The Energy Grid Blog is managed and written entirely by Seed editors and expert guest bloggers. Read more about them Whether or not nuclear power should be a part of the future energy mix, two things seems almost inescapable. First, nuclear power will part of the near-term energy mix because there are plenty of existing plants that have more than a decade left in their lifespan. Shutting them down doesn't make any economic sense nor would that do anything measurable to mitigate climate changes. Second, nuclear power won't be a part of the long-term energy mix unless some way can be found to split atoms a lot more cheaply.
Just about every time I give a presentation on climate change, someone in the audience will ask why I haven't devoted much attention to the potential contribution of nuclear power. After all, it's (almost) carbon neutral and it's one of the only existing technologies that can produce baseload electricity (unlike PV solar and wind). My response is always the same: Assuming we are willing to find a way to deal with the relatively modest waste and weapons proliferation issues, we still have to acknowledge that nuclear power generation is hideously expensive.
Joe Romm's posts are among the best at laying out just how expensive. The cost of a gigawatt of generating capacity for a new plant just keeps going up. From $4 billion, to $7 billion to $10 billion, depending on the technology involved. There's a reason why no nuclear power plants have been ordered in 30 years and Three Mile Island isn't an excuse anymore.
In Canada, Ontario Hydro just last week announced it was giving up on building more nukes for the foreseeable future because "the bid from Atomic Energy of Canada Ltd., the only 'compliant' one received, was more than three times higher than what the province expected to pay" That from the Toronto Star.
Yes, Canada likes to use heavy-water and U-238, rather than ordinary water and U-235, but the same problems are afflicting US power utilities. And in France (where the nuclear utility has the highest debt load in the world), and in Finland.
Wind is now cheaper the nuclear, even though you have to build three times the capacity to account for the fact that the wind only blows strong enough a third of time. And baseload power can be supplied by concentrated solar-thermal plants, in which heat is stored in fluids for release at night. So why spend the extra money when competing technologies are less expensive?
A related problem is the ever-changing regulatory and economic context. In order to invest the huge upfront sums of money required by nuclear plants, utilities need to know what kind of world they'll be operating in for decades to come. They simply don't have that when it comes to nuclear power. Costs are always rising, and environmental restrictions are ever tightening. The Tennessee Valley Authority learned that the hard way and end up mothballing incomplete nukes. Plus, the question of where we're going to put the waste is now completely in doubt thanks to Barack Obama's decision to scrap the Yucca Mountain plan.
By comparison, we know that wind and solar power, as uncertain as the economics are at the moment, will be increasingly favored.
Consider also the news that GE, one the biggest nuclear plant builder, is now asking for more government financial assistance "to help make power generation from new nuclear plants easier." GE would like to try to recycle nuclear waste, presumably using breeder reactors or other "fourth-generation" technologies.
So would we all. But thorium reactors, for one, and the other 4G technologies aren't ready for prime time and won't be for another 15 years or so at the earliest. And we need clean alternatives now.
So maybe, just maybe, nuclear power in some new form will be part of the energy mix in the distant long term. Maybe. But I wouldn't bet on it.
Environment
Technology
Comments
We needn't be so dismissive of thorium power if we were willing to devote the resources we did for the Manhattan Project and Apollo 11.
Posted by: Irving Smith | July 27, 2009 1:46 PM
Could be sooner than you think, Thorium Power just signed a collaboration with Areva.
Posted by: Ken | July 27, 2009 5:58 PM
My, my James. What a strong bias you have!
You point to the "up front" capital cost of a new nuclear plant, but don't address the cost of equivalent forms of generation. Furthermore, you don't point to the operation and maintenance costs over the lifetimes of such plants.
Yes, safe nuclear plants are expensive to build relative to oil or gas or similar, but are fairly cheap to operate over their sixty year lives.
Base load is an important concept not to be treated trivially. So far, only fossil fuels, nuclear, and hydro [where available] have consistently delivered reliable base load energy in sizeable quantities. Point to a single solar-thermal plant in the world that stores sufficient heat energy to generate a thousand megawatts of power all night long. It is true that one would have to build three thousand megawatts of wind power to deliver the energy equivalent of 1,000 megawatts of nuclear power because the fission process provides heat constantly whereas the wind blows strong infrequently. But even with three times as much generation capacity, wind delivers nothing when it isn't blowing. It can never serve as baseload.
Peak load generation is important because the demand for electricity is not flat. Peak consumption is always higher during the day when the sun shines. Except for cloudy days, solar power is great to shave off these peaks. So is wind on the days when it blows. Fortunately, the wind tends to blow stronger on cloudy days so these technologies work well together for peak load shaving.
