Home > Environment > The nuclear option: AP1000 or bust?

The nuclear option: AP1000 or bust?

January 3rd, 2012

I’m going to relax my prohibition on discussion of nuclear energy, and offer a couple of points to start things off. I expect everything to go in circles as usual, but there’s plenty of room on the comments page, after all.

First, I have a piece coming out soon in the National Interest, arguing that 2011 marked a watershed in the development of energy sources to replace fossil fuels, with nuclear power finally ceasing to be a relevant option, and solar PV finally becoming a serious contender. I’ll add a link when it appears.

I’m sure not everyone will agree, so I’ll also offer a subsidiary claim which, if accepted, would at least simplify the debate. The claim is that the only nuclear technology with a serious chance of substantially reducing CO2 emissions before 2030 is the Westinghouse AP-1000.

To clarify my reasoning, I’ll begin with a definition. I’m going to define “substantial reduction” to require at least 100GW of operational installed power. If that exclusively displaced coal, it would save around 600-700 million tonnes of CO2 a year, a bit more than Australia’s current emissions, or about 2 per cent of the total. That seems like vey modest goal.

Next, I assert that achieving that outcome will require at least 20GW of capacity to be operational or very close by 2020. Otherwise there simply won’t be any chance to achieve the economies of scale needed to make nuclear economically viable, and the operational experience required to convince decision-makers to invest the necessary capital.

That in turn means that the decision to build the power stations has to be taken very soon. Even in China, where construction schedules are much faster than elsewhere (with implications for safety standards that have been seen in other sectors), a power plant takes a minimum of five years from the first concrete pour, and planning and site preparation takes two or three.

My final point is that any option that can’t make a real difference before 2030 should not play a major in current policy discussions. Obviously we need substantial reductions in emissions well before then, and that should be the primary focus.

The case in favor of the AP-1000 as the only feasible nuclear option is based on two simple facts.

(i) The US NRC has just approved it for the only four plants currently under construction in the US
(ii) China has adopted it as the main design for its program.

The argument that there are no other serious contenders remaining can be approached in two ways. First, by considering the options. These can be divided into three groups
(i) Currently existing competitors, such as CANDU, Areva and so on. None of these have any real chance of achieving the necessary scale, either now or in the future. The only possible competitor is the CPR-1000, a Chinese adaptation of a French design, which is at least a generation behind the AP-1000

(ii) Small modular reactors, still being touted despite having gone nowhere. The most promising version, the ‘pebble-bed’ design being developed in South Africa, was abandoned a couple of years ago. The only contender that has any serious chance of being operational by 2020 is a cut down version of the AP1000. Apart from that, there are a bunch of designs with no prospect of being built, even as demonstration projects, much before 2030.

(iii) Breeder reactors, like small modular reactors, only with a much longer track record of consistent failure. The only actually existing technology, sodium-cooled fast reactors, has been tried in quite a few countries, but has invariably failed. There are a handful of efforts still continuing, but none on a scale large enough to make a difference by 2030. Then there are a bunch of conceptual designs, with no firm plans for construction, even of prototypes.

So, if the AP-1000 is the only real contender, how good are its prospects? Not very good, in my view. The claim that modern designs will produce big cost reductions is so far, just a claim. Even a substantial deployment in China is unlikely to allay safety concerns, given China’s poor track record in this respect. So, a lot depends on the US projects being completed on time and on (or under) budget. Maybe this will happen and maybe not. Even if the AP1000 is as successful as possible, nuclear is still a backstop option at this point. Our primary hopes have to be placed on some combination of energy efficiency, renewables and lower energy consumption.

A final point about the Australian debate. Given the most optimistic possible projections, the AP1000 design might have achieved a couple of hundred reactor years of successful operation by the late 2020s. That would be a good time for Australia to start looking at the technology. The idea that nuclear power is a relevant option for Australia any time soon is just silly.

Categories: Environment Tags:
  1. Ikonoclast
    January 3rd, 2012 at 07:46 | #1

    There doesn’t seem to be (to my knowledge) any logical, empirically based evidence that would refute JQ’s conclusions above. Supporters of nuclear energy will need to do better than the usual pie in the sky arguments about thorium reactors and other breeder reactors. Such plans (in the case of thorium reactors) are still purely hypothetical wish-designs, thirty years in the future IF they can ever eventuate. Breeder reactors as a class have never been safe nor have they been even remotely near to commercially viable.

    Energy frugality is upon us whether we would wish it or not.

  2. CM
    January 3rd, 2012 at 08:12 | #2

    What is the problem with thorium reactors or other breeder reactors? I ask not to make a rhetorical point but to try to parse the arguments. The material that I have seen on thorium reactors seems mostly evangelical, which fuels my skepticism, but it also seems reasonable prima facie. I can see that a technology that is commercially untested and has no scale will not be a solution in the next decade, but might further research be a bet with a positive return in expectation? In what way are they not safe? What would be required for them to be commercially viable?

    I’d really appreciate someone with good knowledge and background guiding me through this…

  3. Chris Warren
    January 3rd, 2012 at 08:19 | #3

    The AP1000 is a cheaper version in part due to standardisation but also because its standards are not as robust as more secure models.

    The concrete containment dome is 0.9 metres thick and could easily be breeched by a terrorist attack.

    Most pundits peddling nukes, never give an authoritative listing of the isotopes in the spent fuel.

    The risk of a nuke accident is small, but the consequences are huge. As the number of plants increase, humanity thereby sets itself its own generational death-trap.

    The real lust for nukes is more economic than rational – Nukes result in an almost immediate monopolisation of energy supply in a region. It generally represents a corrupt business-political back channel driving government policy. It will also create a ‘too-big-to-fail’ possibility that street-wise commercial companies will use to extract subsidies and concessions from government.

    Electricity from a nuclear reactor is filthy cheap in the short run, but cataclysmal in the long-run.

    If nuclear waste is so perfectly contained and safe – why not store it in the basement of parliament house?

    1 Chernobyl or 1 Fukushima every 40-50 years plus acres and acres of leaking waste depository building-up over time will jeopardise the ecosystem for future generations.

    Nations constructing AP1000’s will have little continuing interest in developing fusion and other saner energy sources.

  4. John Quiggin
    January 3rd, 2012 at 08:40 | #4

    @CM The problem with breeder reactors is that they have never worked well, and there’s no reason to suppose that will change in time for them to have a significant impact. The existing sodium-cooled designs have been abandoned just about everywhere. The most optimistic schedules for thorium would have the first reactors coming on stream around 2030, which puts a really large scale deployment somewhere around 2050.

  5. conrad
    January 3rd, 2012 at 09:01 | #5

    At least in China’s case, I think you should increase the real price of coal for comparitive purposes — This is because it is well known that (1) particulates that coal plants belch out reduce rainful, and this is presumably a huge indirect cost since it would affect crop yields etc.; and (2) unlike Australia, many people have the enjoyment of breathing in the filth that comes out of Chinese coal plants, and for this reason, I’m sure many would rather be living next to a nuclear plant than a coal one. Apart from the unpleasantness of breathing in filth all day, there are also large medical costs involved that range from the immediate and obvious (cancer), to long term chronic health problems due to it permanently affecting children’s lung development. Personally, I’d take nuclear just for health reasons.

  6. rog
    January 3rd, 2012 at 09:06 | #6

    And then there is Terrapower, which is being touted by Bill Gates.

    To my mind this is just wishful and fanciful thinking and symptomatic of avoidance of dealing with the existing problems associated with the nuclear industry. The tsunami and subsequent meltdown at Fukushima has somehow been swept to one side however the costs continue to mount. The once well capitalised TEPCO is now to all intents and purposes broke and the cost of compensation to the families and businesses in the vicinity of Fukushima will now have to be borne by the taxpayer.

    Despite these once hidden now assessable costs nuclear power continues to be attractive to those that promote free markets and decry regulation taxation and big Govt.

  7. Hermit
    January 3rd, 2012 at 09:08 | #7

    It seems Australia can produce more thorium than uranium without really trying as a byproduct of rare earths and heavy sands extraction. To my knowledge the AP 1000 can’t use thorium unlike the CANDU. After using enriched uranium for startup I understand the CANDU can use unenriched fuel, both uranium and thorium. Until Australia has an enrichment industry we’d have to buy in ready-to-use fuel, perhaps ironically sourced from Australian uranium.

    Apart from China it seems the UAE got a remarkably good deal on four units to be built by a South Korean firm, about $3 capex a watt being the latest figure I believe. It is often claimed that we cannot build that cheaply in Australia but that seems to include a stiff risk premium in financing costs. Note UAE has plenty of sunlight but they seem not to think solar baseload is practical.

    Where will the money come from? The Feds could cut the $36 bn NBN budget in half and lend the money at 1% to any aspiring nuclear builder. Any AP 1000 complex would have to be on the coast for seawater cooling. Perhaps the desalination plant now being built at Wonthaggi Vic should have been co-located with nuclear plant.

  8. John Quiggin
    January 3rd, 2012 at 09:43 | #8

    The UAE deal looks remarkably good, as you say. In fact too good to be true

  9. January 3rd, 2012 at 10:33 | #9

    I really don’t even understand who in this country is advocating nuclear. 1. It’s hugely expensive. 2. Australians are extremely nuclear phobic. 3. How do these people seriously think the plant would be built, you can’t even get a pulp mill built in Tasmania.I’m of the view that at present nuclear is the only source of non-carbon base load power available but I can’t see it ever happening in this country.

