Is Rocketing Nuclear Waste into the Sun a Viable Solution?

[mheslep]

From MIT's 'Future of Nuclear Power':
http://web.mit.edu/nuclearpower/img/summary.gif

Proliferation Summary

Proliferation. The current international safeguards regime is inadequate to meet the security challenges of the expanded nuclear\ deployment contemplated in the global growth scenario. The reprocessing system now used in Europe, Japan, and Russia that involves separation and recycling of plutonium presents unwarranted proliferation risks.

Under unresolved problems:

Proliferation: nuclear power entails potential security risks, notably the possible misuse of commercial or associated nuclear facilities and operations to acquire technology or materials as a precursor to the acquisition of a nuclear weapons capability. Fuel cycles that involve the chemical reprocessing of spent fuel to separate weapons-usable plutonium and uranium enrichment technologies are of special concern, especially as nuclear power spreads around the world;

Statement:

Nuclear power should not expand unless the risk of proliferation from operation of the commercial nuclear fuel cycle is made acceptably small. We believe that nuclear power can expand as envisioned in our global growth scenario with acceptable incremental proliferation risk, provided that reasonable safeguards are adopted and that deployment of reprocessing and enrichment are restricted. The international community must prevent the acquisition of weapons-usable material, either by diversion (in the case of plutonium) or by misuse of fuel cycle facilities (including related facilities, such as research reactors or hot cells). Responsible governments must control, to the extent possible, the know-how relevant to produce and process either highly enriched uranium (enrichment technology) or plutonium.

Three issues are of particular concern: existing stocks of separated plutonium around the world that are directly usable for weapons; nuclear facilities, for example in Russia, with inadequate controls; and transfer of technology, especially enrichment and reprocessing technology, that brings nations closer to a nuclear weapons capability. The proliferation risk of the global growth scenario is underlined by the likelihood that use of nuclear power would be introduced and expanded in many countries in different security circumstances. An international response is required to reduce the proliferation risk. The response should:

o re-appraise and strengthen the institutional underpinnings of the IAEA safeguards regime in the near term, including sanctions;

o guide nuclear fuel cycle development in ways that reinforce shared nonproliferation objectives.

Recommendation:

Accordingly, we recommend:

o The International Atomic Energy Agency (IAEA) should focus overwhelmingly on its safeguards function and should be given the authority to carry out inspections beyond declared facilities to suspected illicit facilities;

o Greater attention must be given to the proliferation risks at the front end of the fuel cycle from enrichment technologies;

o IAEA safeguards should move to an approach based on continuous materials protection, control and accounting using surveillance and containment systems, both in facilities and during transportation, and should implement safeguards in a risk-based framework keyed to fuel cycle activity;

o Fuel cycle analysis, research, development, and demonstration efforts must include explicit analysis of proliferation risks and measures defined to minimize proliferation risks;

o International spent fuel storage has significant nonproliferation benefits for the growth scenario and should be negotiated promptly and implemented over the next decade.

Design recommendations: some particular designs and methods as realizing the lowest proliferation risk:
-a uraniuim once-through and dispose fuel cycle vs a close-thermal or closed-fast cycle. They specifically mention the use of the PUREX/MOX closed cycle used by Europe and Japan as inferior to open cycles for non-proliferation purposes.
-gas-cooled vs LWR

 

[vanesch]

Just caught this on 2nd pass. I disagree that its entirely unthinkable. Even with wind power alone its very conceivable: 1.5MW turbine farms now give us 10MW/km^2. US electric capacity is ~1000GW, so 10^5 km^2 (25M acres). That compares to about 20M acres currently in use for US corn ethanol (a mistake).

This is the straw man argument given by a lot of anti-nuclear people: wind is better. But there is not to be chosen between wind (or solar, or whatever) and nuclear: both are possible. Wind is not hindered by nuclear, and nuclear shouldn't be hindered by wind. There are many countries NOT using nuclear power, but they have not developed wind energy (if it is so possible) into a significant fraction of their production. My opinion is: first SHOW ME that wind CAN provide a country with, say, 60% of its electricity (without cheating). If that's done, and we can evaluate this, then do this everywhere. And once half of the world turns for 50 or 60% on wind, then we can consider the option of dropping nuclear, eventually. If wind cannot do that (and until it is demonstrated, I don't believe it), then wind is all right, but no substitute for nuclear or coal or gas. It is then just a nice auxiliary minority means of generating power - against which I don't have anything, btw.
So let us give wind all of its possibilities, but let us not stop nuclear before wind has shown to be capable to do what nuclear can do. Personally, I don't believe it, but I can readily accept to be wrong on the issue.

