Prescription for the Planet – Part IV – Show me the money!

We’ve now covered all the major technologies proposed by Tom Blees in the book Prescription for the PlanetIntegral Fast Reactor nuclear power for electricity generation, boron-fuelled vehicles for transport, and plasma burners for recycling of waste. Set in the context of a legacy of ongoing problems, with stockpiles of nuclear waste and weapons, a rapidly degrading environment and climate system, world energy security standing on a knife edge, and a future of looming shortages as we struggle to make up shortfalls using zero-carbon energy, Blees says we either founder as a civilisation, or choose to re-invent ourselves and emerge renewed as an equitable and sustainable society.

Now that all sounds fine and dandy, but in the cold hard light of day, it’s not all that realistic… is it? I mean, the staggering cost of retooling our entire energy and transport industry, on a planetary scale, is just too high and too difficult, and the pay offs such change might yield are just not worth the pain of adjustment. Business-as-usual is surely the better, cheaper course, at least for now; we should let future generations sort out the energy problem since they’ll all be richer than us thanks to the magic of economic growth. Well, that, in caricature, is what the ‘fossil fools forever’ mob will try to tell you. Blees says they’re wrong — on both counts.

This post, part IV of VI, reviews chapters 8 and 9:

— Chapter 8: Check, Please! (pg 197-240)

— Chapter 9: Cui Bono? (pg 241-262)

I’ll give you the bottom line of these two chapters up front: it will almost certainly cost us less to power the world with IFRs (and whatever contribution renewable energy ends up making — sizeable, I hope) than it will to try and reinforce our creaking, aging fossil fuel infrastructure. Economic arguments pass muster.

Let’s start with the basics. If you are seeking solid ground on relative costs of delivered energy, the assessment of the real-world price of any energy technology can be grounded on three basic principles [you need all three]: (i) discounted cash flow analysis, (ii) scale-up capacity assessment, and (iii) recent experience.

Step (i) is always possible, but  can be prone to wild speculation when hard data on the model parameters are lacking or difficult to estimate in context. For instance, you really need all of the following: interest/discount rate for future value of money estimates, lifetime of infrastructure, capital recovery factor, installed capital cost including interest during construction, average capacity factor of power delivery, annualised capital and decomissioning cost, fuel cost, and cost of operation and maintenance.

Phew! Of course there are, in truth, many details besides these that are involved, such as relative baseload vs peak usage (usually subsumed with the capacity factor and annualised cost estimates), transmission connection (usually within installed capital costs), etc. But putting these complexities aside, the end product of such accounting really depends on whether you are able to get accurate operational data on each of these inputs, or alternatively, whether you pull some parameter values out of certain dark orifices.  Indeed, you can basically come up with anything, from quite reasonable estimates through to wildly under- or over-blown costings of long-term delivered energy (usually expressed as c/KWh). My advice — check assumptions carefully!

Step (ii) is really a topic for another day, but involves the costs and logistical challenges associated with going from small-scale to large-scale operation. For instance, wind power operates perfectly well on a fossil-fuel-based grid infrastructure when it contributes a small fraction of total power, but as it constitutes more and more of the total installed capacity, issues of energy storage or backup from other generating sources come into play. Such problems are not insoluble, but they do impact on the bottom line in ways that cannot necessarily be anticipated using simple ‘bottom-up’ approaches that work by costing each unit and then multiplying by the amount that would be needed.

The guts of Chapter 8 is lots of numbers — really big numbers. It even comes with a vertigo warning. There is a lot of material here, and so I can’t cover all of the key details in a brief review like this. Which is kind of a shame, because you really have to assess the logic of the chapter in full to appreciate the validity of the fundamental argument Blees is trying to make: it can make real financial sense to build IFRs on a massive scale.

Step (iii), recent experience, is the primary basis upon which Blees bases his costings. Perhaps some of the details can come out in the comments of this post, when folks here have specific questions. So, just the core points then:

1) Repeated studies from authoritative sources such as the OECD and IEA show nuclear power is highly cost-competitive with coal (indeed, cheaper 10 of 12 countries assessed).

2) Recent experience shows that modern nuclear reactors can indeed be built to a tight budget — in the range of $1.4 to $2.5 billion per GW capital cost.