We will need most all technologies in our future energy mix. Please don't confuse readers by ignoring the important fuel type distinctions that serve base load and peak load requirements well.
Posted by: Sid | July 28, 2009 12:42 AM
Two of you -- James and Sid -- have misused the term "baseload", basically as a synonym for "uninterruptedly reliable". It means no such thing, and nuclear isn't uninterruptedly reliable, either.
By definition, baseload power is the power that an electricity system's dispatcher -- these days, usually an independent system operator -- "loads" onto the grid first, at the base of the "stacking order". That decision hinges primarily on short-run avoidable costs (without reference to capital costs like construction or major refurbishments), though it is also influenced by technical factors like the hazards of throttling a nuclear plant up and down to follow the grid's changing demands.
In most systems, the bottom of the stacking order -- the "basest" of the baseload capacity -- is made up of nuclear power, run-of-the-river hydroelectric capacity (the stuff that can't be held back until it's needed), wind, and solar (if any). Get used to it, wind and solar are baseload! (What did you think they were, peaking?!?)
Contrary to the confused perception that is fostered by the world's nuclear industry, a reliable grid doesn't need reliable baseload plants, NOR are baseload plants especially reliable. In order to be reliable, a grid DOES need PEAKING plants that are especially reliable, or the lights will go out when demand is high, or when the baseload plants go on the fritz.
In the "artists' conception" of an electrical grid, baseload capacity is BUILT (not run, built), not to increase the reliability of the grid, but to lower its costs. In the real world, especially in North America, nuclear generators haven't done that at all, because their total costs, including their capital costs, are too high.
But nuclear is still overwhelmingly baseload, partly because of its special hazards when load-following, and partly through special deals. E.g., in Ontario, the Pickering and Darlington plants are owned by Ontario Power Generation, which is owned by the Ontario government, which guarantees OPG a market for its electricity, essentially regardless of cost.
The two other nuclear stations, Bruce A and Bruce B, are operated by a private consortium called Bruce Power, which leases the reactors from OPG. Bruce Power, not being owned by the government, doesn't get that special guarantee, except that it negotiated for a guarantee for Bruce A's power (and the government foolishly granted it) before agreeing to refurbish two reactors at Bruce A that should have stayed shut down forever. But Bruce B still has no guarantee, so poor Bruce Power has had to sell some of Bruce B's power at NEGATIVE prices recently, and has twice now been asked to shut down a whole reactor at Bruce B, because demand has been so low in the "valleys".
Baseload means INFLEXIBLE. It means power that tends to be "take or pay", needing (technically and/or economically) to be used, whether or not it's needed. Think nuclear, think wind. There's a limit to how much of that kind of power can be accommodated in an electrical grid -- as Ontario is finding out right now, as we speak! There is no limit to how little of that kind of power we can make do with. And don't forget that the main rationale for accommodating inflexible capacity is to lower costs, which nuclear capacity doesn't do, at least until its original owner has gone bankrupt or taken huge "stranded asset" losses and sold its reactors for a few cents on the dollar. (In Ontario, we have a separate item on the electricity bill for the nuclear-plant mortgages, euphemistically called the "Debt Retirement Charge".)
It is true that nuclear reactors can sometimes attain annual capacity factors that wind generators can't approach -- 100% (or even a few percent higher!) vs. around 30%. And it's also true that an operating nuclear reactor has a relatively low probability of suddenly ceasing or decreasing generation, compared to a wind farm. But it's also true that whole nuclear stations can stop generating ANYTHING for YEARS, through windy days and sunny days, while wind and solar plants are virtually immune from that brand of unreliability. Again, here in Ontario, we had 8 full-size reactors that generated absolutely nothing for almost 7 entire calendar years. And soon after two reactors at Pickering A were finally refurbished, big bucks and years over the industry's promises, they had to be shut down through their first entire summer -- Ontario's peak season for electrical demand! During those shutdowns, the wind blew as reliably as ever, and the sun shone, too.
Instead of hiding behind misleading and poorly understood words like "baseload", you should just try to make your claim in plain English. Like "a modern grid has to be around 99.99% or so reliable, and only nuclear plants can attain that level of reliability." Unfortunately, the statement is laughably false, and the error is obvious when the misleading and misused jargon is stripped away.
Posted by: Norm Rubin | July 28, 2009 2:08 AM
Norm -
Properly speaking, wind and solar are opportunity sources, not baseload, since they can't be planned for. Indeed they are best classed as negative load.