The country that gets most out of wind, is Denmark. Denmark actually has two non-connected grids, and over the whole of Denmark, 16% is wind energy, while over the "windy" grid, it is 20%... but that grid is connected to Sweden's grid (and I think, also to Germany's grid). Sweden has no problem with the fluctuations of Denmark, because Sweden, because of its particular geography, has about 50% hydropower. So each time that the wind blows harder in Denmark, the hydropower in Sweden is diminished, and vice versa. When Denmark has no wind, then Sweden cranks up a bit more its hydropower, to compensate.
So Sweden plays Denmark's buffer. And we see here BTW that Denmark's wind turbines DON'T displace much CO2 production: what it wins, is mainly compensated by LESS hydropower in Sweden.
But to be fair, we should now look at what is Denmark's wind energy fraction, over the production of Denmark and Sweden combined, and then we get below 10%.

So the Danish experiment shows us that, combined with a very suitable wind placement, and a very suitable geography (Sweden's, with a lot of hydro), they managed to get ~10% wind energy installed.

Also, realize that the Danes are extremely well placed for wind energy. Their offshore farms are on the better places in the world, with a high wind offer, that is rather steady. On land, this is much worse.

The Germans have much less than 10% wind installed. In march 2007, they decided to build 27 brown coal power plants. They have suffered a few major grid breakdowns, which some analysts pointed out to be their wind turbines - although others disagree with that. Now, in Germany, the atmosphere is pretty anti-nuclear (they want to get rid of their 36% nuclear) and the public opinion is very "green", so, if wind was such an attractive option, why do they forego this then, and build *brown coal* power plants ??

Nevertheless, wind turbines are a pain for the network management, because of their erratic character. THAT is the main reason why I think that wind cannot be, in the next decades, be the main power source of any country. Most network analysts think that a network containing much more than 20% of wind energy becomes essentially unpilotable, unless you associate with each farm, also gas turbines which compensate immediately and locally, or when you have a very big and distributed network of hydropower plants. Then you can get to higher fractions, but still less than 50%. A ratio of 1/3 of installed power, and average power, is considered a very good performance (usually only with offshore farms on a good place).

As I said, I have nothing against wind energy, but when you look at its technical issues, it is difficult to conceive it to be a majority provider (and that's an understatement).

It's not the price of the wind turbine, or the place it takes up which puts the ultimate limit to wind energy: it is its erratic power. If we don't have (and we don't) compact and cheap electricity storage (apart from hydro pumping stations), and a large variable at will load (like, you said it, hydrogen production), then I don't see this change. This is why I don't think that this is going to be different in the coming decades.

 

[vanesch]

Nuclear Cost:
I like the http://web.mit.edu/nuclearpower/" with a cost of $14B + $3B transmission, or ~$7/W; no chance of $0.06/kw-hr power coming from Levi at that cost.

I don't understand that $14B. The new EPR, which is still in "pilot" stage (no mass production yet, so more expensive), is estimated to cost ~ 3.3B Euro
http://en.wikipedia.org/wiki/European_Pressurized_Reactor
(although the Finnish project will probably be more expensive, mainly for extra regulatory issues - red tape costs). It is a 1.6 GW e plant. So even with a serious cost overrun (say, 50%), and knowing it is still a pilot plant, $14B seems extremely expensive compared to this.

Wind Cost:
UK BWEA report, 2005, with 2003 costs.
http://www.bwea.com/pdf/briefings/target-2005-small.pdf
Average onshore: $0.06 / kw-hr
Average offshore: $0.11 / kw-hr

Yes, but *average* doesn't mean: "when needed" ; as such, average erratic power is less than half the price of "available".
Taking the average price equal to the total price can only be the case when wind is "within the noise margin" of power production, as it is in all countries but Denmark. So the marginal cost of wind is relatively low for the first 10-20% of installed wind power, and once it becomes an important component in the network, and starts causing problems because of the fluctuations, costs rise dramatically, because of the need of compensation.