3) Reactors become far cheaper if they have the following characteristics: (a) standardised blueprints, (b) simpler design, (c) factory-built modular units, which can be trucked to site, (d) a system within which legalistic impediments are surmounted by sound legislation (one of the big problems in the US).

Note: Standardisation and modularity are the game-changers for the nuclear power industry. For instance, Generation III and III+ light water reactors, which follow these principles, such as France’s European Pressurised Reactor (EPR), GE’s Advanced Boiling Water Reactor (ABWR) and Economic Simplified Boiling Water Reactor (ESBWR), and Westinghouse’s AP-1000, cost around $1 to 2 billion per GW installed. Two ABWR were built in the late nineties in Japan for $1.4 B/GW within 36 months, and China has ordered 100 x AP-1000 units, carrying a price tag of $1 B/GW. These are likely to most closely reflect the price tag attached to an S-PRISM (Super Power Reactor Innovative Small Module — a sodium-cooled fast spectrum reactor with metal fuel). GEH estimates the cost of an S-PRISM at $1.3 B/GW, with a high-end estimate for fast reactors of $2.5 B/GW. As to a nation going nuclear in a big way, and benefiting, France is the stand-out real-world example on a per captia basis, with 59 light water reactor plants generating over 63 GWe (80% of supply). Their electricity costs are among the lowest in Europe, at around 3 eurocents per KWh.

4) If the world was to replace it’s energy supply with IFRs, the cost would be roughly $28 trillion (including transmission lines). That’s on the basis of a capital cost of $2 B/GW (so ignoring likely economies of scale that will bring prices down) and the need to supply around 8.75 terrawatts (TW) of generating capacity (for all energy use). For this, you’d need about 3,500 power plants of 2.5 GW each (using 8 x 380 MW modular reactor vessels within 4 power blocks). Any further energy supplied by renewables, geothermal and syngas (e.g., from plasma burners) would be a bonus. For reference, global electricity production in 2005 amounted to 2.3 TW, 16% of which came from nuclear.

5) What may be surprising to many is that the cost of business-as-usual energy development, or a gradual path to de-carbonisation, is about $26 to 35 trillion from 2010 to 2030! These are estimates that come from credible sources such as the International Energy Agency’s World Energy Outlook 2008 and Stern Review on the Economics of Climate Change 2006. This is for the cost of shoring up our fossil fuel infrastructure and upgrading/replacing transmission infrastructure (IEA), or investing 1% of GDP per annum on carbon mitigation (Stern).

6) At the height of its nuclear build-out phase, France was rolling out 6 plants per year. Six countries have a GDP higher than France and all already possess the technology to build fast reactors: USA, China, Japan, India (building one now), Germany and the UK. At France’s historical rate, these countries could together build 117 IFR plants per year, with no greater urgency than the French brought to bear on their road to energy independence. Indeed, China is rolling out over 50 large coal-fired power stations of equivalent size each year. So at this quite feasible rate, it would take 30 years to build 3,500 plants in 7 countries. For less than the cost of reinforcing our fossil fuel infrastructure.

7) There are further paybacks from an IFR roll out: we solve the multi-billion dollar nuclear waste problem fully, and we also save big $$ (and lots of lives and avoidable misery) by drastically cutting air pollution (renewables also achieves this) — a best estimate of $167 billion per year, each year, from the US alone (and a whole lot more from China — I can attest personally to the health effects of that Asian Brown Cloud).


graphic-2largeGo and read the extended version in P4TP and see what you think about the credibility of the above analysis. It certainly looks robust to me.

Chapter 9, Cui Bono? (who benefits?), explores the multitude of problems with the privatised portion of the American power industry, with special attention to the past litany of misdemeanours and cover-ups of some nuclear utilities and related problems with its regulator, the Nuclear Regulatory Commission (NRC).

Basically, due to the regulatory framework in the US, energy utilities have the ability to gouge their customers with exorbitant pricing (under certain conditions), and most take full advantage of that opportunity whenever they can. Energy is like a societal drug addiction – we can’t do without it (even for a short while), and so we are acutely vulnerable to being exploited by ‘dealers’ when energy is not in abundant supply. Indeed, this goes right through from spot prices of energy delivered to cost estimates of new nuclear power stations. Remember Step (i) above? Well, in the good old US of A, nuclear utilities can have a field day with that number crunching, and charge customers for current power costs on the basis of these pick-and-choose models!

