For the record

Another quick post on nuclear power, probably the last for a while. Most of my discussion about nuclear power has been on the question of whether expansion of nuclear power is, or is likely to be, a cost-effective way of reducing CO2 emissions. The answer, as revealed by the failure of the heavily subsidised “nuclear renaissance” in the US, is “no”. But, for the existing (mostly Generation II, see over fold) plants, there’s a separate question – does it make sense to close them down early, or, alternatively to seek to extend their lives.

Since this issue comes up a lot, I thought I would state my position clearly. Nuclear power is an almost exact substitute for coal, has no CO2 emissions and (except where particular vulnerabilities have been demonstrated) comparable or lower health and safety risks (these numbers can be played with in various ways). The marginal cost of generating power from existing plants is low. Problems like waste disposal will have to be addressed anyway, and a few more reactor-years worth won’t make much difference.

So, except where there are particular vulnerabilities that are too costly to repair, I favor keeping existing plants open as long as they can be kept in good repair.

A quick typology of nuclear plants (cribbed from Wikipedia, where you can get more detail if you want)

Generation I refers to the first experimental plants, now all retired
Generation II refers to designs from the mid-60s to about 1990. Nearly all operational nuclear plants, including some now being completed after long delays, are in this class
Generation III refers to designs with improved safety features and other advances, developed from 1990 to the early 2000s. Virtually none of these were built
Generation III+ refers to the current state of the art for plants now being built. In practice, the Westinghouse AP-1000 is the only serious contender (some prospect of 20+ being built)
Generation IV refers to hoped-for future evolutions on this path

Finally, there are various vaporware proposals, such as thorium reactors, the Integral Fast Reactor, small modular reactors and so on. None of these has got past the prototype stage, and most not even that far.

44 thoughts on “For the record

  1. Hermit :
    I think we could see an explosion (in the metaphorical sense) in small modular reactors after 2020.

    Highly unlikely. The most developed design (the B&W mPower) has a planned first deployment date of 2022 which, based on past history, will probably be pushed back. And so far the only company that has shown any real interest in building the proposed design is the state-owned Tennessee Valley Authority. The other two designs that have some viability (the NuScale MASLWR and the Westinghouse SMR) are much less developed and don’t have firm customers. And with all three, the main advantage is the capital cost. They are not “advanced” reactors; they are PWRs with a uranium fuel cycle, which means they will face the same fuel costs and disposal problems that large reactors would face if either gets built at a higher rate. But a lower capital cost is a benefit that is entirely unproven at this point, so private companies won’t sign up to build them until enough are built for the cost reduction to be proven, leading to a catch-22 sort of problem. Short of a collapse of natural gas production, SMRs are highly unlikely to be deployed en mass. I suspect the reactors at the TVA plant will be the only ones of that particular design constructed. NuScale might get lucky and get enough funding to build one at Hanford or INL, but no more than that because the Northwest has plenty of hydro and wind resources and doesn’t need nuclear. If the Westinghouse design ever gets built it will be in China or some country on a nuclear ego trip. (Brazil? Indonesia?)

  2. Solar PV demand is going through a bumpy phase right now, as Germany backs away from generous incentives. But production capacity is still growing – could easily approach 100 GW a year in the next few years.

  3. @PeakVT
    You touched on the key variable of the gas price. South Australia’s biggest project the Olympic Dam expansion was mothballed to due to high ‘input costs’. One component was the construction of a 250 MW gas fired power station at OD with other energy inputs such as coastal desalination to be supplied by the state grid. That’s a perfect candidate for SMRs either air cooled at the mine or with the desal on the coast. Store the ‘spent’ fuel in disused parts of a uranium mine if that’s acceptable.

    There’s a nuclear irony in that a key driver of the increasing gas price is Japan importing over a third of the world’s LNG (note all LNG is exported, domestic gas is piped). So well before 2020 there will be reluctance here to fire up more gas plant due to expense. The Germans twigged to this early which is why they are building more coal fired capacity. Therefore I see no proven alternatives to SMRs for low carbon dispatchable power. If this is correct it could turn into a technology export bonanza for the US.

  4. Hermit, Olympic Dam was mothballed because BHP chose to (stupidly) put its money into US shale oil and gas plays instead. Low expectations on returns were due to declining commodity prices – including uranium – not the cost of gas, or a gas-fired power station. Wouldn’t BHP have just been piping its own gas to the station anyway? I’m sure it could have managed to come to a satisfactory price agreement with itself.

