Monday, April 17, 2006

Is Nuclear faster?

LeoPetr, in comments:
Yes, but we can get nuclear up faster than we can get solar or wind. The manufacture of solar and wind plants requires a much larger use of oil than does nuclear, and thus results in a lower return on energy invested. Replacing both coal and nuclear with wind at the same time is a tremendously expensive and difficult option. Getting enough wind up so that we are able to make more wind plants using only wind power is the big hurdle. I think we should be building both for the time being.
Well, first of all I wouldn't advocate trying to replace coal and nuclear at the same time. My ideal policy would phase out coal first (for environmental reasons) then natural gas (for economical ones) then nuclear, while building up wind, solar, and small hydroelectric plants. I don't even particularly need to see nuclear plants shut down before their operational lifespans are up - most of Ontario's plants will need to be shut down by 2020, so there's no need to hurry them out the door.

(Parenthetical: While I'll be writing mostly about wind here, I should point out that any jurisdiction you could name - including Ontario - could easily conserve and "efficient" so much as to make any new generation, icluding nuclear, unncecessary for the time being.)

However, it's a mistake to say we can build nuclear plants faster or cheaper. The last nuclear plant that Ontario built was Darlington, which was literally seven times more expensive, and years later than original projections. In the end it cost almost $15 billion, for 2.4 gigawatts of electricity, or $6.25/watt. This is well above what wind costs today, and only slightly lower than solar costs. We have seen, over and over, a history of cost overruns that are second only to the Pentagon in terms of waste. People of my generation - who had no say in whether or not these monstrosities were built - will nonetheless be paying for their construction for the rest of our lives, thanks to the massive debt that was incurred.

Meanwhile, Spain and Germany are each building one Darlington's worth of wind power every year. The latest plans for Darlington include adding two more 700-megawatt reactors - something even the plans proponents concede will take at least a decade. In the same time, Spain and Germany will have each built 20 gigawatts of clean, radiation-free electricity. And this assumes that their construction rates don't accelerate further, which would defy recent history - wind and solar have been growing exponentially at double-digit rates.

The US, thanks to the recent renewal of the wind-power tax credit in Congress, is seeing a major resurgence after two years of doldrums. And before you scream "subsidies are evil!" consider that the nuclear industry (in both the US and Canada) gets the largest subsidy of all - public insurance. In Ontario, the nuclear liabilities act limits the damage to the nuclear industry to a paltry $75 million, with taxpayers picking up the rest. If anything ever goes wrong at Pickering, you can bet it'll be worse than $75 mil. (Ironically, the American government provides far broader and better insurance to the nuclear industry than it does for the health care of it's own people.)

So, in sum, nuclear isn't cheaper, isn't faster, and we shouldn't be building more of it. However, LeoPetr added:
Citing ultracapacitors as something we should bet the pot on reduces your credibility because they are not [sic] after all, I could be arguing that we have become spectacularly better at reducing pollution at nuclear plants thanks to recent advances, that we could be disposing of waste in the same mines we dug it out of, and that breeder reactors are much more fuel efficient and thus produce far less waste than conventional ones.
There are two different arguments going on here, and I'll try and address them separately:

First: Do ultracapacitors lessen the force of my argument? Fine, we don't need them. There are more than enough electrical storage technologies out there - lithium ion batteries, flow batteries, even fuel cells (though I'm not a fuel cell fan) - that we don't need to pick one technology. However, given that EEstor is going in to third-party verification this summer, I don't think it's unfair to suggest their technology is realistic in the short term.

Cheap electrical storage will revolutionize the electrical grid, however it arrives. We currently need to over-build the grid by more than half because of the need to generate peak electricity. Stored electricity could be generated overnight, lowering costs. It could be stored where it is used, meaning less strain on the grid. Even if EEstor's ultracapacitor's don't pan out, this makes too much sense for either the market or government to ignore. Storage even makes nuclear more attractive, while the plants still operate.

Secondly, advocating for the construction of breeder reactors is dangerous. I know of no breeder design that doesn't create weaponizeable by-products. They're more fuel efficient, but the increased risk of terrorism, or simple proliferation among nuclear states, far outweighs the efficiency. Consider that Japan is only about a year away from having a nuclear bomb at any time, thanks to its breeder reactors.

And - this is important - if we don't rely on breeder reactors, than the energy-efficiency argument for nuclear starts to go out the window. Unless we rely on the most fuel-efficient (albeit dangerous and proliferating) breeder designs, we start to see a shortage of fuel-grade uranium ores before the end of this century.

I literally do not know of a single good argument in favour of starting new nuclear construction. This isn't to say that I believe all existing plants should be mothballed right away, or anything so dramatic. I'll even reserve judgement (for now) on the new reactors they want to add to Darlington, because Darlington is basically half-built anyway - there's room for those reactors within the existing plant. But no government should be considering nuclear construction when cheaper, faster ways of spending taxpayer money exist.

In short: If we need electricity now, then nuclear won't do because we can't have it for ten years time. If we need it in a decade from now, than we might as well buy the wind and solar five years from now, when they'll both be cheaper than nuclear.

4 comments:

Mark Richard Francis said...

Other points to add:

Nuclear plants in Ontario have only something like a 65% uptime. That's worse than wind power.

Uranium is limited, and will get more costly as many countries start using even more nuclear power generation.

Transmitting power a long way is expensive and results in greater inefficiency as there is power loss over the lines.

Nuclear generators take a long time to restart (days, not hours) - at least the ones in Ontario do.

Alternative power sources are: tidal, hydro, wind, solar, biomass AND conservation.

Biomass plants are fine (to a point) if they respect the carbon cycle.

Declan said...

I agree with your post. One question though, you don't mention large hydroelectric plants.

Is this because you consider them a worse option than nuclear / coal / gas, or just because you figure we don't need them if we build enough wind/solar/small hydro/conservation?

AJSomerset said...

Which is more environmentally sound: one large hydroelectric plant, screwing up one large river, or two dozen small hydroelectric plants, screwing up two dozen small rivers?

A lot of people (not necessarily here) treat hydro as benign, and it ain't.

Anonymous said...

I'm interested in the "large hydro" question as well. I assumed you omitted it because we've already tapped all the best/most immediately available sites for large hydro plants, but I have no reason to actually assume that's the case. So, uh, why?