Being rather sympathetic myself to the benefits of publically owned service providers, I think Blees makes a strong case for full public ownership of nuclear power. It’s a form of socialism, to be sure, but before the free market ideologues start frothing at the mouth, consider where society would be without public ownership. Transportation infrastructure, public education, national defense, standards authorities, etc. It’s really a matter not of if public ownership is appropriate, but when. And when we are talking about something as crucial as oversight of nuclear power and a secure energy supply, well… I’ll leave you to judge (but after you’ve read the chapter, please!).

The message about ownership and oversight has a deeper purpose than merely bringing energy costs in America under control. Without something similar operating worldwide, the risks of rampant unmanaged global nuclear energy deployment could well outweigh the benefits it brings to individual countries — perhaps catastrophically so.

We need something GREAT to ensure a global rollout of IFRs is safe and equitable. That’s just what the next part of P4TP is all about…

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to Furl


By Barry Brook

Barry Brook is an ARC Laureate Fellow and Chair of Environmental Sustainability at the University of Tasmania. He researches global change, ecology and energy.

36 replies on “Prescription for the Planet – Part IV – Show me the money!”

Nuclear power may be an important source of energy in Australia’s future. However, at the moment it is very expensive. Currently new reactors in the United States cost about $7,300 Australian per average kilowatt of output and up. A new Finnish reactor under construction is three years behind schedule and costs about $6,200 per average kilowatt of output. A recent tendering process in Turkey resulted in a consortium offering to supply nuclear energy at a cost of 32 cents a kilowatt-hour. These prices make nuclear energy more expensive than solar thermal power in Australia and much more expensive than wind. At this time investing in wind power and energy efficiency are the cheapest ways to cut our CO2 emissions in this country. Fortunately, due to Australia’s large size and relatively low population, there is no shortage of suitable sites for wind turbines and there is currently no economic reason why wind can’t supply 20% of Australia’s electricity. If in the future the cost of nuclear energy drops below that of other alternatives, then it would make sense to use it. However, as the cost of other low emission sources of energy are rapidly decreasing, nuclear energy may never be competitive in Australia.


Ronald, please re-read the above to understand the context of those US and Turkish cost estimates, and follows those hyperlinks to see the real-world experience, which says US 2.5 to 4.5 c/KWh is the norm, not 7.3 to 32 c/KWh.


As a further point to the above costs, a back of the envelope calculation is enough to show how silly numbers like 32 c/KWh are for nuclear (those Turks were being taken for a ride, I presume).

For a 2.5 GW plant costing, say $9 billion (at the top of the quoted range in US), you would yield 8760 (Hours in year) x 0.9 (capacity factor) x 2500 (MW output) x 0.32 (c/KWh) x 1000 (KW to MW) = $6.3 billion in income each year. At that rate you’d have paid off your capital in less than 2 years, your interest in another year or two (accepting O&M costs) and then be making money hand-over-fist for the next 36 to 56 years lifetime of your reactor.

Really I’d need to add more detail to make the above precise, but it’s clearly such an absurd calculation it doesn’t bear breaking down any further than this.


Tom Blees has told me that the marginal cost of nuclear power in the US is 1.68 US cents. So if the capital costs of a nuclear plant are $1.4 billion US a gigawatt and the cost of capital is low, then it could produce electricity at about 2.75 US cents per kilowatt-hour, or about 4.26 Australian cents, assuming there are no other costs. If we could get nuclear reactors built at that price it would be great. However, I doubt that anyone will currently offer to build a reactor for that much in Australia. I think they considerably more than that and I don’t think we currently have any special ability at containing prices compared to people in Finland and the US. But I could be wrong about this, and as I don’t want my pessimism to result in Australia missing out on cheap, low emission electricity, I suggest that Australian make it known that it will welcome any company or consortium that wants to build a nuclear plant here and sell electricity at market rates. There are some issues that would have to be addressed. The plant would have to meet levels of safety expected for modern reactors, a minimum level of security would have to be met, insurance would have to be paid for, and a decommissioning fund would need to be set up, but overall it should be treated much the same as any other power plant. As the average wholesale cost of electricity in 2007 in South Australia was 7.5 cents a kilowatt-hour, a nuclear plant that could produce electricity at 4.26 Australian cents a kilowatt-hour would make a huge amount of money, so companies should jump at the chance if they can produce electricity that cheaply. But in the meantime I think we should cut CO2 emissions using energy efficiency and wind power, which is currently the cheapest source of low emission electricity in Australia.


