March 20, 2005
American Electric Utilities Planning New Nuclear Reactors

The high cost of natural gas is making nuclear power more attractive. The New York Times reports that nuclear power plant operators Entergy, Exelon, and Dominion have applied for approval for sites where nuclear reactors might be constructed and Duke Power has informed the Nuclear Regulatory Commission it plans to apply for a reactor license. None of the new reactors are expected to be radical departures from previous designs.

On the drawing boards are all kinds of exotic designs, using graphite and helium, or plutonium and molten sodium, and making hydrogen rather than electricity. But the experts generally agree that if a reactor is ordered soon, its design changes will be evolutionary, not revolutionary.

The utilities are not ready for a giant technology leap; they want a plant that does what the existing ones do, but slightly better. So if new orders materialize in the next five years, it will be the mechanics and engineers who will get to show what they have learned. The physicists will have to wait.

The Westinghouse AP1000 is considered typical of the new reactors that incorporate many improvements.

Westinghouse is one of the companies trying to market a reactor, the AP1000, with more modest technical goals. It has an output of a little over 1,000 megawatts with what is called a passive approach to safety. It requires only half as many safety-related valves, 83 percent less safety-related pipe and one-third fewer pumps.

Unfortunately the article does not provide cost estimates. However, two GE reactors are under construction at Yenliao, Taiwan and an Areva European Pressurized Water Reactor (EPR) is under construction at Olkiluoto, Finland. I can't find any total reactor cost information on it in Areva's press releases (anyone who wants to look through those press releases go here) but Areva claims that the new reactor in Finland will provide cheaper power than previous nuclear reactors.

With a capacity of about 1600 MWe, the EPR has a number of major innovative features making it safer and more competitive. The electricity generated is 10% cheaper than that generated in the nuclear reactors currently in operation. It uses 15% less uranium to generate the same amount of electricity and so produces less spent fuel. Maintenance operations are simpler and therefore shorter, increasing availability to over 90%.

But what is the total cost of the Finnish project? Anyone know?

Even if we leave aside the hard-to-calculate clean-up costs it is hard to find good information on the real costs of nuclear electric power for new reactors today as compared to natural gas and coal. If anyone has some good sources for comparative costs please post them in the comments. One complicating factor on clean-up costs is that old reactors are being kept in operation for longer htan originally planned and hence their clean-up can be amortized over a longer than expected operating life. Double the life of a reactor and the effect is to greatly decrease clean-up as a fraction of total costs.

The Taiwan project has a total cost of $6.5 billion (presumably US dollars).

The Lungmen (Dragon Gate) nuclear project is Taipower's fourth and largest single investment in a broad building programme. Pressed by demand and the country's uncomfortable exposure to foreign fossil fuel supply disruption risk, Taipower is determined to proceed. This is despite unprecedented protests against its plans and the total price tag of an estimated $6.5 billion. Its opponents, mostly from the site area and opposition parties, are equally determined to stop what they see as an unnecessary investment with very high risks.


The project involves a 2,700MW plant consisting of two 1,350MW units at Yenliao, on the north eastern tip of Taiwan, near the capital Taipei. The advanced boiling water reactor technology marks a departure from the utility's three existing nuclear plants, which are smaller units completed in the late 1970s and early 1980s. US technology, however, will be a common thread from the first to the fourth plant. The units will come on-line in 2004 and 2005.

Does 2,700MW of reactor capacity for $6.5 billion strike anyone as pricey? I've read previously that Westinghouse's 1,100 MW AP1000 reactors might cost $1 billion each. But perhaps that figure is a price for a smaller subset for the reactor itself and not including the electric generators and surrounding complex? How to compare nuclear electric power costs to fossil fuel electric power costs?

Share |      Randall Parker, 2005 March 20 01:03 PM  Energy Nuclear

Wellington said at March 20, 2005 2:08 PM:

Something rarely factored into the cost of fossil fuel generation is the cost in pollution gases and other human costs. If you are asthmatic and your medical bills are thousands of dollars higher because of that coal plant upwind, the cost to you is not included in the calculations. If your son is killed fighting in the middle east trying to safeguard fuel reserves there, your personal cost is not included. And in China, if you're a coal miner certain to die in the mines in the next few months or years, your personal costs and those of your family are not included.

Nuclear power is risky mainly because of the dual uses the technology can be put to. In the Star Trek universe, they look back on the 20th and 21st Earth centuries with relief, at having escaped the great nuclear holocaust. In the real world we haven't gotten past that point yet.

undecipherable said at March 20, 2005 4:23 PM:

America could put OPEC out of business in 10 years if it's citizens could somehow overcome their wildly overblown fantasies about the risks of nuclear power.

I do not mean to suggest that the current technology poses _no_ risk, but in truth it is a much safer and _greener_ technology than burning fossil fuels. To the previous poster's point about coal, beyond asthma, consider acid rain and (purportedly) childhood autism now. Come to think of it, burning coal releases radioactive pollution too.

Anyway, the point I want to make here is that there are considerable beneficial externalities to nuclear power which are not reflected in its price; to wit, increased energy independence and less (expensive) trouble around the world. When your utility uses nuclear power, you're buying the electricity and getting a side of _freedom_ for free -- It's like a power happy meal.

Think of all the economic growth lost to the high cost of oil -- Anybody else here old enough to remember the Oil Shocks of the 70s? -- the economic and humanitarian costs of suffering all these terrible governments in the ME, the billions spent fighting wars over there, and the all the people (soldiers, civilians) killed. And no, I'm not one of these people with the fantasy that we went over there to "get their oil," but if not for oil, the ME certainly wouldn't be as much of a hotspot as it is today and guys like Saddam Hussein wouldn't be around either.* Hell, think of _all_ the freaks we could put out of business by reducing our dependence on foreign oil-- Its a lot, and they're not just in the ME either. And it's not just oil, but LNG and other forms of this crack too.

Anyway, we're talking trillions of dollars here over long periods of time. If the "true" cost of oil, or coal for that matter, was really reflected in its price, it would be very, very expensive indeed. Imagine if the government had imposed a tax on oil for the last 60 years to finance everything it ever did in the ME since the beginning of the Cold War. It'd be a prety big tax.

Nuclear is also more expensive in the US than it really needs to be because of the awful regulation. And I don't mean the stop-them-from-dumping-contaminated-waste-behind-a-school-type regulation, but the type that makes them build very few of these things and on a very massive scale. Very big projects are inherently more costly and more risky (e.g., Boston's "big dig"). Not to mention the fact that American nukes are required to utilize fuel (small cost) inefficiently and actually generate _more_ radioactive waste (very big cost to manage, dispose, store, secure) because of the government's ban on using plutonium (for those who didn't know, as originally designed a breeder reactor could keep reusing its own waste (which would include plutonium) as fuel, instead of burying it under a mountain in nevada).

People in France, a country with no significant reserves of oil or coal, don't freak out about nuclear the way Americans do, and they're big users of the technology. They understand the nuclear keeps their country clean and increases their independence from the ME. (The Germans, BTW, might be even worse than the Americans w/respect to nuclear, they're in hysterics over nothing there and it will cost them dearly -- is already)

China gets the big picture too. They are industrializing so quickly, they can't get coal into the country fast enough to keep their generators firing 24 hours a day. And where they have got it, its a polluted mess because their technology is dated. China is going to be going nuclear in a big way. They are already doing it and are also looking into some new technologies which may allow them to build more, simpler, smaller-scale facilities.

