keep expanding so we have to be able to meet those energy goals is bogus. Population will not keep expanding. The question is whether we can keep it from crashing to a new, much-lower carrying capacity, and whether we can avoid extinction in the next thousand years or so. We will not keep increasing energy use because without new technology we don't yet know how to get, we can't.
There is lots of uranium. As soon as the USA gives up the idea that it's more important to stop other nations and terrorists from getting nukes than to have lots of nuclear power, we will build breeder reactors to use that uranium. And it's a lot cheaper to make plutonium than it is to purify U235. Plus there's something like 3 times as much thorium. Maybe part of the reason the USA has ignored thorium is we don't like its distribution in the world.
If we can't get anything better soon, and nuclear reactors give us 100 years breathing time to find something better, we might have to do that.
To my way of thinking, there are two big issues with nuclear power. The smaller of them is the scale. Given the problems of nuclear contamination, there's mostly no point building small nuclear reactors. They need to be big, with a giant slow construction process and giant corporations to run them. Or if somehow they don't need that, they will get it anyway. These are not the circumstances that give us our best innovation. It will promote the sort of ownership problems we have been having in other contexts in recent years. TBTF, etc.
The bigger problem is nuclear pollution. Currently we're averaging one medium-size nuclear accident every 25-30 years. Fukushima. Chernobyl, etc. With 10 times as many reactors we can expect a medium-size nuclear accident every 2.5-3 years. With 100 times as many reactors that would be one every 3-4 months.
We have never yet had a big nuclear accident so we have no track record at all about them. We would have to find out how often they happen and how bad they are.
One good thing about having a lot of reactors and a lot of accidents is that we would quickly learn how to build them better, and operate them better. But we need to factor that into the expense. Also we can't get significant private insurance for nuclear reactors because there isn't enough experience with them to figure out what the rates would be. Once we have had enough expensive accidents then private insurance will become available to some extent.
We would develop that experience quicker if we built lots of small nuclear power plants instead of a few big ones. We would quickly learn how to build them cheap and reasonably safe, and operate them safely. But we won't do that. Maybe just as well, a big accident at a small nuclear power plant might turn out worse than we can handle, just as much as a big accident at a big plant.
There are people who claim that "radiation hormesis" will mostly prevent problems from nuclear contamination. They think that small amounts of nuclear radiation is good for you. So even when we have hundreds of nuclear accidents, none of them will actually cause trouble beyond, say, a 10 mile radius. We would get a big break if that's true. But the evidence is not there. They claim there's plenty of evidence and the government is suppressing it. Maybe. We don't know. If they're wrong, we really can't afford to have radioactive stuff getting bioaccumulated through our crops on top of everything else.
It's just too risky. I don't want to try large-scale nuclear power until I'm sure all the other choices are worse.
]]>