    There maybe scope for older coal plants to be replace with gas. However, there are no easy solutions. No one is going to invent cold fusion sometime this year. I think realistically we can mainly hope to increase our efficiency and cap emissions at our current levels. With the hope that all the research and market incentives for renewable energies payoff in the next 20 years. The other thing we can do is restore some of the environmental systems we have destroyed over the last 100 years.

    Anyway I’ve done my part, enjoying summer in my hot unairconditioned house.

  10. Chris Warren
    January 3rd, 2012 at 10:37 | #10


    Now we see the capitalists roll out their cries for concessions, cutting provisions, and gifts:


    Australia risk premiums are “stiff” and are not “that cheap”

    Nuke builders to get funding gifted at 1% (!)

    Government sponsored loss-leading.

    and on, and on and on ……

  11. Fran Barlow
    January 3rd, 2012 at 10:46 | #11


    If you’re interested in a detailed discussion of thorium reactors, Brave New Climate is a good lace to start as there is a focus both on the engineering and economic questions there. Some are keen, others less so. This is a recent topic:

    The Guardian questions: thorium, shale gas, off-grid renewables, and much more…

    but if you piut thorium into the search box at BNC there’s plenty more.

  12. January 3rd, 2012 at 10:54 | #12

    I will just quickly explain why thorium reactors are almost completely useless. Then no one who reads this will ever suggest that thorium reactors are a significant improvement ever again. I’m sure that’s how it works.

    Thorium reactors are almost completely useless because the cost of nuclear electricity almost completely comes from factors that aren’t the cost of uranium, so using a cheaper fuel only helps a little bit. Now every little bit helps, but nuclear power needs more than a little bit of help to become competitive. As we currently have low emission sources of electricity that are both cheaper than the cost of new nuclear and cheaper than the insurance cost alone of new nuclear, a little bit won’t be enough to make thorium reactors competitive. On top of this, thorium reactors aren’t free or ready to roll out. They will require a lot of money and brainpower to develop and would take decades to be ready for a large scale build up, while other cheaper alternatives are ready now.

  13. Fran Barlow
    January 3rd, 2012 at 10:59 | #13

    I want to avoid restartingparticipating in the whole mery-go-round on nuclear here. Stipulated then — nothing that has happened in the last year has changed my mind that as things stand, the vast majority of industrial scale energy systems are going to need nuclear power if they are to achieve near zero net emissions.

    On the timeline question raised by PrQ …

    The point is moot. Either “we” (in this case we is “the world”, but not necessarily Australia) get our fingers out and develop nuclear power or something equivalently good at adequate scale on the needed timeline, or there will inevitably be one kind of serious mess or worse or another. The short timeline implies resort to existing technology — like the AP1000 — but that doesn’t mean we ought not to be working on the more advanced technologies now. The key question remains — how do we take coal, gas and oil out of the system on and adequate timeline? Nuclear power remains the most plausible vehicle for doing that right now.

    “We” clearly have the resources to build all the reactors we need, if that’s a priority. In four insane years, the US alone ran up $10bn per month occupying Iraq — enough to build two reactors every month and give them away.

  14. Hermit
    January 3rd, 2012 at 10:59 | #14

    @Chris Warren
    1% is better than free capital that requires no repayment at all. The ‘solar flagship’ projects at Moree and Chinchilla will get about $300m out of $750m gratis from the government. Plus ongoing subsidies for each megawatt hour. Note the biggest US energy loan guarantee default was solar firm Solyndra, not a nuke operator.

  15. John Quiggin
    January 3rd, 2012 at 12:08 | #15

    Actually, a long-term loan at 1 per cent amounts to a gift of well over half the principal (assuming a 20 year term and a commercial rate of 8 per cent).

  16. Sam
    January 3rd, 2012 at 12:21 | #16

    How much would uranium have to cost in order for breeder reactors to be commercial? Also, given peak uranium, when will we get there?

  17. PeakVT
    January 3rd, 2012 at 13:25 | #17

    I agree with the general point about SMRs, but I have a few comments. The KLT-40S is frequently put in the SMR category, and it is in production. However, it’s highly unlikely that anyone other than the Russian government will order one. The two that are being built will probably go critical in early 2013. Site prep is under way in Argentina for a demonstration-scale CAREM reactor, but given the country’s history with nuclear projects, completion could be decades away. Construction on the HTR-PM demonstration project in China is under way according to a report I just downloaded from the IAEA, but current information on the project in English is scarce.

  18. Ikonoclast
    January 3rd, 2012 at 15:34 | #18

    @Fran Barlow

    Fran Barlow writes, “…. the vast majority of industrial scale energy systems are going to need nuclear power if they are to achieve near zero net emissions.”

    This statement contains two seriously flawed assumptions. Second in order but first to be taken issue with is the assumption that nuclear power achieves near zero net emissions.

    Kurt Kleiner in Nature Report at nature.com notes the following; “Nuclear plants have to be constructed, uranium has to be mined, processed and transported, waste has to be stored, and eventually the plant has to be decommissioned. All these actions produce carbon emissions.”

    From the same source, referring to work by Benjamin K. Sovacool, a research fellow at the National University of Singapore;

    “According to Sovacool’s analysis, nuclear power, at 66 gCO2e/kWh emissions is well below scrubbed coal-fired plants, which emit 960 gCO2e/kWh, and natural gas-fired plants, at 443 gCO2e/kWh. However, nuclear emits twice as much carbon as solar photovoltaic, at 32 gCO2e/kWh, and six times as much as onshore wind farms, at 10 gCO2e/kWh. “A number in the 60s puts it well below natural gas, oil, coal and even clean-coal technologies. On the other hand, things like energy efficiency, and some of the cheaper renewables are a factor of six better. So for every dollar you spend on nuclear, you could have saved five or six times as much carbon with efficiency, or wind farms,” Sovacool says. Add to that the high costs and long lead times for building a nuclear plant about $3 billion for a 1,000 megawatt plant, with planning, licensing and construction times of about 10 years and nuclear power is even less appealing.”

    The University of Sydney published a study which notes; “Whilst the findings show that greenhouse gas emsisions are much lower than conventional fossil fuel powered generation systems (around 10-120 g CO2e/kWh depending on the reactor type), nuclear power is by no means a “Zero emissions” generation option.”

    Sovacool’s finding is in the middle of the U of Sydney range. Some of this variability probably occurs depending on whether or not the study is fully comprehensive of the entire mining, fuel processing, plant building, decomissioning and waste storage life cycle of nuclear power.

    Fran’s other assumption is that other technologies cannot deliver “industrial scale energy systems”. This is outside of nuclear and fossil fuels thus meaning essentially renewables like solar, wind and tide. I exclude biomass as most farmed biomass production is going to have to be for food production or world starvation will result. Trash biomass will always only be niche at the industrial scale.

    Whether solar, wind and tide can or cannot deliver industrial scale energy systems for about 7 or 8 billion people is still an open question so far as I can tell from my amateur research. Anyone who can bravely state that theses can or cannot do so has access to better research or a better crystal ball than I have. On the raw numbers there is adequate insolation hitting the earth’s surface daily such that the harnessing of a very small fraction of this would supply all our energy needs. Figures I have seen vary but indicate that if we coould harness between 1/7000th and 1/20,000th of this energy we would meet our current total energy needs.

    There is enough energy there to be harnessed. The moot questions are;

    (a) is the energy density high enough to satisfy EROEI (energy return) and financial return requirements

    (b) can enough raw (and then manufactured materials) be assembled to build the extensive infrastructure required?

    Already there is enough data to answer (a) in the affirmative. I wont rehash it but I have posted on this before. Given that the main raw materials for (b) will be silicon, cement and concrete aggregates, iron (or carbon fibre) and copper we are talking superabundant materials except for copper. So as I said, still an open question but far from out of the question at a first raw assessment.

    Utlimately we will have no choice. Either renewables will work on a large scale or civilization will have to retreat to a 17th C energy use level (at best considering the other damage we have done and are still doing to the biosphere). Fossil burning and nuclear fission of earth sourced fuels are non-renewable processes and must deplete. This is obvious except to cornucopian fantasists who believe the finite earth holds an infinite store of these fuels.

  19. frankis
    January 3rd, 2012 at 15:59 | #19

    My complex two-step plan to a better energy future for the world is:

    First build no more power plants that are not designed to run on renewable fuels such as solar, wind, biomass, ethanol from lignocellulose, wave, tidal, or on nuclear.

    Second, for the next decade at least concentrate on building 100% renewables plants plus conservation plus improved systems and networks, all the while retiring coal plants and improving those cleaner technologies including nuclear. There’s comfortably plenty of waste and fat in present practices, plus miles to go before we reach any realistic “baseload” limitation to a renewables fraction of world energy supply.

    After a decade of that it’d be as inconceivable that we’d fall back to coal as it would that we’d fall back from automobiles to horses; although bicycles could be looking good.

    The best way to encourage this diabolically complicated program would be with internationally synchronised carbon taxes which could be used to fund research and to reduce other taxes on good things such as income and payrolls.

    The designed minimal degree of legislation and regulation necessary to make this happen is just details. The effect on the economy would be nett beneficial (just as all advances of civilisation wind up being nett beneficial in economic terms).

  20. frankis
    January 3rd, 2012 at 16:10 | #20

    Ethanol? … I may have intended “geothermal etc etc” or something where I’ve mentioned ethanol.