 

[vanesch]

Design recommendations: some particular designs and methods as realizing the lowest proliferation risk:
-a uraniuim once-through and dispose fuel cycle vs a close-thermal or closed-fast cycle. They specifically mention the use of the PUREX/MOX closed cycle used by Europe and Japan as inferior to open cycles for non-proliferation purposes.
-gas-cooled vs LWR

We've been through that, and my opinion (shared by a lot of people in the nuclear world) is that, although it is true that the current *technological knowledge* of the fuel cycle is a proliferation issue, its actual application in most western countries is not.
Personally, I find the open fuel cycle a huge waste. You throw away 95% of the energy content of the material, and you have much longer-lived waste than needed.

The real solution is the use of fast breeders (or fast reactors, not necessarily breeders), and reprocessing which is less "picky" and separates the actinides from the fission products, without separating with high purity the plutonium. There's a lot of development under way, but, contrary to the PUREX process which has an industrial experience of more than 40 years, these things are pretty new, and still in the research or prototype phase.

There's a point I don't understand in the MIT recommendation. Although it is true that the PUREX process (the technological knowledge) is a proliferation issue (and then, it is not, because you can now even find it in books, I have one), the MATERIAL itself is, I would think, not so much of an issue. The plutonium that comes out of a PWR (so in the spend fuel of the open cycle) has a bad isotopic composition to make bombs out of. It is not impossible, but it is already difficult to make a plutonium bomb with "good" plutonium (Pu-239), but it is still much more difficult to make a plutonium bomb with reactor plutonium. The probability to have a fizzle is bigger, and in any case the yield is very low (it will be a low-power bomb of at most a few kT). Now, MOX fuel is again an U-Pu mixture, so if you make immediately MOX fuel of the separated plutonium, you have not much more of a bomb material than in an open fuel cycle. In fact, making MOX fuel and putting it again in a reactor makes it hopelessly useless as a bomb material, so in fact one of the reasons to use MOX fuel is to diminish the presence of "bombable" plutonium.

In other words, the usability of MOX fuel as bomb material (before use in a reactor) is of the same order as spend fuel, and is in any case very low (you still need to separate it, and it is "bad" material of which it is difficult to make a working bomb). USED MOX fuel is hopeless bomb material. The PUREX knowledge is out, in any case. People know now how to separate Pu from U.

And, in any case, if you really want to make a bomb, and you put enough money and engineering to it, you will succeed.

 

[russ_watters]

Wind also installs at ~$1.7/W. Nuclear has to fight the hippies to get that cost down now, and would have to keep work hard to keep it there w/ a big mining load and a big waste load.

Wind is definitely a viable technology now. It isn't a single-source solutin, but it does help. But due to it's unique limitations, it can only ever be a suppliment to reliable single-source solutions like nuclear power.

 

[russ_watters]

I like the http://web.mit.edu/nuclearpower/" for my gold standard. It is not without criticism, but everyone pro and con seems to use it as a baseline for discussion.

Well, I suspect it is pretty good on the facts, it's just the opinions that are suspect. The author seems to have a pretty strong left-wing bias that reflects in his work:
http://web.mit.edu/chemistry/deutch/policy.html

Simply put, the opinion parts of his position seem based on the obsolte/incorrect hippie view of nuclear power. I must admit, I've only read his summery so far - I'll have to read more of his analysis.

 

[vanesch]

BTW, in this document http://web.mit.edu/chemistry/deutch/policy/80TheNuclearOption2006.pdf
you also find a balanced approach, with similar recommendations.

Now, personally, I would find it a huge pity to throw away all that spend fuel. That said, we can keep it in a temporary storage until people will realize (or not) that it is full of good fuel, and that we can use all that plutonium to start a fleet of breeders (50 years from now, who knows). The later the reprocessing occurs (say, 60 years from now) the easier it will be (except for the Pu-241 which will have decayed into Am-241 by then, and which is a pain).