One worry from the news: what’s really going on now.

“… Siemens, Germany’s engineering giant, would divest the 34% stake it has held since 2001 in Areva NP, a Franco-German joint venture in nuclear reactors…. Nuclear power is in the midst of a global comeback, and Areva NP’s new design, the European Pressurised Reactor (EPR), is leading the field against competing blueprints from American and Japanese rivals…. a Finnish utility, is demanding €2.4 billion in compensation from the two firms for delays in building the EPR at Olkiluoto ….”

The Finnish story is a management/design/execution disaster. Design info here:

(EPR is a “Generation 3+” design — definitely NOT the advanced type.)

Aside: ever notice how every large technology seems to go one step beyond the point at which it should have stopped, and build one last version?

That happened with steam locomotives.

It happened with plywood aircraft.

It’s probably already happened with boiling water reactors.

Here’s Charlie Stross on visiting a currently operating Advanced Gas-cooled Reactor (AGR) (“filled with carbon dioxide, circulating at a temperature of 700-800 degrees celsius”):

That’s more than needed to retrofit the latest supercritical coal plant:

We’re now backing away from the really high temperature plant designs. That’s hopeful in one sense — we’ll have results from materials used in those. But it leaves the high end most thermodynamically efficient models heated by coal.


PS — Barry, someone ought to cost out the benefits of removing coal trains from the railroads! John McPhee did a wonderful book on them. The length (tying up crossings and lines, halting other traffic both on roads and rail); the weight and maintenance costs.

Another item is the value of the railroad rights-of-way (which is absolutely enormous) — there are plenty of 21st century uses for cross-country uninterrupted rights of way, both for data and for transportation — and it’s often pissed away and broken up forcing new uses to go to great cost to reacquire it.


Barry, I have no idea why the Turkish bid price was so high. That’s simply the price the Atomstroyexport-JSC Inter Rao Ues-Park Teknik joint venture asked for. But the capital costs of the new Finnish reactor, which are lower than US new reactor capital costs, are about 40% higher per average kilowatt of output than for wind power in Australia. Their costs may have been lower if they had used a less advanced reactor design, but even so, I doubt the French/German companies involved would offer to build a nuclear plant in Australia at a cost that competes with wind’s capital costs combined with its very low operating costs. Of course, if they are willing to do so, we should welcome them. I just doubt it will happen.


Barry, I’ve received some new information from Tom Blees. He told me that GE says they can build reactors that produce electricity, including all costs, at 7.13 Australian cents per kilowatt-hour. If they could produce electricity for this price in Australia it would be competitive as in South Australia the average wholesale price of electricity was 7.5 cents a kilowatt-hour in 2007. However, as there are increased costs and risks involved for GE in building reactors in Australia, combined with the fact that since reactors are large bringing one on line tends to push down the price of electricity, I doubt that GE would think it would be profitable to do so. And I have doubts that it would be profitable even after our mild carbon trading scheme is introduced. However, if GE wants to build a reactor in Australia we should of course let them.

Most countries have higher wholesale electricity costs than Australia, so I would expect GE to build reactors in those countries first to take advantage of higher profit margins. However, this doesn’t seem to be happening at the moment.


Barry Brook – “Two ABWR were built in the late nineties in Japan for $1.4 B/GW within 36 months,”

From my original reference:

“Under a DOE program for promoting building of new-generation nuclear plants, a $4 million feasibility study on building two ABWRs at Bellefonte in Alabama was undertaken in 2004 by the Tennessee Valley Authority (TVA) plus vendor GE as well as Bechtel and others. The study showed that twin 1371 MWe ABWRs would cost $1611 per kilowatt, or if they were uprated to 1465 MWe each, $1535 /kW, and be built in 40 months.

Based on this study, Florida Power & Light in February 2008 released projected figures for two new AP1000 reactors at its proposed Turkey Point site. These took into account increases of some 50% in material, equipment and labour since 2004. The new figures for overnight capital cost ranged from $2444 to $3582 /kW, or when grossed up to include cooling towers, site works, land costs, transmission costs and risk management, the total cost came to $3108 to $4540 per kilowatt. Adding in finance charges almost doubled the overall figures at $5780 to $8071 /kW. FPL said that alternatives to nuclear for the plant were not economically attractive. ”

With the relevant passage being:

“These took into account increases of some 50% in material, equipment and labour since 2004.”