  21. Fran Barlow
    January 3rd, 2012 at 17:20 | #21


    Nuclear plants have to be constructed, uranium has to be mined, processed and transported, waste has to be stored, and eventually the plant has to be decommissioned. All these actions produce carbon emissions

    That’s true of course, but the precise CO2 intensity of this process is at issue. Firstly, and most obviously, while uranium does indeed have to be mined, (I’m putting thorium to one side here) the precise source of the feedstock is obviously key in assessing mining-related emissions. Olympic Dam for example is a multi-ore mine, and so proper hypothecation means that the CO2 generated ought to be spread across all of the mineral ores either by mass or perhaps by value. In either case, the CO2 associated with the uranium would be quite small. Most would attach to the copper. Copper of course is an essential mineral in all industrial scale energy systems, and so bearing in mind the copper required for highly reticulated geographically dispersed energy systems, nuclear plants (because they can be placed near load centres and don’t require overbuild) come out relatively well. Because it is very energy dense, transport of uranium ore is also low CO2 intensity.

    One might add that if and when 4th Gen technology becomes the standard, then what is currently hazmat — once used fuels for example — will become the new feedstock post refabrication, and these fuels’ CO2 intensity will be no greater than the refabrication cost — less the CO2 cost of storage. It might also be that with modularisation, not all of the existing plant will require decommissioning.

    The broader point though is that as the economy as a whole becomes less CO2 intensive so too does the lifecycle cost of all the feedstock harvest, transport, fabrication and decommissioning. In a system which was 80% supported by nuclear power, all the these Co2 burdens would fall towards the marginal CO2 intensity of plant operation.

    According to Sovacool’s analysis, nuclear power, at 66 gCO2e/kWh emissions is well below scrubbed coal-fired plants

    I’ve seen a range of figures for this — some as low as 5gCo2e/kWhe for nuclear power. Both this (and the coal figure — up to 1.4kgCo2e/kWhe for brown coal) are clearly marginal CO2 intensity rather than lifecycle over the life of plant. In the case of coal, the lifecycle analysis is much worse when the plant is not near the feedstock source, as is the case also for biomass.

    On the other hand, things like energy efficiency, and some of the cheaper renewables are a factor of six better. So for every dollar you spend on nuclear, you could have saved five or six times as much carbon with efficiency, or wind farms

    Specious. Yes it is true that energy efficiency at the margin (along with energy usage avoidance — EUA) can be far lower cost abatement than any other strategy. I’m strongly in favour of picking all this ‘low hanging fruit’ and as soon as possible. The trouble is that once you pick it, you don’t get to pick it a second time. It’s a one-time deal. At some point, the cost of abatement using energy efficiency and EUA starts to exceed low-carbon energy production. We aren’t close to that point now, but if we do the right thing, one day we will be. We can put off the day of reckoning, but when that day comes, we will need something with which to reckon. It’s likely that world population will reach 9bn and possibly 10bn before stabilising, and given that about 3-4 billion of the world’s current population want to live something like us first worlders, our energy efficiency and EUA isn’t going to change the calculus much for them. They need new low-CO2 intensity highly available/dispatchable and cheap power to supply food, shelter and the components of industry.

    Whether solar, wind and tide can or cannot deliver industrial scale energy systems for about 7 or 8 billion people is still an open question so far as I can tell from my amateur research.

    It’s “open” only in the sense that one cannot say with certainty that some combination of technological breakthroughs on storage efficiency and cost-effectiveness will not change the calculus associated with them greatly for the better. Personally, I’d love that to occur, because that would give us far more options and flexibility. As things stand though, there is no basis for thinking that anything more than modest incremental improvement will occur on the timelines we are discussing here.

    Figures I have seen vary but indicate that if we coould harness between 1/7000th and 1/20,000th of this energy we would meet our current total energy needs

    Assuming 100% conversion to despatchable energy where it is needed, which currently, one can’t. It is not so much energy we need but the work that comes from harnessing it with value — i.e. power.

  22. Chris Warren
    January 3rd, 2012 at 17:25 | #22

    Nuclear power is zero emissions is typical of nuclear spin-doctors.

    Normal cement emits almost its own weight in CO2. The outside surface of a fresh concrete block will reabsorb some CO2 but this is negligible.

    There is sustainable cement but this has been patented by capitalists.

    CO2 emissions (now 392) will hit 400ppm in a few years and then go one to 450 and higher.

    So all we can do is wait until they work out a waste free nuclear reactor made with wood and recycled plastic.

    Unless emission-free cement is available in bulk in China, India, and Africa to replace standard cement (Portland), then CO2 emissions will only skyrocket, and nuclear plants will be one of the main contributors.

  23. Sam
    January 3rd, 2012 at 17:45 | #23

    @Fran Barlow

    “As things stand though, there is no basis for thinking that anything more than modest incremental improvement will occur on the timelines we are discussing here.”

    I seem to remember John Quiggin doing a post on the solar miracle not long ago. He was talking about a Moore’s law-like reduction in panel cost. Also, there were forecasts of unsubsidized grid-parity with coal in most locations before 2020. You can talk about power availability, storage and the like, but this really is a big change.

    “It is not so much energy we need but the work that comes from harnessing it with value — i.e. power.”

    That’s not what power is.

  24. Hermit
    January 3rd, 2012 at 17:48 | #24

    The world’s oceans contain many times the amount of water needed to grow bananas in the Simpson Desert. Now that problem is solved it can’t be too hard replacing coal.

  25. Fran Barlow
    January 3rd, 2012 at 18:23 | #25


    I said:

    It is not so much energy we need but the work that comes from harnessing it with value — i.e. power

    Sam responded:

    That’s not what power is.

    I was being somewhat rhetorical — technically, power is about potential output e.g a 20kW turbine, a 200kW engine (as opposed to energy — measured over time e.g. kWh). In the end though what we are interested in is what we can do with the output of an energy system. Can you stop your food from spoiling, run your machines as specified and so forth?

    Notionally, one can replace a 1.6GW plant like Hazelwood with a 1.6GW windfarm. In practice, even a 20GW complex of windfarms spread across the country may not be enough all the time or even most of the time. IIRC the capacity credit for wind in Victoria is about 3% even though the system is linked over an area spanning about 1100kms of Southern Australia. So yes, what is needed is predictable available power rather than energy per se.

    That’s one reason why the CO2 figures for wind can be so flattering. They are notional rather than actual.

  26. Ernestine Gross
    January 3rd, 2012 at 21:23 | #26

    Ikonoclast @17 your post makes sense to me. I am a bit more optimistic than you seem to be. New construction and building methods, more local food production, comfortable rather than ostentatious houses are other sources of reducing energy consumption. These ideas seem to become very popular with youg people.

  27. iain
    January 3rd, 2012 at 21:44 | #27

    I’ve made this observation before – the best responses on this topic have come from commentators “Salient Green” and “nanks” who have previously written the following:

    “Personally I’d rather see renewables for a range of reasons … decentralisation and autonomy – another being the expansion of research across a much broader range of disciplines than that required by fossil fuels or nuclear.”

    Decentralisation of energy supply, and autonomy in producing energy, are not really part of the nuclear equation. Nuclear doesn’t solve a lot of the existing externalities currently associated with energy (including energy resource wars).

    “Unfortunately, the nuclear vs renewables debate is mostly based on how to continue with business as usual. There is no doubt that either or both could be used by the human race to continue BAU [in the short term] …What most people don’t get is that we can’t have BAU…”

  28. Fran Barlow
    January 3rd, 2012 at 22:03 | #28


    Unfortunately, the nuclear vs renewables debate …

    This is the chief mistake the renewables only crowd make. It’s not about nuclear versus renewables. Its about fossil fuels versus everything else or fossil fuels versus everything except nuclear.

    That’s playing with one hand tied behind your back against a behemoth.

  29. January 4th, 2012 at 02:48 | #29

    “The only actually existing technology, sodium-cooled fast reactors, has been tried in quite a few countries, but has invariably failed …”

    The IFR didn’t fail. It was shut down by the natural gas interests.

    If that counts as failure, it’s a sort of failure that indicates something we need to revive.

    CANDUs such as those powering more than half of this computer don’t seem to have any logistical difficulty being built up to hundreds or thousands of GW.

  30. John Quiggin
    January 4th, 2012 at 07:30 | #30

    CANDU is the deadest of dead ducks

    As regards IFR, it doesn’t really matter who killed it. Even if it were revived tomorrow, there’s no way it could be commercially deployed before 2030. And, of course, it isn’t going to be revived in the US because fast breeder technology has been a total failure where it has been tried, most notably in France and Japan.

  31. Robert Merkel
    January 4th, 2012 at 08:41 | #31

    There is one technology that, might – even allowing for our host’s conservatism – have a chance of commercial deployment before 2030 – the small scale PWR being designed by NuScale Power.

    PWRs are known working technology, and should theoretically pose less of a challenge to get past the regulators.

    Ruling out the EPR seems a little unfair, as well. Yes, Areva screwed up construction of Olkiluoto-3 royally, but they should have learned their lessons…

  32. Hermit
    January 4th, 2012 at 08:42 | #32

    If the Sellafield reprocessing plant in the UK uses two S-PRISM reactors to burn plutonium stocks it will be a massive boost to the IFR approach. I understand the deal is that GE-Hitachi says if it doesn’t work there will be no charge for the use of the machines. I wouldn’t worry about Australia running short of enriched uranium late in the century since we should be able to cut a favourable deal as we have most of the raw product.