The key concept is despatchability, which crudely means can you turn it on if you need it; for both wind and solar the answer is no. That is where the grid reliability comes from; you don't need 99.99% capacity factors (the 95% or so the nuclear plants get is sufficient, since much of that 5% is scheduled downtime). But you do need the downtime to be independent and predictable. Wind in particular has issues with both.. and here in the UK, for the 4 months of the year when electricity demand is highest, solar generation would be trivial (we are a long way north) and we will get multi day periods of light to zero wind.
Posted by: Andrew Dodds | July 28, 2009 3:45 AM
Norm, the LFTR was developed specifically because it could be load following, as a prototype for an aircraft engine. As nutty as this sounds, it was developed as a power plant for a doomsday nuclear bomber, and one of the constants was that the power plant needed the ability to be throttled.
That was a lot of verbage that revealed your ignorance, but little else. For example, nuclear reactors routinely achieve 95% on line performance, and the operators of the grid can rely on them as a constant within the network. The energy in the flow of electrons is utilized as it is produced and the stability of the grid relies on the predictability of both production and consumption.
What no one pays for out of pocket is the externalities of other forms of power generation. Coal for example kills 12,000 Americans a year, according to the Bush administration. These are costly deaths, it poisons the seas with its heavy metals, the streams and rivers with its acid rain and melts the ice caps. Relatively speaking, Nuclear waste is not a problem, I repeat for emphasis, nuclear waste is not a problem, I will type is slower if you like so that people will get it. You should be worrying about Mercury, spewed out by the ton from your local coal fired plant, but you are instead worried about kilos of Np237, distributed within the matrix of intact fuel rods sealed in multi-million dollar containers, waiting eventually to be burned in future reactors where it will become short lived fission products. You should be worried about microparticulate pollution, but instead you are worried about Plutonium that is unsuitable for weapons production. You are worried about Chernobyl, while all the plants being built are far far safer than three mile island, which killed no one, unlike wind mills, whose construction resulted in over 300 deaths last year, and that was human deaths, not bats and birds and acres of trees cut down to place the wind mills and access roads. You worry about mining Uranium, but it does little damage compared to mountain top removal, where the tops of mountains are literally pushed over into the adjacent valleys to reveal an underlying coal seam. It really seems to me that paying a little more for nuclear is a bit more honest accounting.
Posted by: Ken | July 28, 2009 9:00 AM
Also, Shell sponsors this blog, think about it.
While I recognize the lousy optics of an energy blog sponsored by one of the largest oil companies in the world, I can assure you that Shell has absolutely no say over the content. None. -- JH
Posted by: Ken | July 28, 2009 9:03 AM
Wow James, Norm et al I guess all this goes to show how complex power generation is and how strategically important it is to get it right.
I believe in a well balanced grid that is designed to deliver power 100% of the time. That grid has to have a range of production techniques available to it. Presently I am working in the Nuclear Industry but would be equally happy working in other generation industries.
What I would like to see is a debate based upon facts. This way we have the best chance of optimising the mix for the future.
Already in this blog we have seen a number of simplifications or summaries that end with a misleading result. Norm correctly points out an issue around the undersatnding of baseload. But he then goes on to propose that windmills have a 30% capacity factor and may be they do plan on achieveing that...but it is not what they achieve in Ontario.
Then James does some spurious maths to say that if they run a third of the time then if we build three times as many we will have the power we need. But this would only be true if the wind were kind enough to ensure that it is always blowing on a third of the windmills. Sadly it doesn't oblige. The peak power days in Ontario are a hot sultry summer day characterised almost by definition by having no wind over a large area and the very clear crisp cold winters day characterised (yes you have guessed it) by having no wind over a large area. So basically in Ontario you have to back up pretty much every windmill with an equal amount of some other power.
Now a few other errors.
1) reactors have been ordered over the last 30 years, China, France, Finland etc have all been building commercially purchased plants. Canada hasn't bought any because it did not need the power. It has nothing to do with cost.
2) OPG did not make the decision to suspend the nuclear procurement process in Ontario, the Provincial Government did it.
3) No one said they were giving up building for the forseeable future just that the RFP process was suspended. In fact the rhetoric confirmed that new build was still expected.
4) There has been no indication of the actual prices quoted. It has been confirmed that they were higher than expected but not by how much. No one has identified what the apparently leaked figures were meant to cover and so they cannot be used in any calculation.
I could go on.
We owe it to people to represent information accurately so that we end up with a workable energy mix.