You have to realize that in spend fuel, still 95% of the energy content is present, which is available in a fast reactor. That means that if you have been running on nuclear power for, say, 30 years, you can still extract 20 times more of it, that means, 600 years on the same material at the same power output (and still 10x more if you use all the impoverished uranium, so 6000 years), just with the spend fuel and the "waste". You don't even need mining of fresh uranium ore anymore.

Don't you think that it is totally wasteful to throw away 200 times the energy content of material you possess, just because of some idle hope that some bearded maniac will have a slightly little harder time (you think) to make a bomb ?

Also, when you have breeders, and reprocessing, you can drop not only mining, but also enrichment. So if you don't like enrichment, then you should go as fast as possible to a fleet of breeders and reprocessing.

 

[Phy6explorer]

BTW, in this document http://web.mit.edu/chemistry/deutch/policy/80TheNuclearOption2006.pdf
you also find a balanced approach, with similar recommendations.

Now, personally, I would find it a huge pity to throw away all that spend fuel. That said, we can keep it in a temporary storage until people will realize (or not) that it is full of good fuel, and that we can use all that plutonium to start a fleet of breeders (50 years from now, who knows). The later the reprocessing occurs (say, 60 years from now) the easier it will be (except for the Pu-241 which will have decayed into Am-241 by then, and which is a pain).

You have to realize that in spend fuel, still 95% of the energy content is present, which is available in a fast reactor. That means that if you have been running on nuclear power for, say, 30 years, you can still extract 20 times more of it, that means, 600 years on the same material at the same power output (and still 10x more if you use all the impoverished uranium, so 6000 years), just with the spend fuel and the "waste". You don't even need mining of fresh uranium ore anymore.

Don't you think that it is totally wasteful to throw away 200 times the energy content of material you possess, just because of some idle hope that some bearded maniac will have a slightly little harder time (you think) to make a bomb ?

Also, when you have breeders, and reprocessing, you can drop not only mining, but also enrichment. So if you don't like enrichment, then you should go as fast as possible to a fleet of breeders and reprocessing.

You are completely right.That's exactly what I read about spend fuel. What about constructing factories for producing energy from spend fuel side-by-side the nuclear industries. In that way we can avoid too much transportation of the spend fuel and of course increase the amount of energy produced. But actually who said it is being thrown away?According to a UN census in 1997, in the 20 countries which account for most of the world's nuclear power generation, spent fuel storage capacity at the reactors was 148,000 tonnes, with 59% of this utilized. Away-from-reactor storage capacity was 78,000 tonnes, with 44% utilised.But the fact is that even after using up the spend fuel, that is after the reprocessing and vitrification of the 25-30 tons of spent fuel produced per year by a typical large nuclear reactor, waste is produced which amounts to about three cubic meter per year. I read that it has been accepted that this final waste will be disposed of in a deep geological repository.But my question is, what will happen to the final waste there?

 

[vanesch]

What about constructing factories for producing energy from spend fuel side-by-side the nuclear industries. In that way we can avoid too much transportation of the spend fuel and of course increase the amount of energy produced.

In fact, the US had such a pilot plant, with integrated fuel reprocessing (which was moreover much more proliferation resistent, because it didn't isolate the plutonium specifically, using pyro-processing). It was the IFR project (Integral Fast Reactor). http://en.wikipedia.org/wiki/Integral_Fast_Reactor
For totally incomprehensible reasons, this has been abandoned when it was almost finished (sounds like the Superphenix and Kalkar debacles).

But actually who said it is being thrown away?According to a UN census in 1997, in the 20 countries which account for most of the world's nuclear power generation, spent fuel storage capacity at the reactors was 148,000 tonnes, with 59% of this utilized. Away-from-reactor storage capacity was 78,000 tonnes, with 44% utilised.

Yes, temporary storage is OK with me. But when one talks about the open cycle, one means: geological (irreversible) storage for good of the fuel elements "as is".

But the fact is that even after using up the spend fuel, that is after the reprocessing and vitrification of the 25-30 tons of spent fuel produced per year by a typical large nuclear reactor, waste is produced which amounts to about three cubic meter per year. I read that it has been accepted that this final waste will be disposed of in a deep geological repository.But my question is, what will happen to the final waste there?