  5. @Hermit
    “If wind and solar were up to the heavy lifting task we’d already see smelters and affordable SUVs powered by them.”

    In the past, renewables have been more expensive than fossil, hence they made up a small fraction of the energy mix. There is reason to believe this situation will be reversed in the future. Nothing you’ve said here or anywhere convinces me that renewable intermittency is an insurmountable problem.

  6. @ Nick
    The OD expansion plan involved building a 400km gas pipe that originates at Moomba run by Santos, not BHPB. Santos have just renewed old gas contracts that were $4 a GJ at the new price of $9. Couple that with pending CSG drilling restrictions and SE Australia will face a gas shortage, fracking or otherwise.
    @ Sam
    The day an aluminium smelter gets more than a few percent of its power from wind and solar I’ll recant. If the intermittency problem is easily solved how come Germany is building new coal fired power stations? Google the Bloomberg articles on Germany and coal.

  7. The nuclear industry has had the ability to build small reactors for a long time. The first reactors were quite small and small reactors have been built for ships, submarines, satellites, and some were even built that were intended for airplanes. So it’s been possible to build small nuclear reactors for a very long time. Economies of scale, which are supposed to make small nuclear reactors competitive, have also been known about for a very long time. So if small nuclear reactors will be cheaper to build than large nuclear reactors, then the nuclear power industry has really been very very dumb indeed to continue building larger and larger nuclear reactors through the 50s, 60s, 70s and right up to the modern day, when they could have instead been building smaller, presumably cheaper reactors. It’s really hard to grasp how phenomenally dumb they have been. Think of all the countless billions of dollars they left on the table simply because they were too dumb to build small nuclear reactors instead of building gradually larger and larger reactors as they actually did. Really astoundingly dumb. A Brobdingnagian dumbness of immense proportions. And you know what? I don’t want people that dumb building any sort of reactor at all.

  8. Sorry Hermit, I didn’t realise you’d replied. See:

    “The natural gas required to operate the CCGT plant would be delivered to Olympic Dam from the Moomba hub via a new gas pipeline. The natural gas supply may come from one or several of the major production wells connected to the Moomba hub (i.e. the Gippland Basin in Eastern Victoria; the Otway Basin in Western Victorial; the coal bed methane fields in Queensland (see Figure 5.40)).”

    I’m pretty sure BHP has interests in all of those locations.

  9. Ronald, on technological developments, something I meant to post the other week. Did you catch that graphene was recently given a one billion euro research program by the EU? Such a fascinating material, and widely regarded now as having the potential to be more important to the world than plastic in the 20C.

    One property it has is that it will allow water and water vapour/gas to pass through, but almost nothing else. Applications for coal and gas flue filtering immediately come to my mind. Osmotic power is another one.

  10. @Nick
    Nick, gaphene certainly seems to have a lot of applications. People are looking into using it for both solar cells and for supercapacitors which means it could be used to both gather and store energy. And with one nanometre holes poked in it it can let water molecules through but not chlorine and sodium ions. (Actually, I think the sodium ions can go through, they just don’t like it on the other side as it’s a water fest, so they head on back over to hang with the chlorine ions.) Graphene might be to the 21st century what low cost steel was to the latter part of the 19th century. But the really good thing is that while further advances are welcome and make things easier, we don’t need to rely on them to save the planet. We’ve got everything we need right here. We just need to use it.

  11. @Hermit
    New coal in Europe is a problem, but it’s been caused by three things.

    1) A foolishly depressed carbon price, partly because of over allocation, and partly because of extreme demand-side insufficiency induced by austerity.

    2) Looming policy changes, which decree that by 2016 new coal must be cleaner (and therefore more expensive). This is resulting in a “coal rush”, before construction gets more expensive.

    3) The american exodus into gas, resulting in more exports from the US, and hence a lower price.

    As for aluminium smelting, I must say you sound like a broken record on this. It’s been dealt with ad nauseum on this blog. I’ll just repeat the standard points (which I’ve never heard you refute convincingly)

    Aluminium smelting makes up a small part of our national electricity demand. It’s an example of the electricity consumer which would be most expensive to do demand side shifting on at this time, but we don’t have to worry about that last 10% type problem until we get down to our last few coal stations. That’s a long way into the future no matter what we do.