    More immediately on the question of low carbon baseload (or the recurring 50% of peak demand) it may be that carbon tax has killed new coal fired plant in Australia. That’s unless they go for the ‘carbon capture ready’ ruse. The problem with combined cycle gas is that really only Western Australia and Queensland have reliable gas supplies long term. That’s why talk has petered out of a big gas fired plant to replace a brown coal station in Victoria. Regional imbalance, LNG exports and future gas demand (eg as a diesel substitute) make steady price rises inevitable despite the modest capex and fast build times of gas fired plant. If the government takes climate scientists seriously we eventually have to cut CO2 by 80%, a bridge too far for gas fired plant or the wind-gas combination.

  33. Ikonoclast
    January 4th, 2012 at 08:47 | #33

    Fran Barlow says, “nuclear plants …. can be placed near load centres”.

    Japan tried that. Remember Fukushima?

    Anyway, corporate capitalism never passes up a concentrated monopoly profit opportunity with corrupt political influence and federal money to assist. If making money out of nuclear power stations was that easy they would have sprouted them everywhere… like mushrooms.

    The fact is that nuclear power stations are devilishly dangerous, devilishly expensive, devilishly difficult in an engineering sense, in the safety sense, the insurance sense, in the public image sense etc. etc.

    On top of all this, they run on a limited, exhaustible fuel. Peak uranium is about now just as peak oil and peak good quality coal are about now. Peak gas is due in as little as 5 years. And no doubt the claim for large recoverable thorium reserves for thorium breeder reactors (which have not got off the fantasy CAD drawing yet) is a specious as the claims for large uranium reserves.

  34. John Quiggin
    January 4th, 2012 at 09:17 | #34

    @Robert Merkel

    EPR is not quite as dead a duck as CANDU, but still moribund. Here’s the outlook from Wikipedia


    Even if the Chinese plants go to plan, they aren’t ordering any more as far as I can tell – as I said in the post, they’ve gone for either AP-1000 or their own knockoff of EPR, which is a generation behind.

  35. John Quiggin
    January 4th, 2012 at 11:32 | #35

    @Robert Merkel

    As regards NuScale Power, their website (unlikely to be over-pessimistic) suggests they could have their first plant in operation by 2020, if all goes well. Assuming that’s a maximal 12-unit plant (each unit being 45 MW), that gives them a total output of about 500 MW, and a construction capacity of 100 MW/year over the next decade. If they doubled that every 2 years thereafter, they’d be producing 10GW/year by 2040.

    For comparison, installations of solar PV last year were about 25GW, which, allowing for the different availability factors and the better match of solar with peak demand, is close to the equivalent of 10GW of nuclear.

  36. may
    January 4th, 2012 at 12:49 | #36

    renewables as a lumped together field is a misnomer.

    each discrete energy source has an area of strength and an area of weakness.

    the underwater wave power system up and running off Garden Island in Western Aust is a continuous not an intermittant energy flow(running a turbine or running a desalinator).

    sun and wind are intermittant.
    weaving of the sources into the grid is the challenge.


  37. Fran Barlow
    January 4th, 2012 at 14:37 | #37

    @Ikonoclast quotes me:

    Fran Barlow says, “nuclear plants …. can be placed near load centres”.

    then continues:

    Japan tried that. Remember Fukushima?

    I do, but the salience of the observation is unclear. In what ways, if any, would nulcear plants now built near load centres resemble the plant complex completed in the mid-1960s at Fukushima? In what ways would the regulatory regime be similar? In what ways would the threats to the plants be similar? Would the SCRAM systems be the similar or greatly different?

    Here’s another thought — had Fukushima hosted a complex of coal (instead of nuclear) reactors, what would the morbidity and ecological footprint have been? Had the tsunami come 12 months later, would the nuclear hazard have made a front page outside of Japan?

    Peak uranium is about now …

    Simply laughable on a number of grounds. One can obtain uranium from seawater. One can fabricate fuel from waste. There’s also thorium. Right now, uranium is cheap and nuclear fuel is not a significant factor in nuclear power prices.

  38. Ikonoclast
    January 4th, 2012 at 15:57 | #38

    @Fran Barlow

    Fran, once again, like all pro-nuclear advocates, you show you are immune to empirical evidence or suffer from empirical-evidence-alzheimers. Really, I do have to put it that strongly. Five minutes, five hours, five days or five months after being presented with conclusive refuting evidence, pro-nuclear advocates conveniently forget it all and advance the same old refuted arguments. I’ve noticed libertarians and AGW-denialists do the same thing. This is what makes arguing with all such evidence-immune zealots such a futile and circular business.

    The “one can obtain uranium from seawater” furphy is the one that really pegs you as a selective evidence denier or a selective evidence forgetter. The energy cost of separating uranium from seawater (at a concentration of 3 parts per billion!) is much greater than the energy that can be recouped by splitting that uranium in a reactor. This was covered in a nuclear sandpit debate on JQs blog some several months ago. I posted links to reputable scientific papers which covered this issue and removed any doubt that such a process of uranium recovery from seawater for splitting was anything but a big energy sink process and not an energy gain process. I am pretty certain you read that debate.

    Also, the fact that you don’t question such claims and go and check on them before repeating them indicates a certain level of naievity about physics and energy physics not to mention naievity about the incidence, psychology and sociology of zealotry and charlatanism in terms of both perpetrator chicanery and self-delusion as well as target audience credulity.

    I think the salience of my observation about Japan’s reactors is perfectly clear. You are essentially proposing the “we know better now” argument. We might know better (to some extent) but are we acting better in empirically demonstrable ways? Are all the reactors being proposed now of as genuinely disaster-proof, failsafe and fool-proof design as you imply? Are no new reactors due to be commissioned in tsunami zones and earthquakes zones anywhere in the world? Are current reactors in these danger zones plus all those of unsafe design still operating due to be decommissioned immediately? Are matters of political and financial expendience and advantage never now being allowed to influence nuclear reactor design, development and siting decisions anywhere in the world? Puh-lease!

    Such empirically weak arguments don’t wash with a hard-nosed realist like me.

  39. Hermit
    January 4th, 2012 at 16:06 | #39

    Dare I suggest PV is more of a middle class fashion statement than a serious source of electrical power. It helps when pensioners pay the feed-in tariffs. Admittedly if I’d spent the $20k on it I spent in 2005 but in 2010 I would have got more watts. Through subsidies and fashion it now adds up to many gigawatts nominal capacity (as Pr Q mentions) world wide.

    However aluminium smelters operate on dark nights and don’t want to pay more than about 3c per kwh. A recent estimate of a PV and battery storage system gives a levelised cost of about 30c per kwh. See nearly halfway down in this link
    Interestingly lead acid batteries are still the cheapest go-anywhere electrical energy storage system after 200 years of research. If we think PV can power heavy industry then we’re handing jobs to countries with no AGW qualms.

  40. Fran Barlow
    January 4th, 2012 at 17:34 | #40


    The EROEI calculations assume once through fuel usage (thus underestimating the potential output of energy by orders of magnitude). Long before we would do this we’d simply use breeders, or thorium, both of which will serve us for a very long time to come, rendering present-day calculations moot. Folks like Willem Storm Van Leeuwen came to mind as you ranted at me.

    The main reason breeders are uneconomic now is precisely because uranium feedstock is so cheap.

    I think the salience of my observation about Japan’s reactors is perfectly clear. You are essentially proposing the “we know better now” argument. We might know better (to some extent) but are we acting better in empirically demonstrable ways?

    New reactors are already far safer than those built in the 1960s. Nobody would contemplate revisiting the specifications applying then. Had Fukushima had independent and isolated cooling system backup there’d have been no problem. This was proposed AIUI in 1999 but the plant was due for retirement and a bad judgement call was made. That call would not be made now.

    You are not, as far as I can tell ‘a hard nosed realist’. You are someone with an obsessive fear of one particular technology, and unrealistic visions of how humanity will respond without that technology.

    As far as we know, there has been no morbidity associated with the plant failure at Fukushima, but there would have been very considerable morbidity if, in 1960, a decision had been taken to build coal reactors there. Your indifference to that is troubling.

  41. quokka
    January 4th, 2012 at 17:49 | #41


    You seem to be demanding that all nuclear power plants in earthquake zones be de-commissioned immediately. Why? The Japanese NPPs, some of quite old design, withstood a monster earthquake remarkably well. It was station blackout caused by the tsunami disabling the diesel generators that was responsible for the accident – not earthquake damage. All reactors scrammed properly as designed. In all likelihood if the diesel generators were properly protected from inundation, the outcome at Fukushima Daiichi would have been much the same as at Fukushima Daaini just a few klicks down the road – extensive site damage, but no core damage and no large radiation release. And trivial compared to the devastation of the NE coast of Japan by the tsunami.

    Your call for de-commissioning reactors in earthquake areas appears to be evidence free.

  42. John Quiggin
    January 4th, 2012 at 18:04 | #42

    “aluminium smelters operate on dark nights and don’t want to pay more than about 3c per kwh”

    Equally, I’d like coffee shops to open at 5am and charge 20c for a double espresso.

    There’s a long tradition of corrupting Australian electricity policy to make sweetheart deals with aluminium smelters, one that has been ruinous to taxpayers. The bad effects on the viability of publicly owned electricity systems have been one of the levers used in the push for privatisation. If a switch to renewables helps to kill all that off, that’s a free bonus.