The writing is already on the wall, has been for a long time. Take a look at this chart:

The little green line on the bottom is the S&P 500 over the last 12 months. The other two lines are two companies that are big in the uranium business. (USEC, btw, gets uranium from decomissioned Russian nuclear weapons and turns it into usable commerical reactor fuel).

The smart money already knows that the world is going to _have_ to go nuclear in a big way. Now, oil may get cheaper in the short term -- Less trouble in the ME maybe, additional refining capacity brought online, breakthrough in extracting from shale or something -- however, as less-developed nations (finally) begin to mobilize their economies, those endless needs are only going to impinge more urgently on the world's relatively inelastic (Yes, perhaps limited) supply.

On the environmental side, the now-developed world may have polluted its way through industrialization exclusiveley on fossil fuel (we didn't know better), but that's just not going to work if _everybody_ on the planet is doing it, especially since we're talking _billions_ of people now, not just a few million in certain parts of England, the US, Japan and Germany (1st, 2nd, 3rd industrial revolutions). Some of these later-day industrializers are already too crowded for that style of industrialization anyway.

* Yes, I know it was us that created him, but this just goes to prove the same point -- Iran might not have become a theocracy either if not for out intriguing over there against the Soviets, and they wouldn't have been there but for oil.

Kurt said at March 20, 2005 4:54 PM:

Its about time that nuclear power comes back into vogue again. The more nuclear power plants, the less dependence of at least the electrical supply on middle-eastern politics.

The problem with the Taiwan nuclear power plant and why it is so expensive has nothing to do with nuclear power, per se. It is the corrupt manner that large projects in Taiwan are financed that involve government funding. The shinkansen train as well as Taipei's subway system involved the same kind of kickbacks to politicians and sweet deals to the contractors. The opposition to the new nuclear power plant in Taiwan is based on opposition to this kind of corruption. I lived in Taiwan in 2000-2001 when this nuclear power plant initiative was defeated the first time.

the major problem with nuclear power in the U.S. is that it was developed and promoted as a government program similar to that of defense contracting, along with the excessively expensive and low-efficiency designs. Since the nuclear power industry was "de-regulated" in 1991 (when management of the plants turned over to the utilities), plant operating efficiency improved from around 65% up to 90%.

The next generation plants being built by GE will be more efficient and lower cost to operate, even if they are not the IFR design that would be most ideal.

Unless there is a breakthrough in fusion power (either hot or cold) or ZPE, I expect much of the world's electricity to come from nuclear power by 2050-2100. There is simply no other method of generating the gigawatts and terawatts demanded by modern civilization.

Engineer-Poet said at March 20, 2005 6:28 PM:

When even James Lovelock is touting nuclear power as the way to save both technological society and the world, you have to figure that something is up.

$2400/kW sounds mighty expensive to me too, but if all Taiwanese projects have the same sort of markup it would be a wash.  It would take more information about the local situation to be certain.

Unfortunately, it will take a serious evolution of vehicle technology to be able to replace oil with nuclear power.  Without plug-in hybrids or hydrogen fuel cells, the biggest role nukes could play is helping to hydrogenate coal.

Tom DC/VA said at March 20, 2005 9:28 PM:

My main point in the last big energy thread was that given how much better GenIV plants look w.r.t. inherent safety and fuel utilization, any investment in GenII+ plants is pointless. I think I'll stand by that.

I see some nuclear power = energy indpendence rhetoric popping up. For reasons E-P points out, this is far being true. It also means that not enough people look at the data on existing electrical generation or at data on consumption by source. It is not our light bulbs that are dependent on Middle-Eastern oil sheiks, it is our SUVs - at least until E-P's dreams of plug-in hybrids come true.

Randall Parker said at March 20, 2005 11:01 PM:

Tom DC/VA,

There is a fair amount of substituteability in energy.

If we were to build enough nukes to replace all the natural gas electric plants then all that natural gas could be used in transportation. Some could be used as compressed gas. There are fleet vehicles running off of natural gas.

Also, natural gas and propane could substitute for oil in areas where oil is still used in heating.

Also, yes, we could migrate to pluggable hybrid vehicles.

Do you know what the fastest growing US produced source of liquid fuel is? Natural gas condensates. See, for example, table 2 which shows US crude oil and total oil production. Some of those condensates are larger hydrocarbon molecules removed from natural gas fields as liquids. I can not figure out whether natural gas is being converted into liquids. But surely this can be done. With what energy loss, though?

We need to use less natural gas because domestic natural gas production is declining more rapidly than expected.

Driving the concern are data suggesting that natural-gas production in the US is tumbling faster than anticipated. As a result, environmentalists, industry representatives, and others are marshalling their arguments about what needs to be done to stave off a crisis. Ideas range from pushing for greater conservation and efficiency to opening new areas for drilling to importing more liquid natural gas (LNG).

Mr. Simmons and others say the US confronts a problem not only in the immediate future, but is likely to see natural-gas shortages for years to come. In a September report, the National Petroleum Council warned that the US is on a course to pay an additional $1 trillion in natural-gas costs over the next 20 years as a result of shortages.

What is the best way to free up natural gas for other uses? Stop using it to generate electricity. How to do that? The two choices are coal and nuclear. I'd be a bigger advocate of coal if the coal electric power generators were not so resistant to lowering their mercury and particulate emissions. As it stands now I'd rather live near a nuke than near a coal burning plant.

AK600 said at March 21, 2005 1:28 AM:

A Framatome press release says:
"The overall Olkiluoto 3 project cost has been estimated by TVO at around € 3 Billion."

J. D. Enox said at March 21, 2005 2:13 AM:

Yes the writing is on the wall now. I still become a bit discouraged when reading articles expousing pure solar or wind. Those solutions just aren't practical due to placement, variability of conditions, land use and just the awfull burden of logistics support to keep projects as gargantuan as those would be operating. Of course when it makes sense use them, I'm all for that. But only nuclear has the feesability for large baseload power.

Fusion is still problamatical at best. After 40 years still no reliable breakeven on power output and no one has yet to address the matter of the heavily irridiated equipment from such plants. Also few realise how positively massive a power plant using the current fusion technologies would be. Fusion isn't necessarily a deadend, perhaps there'll be a breakthrough in the future or else exploration of say colliding beam fusion might work out. But as of now dirty old fission is the only practical course. Also I'm in agreement with the previouse poster. I'd much rather live next door to a nuke breeder plant than even a clean coal plant.

Philip Sargent said at March 21, 2005 3:13 AM:

Another plug for Bill Nuttall's book on this...
The AP1000 from Westinghouse is not just designed for simplicity and passive safety, but is also modular so that much more of it can be constructed off-site, thus speeding construction time. "Westinghouse confidently reports that complete construction can be implemented on budget within a 3-year envelope".

Reliable and fast construction, and the associated increase in the capital cost element and regualtion risk, is one of the biggest commercial problems for new reactors. AP1000 costs are predicted by Westinghouse to be $36/MWhr over a 60 year life at 93% availability (in $/MWhr: fuel 5, ops & maint. 5, disposal 1; in $/MWe capacity: 1.15).

BNFL bought Westinghouse in 1999. Unfortunately it is 20 years since Westinghouse built any kind of nuclear reactor, anywhere.