It will decay, and its radioactivity will decrease. The radioactive components of spend fuel are of 3 orders. You also have to know that the shorter the half life, the higher the activity, but the faster it decays, while the longer the half life, the lower the activity, but the longer it takes.

There are 3 components to the spend fuel:
- fission products: the most active, but after a few hundred years, they have decayed. It is the "essential waste" because it is the "ashes" of the fission process.
A ton of spend fuel contains about 50 kg of fission products.

- minor actinides (americium, neptunium, curium). They are undesired products produced in thermal spectrum, and they remain active for a few thousand years. They are much less active than the fission products (except for the curium, but which has a half life of 18 years, so decays quickly), but are nevertheless sufficiently active to "consider them a hasard" for several thousand years (although, as I said, much less so than the fission products).
A ton of spend fuel only contains a few kilogram of minor actinides.

- the plutonium. Similar to the minor actinides, but we have an activity for about 100 000 years. A ton of spend fuel contains about 10 kg of plutonium.

So, the fission products need to be contained for a few hundred years, the minor actinides for say 10 000 years, and the plutonium for 100 000 years. These are in fact the times it takes for the radio-toxicity to decrease to the level of natural uranium ore, at which one considers that geological presence is not much more of a problem than actual natural uranium ore.

So the problem of geological disposal is to ensure that no significant amounts of the material can get back to the biosphere and ground water before stated times.
This is partly accomplished by the human structure (canisters, fillings, ... ) and partly by the geology itself (for the longer times). The point is also that the longer one waits for a leak, the less severe are the consequences.

If we take out the plutonium (PUREX or another technique), then the last component is not part of the waste. This diminishes the necessary "containment time" from 100 000 years to 10 000 years. If we take out also the minor actinides, we arrive at a few hundred years.

The last point is important, because it is possible to design canisters that will last that long. However, it is difficult to design canisters (although the Swedish did so) that are supposed to remain intact for 10 000 years or longer.

But "taking out" is only part of the story: what do you do with it next ? With plutonium, that's easy: its a good fuel for fast reactors. So you burn it in a fast reactor. You can burn it partially in thermal reactors (MOX), but that's limited. You will always remain with a certain fraction of unusable plutonium that way.

The minor actinides are part of a discussion. You can burn them (in small amounts) in fast reactors, or you can build ADS systems which try to burn them on purpose. The discussion is whether this is worth the effort - we'll come to that.
Fast reactors don't produce minor actinides (or in very very very small quantities, that is). So this problem is purely a difficulty of thermal reactors.

Now, if you reprocess fuel, then you vitrify the essential waste (let's say, minor actinides and fission products), and these go into a stainless steel canister, which will go into a bigger "repository" canister. Around that, you put some filling material, mainly clay or concrete. Normally, that's it.

But people study what's going to happen if there is some ground water flow through the repository. The stainless steel will rust away over about 1000 years, but in doing so, it will generate iron oxide, which is a strong reducing agent. The glass will very slowly dissolve in the ground water, which will also take a few thousand years. At that point, the waste is now "free" but the fission products are gone by now. We only have the minor actinides to worry about. Turns out that minor actinides don't migrate easily through a reducing atmosphere, and the dissolved glass also forms chemical migration barriers. After that, the clay sorbs actinides very easily. So it will take several thousands of years for the actinides to even be able to migrate outside of the "human structure" in small quantities. After that, they are confronted with the actual geological barrier.

One studies several possible scenarios and tries to estimate what will be the release of radioactive material after several tens of thousands of years. By then, most has decayed to very low activity levels. From this results then the final potential "contamination" of the repository. It is usually orders of magnitude below the natural background radiation level.

EDIT:
In fact, in the case of (tiny) releases in the far future, it turns out that the culprit is mostly Sn-126, with a half life of more than 100 000 years. This is one of the few fission products which live longer than a few hundred years, are produced in very tiny quantities, and ARE able to migrate. But, as said, the doses they can deliver are orders of magnitude below background doses. The actinides never seem to be able to migrate out, as they are chemically bound so easily to the local material (clay or other).