  43. January 4th, 2012 at 18:54 | #43

    While people are talking about nuclear power, I thought I’d mention that the cheapest solar PV module is down to $1.14 US a watt. This means that for a low cost installation it may now be cheaper to install solar panels than to buy grid electricity in some parts of Australia. The lowest cost installations are likely to be on large flat roofs or on the ground. This is an interesting development, because with further cost reductions, it will result in Australia getting a significant fraction of its electricity from the sun. If PV saves money businesses will install it.

    ATTENTION CRAZY PEOPLE: Electricity generated by solar PV does not need to be stored.

    ATTENTION CRAZY PEOPLE: Electricity is more expensive during the day than at night, so what would the business model of storing electricity from solar PV be? Buy high, sell low?

    ATTENTION CRAZY PEOPLE: People will notice if the price of electricity during the daytime drops significantly and respond appropriately.

    ATTENTION CRAZY PEOPLE: Just because we are likely to get some of our electricity from solar PV does not mean we have to get all or even most of our electricity from solar PV.

    ATTENTION CRAZY PEOPLE: The amount of solar power we end up with will depend upon the cost of solar power and the cost of alternatives. We are unlikely to end up living under the fist of a solar obsessed fascist state determined to run everything off solar power regardless of the cost.

  44. quokka
    January 4th, 2012 at 19:03 | #44

    As a matter of fact, there are ten distinct small modular reactors under development in the US at the current time:

    Integral Light Water Reactors:

    SMR (Westinghouse) 225 MWe
    HI-SMUR (Holtec) 140 MWe
    mPower (B&W) 125 MWe
    NuScale (NuScale) 45 MWe

    Gas Cooled – Intended for Process Heat

    MHR (General Atomics) 280 MWe
    PBMR (Westinghouse) 250 MWe
    ANTARES (Areva) 275 MWe

    Fast Spectrum:

    PRISM (GE-Hitachi) sodium cooled
    ARC-100 sodium cooled
    Hyperion lead-bismuth cooled

    I would dispute Prod Q’s assertion that the first SMR online in the US will be the Westinghouse effort. My money would be on the B&W mPower. The dedicated test centre built for it’s development has just been opened, there is a memorandum of understanding with TVA, B&W is a long standing incumbent supplier for TVA, mPower is evolutionary light water technology using standard LWR fuel and B&W has a long history and much engineering experience with small reactors for the US navy.

    We may also see a pair of PRISMs (Integral Fast Rector) built in the UK to dispose of plutonium stocks. The GE-Hitachi proposal is under consideration.

  45. John Quiggin
    January 4th, 2012 at 19:23 | #45

    Do any of these contenders have a serious chance of commercial deployment on any significant scale before 2020 (B&W are hoping for this, just like NuScale discussed above). If not, how can they possibly scale up to make a substantial contribution by 2030, or even 2040?

  46. Salient Green
    January 4th, 2012 at 19:55 | #46

    [email protected], PV a fashion statement? Not a serious source of electrical power? Come on!!!
    Households are a serious user of electrical power which can be generated on a fraction of residential roofs by PV or replaced by Solar thermal.

    Add PV to another fraction of residential roofs and generate the electrical power for commuting. Add a little more PV, live a simpler life, localize, recycle, ignore fashion, buy quality and not only are your energy needs met but the need for industrial energy is reduced markedly.

    It’s groundhog day for me and aluminium as it is with Iconoclast and nuclear. Aluminium is too cheap as I have shown before in previous blogs.

    iain @27, you honor me with that quote which I had forgotten, thanks.

  47. Sam
    January 4th, 2012 at 20:12 | #47

    If we accept John Quiggin’s technological pessimism, we should be even more anti-nuclear than he is. Here’s the way I see it.

    1) Peak uranium is coming soon.

    2) A nuclear renaissance would only exacerbate this.

    3) Plenty of uranium exists (for example in seawater), but it is thermodynamically impossible to extract and burn it in a once-through reactor with energy surplus. It’s probably possible to do this in an efficient breeder (the margins are not great but at least it’s not a physical impossibility).

    4) Once-through reactors then, will not form a big part of the world’s energy future. If there is to be a nuclear renaissance of any consequence, it will be through breeder technology.

    5)If John Quiggin is right (I have no private knowledge about this at all), it will not be through breeder technology.

    6)Therefore, there will be no nuclear renaissance of any consequence.

    Two final points: It doesn’t matter that fuel is now a small part of overall running costs; price insensitivity in the face of non-increasable supply will just lead to ever-higher prices until fuel is a large part of costs. It won’t matter how good/cheap/safe the AP-1000 is; if there are lots of them, they won’t have any fuel to run. So if it’s AP-1000 or bust, it’s just bust.

    If I was someone like Fran or Quokka, I’d be spending all my time trying to rebut JQ’s technological pessimism wrto breeders. I don’t see another way out. That’s not supposed to sound belligerent; I’d really like to know if there are any reasons to be optimistic on this score.

  48. quokka
    January 4th, 2012 at 20:15 | #48

    @John Quiggin

    How long would it take to scale up production of Boeing or Airbus commercial aircraft? It’s inherently no more difficult to make SMRs in factories than aircraft. Probably easier. If the economics are right, why should this not happen and why would this not happen?

    While we are on the topic of rate of deployment, it should be observed that the world PV capacity is about 50 GW. Assuming an average capacity factor of say 15% (which may be generous in view of the large deployment in northern Europe) that’s about the equal of about 7 AP1000s and quite insignificant compared to world nuclear capacity. It’s also insignificant in reducing emissions. After how long?

    The biggest deployment of PV in the world supplies a little over 3% of Germany’s electricity.

    Claims that (non-hydro) renewables can be deployed faster (over decadal time frames) than nuclear really don’t stack up in the historical record.

    There is every indication that deployment of low emission generation capacity is pitifully inadequate to have any serious impact on the climate problem. It does seem to be just a wee bit premature to declare that what has proven (along with hydro) to be by a long way the most successful and reliable low emission technology to be not needed. James Hansen thinks so too in his essay on drinking the sustainable energy kool-aid.

  49. quokka
    January 4th, 2012 at 20:26 | #49


    Please go and read the recent “The Future of the Nuclear Fuel Cycle” report from MIT before making claims about “peak uranium”. They estimate that there are sufficient uranium reserves for a ten fold expansion of conventional once-through nuclear power fuel cycle with fuel cost no more than doubling for most of this century. The cost of mined uranium is only a portion of the cost of nuclear fuel. Enrichment and fabrication form a major part of the cost.

    This is a core finding of the report and one of the fundamental reasons that they do not believe that moving to a closed fuel is a matter of urgency. I do not agree with their conclusion about advanced fuel cycles but for different reasons than an impending uranium shortage.

  50. Ikonoclast
    January 4th, 2012 at 20:37 | #50

    @Fran Barlow

    Fran says; “Long before we would do this we’d simply use breeders, or thorium, both of which will serve us for a very long time to come, rendering present-day calculations moot.”

    So Fran still assumes we could use seawater uranium as some point but also assumes that a long honeymoon interregnum (to gruesomely mix metaphors) with thorium reactors will precede this. Let’s dispense with the seawater-uranium-on-breeder-steroids furphy first and then get back to thorium reactors.

    I am not aware of any realistic order of magnitude improvement even by fuel breeding that could make uranium extraction from seawater at 3 parts per billion EROEI positive whilst still obeying the known laws of physics (not to mention basic maths). Breeder reactors can theoretically decrease fuel requirements by nearly two orders of magnitude for the same useful energy output; that is to 1/100th. This changes 3 parts per billion to 3 parts per 10 million effective if I accept Fran’s argument. This is .000003% concentration if I have got my decimal point in the right place. Non-aqeuous (solid ores) which tail down to 0.05 % are usually considered very low grade ores. I am not sure they would be considered currently commercially viable. You are still four orders of magnitude short of this. I think I am safe in saying this breeder process would still be very EROEI negative. As even the change of one order of magnitude is sufficent in many cases to transition from a commerically viable energy project to one that is not only NOT commercially viable but is an actual energy energy sink (negative EROEI) then I think falling four orders of magnitude short knocks the pitch right of the ballpark and into the impossibility bleachers.

    Also, Fran ignores the facts that all current evidence indicates the breeder reactor program is a failure; neither safe, nor stable operationally nor commercially feasible. Fran ignores the facts that truly commercial Thorium reactors are merely theoretical design board exercises at this stage; that the largely experimental “thorium” breeder reactors (I am not even sure they all deserve that thorium appellation) have had very chequered and short-lived operational lives; and that the thorium fuel cycle presents considerable technical challenges and difficulties. Does Fran think it is “simple” to turn such theory into safe, scaleable and commercially viable practice in the highly challenging fields of nuclear physics and nuclear engineering? Tell that to the nuclear physicists and engineers. It’d be a bit like me berating the Wallabies and telling them it should be simple to run through the All Black line.

    But what am I to expect when lay nuclear proponents obviously get all their “facts” from a crackpot site like Brave New Climate?

    I admit that one should never say never. It is not impossible (though I think it is highly unlikely) that all the theoretical, design and practical issues of commercial thorium reactor deployment could eventually be substanially and satisfactorily resolved. But NOT (and this is a key point in the terms of JQ’s proposition) in a timeframe which allows timely and sufficient abatement of greenhouse gas emissions such that we are saved from dangerous climate change.

    A safe, commercial, operating thorium reactor (if it ever happens) is at least 30 years away. Deployment of sufficent reactors to mitigate climate change, at least perhaps double that time distant. That is 60 years too late and a very low probability otcome. We need substantial mitigation right NOW. Actually, we probably needed it to start about a decade ago. It makes sense to go with what is demonstrably working now, renewables deployment (albeit the main problem is an under-deployment which is well below what is possible) than to plump for a fix that is a low probability for deployment ever and certainty to be of the order of 60 years too late even if deployed.