    We can increase grid connectivity, and use HVDC to link remote places together, reducing variablity.

    We can increase demand side participation quite a lot, especially starting from the low base of almost nothing. Smart grid technologies significantly smooth out the bumps. Preferentially charging electric cars at times of low demand also come to mind. Despite the fact you presented electric cars as a problem for grid stability, they’re actually part of the solution.

    Grid storage exists, from pumped hydro, to upscaled hydro, to compressed gas, to newly emerging flow and grid batteries, and others.

    Yes there hasn’t been much investment in these sort things in Australia yet, but that’s because variability hasn’t been much of a problem yet. As renewables increase their contributions to the grid, rational private actors will make the required investments.

    Smelters can pay more money for a reliable supply. Any level of uptime can be provided, for a price. If smelters have got used to not paying much and now face a new reality, so what? Life changes, get used to it.

    Yes, that would mean a higher price, but we can reduce our total use of aluminium. It’s a substitutable good. This would make almost no difference to most consumers. Yes it would be a real cost, but such a small one that most people would hardly notice.

  12. Looking at a page written for some strange reason in German, it looks like Germany might bring 8 or 9 gigawatts of coal power online by 2015 or so. But this all appears to be due to the high price and potential unreliability of Germany’s gas supply, combined with the success Europe has had in cutting emissions resulting in a low price for carbon credits. Just like here, coal plants take a long time to build and and as far as I can tell they were all planned before Fukushima and the nuclear plant shut down, but of course the nuclear shut down certainly would have helped the prospects of those that were still on paper. Having the competition closed down is handy.


    It took six years to build the plant, meaning that the process started in 2006. It is by no means a reaction to the nuclear phaseout of 2011. And as Altmaier himself points out, the new plant can ramp up and down by 150 megawatts within five minutes and by 500 megawatts within 15, making it a flexible complement to intermittant renewables. In the area, 12 coal plants more than 40 years old have been decommissioned, and the new 2,200 megawatt plant is to directly replace 16 older 150 megawatts blocks by the end of this year, so 2,200 megawatts of new, more flexible, somewhat cleaner capacity (the new plant has an efficiency of 43 percent, whereas 35 percent would be considered ambitious for most old coal plants) is directly replacing 2,400 old megawatts.

    Germany has a target of 35 percent renewable power by 2020, rising to 85 percent by 2050 – meaning that 65 percent of its power supply will be conventional in 2020, and the country will still have 15 percent conventional power by mid-century. Obviously, Germany needs to build some new conventional power plants to reach even that ambitious goal for renewables.

    There have been reports that Germany plans to construct some 23 coal plants, but as in Cologne these plans predate the nuclear phaseout of 2011. The question is how many of these will be built. German environmental organization BUND has a map (in German) of the power plants planned and those already blocked. In addition, Germany’s Energy Agency (Dena), which is not considered a blind advocate of renewables (on the contrary, the renewables sector considered its Grid Studies subservient to grid operators’ needs), estimates in a recent study (PDF in German) that 18.5 gigawatts of coal power capacity (both hard coal and brown coal) will be decommissioned by 2020, whereas only 11.3 gigawatts will be newly installed by that time. Most of the new capacity is expected to be gas turbines, with 20.7 gigawatts going up by 2020.

    Finally, it is simply not possible for Germany to increase its carbon emissions from the power sector because the country has emissions trading, which sets a limit on emissions. If anything, the phase-out of nuclear will remove a chunk of low-carbon generating capacity, thereby raising the price of carbon, which will make future investments in coal plants expensive — but the effects will not be felt for years because it takes years to build these plants.

  14. The Fukushima-design, non-failsafe GE Mark I reactors need to be shut down; all of them. Likewise, the Chernobyl-design, non-failsafe carbon-cooled reactors need to be shut down; all of them.

    This still leaves quite a lot of “second generation” plants which do not have fatal design flaws and can be allowed to run for some time.

  15. @Nathanael
    Unfortunately the Fukushima and Chernobyl reactors were only regarded to be fatally flawed after they flawed fatally. There could be hidden surprises waiting in design and programming of other types of reactors. Of course this does not mean that reactors with obvious flaws should not be shutdown or modified. (And if they had to pay the full cost of insurance I am sure they would be shut down.)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s