Alternate Energy Blog said at March 21, 2005 7:08 AM:

Like Randall I agree that it is hard to find good information on the real costs of nuclear electric power.

Doug Koplow sent me this link to a powerpoint presentation on nuclear subsidies:

from his website (

He is clearly against nuclear power and raises a number of objections which need to be answered.

I think both sides of the debate on nuclear power need to come up with better data.

Meanwhile I remain sitting firmly on the fence on this issue.

Alternative Energy Blog

Engineer-Poet said at March 21, 2005 7:41 AM:

CNG vehicles still waste the byproduct heat of combustion, and it is my understanding that they also increase methane (greenhouse gas) emissions.  They are never going to be as efficient as the combination of plug-in hybrids and cogeneration, which I looked at in cogeneration@home.  Our policy should lean against them.  (Unless I'm very wrong, cogeneration can also cut total natural gas use while displacing nearly all petroleum consumption.  Things get even better if you assume some supplementation by wind, solar or nuclear.)

J. D. Enox writes:

I still become a bit discouraged when reading articles expousing pure solar or wind. Those solutions just aren't practical due to placement, variability of conditions, land use and just the awfull burden of logistics support to keep projects as gargantuan as those would be operating.
Unfortunately, you've got a few things wrong.
  1. The land-use argument is nonsense; we're already shingling more roofs than we need for solar cells, and wind-turbine pads between fields takes little away from farmland.
  2. Logistics support may be an issue for wind, but it's the opposite for building-integrated solar.  Not only are the logistics of solid-state systems a lead-pipe cinch, but you avoid the need for investment, construction and maintenance of transmission and distribution systems roughly equal to the cost of the displaced conventional generation.
Pure solar or wind isn't a solution, true.  However, using them along with cogeneration can slash fuel consumption, carbon emissions and total costs while increasing reliability; if your heat is coming from your cogenerator and a winter gale comes up, you could switch to heating with your dump load (pure electric) and let the wind heat your house as well as lighting the lights and charging the car.  When the wind dies down, you switch back.

NB:  Natural gas is about $.60/therm where I am; this is about 2 cents/kWh.  If gas goes to $1.00/therm, off-peak wind power at 3 cents/kWh will be cheaper than natural gas.  Being able to make productive use of large amounts of wind power makes it feasible, even desirable, to install much more of it than with present consumption models.

Randall Parker said at March 21, 2005 9:45 AM:


If Philip Sargent's source on the AP1000's cost is correct then his figure of $36/MWhr translates into (and correct me if I'm wrong) .36 cents per KWh. Correct? Then isn't that a sixth of what your area pays for natural gas electric?

On pure solar: Drive down the cost far enough and do the same for energy storage costs and it strikes as having the potential to become a complete solution. Why won't that happen eventually? Seems to me that in 20 or 30 years time advances in nanotech make very cheap solar a complete energy source.

Though there is one scenario under which I would not want to be dependent on solar: A massive volcanic eruption.

Update: Michael Vassar points out I made a division error. For some reason I was thinking that a dollar had 1000 cents rather than 100. So the correct figure is 3.6 cents per KWh.

So the AP1000 nuclear reactor cost figure I've read still leaves it an expensive source of electricity. Whatever happened to nuclear power that was going to be "too cheap to meter"?

Invisible Scientist said at March 21, 2005 12:32 PM:

According to the article below, there will be a shortage of uranium in the world
in a few years, meaning that AT ANY PRICE, the uranium will not be available if the
world builds many nuclear plants.

I don't have the exact figures, but you can do a google search for the latest
Westinghouse reactors, which are said to generate electricity at a price slightly
more expensive than coal, which is not too bad, but this is with the current price
of uranium. I would guess that even at twice the price of uranium, the price of electricity
would be acceptable, because it is the cost of operating the nuclear plant that is high.

Thus the problem in the future, if we want to build hundreds, or even thousands of nuclear plants,
is to face the reality that we need fast breeder reactors, which
would generate enough fuel for hundreds of years. And this means that a political decision
will have to be made.

>> Kazakhstan aims to be world's top uranium producer
>> Monday, March 21, 2005
>> ASTANA —Kazakhstan has launched a program designed to make it the world leader in uranium production.
>> The state-owned National Atomic Co., or Kazatomprom, has started a program to reduce the expected shortage of uranium in the world market, Central Asia Newsline reported.
>> Officials said Kazatomprom plans to develop seven new uranium mines by 2010.
>> By 2007, Kazakhstan hopes to produce more than 7,500 tons of uranium. The country is said to have 1.5 million tons, or nearly 20 percent of the world supply of uranium.
>> Officials said the European Union and the United States would hold a stake in the project. They said Kazakhstan's leading clients for uranium include China, Japan, Russia and South Korea.
>> The International Atomic Energy Agency has forecast a shortage in the uranium market by 2010. The IAEA said the market supply would decrease and reach a deficit of 16,000 tons by 2015.
>> Kazatomprom president Mukhtar Dzhakishev said a uranium shortage could harm Kazakhstan. Dzhakishev said such a shortage could prompt an international effort to develop alternative sources to nuclear energy.
>> The company has assessed that the uranium mining project would recover its expenses by 2013. By then, uranium profits would total $830 millio

J. D. Enox said at March 21, 2005 12:47 PM:

In reference to Engineer-Poet I was speaking of the vision proposed by some of vast fields of solar collectors and wind turbines as proposed by some groups. Nothing wrong with using them when it makes sense. However even solar is prone to degradation. The photovoltaic elements begin wearing out and losing effeciency, relays and circuitry breakdown, the outer transparent covering becomes dirty and loses transparency due to windborne abrasives so no they're not a lead pipe cinch.

Solar will only be practical as a supplement when and if the promised "cheap" effecient systems are produced. Wind works in certain locations and situations (though apparently not off the coast of Massachusettes). To support a tech heavy population base though you need a reliable and massive baseload system which none of the renewables achieves.

Engineer-Poet said at March 21, 2005 8:22 PM:

Invisible Scientist:  If we wish to expand the use of nuclear energy ten-fold, IMHO we have more than enough time to decide how to get the fuel to do it.  Reprocessing the existing inventory of spent fuel into MOX would buy us at least 10 years at today's consumption rates.

J. D. Enox:  Who cares if solar systems eventually deteriorate?  So does everything else.  Current panels bear 25-year warranties, which is longer than most roofing materials.  If the manufacturers consolidated their feces, they would make panels with raised-seam junctions and overlapping top and bottom edges so that they could be used as roofing.  Adding structural connections so that the panels could also replace decking material would pay off twice, first by saving the expense of plywood beneath, second by allowing easy access for electrical connections from beneath.

IIRC, Nanosolar thinks they can eventually make panels for 80 cents a peak watt.  The folks who made the 1-micron-thick silicon cells were being touted as being capable of 1 Euro ($1.34) per peak watt.  It doesn't matter how efficient those things are, if they have good light they will make cheap power.  They will also overturn some existing truths; they will invert the current day/night price structure, among other things.  Instead of power being costliest on hot, sunny afternoons, it will probably be more expensive right after sunset on nights requiring air conditioning.  Winter cogeneration would make power very cheap during cold weather.

To support a tech heavy population base though you need a reliable and massive baseload system
I don't think that's true.  You could do it just as well with intermittent main sources and a heap of backups; as long as the backups have sufficiently low capital costs, you can afford to leave them idle most of the time.