The estimates in 2004 were 50% lower than those for 2008. The has been a huge increase in the cost of materials etc in the nuclear industry pushing prices much higher.

Despite what Tom says the reactors built in Japan were costed before 2004 so those figures cannot be taken as anything like realistic for 2008. The Finnish reactor is closer to the mark as it is now experiencing the shortages in materials and qualified personnel hence the delays and cost escalations.

I asked Tom for a recent cost breakdown of the Japanese reactors that were built. I cannot magically conjure a copy of his book to read so I thought in the meantime he would be able to supply detailed figures seeings as his cost estimates seem to be based around these figures.

I have supplied what I believe are accurate figures from the World Nuclear Association on costs in 2008.

“Step (ii) is really a topic for another day, but involves the costs and logistical challenges associated with going from small-scale to large-scale operation. For instance, wind power operates perfectly well on a fossil-fuel-based grid infrastructure when it contributes a small fraction of total power,”

Mark Diesendorf has peer reviewed work on this subject and has found that wind has a capacity credit so it can displace baseload equal to it average output. I posted links to his work.


Hank #6 – thanks for the interesting links, and I agree with your point about taking old tech to the nth +1 degree. I’ve seen discussion that France are not so sure their EPR is the best way to proceed since it’s been superseded by the AP-1000 and ESBWR. The EPR is what was being built in Finland (re: Ron #8) — first one I think, so teething problems are expected, much as is the case for all demo/first order plants (e.g. Nevada Solar One also). I agree that GE are hardly likely to chose Oz to be the first place to build something like the ESBWR — it will either be somewhere cheap (e.g. China) or where power is more expensive and coal is a more limited option (e.g. Japan).

Hank #7 — good point about railroad opportunity costs…

Ender #10: Yes, as the spot price on materials and labour rises, so will the installed cost of a power plant that uses these. This is why the cost of installed wind power also went through the roof in 2008. For instance, look at this table:

Per megawatt, wind uses 11.5 times the amount of steel and 4.5 times the amount of concrete of the new design of nuclear power plants. So you can multiply your 50% cost increase figures by 4 to 11 times to work out what impact this will have on wind power in 2008 vs 2004 wind costs. This is one reason why T. Boone Picken’s giant West Texas wind farm is costing so much and has contributed to the project being delayed until further notice…




Great Blog – great discussion.

Thanks to Joe Campbell for posting a comment / link to this site on my own feeble (and significantly less hit) blog on the consideration of nuclear power in Australia. I’ve been hoping to find a forum like this for some months now. The time I have available to post has decreased sharply as global interest in nuclear technologies has surged in the past few years.

I look forward to joining your discussions after making some time to back-read more of your posts.


> old tech to the nth+1

For anyone who didn’t click through, that’s the Charlie Stross link above. A brief excerpt to make the point — those reactors run as hot as the newer coal plants operate, for the increase in thermodynamic efficiency that allows:

“… the weirdly victorian-looking plumbing around the 12 metre level (where the manual last-ditch controls are available, all brass dials and hand-wheels). What I think I should end with is an explanation of the title of this piece–

As Les explained, “nothing like this will be built again”. The AGRs at Torness are not ordinary civil power reactors. Designed in the 1970’s, they were the UK’s bid to build an export-earning civil nuclear power system. They’re sensitive thoroughbreds, able to reach a peak conversion efficiency of 43% — that is, able to turn up to 43% of their energy output into electricity. By comparison, a PWR peaks at 31-32%. However, the PWRs have won the race for commercial success: they’re much, much, simpler. AGRs are like Concorde — technological marvels, extremely sophisticated and efficient, and just too damned expensive and complex for their own good. (You want complexity? Torness was opened in 1989. For many years thereafter, its roughly fifty thousand kilometres of aluminium plumbing made it the most complex and demanding piece of pipework in Europe. You want size? The multi-thousand ton reactor core of an AGR is bigger than the entire plant at some PWR installations.)