  51. Sam
    January 4th, 2012 at 20:42 | #51

    I’ll certainly read the report, thanks. As I said before, I don’t agree that it matters that mined costs is currently a small fraction of overall cost. If the supply of any material vital to a process is non-increasable, it will become the rate-limiting factor as it diminishes. Price inflexibility just causes more price volatility.

  52. Ikonoclast
    January 4th, 2012 at 20:56 | #52


    I can tell you right now with confidence from previous research that the MIT conclusions would be predicated on highly dubious, unaudited and indeed outright specious estimates of world uranium reserves which take all national claims at face value. These national estimates range from the reasonably dependable to the most outrageously and flagrantly inflated estimates imaginable. Often a large slice of claimed reserves comes under the heading “Yet to be discovered”.

  53. quokka
    January 4th, 2012 at 21:01 | #53


    Surely you can do a bit better than “In my opinion” without offering any sort of authoritative evidence or reference.

  54. quokka
    January 4th, 2012 at 22:07 | #54

    Lots of wild claims about breeder reactors that have been thrown about here.

    1. They have been commercial failures.

    There has only ever been one breeder reactor with the designation “commercial”. That was Superphenix in France. It indeed had many problems, but was ultimately closed down by political decision when it had achieved reasonably reliable operation. Hard to draw firm conclusions about the outcome if it had been allowed to run it’s course. A sample size of one is inconclusive and there remains significant uncertainty about future costs.

    As a counter example the Russian BN-600 has run reliably since 1980 and is one of the most reliable reactors in the Russian fleet. It’s newer and larger sibling – the BN-800 – is now under construction and is designated as commercial.

    In total there have been 10 experimental FBRs, the earliest dating back to 1951, 7 demonstration FBRs and 1 commercial FBR.

    2. Costs.

    FBRs are different from LWRs so one off construction and demonstration plants are almost certainly going to cost more. In favour of FBRs is low pressure operation with far lower requirements for the reactor vessel. Also in favor of FBRs such as PRISM is a greater degree of passive safety than LWRs can achieve with potential savings in safety systems. Also in favour of such designs as PRISM is factory construction. Until such time as there are a number of FBRs built to the same design, the economics of FBRs retains considerable uncertainty – just like things such as EGS geothermal. Uncertain does not necessarily equal too expensive.

    3. FBRs have been abandoned.

    Not even a bit true. Several nations have ongoing FBR programs including France (ASTRID), Russia, India, China and Sth Korea.

    4. Time Frames.

    The GE-Hitachi PRISM is not a “conceptual design”. It is a complete design based on many years of work and considerable expenditure at Argonne National Lab. Construction could commence almost “immediately” (ie within a few years). Molten salt thorium reactors are indeed further away. This is no reason at all for not aggressively pursuing their development. Unless somebody wants to make the extravagant claim that all fossil fuel generation will be replaced by renewables by 2030, there will still be a yawning gap between what is needed and what has happened. It is entirely possible, perhaps likely, that by 2030, the amount of fossil fuel generation will be no less than today.

  55. silkworm
    January 4th, 2012 at 22:30 | #55

    However aluminium smelters operate on dark nights and don’t want to pay more than about 3c per kwh.

    Smelters operate at night only because of the cheaper cost of off-peak electricity. There is no practical reason why they can’t run during the day. In fact, it would be better for the workers’ health if they operated during the day. This points to potential savings in public health costs.

  56. quokka
    January 4th, 2012 at 22:58 | #56


    I think you will find that aluminium smelting is a continuous process that must run 24/7. I believe that power outages are very bad news that can damage plant.

    There are many examples of continuous process engineering. Fertilizer production (eg ammonia,urea) is another, though the energy is typically supplied by burning natgas.

    The whole issue of “movable demand” warrants careful scrutiny. The UK Climate Change Committee found that the only significant sources of movable demand by 2030 are likely to be transport and heating. It is not unreasonable to extrapolate this to countries with similar climates. What is important to recognize is that large expenditure and effort will be required to convert from gas to heat pump or resistive heating. Renewable heat, EVs and PHEVs are undoubtedly important, but there is considerable risk in making overly optimistic assumptions about rate of deployment and consequent influence on the feasibility of a grid with large amounts of intermittent generation.

  57. Sam
    January 4th, 2012 at 23:11 | #57

    This isn’t true unfortunately. The molten salt bath has to be supplied with current continuously. If the current stops, a solid crust forms This wrecks the container.

  58. January 5th, 2012 at 03:52 | #58


    The energy cost of separating uranium from seawater (at a concentration of 3 parts per billion!) is much greater than the energy that can be recouped by splitting that uranium in a reactor.

    Marine uranium extraction has already been demonstrated to be much cheaper than the present market price of petroleum. It cannot beat land-based uranium extraction, and as the latter becomes cheaper, it may never be able to do so, but it beats oil and gas by a mile.


    For there to be any genuine doubt of the amidoxime method’s energy positivity, the price would have to much exceed the present price of oil, not fall far short of it as in fact it does.

  59. Ikonoclast
    January 5th, 2012 at 04:55 | #59


    I have done so (provided the evidence) in past blogs on JQs site. I don’t see why I should trawl back through the nuclear sandpit blogs to find the links to papers I posted then when you could do the same. This is particularly so as I recall you were around blogging at that time just like Fran Barlow. If you failed to read or assimilate the papers and data at the time or have conveniently forgotten them since, you will just do so again. I am not wasting a lot more time trying to argue with evidence-impervious nuclear power advocates.

  60. John Quiggin
    January 5th, 2012 at 05:16 | #60

    “How long would it take to scale up production of Boeing or Airbus commercial aircraft? It’s inherently no more difficult to make SMRs in factories than aircraft. ”

    A good comparison. Airbus entered the industry in 1970 as a consortium of existing firms, with a capacity of about 100 planes a year, which remained pretty much static until the late 1980s. Since then, it’s raised output by a factor of four over twenty years. My demonstration that SMRs were too little, too late allowed for a doubling of scale every five years, or a factor of sixteen over 20 years.

    And of course, Airbus in 1990 had a history stretching back decades making a product that has been technologically mature since the introduction of the jumbo jet in the 1960s – the SMR firms are mostly startups, proposing a product that has never actually been built.

  61. rdb
    January 5th, 2012 at 07:44 | #61

    Tom Murphy’s Do the Math blog’s recent Nuclear Options “tutorial introduction and big-picture take” may interest.

  62. Ikonoclast
    January 5th, 2012 at 08:26 | #62

    @G.R.L. Cowan

    Sigh, I will post these links to reputable scientific papers again, more for those who missed the last debate than those who saw the links to evidence last time and did not read them, did not absorb them and/or conveniently forgot them (usually in the space of five minutes).

    Report on the objective situation re uranium reserves:


    I will do one link per post for fear of the anti-spam engine.

  63. Ikonoclast
    January 5th, 2012 at 08:34 | #63

    This article sinks fast breeder reactors comprehensively. Take note, fast breeder advocates. Thus Peak Uranium theory still holds. Game, set and match.


  64. Fran Barlow
    January 5th, 2012 at 08:35 | #64


    We need substantial mitigation right NOW. Actually, we probably needed it to start about a decade ago.

    On this I agree. Actually, in Australia, we probably needed it to occur before the 1960s coal plants started being upgraded in the 1980s. The economic assumptions that go with those presume a 40-year plant life. We are on the lip of major new investments in fossil fuel infrastructure, and renewables are not close to being reasy to step up and do what the next generation of coal and gas promise. I imagine that it’s much the same both in the other major emitting countries and in those LDCs seeking major industrial development. It seems very clear that in the next decade here, absent nuclear power, what we will get if a lot more fossil fuel investment that we won’t be able to retire in a hurry, even if, magically, renewables dropped radically in installed cost and even more importantly, improved radically in availability. Yet both power usage per person and the number of persons on the planet are both certain to rise sharply. No realistic EUA or energy efficiency plan is going to force down total demand realtive to today.

    Luckily, some places — Korea, Russia, China, Canada, US and UK are continuing nuclear development — which will hopefully lead to continuing real decline in prices and therefore more plants and a lower relative CO2 footprint.

    It makes sense to go with what is demonstrably working now, renewables deployment (albeit the main problem is an under-deployment which is well below what is possible) than to plump for a fix that is a low probability for deployment ever and certainty to be of the order of 60 years too late even if deployed.

    Renewables are not “demonstrably working now” if one defines that as “forcing retirement and deferral of fossil capacity”. At most, some niche markets are being served. Given the political cycles within which most regimes work, and the economic cycles within which infrastructure investment is carried out and the seemingly intractable challenges associated with intermittent and low energy-density sources, it seems recklessly optimistic to rely on early large-scale deployment of renewables.

  65. Ikonoclast
    January 5th, 2012 at 08:37 | #65

    More blasts from the past;

    Ok, Ive found the definitive refutation of the “uranium is plentiful” argument.


    Go the bottom of this linked article and find the links for the four chapters of Micheal Dittmar’s report. Once you follow these links you can then look at the PDFs by clicking one more link for each.

  66. Ikonoclast
    January 5th, 2012 at 08:48 | #66

    Darn, I can’t find the link again to that wonderful scientific paper which debunked “uranium from the sea” hopes. I do seem to recall that it calculated, among other things, that the collection and transport apparatus would have to be larger than the world’s current entire fishing fleet both in terms of nets (the net-mesh equivalent used by uranium collection) and in terms of boats and ships and that the net (no pun intended) EROEI would be a negative ratio (an energy sink).