SkyWind power says their scheme could achieve an 80% capacity factor across the northern Midwest (Chicago to Detroit).  Cheap solar would handle most summer days, CHP would handle calm winter nights.  You could deal with what's left with the same cheap turbines which currently handle peaking loads.

I think we're going to have baseload powerplants (nuclear), and as long as we have them we are going to size them to deal with the constant parts of demand.  But are they essential to an industrial society?  That's "inside the box" thinking, and we may find it desirable to change the rules someday.

Tom DC/VA said at March 21, 2005 9:02 PM:

Randall -

I believe a majority of NG usage for electricity production in the US is still for peaking power (I can't find my bookmark for the DOE power plant database ATM, so I could be wrong on this), which means even if the US were to build enough GenII+ nuclear plants - which are not load-following, AFAIK - in the next 5-15 years to cover use of NG for baseload generation, NG would still have a significant role in the grid.

Even if the US did build enough nuclear plants to cover all use of NG for electricity production and used something else as the load-following energy source, and then switched that NG to the transportation sector, the US still would be dependent on a energy source that has declining domestic production. So I don't see how the US ends up that far ahead. Instead of oil being the main globally and marginally priced fuel, NG takes its place. Don't forget that the infrastructure costs and energy losses for inter-continental transport of NG are much higher for that of oil.

Also, NG is not a direct substitute for heating oil in a lot of places, again due to the infrastructure costs (it is not cost-effective to run a gas pipeline to every town in Vermont, for instance, and the one that runs to Burlington comes from Canada).

In the end, energy costs for electricity production are not at a critical stage right now, and thus the immediate commissioning of new nuclear plants is not necessary (and probably is just pork for GE). The US would be much better off spending a few billion a year on both researching GenIV+ designs and improving solar technologies, with a few hundred million thrown at things like E-P's home cogeneration unit and better fuel cells. Wind is almost competitive right now, and a clearly defined (but declining to zero over the next 20 years) producer tax credit would probably push it to be fully competitive in zone 4 and above areas.

What may be approaching the critical stage, of course, is energy costs for transportation, which means oil. The US remains deeply unserious about addressing this problem, which is unlikely to change anytime soon.

Randall Parker said at March 21, 2005 9:37 PM:

Tom DC/VA,

I'm not worried about expensive liquid fuels for transportation. If we have cheap electric power and expensive gasoline that will increase the incentives for development of means to use electricity to power cars. Though I'd love to see more research on batteries.

It is my understanding that a lot of gas fired electric plants were built when natural gas prices were lower because the regulatory hurdles (owing to lower levels of pollution) for building gas fired plants were much lower than for coal fired plants.

Technologies could be developed to make coal burning much cleaner. The coal burners could eventually be required to remove 99.9+% of mercury from coal exhaust for example. So coal could be used for peak load power. But right now more coal plants mean more particulates, more sulfur oxides, more nitrogen oxides, more mercury, and so on.

You say that immediate commissioning of new nuclear plants is not necessary. But if the new electric power plants are not nuclear then large numbers of coal fired electric plants will be built to handle expected load growth. Because the alternatives cost too much coal is staging a comeback. At this point our choice is between more pollutiing coal plants or more nuclear plants. Given that those are our choices today I favor nukes.

As for more research on nuclear and solar: I have argued for more energy research and repeatedly made that argument. So you are preaching to the choir on that one. Still, I encourage anyone to make that argument repeatedly. I want cleaner and cheaper energy and more efficient technologies for using it. I favor much greater quantities of research funding into fast breeders, thorium for proliferation-resistant fuel cycles, pebble bed modular reactors, lithium polymer batteries, coal gassification, other cleaner ways to burn coal, photovoltaics, and anything else that looks promising.

But right now what should we do to meet the growing demand for electricity?

Engineer-Poet said at March 21, 2005 10:32 PM:

Tom said "NG is not a direct substitute for heating oil in a lot of places..."

Ah, but if you have plug-in hybrid vehicles, oil in a diesel cogenerator (which can easily replace an oil furnace) is a direct replacement for the first X miles worth of motor fuel on each trip.

And Randall asks:  "But right now what should we do to meet the growing demand for electricity?"

We don't necessarily need more electricity, we need more of what it can do.  For a little while we can either squeeze more out of it, or substitute.

For the next 5 years:  CF lights and energy-saving computers (today's 5-watt laptop processor is more powerful than the top of the line from 5 years ago).  Better architecture to avoid the need for so much energy in new construction.  Convert some old coal plants to IGCC, which yields about 190% more power and 20% better efficiency.

5-10 years:  New-tech cheap photovoltaics.  Building-integrated solar-thermal generation or even cogeneration (think Energy Innovations).  New tech nuclear.  Ubiquitous cogeneration.  Omnivorous systems which can either produce or consume electricity or other fuels as needed, like plug-in hybrid cars (which can also be storage devices), cogenerating heating plants.  More wind power, with flexible generation providing backup.  IGCC syngas becomes feedstock for liquid fuels to displace petroleum products on the supply side; meanwhile, electricity replaces liquid motor fuel on the demand side.

10+ years:  Hydrogen via high-temperature electrolysis and/or artificial photosynthesis.  Some NG-fired electric plants converted to IGCC, more of them are phased out.  IGCC syngas piped to consumer sites to fire cogenerating engines, replacing NG where it is still required.  Most new construction requires no added heat or light during daylight hours and is self-sufficient in air conditioning.  Most new vehicles require no liquid fuel for local trips.  Some new vehicles sport conformal photovoltaic skins and are self-powered for part of their driving, and can provide power indefinitely for off-grid uses (camping, construction).

michael vassar said at March 22, 2005 10:54 AM:

Randall, I wouldn't take claims of $36/MWH seriously. That's over an order of magnitude cheaper than current nuclear power.

michael vassar said at March 22, 2005 10:57 AM:

Randall, I wouldn't take claims of $36/MWH seriously. That's over an order of magnitude cheaper than current nuclear power.
Here's an overview of the probable impact of traditional (non-Drexlerian) nanotechnology on energy.

Jim said at March 22, 2005 4:19 PM:

the biggest nanotechnology impact on energy is the new materials, such as more efficient solar cells and higher temperature steel (100 deg C increase in operating capacity of new steel would have HUGE impacts on the energy efficiency of any plant - nuclear, coal, gas-fired, whatever. of course much of the research dubbed nanotechnology is just fantasy or a pre-existing technology that includes the prefix 'nano' to attract research funding

sorry but batteries don't have a lot further to improve. they're already pretty high energy density compared with things like C4 plastic explosive, and as the energy density increases there is a corresponding increase in the danger of violent explosion. perhaps metal-ambient oxygen batteries have potential for more energy, but at most a factor of 2 over lithium polymer batteries. chains of ~8 carbon atoms saturated with hydrogen will remain the most efficient storage medium for portable, liquid energy. whether it's refined from crude oil or exxon-mobile builds a nuclear plant next to it's refinery of the future doesn't make a lot of difference.

nuclear energy is so obvious that it's painful. how much generating capacity could we have purchased with the same amount of money used to invade iraq?

michael vassar said at March 22, 2005 4:54 PM:

What are you thinking. Current batteries still have pathetic energy density compared to theoretical limits. Aluminum Oxygen batteries can, in principle, store energy with much higher density than fossile fuels. Just consult any free energy of formation look-up table for Al2O3 and do the math.