It is this observation which makes one put a question mark on the utility of getting rid of minor actinides, as visibly it are not these which get out after a long time. With current knowledge, it wouldn't make any difference in most if not all scenarios whether or not the minor actinides were removed or not.

 

[Phy6explorer]

In fact, in the case of (tiny) releases in the far future, it turns out that the culprit is mostly Sn-126, with a half life of more than 100 000 years. This is one of the few fission products which live longer than a few hundred years, are produced in very tiny quantities, and ARE able to migrate. But, as said, the doses they can deliver are orders of magnitude below background doses. The actinides never seem to be able to migrate out, as they are chemically bound so easily to the local material (clay or other).

It is this observation which makes one put a question mark on the utility of getting rid of minor actinides, as visibly it are not these which get out after a long time. With current knowledge, it wouldn't make any difference in most if not all scenarios whether or not the minor actinides were removed or not.

I am not sure but, I think that the only way of producing lower yields of Tin-126 is using thermal reactors.But it seems that quite a few actinides are found naturally in the Earth. Then don't most actinides share some common components.Why can't they be dumped deep into the Earth, of course, in a depth which is safely above the water -table?

 

[vanesch]

But it seems that quite a few actinides are found naturally in the Earth. Then don't most actinides share some common components.Why can't they be dumped deep into the Earth, of course, in a depth which is safely above the water -table?

Deep means "below the water table". So deep repositories are necessarily wet, but this is not a bad thing. Because "below the water table" also means "saturated" and hence no oxygen: the environment is chemically reducing. This is a major contributor to stopping the migration of actinides and many other metal ions, which will not be soluble in a reducing water environment. The downside of the wet part is of course the slow corrosion of the stainless steel, and the solution of the glass matrix.

The only repository above the water table I know of is the Yucca Mountain, but it is not considered to be the best of repositories (although it might do its thing).

There is a totally different repository which has been studied, and that was: deep ocean sediments. A very complete study was performed outside of the coast of Hawaii. The idea was to have "penetrator" containers which sink to the bottom and penetrate a few tens of meters into the sediments (which are highly binding also for actinides). The soil there is geologically extremely stable, and the sediments will build up over the years, hence burying the waste deeper and deeper. An eventual leak in the far future would be diluted immediately in the ocean, and no ground water or any people would ever suffer from any increased exposure. Moreover, the technique is very cheap. All considerations of access to the waste for terrorist purposes or whatever are of course totally moot in this context.
However, the project was canceled for unclear political reasons.

 

[Azael]

Deep means "below the water table". So deep repositories are necessarily wet, but this is not a bad thing. Because "below the water table" also means "saturated" and hence no oxygen: the environment is chemically reducing. This is a major contributor to stopping the migration of actinides and many other metal ions, which will not be soluble in a reducing water environment. The downside of the wet part is of course the slow corrosion of the stainless steel, and the solution of the glass matrix.

The only repository above the water table I know of is the Yucca Mountain, but it is not considered to be the best of repositories (although it might do its thing).

There is a totally different repository which has been studied, and that was: deep ocean sediments. A very complete study was performed outside of the coast of Hawaii. The idea was to have "penetrator" containers which sink to the bottom and penetrate a few tens of meters into the sediments (which are highly binding also for actinides). The soil there is geologically extremely stable, and the sediments will build up over the years, hence burying the waste deeper and deeper. An eventual leak in the far future would be diluted immediately in the ocean, and no ground water or any people would ever suffer from any increased exposure. Moreover, the technique is very cheap. All considerations of access to the waste for terrorist purposes or whatever are of course totally moot in this context.
However, the project was canceled for unclear political reasons.

WIPP is also a dry repository. I honestly don't understand why americans even bother with Yucca mountain when they already have WIPP running?
http://en.wikipedia.org/wiki/WIPP

The main advantage with the deep ocean sediment deposits if I understood it right was also that the maximum diffusion rate of waste through the sediments was lower than the rate of new sediment deposits. So no matter what happens to the canisters the waste would never leak through the sedimentation. Vanesch would you happen to know where the research results of the deep ocean research has been published? I have been trying to find it ever since I read about it in Gwyneth Cravens book.