  67. January 5th, 2012 at 09:55 | #67

    This whole issue of scalability of nuclear power, and the wildly divergent claims about it and the potential for new types of reactors, drives me crazy.

    Nothing would please me more than some form of international scientific/engineering commission being set up with the specific goal of giving (relatively) independent advice on the best way to rapidly expand nuclear power taking into account passive safety (as top priority after Fukushima), cost, fuel supply, and waste disposal.

    I suppose the problem will be you end up with a massively complex matrix of possible solutions, as the Defence Department seems to do with every major weapons procurement program, and the “best” solution is never obvious. Still, it seems you can hardly rely on any one country or company coming up with a good answer, as you have the hopeless muddle of regulatory differences and political impasses on matters like dealing with waste.

    That’s what I am hankering for, I suppose: some credible independent scientific/engineering advice that said something like: if you are serious about the rapid need for cleaner energy using nuclear, here’s a few options, and you can stop listening to some of the arguments which we demonstrate here as being fatuous. Politicians and public – it’s now up to you to make it happen.

  68. January 5th, 2012 at 10:30 | #68

    If anyone thinks renewables are not demonstrably working now, you are not paying enough attention. South Australia gets 20% of its electricity from wind. It went from getting less than 1% of its electricity from wind to 20% in seven years. I will also mention that point of use solar PV is much cheaper than new nuclear. Note that point of use PV competes with the retail price of electricity while nuclear competes with the wholesale price. Also, solar PV is load following. That is, it better matches demand than nuclear power making the average value of electricity produced considerably higher.

    Also, given the current cost of new nuclear, grid connected solar PV in Australia is cheaper than nuclear.

    This is before the cost of insurance is included, which is quite high for nuclear power. The insurance cost alone for nuclear power appears to be higher than the cost of grid connected solar PV in Australia.

    Nuclear power plants will simply not be built in Australia because of the cost. It is not possible to make money from building them here. If they were built, competition from other sources of electricity will result in them losing very large amounts of money.

  69. Hermit
    January 5th, 2012 at 12:26 | #69

    @Ronald Brak
    SA’s current 1100 MW of nameplate windpower seems commendable. I haven’t checked AEMO figures yet but I suggest it produced well under 100 MW during the hot spell a few days ago. Some 3% of Adelaide homes have a kw or so of PV but I suggest 80% of homes have 2 kw or more of air conditioning.

    The thing that makes possible 20% windpower penetration is the fact that SA also gets 44% of its electrical energy from gas. Torrens Island power station is Australia’s biggest natural gas user. However the two gas pipelines to the Cooper and Otway basins will dry up after 10-15 years according to some data. If that happens there goes not only the windpower but the gas fired baseload. LNG tankers could bring gas around the coastline from WA but it would be at world prices, probably double the current price of piped gas.

    I’ve heard that SA has other major energy resources, not geothermal.

  70. TerjeP
    January 5th, 2012 at 12:40 | #70

    If anyone thinks renewables are not demonstrably working now, you are not paying enough attention.

    Think global and the same could be said for nuclear. 15% of global production. Vastly more than wind and solar. Much safer and cheaper also. At a minimum we should avoid closing nuclear plants where appropriate.

    Having said that there is so much political risk in nuclear investment (in the west at least) that scaling up new capacity will be hard yards. Production capacity for reactor vessels for AP1000 reactors and other LWR designs remains a constraint. Of course we have a reasonable idea the sort of reckless critics that ferment these political risks.

    The presumption in JQ’s article is that nuclear needs to scale up now. Maybe but it also needs some serious innovation to tackle costs. Not because it can’t beat wind and solar on cost but because as yet it can’t beat coal.

  71. Robert Merkel
    January 5th, 2012 at 13:57 | #71

    @John Quiggin

    A good comparison. Airbus entered the industry in 1970 as a consortium of existing firms, with a capacity of about 100 planes a year, which remained pretty much static until the late 1980s. Since then, it’s raised output by a factor of four over twenty years. My demonstration that SMRs were too little, too late allowed for a doubling of scale every five years, or a factor of sixteen over 20 years.

    That’s completely arbitrary.

    The Model T Ford went from 20,000 units annually in 1910 (first full year of production) to a peak of over 2 million a year in 1923 – or, if you’d prefer, a factor of nearly 10 in a decade given production was around 240,000 in 1913.

    Wartime led to even more breakneck expansion in industrial production capability – total aircraft production in the United States went from just over 2,000 in 1939 to 98,000 (much larger and more sophisticated) aircraft in 1944. While that was going on, the US churned out 2,700 Liberty ships, amongst other things.

    In more modern times, you could also look at the exponential growth in production of mobile phones, LCD monitors, and Chinese e-bikes (from 40,000 a year in 1998 to 10,000,000 annually in 2005).

    Not saying that it will happen with SMRs, but I find the assertion that it can’t happen unconvincing.

  72. John Quiggin
    January 5th, 2012 at 14:17 | #72

    @Robert Merkel
    It wasn’t my comparison – I just used the one suggested by a commenter. But a brand-new nuclear power station is much more comparable to a jumbo jet than to any of the other examples you mention.

    And even the other examples you give don’t have anything like the necessary multiplication. The time scale suggests that they’ll be producing at a rate of a handful of units a year by 2020, maybe 100 MW/year. Multiply by 100, as in the case of the Model-T and you’re at 10GW per year by 2035.

    I’m not claiming that a rapid ramp-up would be utterly impossible, given, say, a decision to put the economy on a war footing with the sole objective of producing SMRs. Just that it is so unlikely that it’s not worth considering in a serious discussion.

  73. January 5th, 2012 at 14:27 | #73

    TerjeP, let’s look at nuclear plants under construction in the developed world:

    France – Flamanville unit 3: 1.6 gigawatts. Started December 2007. Current estimated completion date 2016. Cost 6 billion Euros.

    Finland – Olkiluoto unit 3: 1.6 gigwatts, Construction started 2004. Estimate completion 2014. Cost over 5.3 billion Euros.

    USA – Watts Bar unit 2 Tennessee, the only reactor currently under constrution in the US: 1.2 gigawatts. Constrution started before 1973, expected to come online 2012-2013. Cost unclear – Over $2.5 billion US on top of an unknown amount spent in the past, but the EIA, part of the US Department of Energy, estimates a cost of $5,400 US per kw for 2011 for new nuclear in the US.

    In South Australia new wind costs about $7,000 per average kw produced and has minimal operating costs, as opposed to about 3.5 cents US for nuclear per kilowatt-hour. So no, nuclear is not much cheaper than wind. It’s not cheaper than wind at all. And this is before insurance costs which are extremely high for nuclear. Point of use solar PV is becoming competitive with coal and gas in Australia, so that’s cheaper than nuclear too.

    And if the figures I’ve given for the cost of nuclear power are all wrong, and new reactors are so safe that insurance costs are minimal, then where are all the new reactors? Why are hardly any being built? Are people just dumb? And if people are that dumb, should we really allow nuclear reactors to be built and maintained by dumb people?

  74. CM
    January 5th, 2012 at 17:34 | #74

    Just a quick note to say thanks to the ppl who addressed bits of my question from way earlier in the thread.



  75. quokka
    January 5th, 2012 at 20:37 | #75

    As George Monbiot pointed out, the objective is not to deploy renewables or to deploy nuclear. The objective wrt to electricity supply is to decabonize it. Even if the cost of point of use PV reaches parity with grid supplied electricity in some parts of the world, this implies absolutely nothing about the merits or otherwise of nuclear power for supplying the grid. The overwhelming majority of electricity demand is and will continue to be for the foreseeable future met from the grid. Target the emissions coming from this source or abandon any ambitions to avert dangerous climate change.

    Using arguments about the cost of point use solar as compared the cost of grid supplied nuclear is specious bordering on the fundamentally dishonest. The net effect is to generate confusing about the tasks before us.

    As as aside, PV is not load following. There is some correlation with demand – stronger in some parts of the world than others. Load following means what it says – the ability to increase or reduce power output on demand – regardless of weather.

    In the IEA the “2010 Projected Costs of Generating Electricity” lowest and highest estimates for new nuclear capacity are:

    $0.0290/kWh (Sth Korea – OPR1400 – 5% discount rate) to $0.1365/kWh (Switzerland – PWR – 10% discount rate)

    The estimates for the Flamanville EPR are $0.0564/kWh (5% discount rate) to $0.0924/kWh (10% discount rate) based on an overnight cost of $3,860/kW. Due to delays and recently reported overnight cost of EUR 6 billion would increase those LCOE figures by about 20%. This is still competitive with on-shore wind in Europe.

    SCANA corporation in it’s 2011 Analyst Day Presentation provided an estimated LCOE of $0.076/kWh for the VC Summer AP1000 under construction in the US.

    The UK Climate Change Committee provided an estimate of 6-10p/kWh for nuclear in 2011, declining to 4.5-9.0p/kWh in 2040. The respective figure for on shore wind are 8.0-9.5p/kWh in 2011 and 6.5-8.0p/kWh in 2040.

    These estimates are broadly consistent and simply do not support the hand-waving claims that “nuclear is too expensive”.

    Of course LCOE is not the whole story about costs. The UK is certain to introduce some capacity mechanism in it’s electricity market reforms and this will favour reliable generators such as nuclear over intermittent ones. There are costs in backing up intermittent generation and somebody is going to have to pay them.