Randall Parker said at March 22, 2005 5:48 PM:


We could bave bought about 300 1 Gigawatt nuclear power plants for the price (about $300 billion) of the Iraq invasion to date. Or we could have funded 30 years of Nobelist Richard Smalley's proposed $10 billion per year energy research program. I don't think we'd need 30 years of the program.

We also could have spent $300 billion on insulating buildings and doing assorted other things to lower energy usage.

As for your comments on batteries: But we do not yet have lithium polymer batteries. Donald Sadoway at MIT says that lithium polymer batteries would have enough power density to operate cars. If I'm not mistaken lead acid batteries are 35 Wh/kg. So then lithium polymer would be an order of magnitude greater. To quote from a previous post of mine: on Donald Sadoway and lithium polymer batteries.

Is hydrogen the only viable candidate as an energy storage form to replace gasoline and diesel fuel in vehicles? In a word, no. Lead acid batteries have an energy storage density of 35 Watt Hours per kilogram. This leads to electric cars that weigh too much and have too short a range between recharges. MIT professor Donald R. Sadoway believes lithium polymer batteries can be developed that will have over an order of magnitude greater energy density than lead acid batteries.

Niels Bohr, the Danish physicist and Nobel Laureate, once cautioned that prediction is always dangerous, especially when it is about the future. With this disclaimer, then, we speculate on what is in store for rechargeable lithium batteries. In the near term, expect the push for all-solid-state, flexible, thin-film batteries to continue. This is driven by the desire to maximize the electrode–electrolyte inter-facial area while minimizing diffusion distances within the electrodes themselves, in order to combine high capacity with high rate capability. Recent results from our laboratory indicate that in a multi-layer configuration comprising an anode of metallic lithium, a solid polymer electrolyte, and a cathode of dense, thin-film vanadium oxide, it is possible to construct a battery with projected values of specific energy exceeding 400 Wh/kg (700 Wh/l) and specific power exceeding 600 W/kg (1000 W/l).10,11 Another trend is distributed power sources as opposed to a single central power supply. This allows for miniaturization (e.g., the microbattery). Expect also the integration of energy generation with energy storage, for example, a multilayer laminate comprising a photo-voltaic charger and a rechargeable battery. Ultimately, if scientific discoveries prove to be scalable and cost-effective, we should witness the large-scale adoption of electric vehicles.

When the cost of photovoltaics is lowered far enough to compete with fossil fuels then a combination of photovoltaics and lithium polymer batteries may well be the combination of technologies that will lead to
the phase-out of the use of fossil fuels as vehicle power sources.

The article co-authored by Donald Sadoway and Anne Mayes is from the August 2002 issue of MRS Bulletin dedicated to lithium batteries.

Engineer-Poet said at March 22, 2005 6:25 PM:

Current Li-ion batteries already have sufficient energy density, and the Altair Nanomaterials elecrode material (which dramatically improves power density, charging time and cycle lifetime) will do for everything else.  A commuter car with such a battery pack would require less than 150 pounds of such batteries.

In other words, what we've got is good enough; all it has to do is get cheaper.

Jim said at March 24, 2005 4:15 PM:

why didn't politicians tell the truth about the cost of war and present it to congress against 300 GW reactors (or other reasonable ideas Randall presented)? kinda sad.

regarding the batteries, yes the theoretical energy storage associated with the metal - oxygen reaction is high, but how do you get it to be an effective battery? one key leap in the process is using ambient oxygen, so that only the metal must be carried around. theoretical energies of reactions with pure metal will give very big numbers (michael - if you're going to quote a pure metal, why didn't you pick Mg? - what were YOU thinking?) tricks are:
1)all the necessary surface area has a lot of interfacial energy associated with it. adding to energy and cost penalties
2) what is the proper membrane to make the oxygen give up an electron to pass through - you are limited by diffusion, and therefore must either make it much thinner or greatly increase the temperature, both a bit tricky.

batteries are definitely important for energy fungibility, and improvements will come (maybe bigger than 2x in my previous comment, more like 10x in MRS bulletin, but at what cost?), but not exponential growth. new batteries get enormous DARPA funding and the payoff ain't that big (although i would love a bright researcher to prove me wrong - Niels Bohr was right) taking li ion batteries, the amount of cobalt used renders it impossible to use on a widely adopted basis when compared to the world supply of cobalt - just an example of difficulties encountered scaling up.

the point i was trying to make is that gasoline is very convenient and practical for portable, easily used energy, and it shouldn't be discounted. energy/mass and energy/volume are both outstanding. put some batteries in hybrid cars because you can't really turn electricity back into gasoline, but gas will probably be the primary fuel source for many years to come. and when we get to the point of plentiful energy(my fantasyland of solar shingles and nuclear reactors a plenty), why not just have refineries turning co2 or organic matter + much electricity into gasoline? would it be better if the lingo turned to nano-carbon chains as a hydrogen storage media of optimum length to be liquid at ambient conditions when saturated with H?

Randall Parker said at March 24, 2005 5:45 PM:


Lithium ion batteries are manufactured and purchased in large quantities for use in laptops and other portable electronic devices. The market forces already exist for driving down their costs. But years have gone by and they are still way too expensive for cars. What gives? My guess is that making Li-ion batteries cheaper is a difficult problem to solve.


Yes, I'm a fan of the idea of developing ways to run an artificial carbon cycle where photons or electrons are used to power fixing of carbon to hydrogen. As you point out, liquid hydrocarbons are all kinds of convenient.

Could light or electrons be used to power conversion of natural gas hydrocarbons into longer hydrocarbon molecules? With what efficiency?

Or could light or electrons be used to power the fixing of hydrogen to carbon dioxide coming out of coal plants? Then the carbon could serve double duty. The coal could be burned where there is central scrubbing equipment to remove the SOX, mercury, particulates, and other junk. But the carbon dioxide could be converted into very pure long chain hydrocarbons that would be liquid at room temperature.

michael vassar said at March 25, 2005 6:19 AM:

MgO 610 KJ/mol
Al2O3 1675 KJ/Mol

I agree gasoline is useful, and synthetic hydrocarbons (maybe algae derived bio-diesel?) are my best bet for "fuel of the future", but its nice to know what the options are. A 10X increase in energy storage capacity for batteries at current cost would easily make all electric feasible.

michael vassar said at March 25, 2005 6:31 AM:

By the way, if nuclear can be so cheap why doesn't France export electricity and under-sell the rest of Europe? They don't mind using nuclear.

Philip Sargent said at March 25, 2005 7:49 AM:

Well in fact we do: the UK buys 17 TWh a year of French nuclear electricity over the 2 GW 275 kV HVDC 45km channel interconnector which commenced full operation in 1985. The price paid is not public as it is a commercial deal. It was designed for two-way operation and was expected to swap direction twice a day because the 1 hour time zone difference woudl produce peak demands off set by one hour (or so). However, today it is importing from France nearly all the time.

The French continue to build new reactors, the first Framatome-Siemens EPR in France (same design as the one under costrution in Finland) will be built at Flamanville in Normandy with power available in 2012. This design will then be used to replace all the existing 58 reactors in France.