It’s a weird experience, crawling over the guts of one of the marvels of the atomic age, smelling the thing (mostly machine oil and steam, and a hint of ozone near the transformers), all the while knowing that although it’s one of the safest and most energy-efficient civilian power reactors ever built it’s a a technological dead-end, that there won’t be any more of them, and that when it shuts down in thirty or forty years’ time this colossal collision between space age physics and victorian plumbing will be relegated to a footnote in the history books. “Energy too cheap to meter” it ain’t, but as a symbol of what we can achieve through engineering it’s hard to beat. …”


There’s perhaps the challenge for GenIV.


Barry Brook – “So you can multiply your 50% cost increase figures by 4 to 11 times to work out what impact this will have on wind power in 2008 vs 2004 wind costs.”

I think you are grasping at straws a bit here Barry. Wind increased 30% last year:

“In total, the global installed capacity for wind energy went up nearly 30 percent last year, reaching 121GW. The 27GW installed represent an increase in total installations of 36 percent compared to the figures from 2007. The Global Wind Energy Council estimates that the market for new facilities alone is nearly $50 billion dollars.”

Boone Pickens was hit more by the global financial meltdown than increasing costs.

Also the figures that relate the cost of wind V nuclear could be based on 1990 era wind turbines:

This is a comment on a post where the study by Professor P.F. Peterson of UC Berkeley undertaken in 2005 was also posted.

“I recall reading elsewhere in the blogoshere that Dr. Peterson based his material inputs on 1990s vintage wind turbines, because that was data he had available. I’ve looked but haven’t found a copy of his actual study on-line to check the claim.”

I also have not been able find a copy of the study. Prof Peterson may have also used very low capacity factors for wind. As wind increased 30% during a time when materials increased 50% then it seems that this study is out of date.


Ender, I agree that the GFC also hit Pickens’ plans hard. But my simple point was that you can’t on the one hand claim that costs for nuclear will have skyrocketed because of rising material costs, and on the other tacitly imply that this wouldn’t also be the case for other renewables (or coal, etc.), when the material costs for these are higher. A design like the S-PRISM is more immune to material costs than even the figures quoted above, because (for instance), it does not require the manufacture of a specialised steel pressure vessel (its pool design operates at atmospheric pressure).


G.R.L Cowan – “Ender could say of the B-power paper he said he found that it was peer-reviewed, if it was. Hard to say since he never linked it.”

Not sure what you mean here – can you please explain this?


I will point out that commodity price increases have very little to do with the current run up in cost of nuclear reactors or with the capital costs of other generating capacity. One study put the extra cost of nuclear plants due to commodity increases at only $54 Australian per kilowatt. It is very clear that commodity prices have little to do with the cost of nuclear power when you consider how little the bulk material required to build a reactor actually costs. It takes roughly 10,000+ tons of cement to build a reactor containment building. Cement prices vary but are roughly $100 a ton, so 10,000 tons of cement would only cost around one million dollars. I don’t know how much steel would be required, but if it was 2,000 tons it would cost around two million dollars. Together this comes to three million dollars, which is less than half a percent of the cost of reactors under construction today.


Barry Brook – “But my simple point was that you can’t on the one hand claim that costs for nuclear will have skyrocketed because of rising material costs, and on the other tacitly imply that this wouldn’t also be the case for other renewables”

I am not claiming anything. These are real figures from the World Nuclear Association about the real costs of nuclear. Sure the increase in material prices hit the wind industry however obviously it did not hit it hard enough to dent the rampant growth. Nuclear uses some very specialised components that are not used anywhere else. Its steels have to withstand intense radiation and heat. Also the welds and joins have to be of a much higher standard because of the relative dangers of a leak or failure. If a wind turbine fails there is no risk of radiation leak etc so all its welds are subject to normal commercial standards and it’s component materials are not subject the same temperature and radiation regime as a nuclear reactor. It is the certified materials, welds and join inspections that push up the price of nuclear. The S-PRISM is not immune to this as it has to be inspected and certified. It is also subject to much higher neutron flux than normal thermal reactors and higher temperatures. As you know you never get anything for nothing so if you operate your reactor at ambient pressure you have to increase the temperature to retain thermal efficiency.