 

[vanesch]

WIPP is also a dry repository. I honestly don't understand why americans even bother with Yucca mountain when they already have WIPP running?
http://en.wikipedia.org/wiki/WIPP

Well, a salt repository is also a sub-watertable repository, only, the salt has absorbed all of the groundwater present. A similar project (but with much more political resistance) is under examination in Germany (the Gorleben salt dome). The presence of some water is actually necessary in order for the salt rock to tighten up.

But it is true - I heard several experts on this - that Yucca is not really geologically the best kind of solution for a repository. I saw a presentation by a geologist who said that it is ideal as a reversible long term storage facility.

I have been trying to find it ever since I read about it in Gwyneth Cravens book.

I also have this from that book
There is this: http://www.theatlantic.com/issues/96oct/seabed/seabed.htm
I don't know what it is worth. I'm interested too.

 

[Astronuc]

WIPP was primarily for high level waste for the DOE (weapons) program, and the idea of Yucca Mountain was to have a separate facility for civilian (commercial) spent fuel.

Also, the salt in WIPP apparently 'flows', and apparently at a faster rate than was first determined (IIRC).

 

[mheslep]

I don't understand that $14B.

I don't either. I would like to see the cost break out.

The new EPR, which is still in "pilot" stage (no mass production yet, so more expensive), is estimated to cost ~ 3.3B Euro
http://en.wikipedia.org/wiki/European_Pressurized_Reactor
(although the Finnish project will probably be more expensive, mainly for extra regulatory issues - red tape costs). It is a 1.6 GW e plant. So even with a serious cost overrun (say, 50%), and knowing it is still a pilot plant, $14B seems extremely expensive compared to this.

I am not clear what the nuclear / non-nuclear split means here:
http://energies.edf.com/122321i/Acc...-Flamanville-3/the-Flamanville-3-project.html

projected costs and funding
The Flamanville 3 project involves around 3.3 billion Euros of capital expenditure, including EPR development costs. The nuclear part of the facility accounts for 60% of the total amount with the conventional (non-nuclear) part accounting for the rest.

Yes, but *average* doesn't mean: "when needed" ; as such, average erratic power is less than half the price of "available".
Taking the average price equal to the total price can only be the case when wind is "within the noise margin" of power production, as it is in all countries but Denmark. So the marginal cost of wind is relatively low for the first 10-20% of installed wind power, and once it becomes an important component in the network, and starts causing problems because of the fluctuations, costs rise dramatically, because of the need of compensation.

As I said above wind has to be coupled with other forms of power such as pump storage (26GW in the US so far) and perhaps hydrogen from electrolysis and then to gas turbines*. With a widely distributed wind system I don't believe we're looking at that much backup required to have always available power.

*Or power plant scale fuel cells, looks like FCs will arrive there much sooner than vehicles.

 

[mheslep]

Well, I suspect it is pretty good on the facts, it's just the opinions that are suspect. The author seems to have a pretty strong left-wing bias that reflects in his work:
http://web.mit.edu/chemistry/deutch/policy.html

Simply put, the opinion parts of his position seem based on the obsolte/incorrect hippie view of nuclear power. I must admit, I've only read his summery so far - I'll have to read more of his analysis.

Care to specify? The Future of Nuclear Power study has many authors so its hard to credit him w/ specific views found in there. They favor nuclear power but with specific guidelines

I thought I had seen Deutch's name before; didn't realize it was the political/op-ed Deutch. In any case Deutch is just one of many authors of The Future...
Moniz, the co-chair doesn't have any easy to find political views and seems to focus more on energy policy.

PROFESSOR STEPHEN ANSOLABEHERE
Department of Political Science, MIT

PROFESSOR JOHN DEUTCH — CO CHAIR
Department of Chemistry, MIT

PROFESSOR EMERITUS MICHAEL DRISCOLL
Department of Nuclear Engineering, MIT

PROFESSOR PAUL E. GRAY
President Emeritus, MIT
Department of Electrical Engineering and Computer Science

PROFESSOR JOHN P. HOLDREN
Department of Earth and Planetary Sciences, Harvard University.