Having said that, electricity prices in France are not particularly cheap - but that is much more likely to be because they have not achieved price liberalisation and de-nationalisation of their electricity industry - rather than anything to do with choice of technology. With 58 reactors, the safety-case and planning permission overheads are spread over a much larger number; and the learning effect will have improved construction times significantly.

However, it is hard to get properly comparative prices as the Fench give tax holidays to reactors, and the nationalised industry was able to borrow money from the French government at rates half that avilable to the private sector.

"THE 2000 MW cross-channel link high voltage direct current (HVDC) connection between the British and French networks has saved both countries from power supply emergencies over the years. It probably sidestepped the need to build two more large nuclear units on both sides of the channel."

Jim said at March 25, 2005 9:53 AM:

michael - taking the /mol values above, normalize to mol of metal, rather than the mol-compound and you've got 610:838 kJ/mol-atom. Al is 55% denser than Mg, making energy of reaction with Mg:Al 12% greater by mass. furthermore, that reaction requires 50% more oxygen to completely burn the same mol Al vs. mol Mg, which will have plenty of kinetic penalties when trying to design the type of science-fiction battery we're imagining. it's my guess that in the next 30 years, "A 10X increase in energy storage capacity for batteries at current cost" isn't happening.

i agree bio-diesel is a pretty neat idea. from a recent student design class project i saw, looking at the feasibility of such an approach, it seems there is a general lack of ideal enzymes to make these types of reactions completely feasible, but that's just the kind of biotech breakthrough you could start a company on.

Engineer-Poet said at March 25, 2005 3:58 PM:

Jim:  Do we li-ion cells really need cobalt?  I knew that some recent hubbub about lithium iron phosphate wasn't just my imagination.  Even if not, I see nothing to prevent zinc-air primary or secondary cells from doing the job.

I have low expectations for aluminum or magnesium as battery-fuel; you cannot regenerate aluminum with wet chemistry, and the flammability of magnesium is a safety issue.

Randall asks:

Or could light or electrons be used to power the fixing of hydrogen to carbon dioxide coming out of coal plants?
Hydrogen itself has enough energy to "power" the conversion of CO2 to methane and water, among other products.  But the comment about using the output of coal plants raises two questions:
  1. What's the point of burning coal if you have a non-fossil source of hydrogen?
  2. If you do burn coal, turning the products into hydrocarbon fuels does not sequester them.
If we're going to do something about global warming and we're still going to use hydrocarbon fuels, we're either going to need systems which are far better at grabbing CO2 out of the atmosphere than plants are, or we will have to design our systems to capture the CO2 for reprocessing rather than venting it.

Metatron said at March 25, 2005 10:10 PM:

Proof the nuclear power system is a waste of money.

The Question was Asked:

Does 2,700MW of reactor capacity for $6.5 billion strike anyone as pricey?

Well, do the math:

divide 6.5 billion dollars by 2,700 MWH,

and that comes out to 2407407.4 Megawatt hours.

24 hours in a day makes it

100308.6 megawatt days.

365.25 days per year equals

274.63 megawatt YEARS is the amortisation of the cost.

So, if you charge people $274.63 a megawatt, andthe plant construction is paid off in a year. That's a pretty steep price, and in 1999, a megawatt was only $22, so make that more like 12.4 years. And that's just the construction. Then there's the eventualy decommissioning and dismantlement and burial, which will cost billions more, and the expense of keepng such a complex operation operating safely, is billions more.

If something comes along that can make electricity for less, people will buy it. If it is convenient and clean, people will flock to it.

that all said, pitiful hu-mans will need something like 12 terawatts of power each year, and it will only go up until the population goes down. How you're going to make that much electricity in the meantime is going to be interesting.

He3 from the moon?


Fall before the countenance, and beg forgiveness. It will be granted. But you will still cease to exist as you are consumed by the Almighty that Is!

Tom DC/VA said at March 27, 2005 10:11 AM:

"But right now what should we do to meet the growing demand for electricity?"

I strongly disagree with viewing growth in electricity demand as an unalterable exogenous factor. E-P has it exactly right: "We don't necessarily need more electricity, we need more of what it can do." Between reducing demand growth (tighter building codes, higher efficiency standards for new equipment), small-scale retrofits (CF lights, timed and motion-sensitive switches, replacing all CRT monitors with LCDs), large-scale retrofits (replacing all refrigerators and A/C compressors more than 20 years old), and the total geek-out of a smart grid, there are a lot of ways to alter demand.

Still, some new power plants will be needed to meet increased demand over the next 20 years, as well as to replace existing plants (Vermont Yankee, for instance). But saying that because there is industry opposition to cleaner coal plants (IGCC coal plants certainly sound good) we should therefore build new nuclear plants is rather illogical. Vested interests are very real obstacles, but they should be broken down instead of being allowed to drive decision-making. And since new nuclear plants would be authorized and regulated by the same administration that is surpressing the science related to the new mercury regulations, what makes you think that the nuclear plants would be implemented properly?

"I'm not worried about expensive liquid fuels for transportation."

I am for higher prices for transportation fuels in order to drive down consumption and drive forward innovation, but I would like the price increases to be predictible rather than left to foreign governments, terrorists, market sentiment, or some other kind of SHTF scenario. OTOH, I don't want to see higher input costs for industries that use fossil fuels as a feedstock, as subsitution is a lot harder for those uses. And I certainly am not pleased about who is currently capturing the excess profits from today's higher prices (Saudi Arabia is still the low-cost producer at ~$10/bbl, at least until Ghawar gives out). The only way I see to resolve all three issues is a (gasp!) gas tax. And, of course, this is a complete non-starter for cultural reasons, which is why I say the US is deeply unserious about addressing the oil issue.

Somebody upthread dared to use numbers, so here's some WAGs at what $300B could buy ($300B is the current expected cost, right? The counters I've seen put to-date costs at about $160B) (numbers below are guestimates, but hopefully I have the math right):

- Putting 4.5kW nominal of photovoltaics on 7.5M single family homes in the Southwest (30 150W panels at $1000/panel including mounting hardware, $5000 for an inverter and a reversible meter, $5000 for installation). Peak electrical power: 23GW. If A/C units draw 6kW with a 50% duty cycle, then this covers peak daytime cooling for those homes.
- Installing 30M of E-Ps cogen@home devices ($6000 for a 10kW unit, $1000 for a reversible meter, $3000 for installation and other bits). That would probably cover all single family homes in the Northeast and upper Midwest. Peak electrical power: 150GW.
- Building 100 1.2GW nuclear power plants ($2.5B for the plant, $.5B in escrow for cleanup and insurance). Peak (and continuous) electrical power: 120GW.
- Installing 100K 1.5MW wind turbines in the Plains ($2.5M for the installed unit, $500K for grid restructuring). Peak electrical power: 150GW.
- Purchasing 60B CF lights (average price of $5). If the 100M or so US homes each have 60 sockets and the lights last an average of 10 years, $300B should fund lighting purchases (with interest on the unused initial portion) well past the Rapture. Maximum electrical draw: 150GW. Maximum draw of replaced incandescants: 600GW.
- Purchasing 12M Toyota Priuses ($25K each, 57kW gas engine). Peak engine power: 684GW. (!)

Randall Parker said at March 27, 2005 11:09 AM:

Tom DC/VA,

The vast majority of the population are opposed to higher electricity and higher gasoline prices. So any proposed solution that includeds higher energy prices is a non-starter.