You also cannot base your cost projections on outlyers. The two Japanese reactors that came in at $1400/kW were atypical and cannot be used for estimating costs in 2008 onward for any reactor. The cost savings of the S-PRISM design have yet to be proven so you need to estimate costs on the basis of current reactors. The IFR has a bath of liquid sodium that must be kept from any contaminant. This is not an area where you want to apply cost cutting in weld or join quality. How long do you think the the IFR program would last in the face of a sodium explosion? So everything about the IFR core MUST be of the highest possible standard or you are risking disaster for the whole IFR program. This is where I think you will have the most problems with mass production and savings from modularity and again where I dispute Tom’s rollout schedule.

My layperson’s guess for estimating the IFR cost is a standard normal curve centered over $3000.00/kW. This puts Tom’s figures (that I have seen here) of $1500/kW in what I think is the right probability range.


I’ll take General Electric’s and Livermore Lab’s costings over your layperson probability distribution, Ender. See other thread on Response to IFR critique for more on that.

Regarding quality assurance, I agree, but I think you’re trying to tell GEH how to suck eggs.

In what way were the Japanese ABWR prices atypical?

These are real figures from the World Nuclear Association about the real costs of nuclear.

Actually, most are cost projections. Why are the OECD and EC figures (including 2010 cost projections), and other figures the World Nuclear Org and Tom cites, using data for many countries and giving a range of 4-5.5 c/KWh, not real?

The EU 2007 study they cite gives the following figures:
Nuclear = 5.5 – 7.4
Coal = 4.7 – 6.1
Gas = 4.6 – 6.1
Wind onshore = 4.7 – 14.8
Wind offshore = 8.2 – 20.2


> As you know you never get anything for nothing so if you
> operate your reactor at ambient pressure you have to
> increase the temperature to retain thermal efficiency.

This is total non-sequitur. Temperature of turbine determines thermal efficiency. A steam turbine coupled to a 520C HX of a LMFBR such as IFR has obviously higher thermal efficiency than one coupled to 320C steam generators of a PWR.

Pressure of a reactor primary circuit is determined by nature of coolant. LMFBRs operate at ambient pressure because liquid metal coolant does not need any pressurizing, in contrast to PWRs.

Concerning the costs and schedule, given the fact that the major cost and schedule over-runs were either caused by legal action of antinuclear industry, or the first-of-a-kind issues (or both), the cost and 4year build schedule of Japanese plants is indicative of N+1 plant ones, once the regulatory framework and the construction practices are streamlined.


Barry Brook – “I’ll take General Electric’s and Livermore Lab’s costings over your layperson probability distribution, Ender.”

OK fair enough – I do not want to provoke another round of Ender fatigue – we will see how it goes if it gets built.

Ondrech – “This is total non-sequitur. Temperature of turbine determines thermal efficiency. ”

OK I accept your correction. As I have admitted I am a layperson in this topic and learning is a part of why I post.


As wind increased 30% during a time when materials increased 50% then it seems that this study is out of date.

Wind increased 30%? Capacity factors when Peterson did his study were at least 21%. Today the average is no more than 25%, if that (I think I read recently that average is 23%). As for construction materials cost increases, as Ronald correctly pointed out above they are inconsequential for nuclear. 70s-era plants could buy their construction materials today for about $35/kw, and the newer designs like the PRISM will use substantially less material yet.

You also cannot base your cost projections on outlyers. The two Japanese reactors that came in at $1400/kW were atypical and cannot be used for estimating costs in 2008 onward for any reactor.

Outliers? The two ABWRs were the first two Gen III reactors ever built. The newer designs (AP-1000 and ESBWR) are even simpler and use even less material than the ABWR. But since those two reactors are the only ones that represent Gen III that are actually built and operating, it would be pretty ridiculous to NOT cite them and claim that it can’t be done again at that price. (Please don’t tell me about the EPR in Finland, that’s a case of AREVA building an already-obsolete Gen III reactor, and not doing it well to boot). In fact, GE claims that with what they learned building them they figure they could build new ones now for about $1200/kW. Since commodities cost increases of nuclear are immaterial and interest costs are at record lows, while labor costs have (alas) not risen, these would seem to be the best examples of what nuclear power should cost, not outliers to be ignored.


Don’t forget first of a kind (FOK) to nth of a kind construction cost reductions. The AREVA design will now be built in France (two units on the way, with a third being discussed). Those projects will provide more data relative to that design. Progress of the two AP1000s, about to pour first concrete in China, will also be interesting to observe.

The costs savings due to the experience of an implemented project (including real-world feedback given to designers) is discussed on the OECD document I linked to above. Most of the savings come from construction schedule reductions.