  76. quokka
    January 5th, 2012 at 21:00 | #76


    A blast from the past indeed and well out of date. The 2010 IEA “Red Book” reported that known uranium reserves increased by 15% from 2007-2009 alone. It is virtually certain that more exploration will produce substantially further reserves. The amount of money put into uranium exploration has been and continues to be tiny compared to fossil fuels.

    Every reference you provide is to work from known anti-nuclear individuals or groups. Their position is – no nuclear at any cost.

  77. January 5th, 2012 at 21:27 | #77

    Quokka, I compared the cost of point of use solar to the cost of nucear power, but you wrote. “Using arguments about the cost of point use solar as compared the cost of grid supplied nuclear is specious bordering on the fundamentally dishonest.” So am I automatically being dishonest for comparing the two? Does the simple act of comparing the two make me dishonest? Is this some kind of philosophy? Or is it theology? I simply do not understand how you reach that conclusion. This is so confusing I am having flashbacks to Sunday school.

    And Quokka, an overnight cost of 6 billion Euros for Flammanville is still more expensive than wind power. Or at least it is here in South Australia. You gotta remember to add in the operating cost. Those nuclear plants got bills to pay that apparently come to a marginal cost of something like 3.5 cents per kilowatt-hour.

  78. quokka
    January 5th, 2012 at 21:43 | #78

    Prof Q seems to imply that any nuclear power plant design that might be considered for Australia should have an operational record of a “couple of hundred” reactor years. This seems to me a criteria designed to reach the conclusion rather than one founded on world wide practice. Has this ever been applied anywhere?

    LWRs are now very well understood engineering and technology and such a requirement seems much too conservative. First of a kind projects are always at risk of cost over runs, but where are the grounds for believing that there is a risk of outright project failure in the sense of substantially not meeting design requirements? Is there is good reason in principle that for example the GE-Hitachi ESBWR be dismissed out of hand? All indications are that it should be as capable and safe as the AP1000 and the supplier certainly has a very long nuclear track record.

  79. quokka
    January 6th, 2012 at 00:54 | #79

    @Ronald Brak

    Wind power is not free of operations and maintenance costs. Figures reported by the IEA have a wide range. For example from the US $0.008/kWh, from Germany $0.037/kWh and from China as $0.015-0.027/kWh. These are for on-shore wind.

    You have far from demonstrated your claim that wind in SA has a lower LCOE than the French EPR. The two European EPRs under construction are the most expensive NPPs in the world and and it is no secret that these projects could have been better executed. They should be seen as the something like the upper bound on new nuclear cost.

  80. January 6th, 2012 at 02:30 | #80

    Quokka, because of what you wrote earlier about me being spacious bordering on the thunderously dishonest for comparing the cost of point of use solar PV to nuclear generated electricty when I was comparing the cost of point of use solar PV to nuclear generated electricity, I now regard you as a crazy person and so I don’t really want to get involved in a discussion with you. I have a mirror for when I want to do things like that. But you can go ahead and demonstrate to me that nuclear is cheaper than wind in South Australia if you like. I’ll be reading even though I’ll try to avoid responding directly to you.

  81. John Quiggin
    January 6th, 2012 at 04:41 | #81

    “First of a kind projects are always at risk of cost over runs”

    In fact, there’s no risk about it in the case of nuclear First Of A Kind projects. They’ve always had cost overruns, sometimes huge ones. Even the Switkowski committee recommended against Australia going with a FOAK project.

    Remember, we woud be starting from scratch in terms of a regulatory framework, the professionals need to run the plant and provide the external regulation, interactions with the company building the plant and so on. To do this with a design that is itself untried is asking for trouble.

    The fact that you’re willing to advocate such an approach suggests you’ve let commitment to the cause overcome critical thinking.

  82. Ikonoclast
    January 6th, 2012 at 07:48 | #82


    Did you bother to read the papers I linked or did you dismiss them out of hand because they don’t provide the answer you want to believe? They comprehensively debunk the IEA Redbook and its assessment methods. The papers I linked to are scientific papers and the methods used are scientific and empirically based in the precise sense of those terms. The IEA Redbook is a business-managerialist document not a scientific document. The method that the IEA Redbook follows is the standard business-managerialist method. I wonder Quokka if you understand the profound difference between a scientific assessment and a business-managerialist assessment?

    A thorough scientific assessment of this nature uses known and verified data and makes calculated extrapolations on a basis consistent with mathematical probability theory. A business-managerialist assessment (like the politicised assessment to which it is a close cousin) starts with the desired answer ball-park in mind. It then jigs and re-jigs the “data” and the pseudo-formulas it uses until it arrives at the desired answer. It might couch all this in a pseudo-scientific format but it is anything but a scientific document.

    This might sound like an exaggeration or parody of business-managerialist report methods but it is indeed an accurate and consistent picture of what goes on in about 99% of such reports. If you are not aware that this is the case, then there is a large gap in your real world education.

  83. John Quiggin
    January 6th, 2012 at 08:19 | #83

    Regardless of its motivation, the IEA has a lousy track record in forecasting. I don’t see any reason to place any weight on their estimates


  84. quokka
    January 6th, 2012 at 14:10 | #84

    @John Quiggin

    It is not true that FOAK nuclear projects have always had time and cost overruns. Sizewell B. the first and only light water NPP in the UK came in on time and on budget. It was and is radically different from the UKs existing fleet of Magnox and AGRs. But it was established technology in other countries. “Sizewell B (SXB) won the prestigious British Construction Industry Award in 1994 in the civil engineering category and also the Supreme Award selected from all the competition categories.”

    Source: Royal Academy of Engineering – Engineering the Future: Nuclear Lessons Learned

    This is a very interesting document to aid understanding of these issues, what has gone wrong and what has gone right in recent or current nuclear builds and how lessons can, should and will be learned.

    Just one counter example is sufficient to refute “common knowledge” such as “all FOAK nuclear projects overrun” and demonstrate the need for detailed knowledge and analysis.

    I’m not advocating that Australia (or anywhere else) should or shouldn’t do an FOAK nuclear project. Rather I suggest that should we come to such a point all credible proposals be evaluated on their merits. Among such criteria, I would imagine, would be the degree of financial responsibility to be assumed by the vendor for project overruns. I would also assume that more weight might be applied to this metric for proposals for designs without significant track record.

    Don’t know why this demonstrates abandonment of critical thinking.

  85. quokka
    January 6th, 2012 at 16:48 | #85

    @John Quiggin

    Lets compare costs in IEA Projected Costs of Generating Electricity 2010 to Melb U via SMH

    Melb U:

    Wind: $130/MWh
    PV: $300/MWh

    IEA (median costs):

    Wind (Nth America): $62 – 92/MWh at 5% and 10% discount rates respectively
    Wind (Europe): $110-150/MWh
    Wind (Asia Pac): $75-115/MWh

    PV (US): $215-333/MWh

    I don’t see anything in the SMH report to suggest that IEA has it fundamentally wrong. There actually seems to be reasonable agreement.

  86. John Quiggin
    January 6th, 2012 at 17:03 | #86

    “But it was established technology in other countries.”

    My point exactly. Australia, as a beginner should confine itself to technology that is already established overseas, rather than going for a world FOAK. If you want to give a relevant counterexample, point to a world FOAK nuclear plant, in a country with no previous nuclear experience, that came in under budget and worked as well or better than projected.

  87. quokka
    January 6th, 2012 at 18:30 | #87


    If you are referring to the Jan Willem Storm van Leeuwen and Philip Smith paper that purports to show a negative EROEI for uranium extraction from seawater, then look elsewhere. This paper relies on studies from the 1970s that assume pumping of huge amounts of sea water. Hardly surprising that there would be large energy expenditure! The Japanese experiment referenced above relies only on natural water flow from currents and tides. The laws of physics remain intact, and extraction of uranium from seawater remains a possibility even for a once through fuel cycle. The MIT report came to this conclusion – needs more study but certainly not ruled out.

    I take it that we can assume that the EROEI argument against uranium from seawater is debunked and we do not have to engage in this whack-a-mole game again sometime in the future?

  88. Ken Fabian
    January 7th, 2012 at 16:33 | #88

    The choice of the Right in fossil fuel rich and dependent nations like the US and Australia to deny science based reality stole the opportunity 2 decades ago for nuclear to be the dominant solution for bringing down emissions. Most pro-nukers still think it’s the fault of Greenies – because that’s who the Right continue to try and blame in order to divert attention from their own unwillingness to back nuclear. Even the truth about climate change was never enough to bring the Right to dump fossil fuels for nuclear. In the process they allowed solar a big opportunity to advance, sure in the knowledge that solar would never be able to rise to the challenge whilst nuclear was the real threat.

    The Right should get credit for their key role in giving renewables the big break they needed!

    In places like the USA and Australia nuclear will get no serious political backing until and unless climate denial collapses; the same lot that stabbed nuclear in the back will be knocking on nuclear’s door like there was never a bad moment between them. The pro-nukers, so desperate for any kind of support and blind to just how deep a betrayal climate denial was for nuclear, will welcome the renewed attention and still think it was those greenies that stole their big opportunity.

    Meanwhile so many methods of doing solar work well enough that it hasn’t even reached the point of a clear and obvious ‘best’ technology and that well of innovation is far from dry; they can hopefully take advantage of the start that the combination of Right abandoning nuclear and Left backing renewables has given it.

Comments are closed.