I certainly favor requiring all new coal plants to adhere to extrenely strict emissions regulations. I also favor a more rapid decrease in mercury emissions than the Bush Administration has agreed to with the coal electric generator industry.

Newt Gingrich is advocating for a Republican shift toward support for stricter emissions regulations and tougher environmental policies. I hope he manages to move the Republican leadership in that direction. Newt understands something that many of them do not get: upper middle and upper class people who are the natural constituencies of the Republican Party have living standards high enough that they'd like to see more go to cleaner environments. Their marginal dollar preferences for how money should be spent are more pro-environment than is the case with the lower classes.

I think your cost projection on a nuclear electric power plant is high. But I can't find good numbers that are based on real plants that were built in a competitive environment. The high prices for the Taiwanese plants may be due to the regulatory environment there. I'd like to know what the French are spending per nuclear plant and what the new Finnish plant is expected to cost.

Engineer-Poet said at March 27, 2005 1:13 PM:
The vast majority of the population are opposed to ... higher gasoline prices.
Probably because they don't see how higher prices would benefit them, and nobody has ever asked them to sacrifice anything for the cause of military and economic security.
So any proposed solution that includeds higher energy prices is a non-starter.
Any sacrifice that doesn't come with a justification is a non-starter.

I met with a builder the other day.  He was driving a Jeep Wrangler, but he told me that his next vehicle was going to be a hybrid Escape.  (Would he be looking at a plug-in hybrid if Ford offered the option?)  People are coming around, bit by bit; giving them the facts in a simple, digestible format could make all the difference.

Randall Parker said at March 27, 2005 1:37 PM:


People have been hearing for decades about energy and the environment. You can't get very many people to make personal sacrifices by explaining more stuff to them. Look at all the SUVs on the road. I know liberal environmentalists who drive SUVs and who like to take 12,000 mile airplane flights. You think they don't know that they are polluting the atmosphere?

People will buy hybrid vehicles when their costs are low enough to be quickly paid back by spending less at the gas pump.

The American car companies are bringing out hybrid SUVs. Americans want to have their cake and eat it too.

Only technological advances will provide popular solutions because only technological advances will make things cheaper while also reducing pollution.

Tom DC/VA said at March 27, 2005 7:35 PM:

"The vast majority of the population are opposed to higher electricity and higher gasoline prices. So any proposed solution that includeds higher energy prices is a non-starter."

Higher prices are coming, one way or another. I think this is an important enough issue that it should be pre-emptively managed rather than letting the market work, which I think is where we most significantly disagree. In the short term I am for motor vehicle fuels tax (preferably one that increases predictably each year), though maybe in 10 years or so a general carbon tax should be implemented. I would use the extra revenue to (in order) fund sufficient energy research, to cover the current gaping budget deficit (to spend is to tax), and finally to lower taxes on labor. For electricity, I think efficiency regulations have worked moderately well, so continuing down that path instead of raising costs makes sense for now.

"People have been hearing for decades about energy and the environment."

They've also been hearing plenty of opposing information, often dishonest. Since the media mostly presents the two sides in a he said/she said manner and fails to adjudicate the facts, that leaves people to form opinions on the basis of competing 15 second sound bites. It is hardly surprising that most people choose the path of least resistance and greatest short-term gratification (keep driving that SUV) given what they hear. Individuals and societies don't always make good choices.

My costs for all the generating systems are probably high. They're just WAGs for comparison against the cost of the war. Ordering that much solar or wind would push costs down significantly. I believe costs for 1.5MW wind turbines in Europe are already down in the $1.6 million range installed but not connected to the grid. Wind capacity over 20% of local demand requires backup power, so I raised the unit cost to cover those capital costs and forgot to break that out. But not every turbine needs backup: fully developing wind in the Dakotas would swamp the grid, but wind in the Texas/Oklahoma panhandles is adjacent to enough demand that it could not become the main power source in that area.

This article in The Economist says the cost for the Finnish reactor is $3.6B. Since it is essentially a one-off in a cold location well removed from the companies producing most of its parts, I think it is safe to say costs are higher there than for a plant in the continental US. I cannot find a cost estimate for the new reactor (reactors?) at Flamanville (which has existing reactors, as does the Finnish plant - how does that change costs?), except a quip that said ENL was taking a 25% stake for €750M, which would put costs at €3B. I also found something that said construction is expected to take 57 months. I will poke around some more and follow up if I find anything definitive.

Philip Sargent - do you have link to more information about Flamanville handy?

Engineer-Poet said at March 27, 2005 9:53 PM:

Tom:  I think your cogenerator cost figures are way high (unless they include a very expensive installation, which proper design could preclude); $6000 is about twice what a car's drivetrain costs, and the drivetrain has more parts and must endure much more difficult conditions.  If a 4-cylinder engine and transaxle can be built for $3000 with all required systems, a 2-cylinder generator with no fuel tankage (natural gas line), similar cooling system and improved silencing should cost no more and would probably cost less.

I'm going to have to do a quick comparison between cogenerators and nuclear per annual MWH and per ton-CO2 eliminated.  It may be that cogenerators aren't all that great compared to alternatives, although the political opposition to them is going to be near-zilch.


People have been hearing for decades about energy and the environment. You can't get very many people to make personal sacrifices by explaining more stuff to them.
I see people voting taxes for parklands and open-space preservation.  They obviously see a benefit from it.  This is different, but not extremely so.
Look at all the SUVs on the road.
Look at the market segment taking the biggest hits due to today's (still rather cheap) fuel prices.  Do you think that the SUV segment would be so big at $3/gallon?  Would it even exist at $4?  Don't forget that US wages are stagnant; something is going to give.

A hybrid which is twice as heavy is going to need a battery pack twice as expensive.  Even the Escape isn't all that serious a vehicle, holding less than station wagons and even many passenger sedans; the SUV is going to be priced out of the market whichever way you cut it.

Randall Parker said at March 27, 2005 10:40 PM:


Parklands are cheap. So many are already owned by governments that not much has to be spent now to create new ones. Try adding a $1 per gallon to gasoline via a tax increase and see if that gets support the way that parklands do. I'll tell you in advance that it will get far more opposition than support.

Sure, as gasoline prices rise fewer people buy SUVs. But, yes, the SUV segment would still exist at $4 per gallon. Detroit would sell more expensive hybrid SUVs to get the cost per mile down low enough that the buyers would return.

Yes, certainly gasoline prices could go to $3 per gallon if oil prices get high enough. That might even happen in 2005 or 2006. But the gas prices will not go that high in America as a result of taxes.

As for Tom's suggestion of a hydrocarbon tax to fund energy research: Nobelist Richard Smalley thinks $10 billion per year would be enough to fund a very aggressive and wide ranging attempt to develop new energy technologies. That works out to about $35 per American per year. A very small fuel tax would be enough to raise that amount of money. I'm for a $10 billion per year energy research effort regardless of whether it is funded by new taxes or borrowing. But I am pessimistic that such an effort will be funded any time in the next 4 years.

Randall Parker said at March 27, 2005 10:52 PM:

Tom DC/VA and E-P,

The Taiwan reactor works out to $2.4 million per MW. The Finnish reactor is 1600 MW and given Tom's $3.6 billion cost estimate it works out to a slightly lower $2.25 million per MW. I'd love to see a cost estimate for a fairly new reactor built in France.

What I'd like to know: What would the cost of nuclear reactors be if dozens were being built?