For example, in Korea the 1048 MWe (each) ULCHIN-5 and ULCHIN-6 reactors, were connected in 2003 and 2005 respectively, not much more than four years after the start of construction.

Similar experience may be found in Japan at the 1100 MWe HIGASHI DORI-1 (2005 connection; 4 years, 4 months construction), and the 1267 MWe HAMAOKA-5 (2004 connection; 3 years, 9 months construction).



Hank, do you mean the facility to process the oxides from spent fuel, or an IFR pyroprocessing facility? If it is the former, then yes, it makes perfect sense to build this independent of the Gen IV reactor. There has to be a need to transport waste and fuel to an IFR when it is being set up. It is just that nothing needs to come out, except vitrified fission products which can go to storage when the IFR facility is eventually decomissioned.


Tom Blees – “Capacity factors when Peterson did his study were at least 21%. Today the average is no more than 25%, if that (I think I read recently that average is 23%). As for construction materials cost increases, as Ronald correctly pointed out above they are inconsequential for nuclear. 70s-era plants could buy their construction materials today for about $35/kw, and the newer designs like the PRISM will use substantially less material yet.”

Yes but modern turbines generate far more for less materials as they increase in size. If Peterson did his study on 600kW turbines, which was the average size then, then this would skew the result as a 600kw has much less output per unit construction material than the 3MW ones that are in production today. Also most modern turbines are variable speed which uses less material than the constant speed turbines that were dominant then.

“Outliers? The two ABWRs were the first two Gen III reactors ever built. The newer designs (AP-1000 and ESBWR) are even simpler and use even less material than the ABWR.”

OK – however you cannot provide a breakdown for the costs for these reactors. You also have not provided any data to substantiate your claim that newer reactors with streamlined approval and building codes will be cheaper. I have provided figures that show there has been a 50% increase in nuclear costs including the AP-1000.

Yes you could base your costings on the cheapest reactors you can find however imagine that you were in the position to build these reactors. Would you go to the financiers with the cheapest costs from 2004 and ask for 2 billion for your reactor? Or would you average the latest costs for reactors built everywhere and come up with a figure of 6 billion?


Ender, yes, I agree that ongoing technological development of wind turbines has greatly improved their efficiency, and that a 30% average capacity factor across a fairly wide range of good sites [onshore and offshore] is quite feasible (it drops when energy storage is added).

Re: AP1000 costs, it will be interesting to see the numbers that come out of the first two AP1000s being constructed in China right now. As Tom has pointed out, the increase in costs, if they are real, do not reflect materials or labour, and so will stem from other externalities that may not be applicable in countries with a different regulatory system to the US.


Wind has been given every opportunity – hundreds of billions – in what is still an experiment. It would seem reasonable to try something on that scale to solve the baseload problem.
I just don’t get it. How can pundits ever hope to price something as complicated as a reactor or a wind energy system? Let the engineers crunch the numbers and do what is best for the local grid, given the goal of eliminating CO2 (and other environmental factors). When politicians make energy decisions they will always make them based on politics.


It is not finalized as yet, but China’s deal with Toshiba for two AP1000 reactors is estimated at around $12.5 billion Australian. It is known that China is paying the state-owned French company Areva $18.4 billion for two EPRs.


I’d like to see some sociologist study perceived risk — from the point of view of contractors and construction workers — comparing how they feel about cutting corners.

Compare, say, how they’d feel about cutting corners when building a new college dormitory and a new hospital, in a known active earthquake zone, versus building components that will become part of a fission plant.

Excerpt follows from:
“… among the buildings that crumbled or have been designated uninhabitable by the quake are a university dormitory and a hospital, both of which were built after seismic standards had been raised.

Firefighters picking through the rubble of some buildings told state TV Friday night that some of the reinforced concrete pillars they had removed seemed to have been made poorly, possibly with sand. They said that rescuers using saws or other instruments usually split the pillars cleanly. But in some buildings in L’Aquila, the pillars crumbled into dust, indicating that a lot of sand might have been mixed into the cement, they said.
——-end excerpt——

There’s just something fundamentally cattywampus about human risk perception for construction/profitability of longterm infrastructure that needs study.


Leave a Reply (Markdown is enabled)

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

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

Facebook photo

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

Connecting to %s