Neil Alexander
Organisation of CANDU Industries
Posted by: Neil Alexander | July 28, 2009 9:08 AM
JH, it may well be that Shell does not exercise editorial control over this blog, however, the title and the sponsor are a suspicious combination. I trudge through anti-nuclear blogs and many have clean coal ads on them, in fact most. Anti-nuclear is dangerous and irresponsible. Nuclear is a threat to the fossil fuel industry. The fossil fuel industry is killing this planet. The anti-nuclear proponents are their useful idiots.
Posted by: Ken | July 28, 2009 9:50 AM
Ken -
Nuclear is a threat to the Coal industry, not oil and gas. Quite frankly, I can see no viable scenario that leaves much oil or gas in the ground; even if we built absolutely every non-fossil source as fast as possible, it would be at least a couple of decades before oil and gas use started to be dented below the levels imposed by resource constraints.
So I'm not sure how Shell would be threatened in the real world - if I were Shell I'd be taking a very serious look at part and full-synthetic liquid fuels derived from waste products (Paper+Hydrogen->Methanol, for instance), using off-peak electricity. Because in a Nuclear/Renewable grid, variable load applications such as the above may be easier than variable generation. Just a thought..
Posted by: Andrew Dodds | July 28, 2009 10:16 AM
Nuclear power can be linked to fuel production and with the appearance of plug in hybrids and electric cars on the market can erode market share for the oil companies whose margins are going to be increasingly squeezed by the ever increased cost of finding new oil reserves. Syngas production and hydrogen can both be driven by electric power from conventional LWR, high temperature reactors such as the LFTR and HTGR can drive hydrogen production even more efficiently.
Posted by: Ken | July 28, 2009 10:38 AM
Shell must sell a lot of LNG, as I stated before, the LFTR has load following capabilities, this would knock out the need for natural gas fired peaking generators, as well as provide base load generation capacity.
Posted by: Ken | July 28, 2009 11:01 AM
Why do we always look to the past for a solution to our problems? Future energy sources will come from the development of future technologies. When the same amount of money is allotted to development of new energy sources, as is forecast to be the cost of Nuclear power development, we may have our solutions. The biggest obstacle to overcome on the road to new technology appears to be the corporate dominated control of our energy supply(they produce at whatever the cost and we buy from them at whatever the price).
How surprised will people be to discover that their future energy requirements, in their home and office, will be self generated, that there will be no need for an electrical power grid, and no monthly power bill to be paid, just an annual maintenance regime to be followed on their own environmentally friendly power generating system? Sound far fetched? Think of some of mankind's finer accomplishments and how they have revolutionized our world. In the beginning each and every one were no doubt thought impossible and ridiculed.
Mankind has to get back to the idea that we don't know everything there is to know about everything. Educators have to stimulate the minds of the youth in order to allow them to develop new technologies instead of relying on decades old beliefs.
In the words of Thomas A. Edison "I have not failed. I've just found 10,000 ways that won't work."
Give innovators the time and funding and we will be amazed.
Posted by: John Mehlsen | July 28, 2009 2:28 PM
"Give innovators the time and funding and we will be amazed."
allow me to finish is a rational way:
By the new nuclear technologies that will be commercialized and put into production.
Posted by: Ken | July 28, 2009 4:30 PM
I would not bet against it. The direction the US policy will take may have more influence than we assume. The Clean Energy and Security Act may deem nuclear as 'clean'. It's still early but feel that nuclear generated power will continue for some time in Ontario. The high costs of nuclear will continue to scare a provincial postering with the ferderal with the liquidation of AECL. It might be safe to assume the new owner(s) may have a stake in the ground already.
Posted by: Jerrod | July 28, 2009 4:40 PM
"Wind is now cheaper the nuclear, even though you have to build three times the capacity to account for the fact that the wind only blows strong enough a third of time."
Except that it isn't. It is true that one gigawatt of wind is cheaper than once gigawatt of nuclear, but three gigawatts of wind is about 2 to 2.5 times more expensive. According to the MIT update to "the future of nuclear power," wind is $3,000 per kilowatt and nuclear $4,000 per kilowatt. So, nuclear is cheaper. Also, nuclear plants last 60-100 years and windmills 15-20 years! In the long run, nuclear is quite cheap. In 2008, nuclear was 1.87 cents per kilowatt-hour, coal 2.75 cents per kilowatt-hour, photovoltaic solar 38 cents per kilowatt-hour. If windmills run a quarter of the time, and last a quarter as long, that is a 16 fold cost increase. Wind also requires expensive gas imports or pumped storage for load following for when the wind doesn't blow. The claim is often made that Denmark gets 20% of its electricity from wind, but 84% is exported to Norway where hydropower serves as load following.