I'd also like to know what coal plants cost if they are built with SOX and NOX scrubbers. Check out this article about NOX and SOX scrubbers on coal burning electric plants. Note how they greatly reduce mercury emissions too as a side effect of removing the NOX and SOX.

I don't think a wind installation's peak MW can be compared to nuclear MW because wind's average operating level is much lower than nuclear's. Plus, nuclear provides power any time you want it, not when the wind blows.

Engineer-Poet said at March 29, 2005 11:52 AM:

I read a comment elsewhere (which I cannot find at the moment) to the effect that government energy research will be a captive of the politicos' need to buy influence and not the need to look at hundreds or thousands of avenues to see which ones are actually productive.  This view seems to be borne out by experience.  Conclusion:  $10 billion/yr could disappear down the drain, and even push out research into the best options by attracting the researchers to the money rather than the potentials.

It may indeed be better to view wind as a source of energy rather than grid capacity, but that doesn't mean it can't do things more cheaply than today; I think the possibilities are great if we can get the right leverage.  I hope to be expanding on this at The Ergosphere.  (Assuming that Blogger actually allows posting; it's taking forever to post an update to cogeneration@home.)

Engineer-Poet said at March 30, 2005 1:04 PM:

Found the comment here.

Tom DC/VA said at March 30, 2005 10:43 PM:

Here's another link that states a round €3B price tag for one reactor at Flamanville. Really hardcore analysis is probably only in French, and my French skills have deteriorated beyond usefullness, so someone else will have to look. I would say even if dozens are built the cost savings wouldn't be more than 1/3 due to how much is site-built. The publicly stated costs may slope down faster depending on how the engineering costs are spread.

"I don't think a wind installation's peak MW can be compared to nuclear MW because wind's average operating level is much lower than nuclear's."

No doubt. I stated peak numbers because the capacity factors for solar and wind are both geographically dependant and subject to some debate, and unknown for the cogen device. Clearly the nuclear plants look really good once you knock wind down to 25% capacity.

Engineer-Poet said at March 31, 2005 7:26 PM:

Wind doesn't look so bad if it's in a context of systems which can react flexibly to availability (probably solar too, I haven't checked that out in depth yet).  My analysis is here at Forty-two (yes, the reference is deliberate).

Nuclear looks great in the long run, but I can't see the first modular pebble-bed system going on-line in the USA less than 10 years from now; 5 MW wind turbines could be on-line in 2006.  I think we're going to need everything.

schrodinger said at April 4, 2005 3:34 AM:

Nuke costs:

3 billion euros for EPR works out to $1800 / kw at purchasing power parity exchange rate of $1 = 1 euro, or $2400 / kw at todays market exchange rate of 1.30. IMO the PPP exchange rate is the most relevant for guessing costs in the US.

Costs for series production would fall 5% for each doubling in the number of units built, and another 5% discount for twin units on one site.

First of a kind engineering work would raise the cost of the first reactor of any type by 25%. Presumably this has already been paid by the French government.

For their last generation Framatome claimed $1350 / kw for the fourth plant. Those were built in 2 groups of two.

Elsewhere, the Koreans built 6 (I think) System 80 PWRs for $2400/kw.

Costs for first two ABWRs built in Japan are often quoted as around $2000/kw. However the latest one seems to be 10% cheaper. The Taiwanese ones were originally quoted at $2000/kw, but political delays seem to be driving up the cost.

The Canadians say the two CANDUs recently built in China cost $1640/kw.

Conclusion: $2000/kw seems like a good estimate for a couple of nucs built in the US. Possibly it could be done for less, especially if a lot of one design were built in series.

George said at September 25, 2005 4:46 PM:

I have a question. Does anyone know or what are the opinions pertaining to photovoltaics (PVs) being mass produced in China to where they would be a routine commodity at Walmart?

All my life I have heard the annoying whine of how expansive it is to produce them, or how exotic they are... bla bla bla. I am using the 60-70s space age as a reference. They were used in Skylab 30 years ago. Calculators are now cheap, computers set on everyones desk, what up with PVs? Why hasn't some enterprising entrepreneur opened a plant over there and sell them for $200 bucks at Walmart or Home Depot? It seems like this a Space Technology that is intentionally kept impractical. Especially because of U.S. Defense Contractor style manufacturing.

I was watching the Science Channel and they shown how they are manufactured in this plant- ones and twos instead of mass assembly like every other industrial product.

Randall Parker said at September 25, 2005 5:30 PM:


First of all, photovoltaics costs have fallen orders of magnitude. It is just that they were so incredibly expensive to begin with.

Second, the difference between silicon chips and silicon photovoltaics is that chips got cheaper by getting smaller. But photovoltaics can't do that because they need to take up more space to catch more sunlight. So the economic process of shrinking down transistors that made them simultaneously cheaper and faster does not work for photovoltaics made from silicon crystals. The silicon crystal cost is still high.

No, the technology was not intentionally kept impractical. A lot of smart minds have worked on how to make it cheaper. Several venture capital funded start-ups are currently working on approaches that avoid the use of silicon crystals. With cheaper starting materials photovoltaics could be much cheaper. But the challenge is to find alternative materials and find ways to make those materials.

karva said at June 10, 2007 1:11 PM:

I am a French Scientist
I tried to have an estimate of the EPR price from the project decided in France
(Flamanville, Normandie). It is said to be 3.2 BEuros, i. e. 2 euros per watt.
It is the overall construction price. I have been said that this price is high
because it is the beginning, the Finish one was sold at a much lower price, and
the Finish one is 1.5 years late, which will cost 0.5 MEuros to Areva for delay...

It was very difficult to decide this building in Europe, and that was a
debate at the last elections. It semms to be decided now, but I believe that
we (US and Europ) should build a lot of these, and it is still very difficult
to convince people that it is the most ecological way of obtaining energy...


karva said at June 10, 2007 1:44 PM:

As a French, I find funny, and completly false the following remark: (From Philip Sargent)
Having said that, electricity prices in France are not particularly cheap - but that is much more likely to be because they have not achieved price liberalisation and de-nationalisation of their electricity industry - rather than anything to do with choice of technology.
The liberalisation of the prices is now imposed to France, and this RAISES the prices of electricity. Not probably because of the free market, though French often do not feel it very interesting, but because our electricity is less expensive than the production prices of neighbouring countries: they do not use nukes! So EDF began to align its prices for people which have chosen the "free market" with the other companies in Europ, and with their producing techniques: gas, petrol, Coal. The result is funny: people now are free to buy electricity to any company, but NOBODY leaves the controlled tariff from EDF. And if you leave it, you cannot come back, it is an european decision!

This "liberalisation" is ridiculous. And if you compare the price of electricity for individual homes in Europe, you will understand that I do not want to leave the regulated tariff of EDF!

Dogmas are stupid. For the future, we must build nukes, and no matter which method you follow: private or state industriy: The French example of ecological and cheap energy is the result of a clever politics of the state. But we have now to decide our people to continue...


Randall Parker said at June 10, 2007 2:14 PM:


What does electricity cost for French homes per kilowatt-hour?

In the United States the cheaper states have 7 cents ($0.07) per kilowatt hour. In the more expensive states the prices are double that amount. You can look at this table from the US government Energy Information Administration for the average electric cost per state. The states with lots of coal and hydroelectric power have lower prices.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©