"And baseload power can be supplied by concentrated solar-thermal plants, in which heat is stored in fluids for release at night"
This is also not true. The fluids, like molten salts, cool off progressively into the night, producing diminishing power. This technology is suited to peak-load, helping smooth out fluctuations from when a cloud passes overhead. It doesn't work at all in the winter.
"4G technologies aren't ready for prime time and won't be for another 15 years"
This is also not true. The U.S. shut down its Integral Fast Reactor 2 years before completion in 1994. We'd be building `em now if it wasn't for that. According to tom Blees, we could start building Integral Fast Reactors by 2015, and fuel them with our existing nuclear waste. A new generation of Gen-III+ light water reactors today is fine, since their waste can safely be stored right next to the power plant for decades, then be reprocessed in Fast Reactors at 160 times the efficiency.
Posted by: Zachary Moitoza | July 28, 2009 7:06 PM
John Mehlsen -
Technically, given my south-facing roof, I could put sufficient solar panels up there that - assuming I also paid for an awful lot of battery backup - I could go off-grids for electricity at least.
However, I would still be dependent on corporations to supply the solar panels and batteries, and supply replacements as these degrade. I don't see how this is different from a central corporation generating the electricity and supplying it direct, apart from there being greater reliability and lower cost doing it centrally.
As far as making a genuine baseload plant from wind or another intermittant source, there is an equation for this:
n=1+((1-Cf)/(Cf*Se))
Where Cf is capacity factor and Se is storage efficiency. For example, imagine a wind turbine installation where the capacity factor of 0.3 and a H-fuel cell backup system efficiency of 0.6 (Generate hydrogen during high wind periods, generate electricity in low wind periods).
In this case, n=4.89, meaning that you need ~5 1Mw turbines to allow 1Mw of continuous power supply from your pututiave wind-baseload plant.
Posted by: Andrew Dodds | July 29, 2009 4:03 AM
Whether nuclear is economic or not boils down to the following question:
* what discount rate do you apply to the project?
If you assume a low discount rate (say 7.5%) nuclear makes money hand over fist. If you assume a high discount rate (say 15%) nuclear is daft. Generally people assume a high discount rate.
Alternatively phrased, suppose we have a project to build and run a plant. If the project gets executed as intended, nuclear is profitable, but if the plan gets derailed or the plant fails, it's not. Generally people assume the chance of project failure is high.
Why?
I don't know for sure, but these are my general impressions (based partly on "Nuclear Energy: Principles", by Bodansky).
Most of the project risk in nuclear occurs in the planning and construction phases. In the planning phases the big risk is regulatory, i.e. you sink say $15m in a plan to build in location A, and then the government decides that location B is better, and you need to move there. Or when (as happened with Shoreham) over $6bn 1980s dollars was sunk into the reactor only for government to change its mind. Construction risk is also large, in large part because there's not much experience in constructing nuclear plants at the moment. This seems to be the case in Finland.
Construction time is obviously also a big factor. If you're paying interest on capital costs for 150 months (which is the average time to construct a plant in the US) the plant is going to be much less economic than if you're paying interest on capital costs for 50 months (the average time in Japan).
The post-construction risks seem to be on a par or lower than conventional plants, so the same or a lower discount rate should apply for existing plant.
The big cost factors seem to be in regulation and experience, not the fundamental nature of the technology.
Posted by: Patrick Caldon | July 31, 2009 11:53 PM
Connecting weather with climate is a tricky thing. Some thoughts... one very interesting result of GCMs would be the projected locations of low and high pressure areas with a higher tropospheric energy (stored there
Posted by: seks | August 15, 2009 10:59 AM
Connecting weather with climate is a tricky thing. Some thoughts... one very interesting result of GCMs would be the projected locations of low and high pressure areas with a higher tropospher
Posted by: porno | August 15, 2009 11:01 AM
I agree with John Mehlsen comment about giving the innovators and brilliant engineers out there the time and resources to develop and succeeded.
The patent system is a farce. It feels more like a boys club rather than a system that supports and works round problems.
The patent system needs a complete overhaul to allow the free thinkers out there the resources to thrive.
We cannot allow good people to slip through the net.
Posted by: James Booth | August 18, 2009 8:41 AM
The current financial turmoil has hit people and companies really bad. In this situation any organization would think of earning more and more. At a point when organizations are laying off employees heavily, few organizations can actually think of sustainable development, conserving energy and promoting clean energy.
All that comes in mind is the next quarterly results, top line and bottom line and nobody has time to think about triple bottom line. Our short term needs.
We will pay heavy, if we keep neglecting our long term needs. We better get aware and make others aware.
Posted by: Sara Wolcott | September 1, 2009 8:57 AM