PROFESSOR PAUL L. JOSKOW
Department of Economics and Sloan School of Management, MIT

PROFESSOR RICHARD K. LESTER
Department of Nuclear Engineering, MIT

PROFESSOR ERNEST J. MONIZ — CO CHAIR
Department of Physics, MIT

PROFESSOR NEIL E. TODREAS
Department of Nuclear Engineering, MIT
Department of Mechanical Engineering, MIT

 

[Astronuc]

The $14 billion for two Westinghouse AP-1000's seems way too high. I would think it closer to $2-3 billion for the pair, and $3 billion seems awfully high for a transmission line, unless they have to obtain a new right-of-way.

EPR is still relatively new, so it is expected that the cost will be high, but it should decrease if more units are built.

With respect to nuclear/convention (or non-nuclear), those may refer to the primary system of which the nuclear reactor is part and the secondary side and balance of plant. Some of the biggest capital expense is for the reactor pressure vessel and steam generators, which are BIG forgings, and the reactor coolant pumps and high pressure piping.

 

[mheslep]

The $14 billion for two Westinghouse AP-1000's seems way too high. I would think it closer to $2-3 billion for the pair, and $3 billion seems awfully high for a transmission line, unless they have to obtain a new right-of-way.

EPR is still relatively new, so it is expected that the cost will be high, but it should decrease if more units are built.

With respect to nuclear/convention (or non-nuclear), those may refer to the primary system of which the nuclear reactor is part and the secondary side and balance of plant. Some of the biggest capital expense is for the reactor pressure vessel and steam generators, which are BIG forgings, and the reactor coolant pumps and high pressure piping.

All of the several new US reactor projects are coming in at very high costs - no less than $5 and up to $14B per today's WSJ. I'll post the details later ...

 

[mheslep]

Here it is, Monday 5/11 WSJ.

New Wave of Nuclear Plants Faces High Costs
http://online.wsj.com/article/SB121055252677483933.html

Byline Rebecca Smith, the journal's energy reporter.

A new generation of nuclear power plants is on the drawing boards in the U.S., but the projected cost is causing some sticker shock: $5 billion to $12 billion a plant, double to quadruple earlier rough estimates.
...
FPL Group, Juno Beach, Fla., estimates it will cost $6 billion to $9 billion to build each of two reactors at its Turkey Point nuclear site in southeast Florida. It has picked a reactor design by Westinghouse Electric Co., a unit of Toshiba Corp., after concluding it could cost as much as $12 billion to build plants with reactors designed by General Electric Co. The joint venture GE Hitachi Nuclear Energy said it hasn't seen FPL's calculations but is confident its units "are cost-competitive compared with other nuclear designs."
...
Exelon, the nation's biggest nuclear operator, is considering building two reactors on an undeveloped site in Texas, and said the cost could be $5 billion to $6.5 billion each.
...
The latest projections follow months of tough negotiations between utility companies and key suppliers, and suggest efforts to control costs are proving elusive. Estimates released in recent weeks by experienced nuclear operators -- NRG Energy Inc., Progress Energy Inc., Exelon Corp., Southern Co. and FPL Group Inc. -- "have blown by our highest estimate" of costs computed just eight months ago, said Jim Hempstead, a senior credit officer at Moody's Investors Service credit-rating agency in New York.

Just a two part explanation: inflation/dollar hits on materials; suppliers and expertise has faded:

Plants are being proposed in a period of skyrocketing costs for commodities such as cement, steel and copper; amid a growing shortage of skilled labor; and against the backdrop of a shrunken supplier network for the industry.

I had thought regulatory hurdles were partly the blame, but not so much:

The price escalation is sobering because the industry and regulators have worked hard to make development more efficient, in hopes of eliminating problems that in the past produced harrowing cost overruns. The Nuclear Regulatory Commission, for example, has created a streamlined licensing process to make timelier, more comprehensive decisions about proposals. Nuclear vendors have developed standardized designs for plants to reduce construction and operating costs. And utility executives, with years of operating experience behind them, are more astute buyers.

 

[mheslep]

Then the US CBO just released its report
"The Role of Nuclear Power in Generating Electricity"
http://www.cbo.gov/ftpdocs/91xx/doc9133/toc.htm

which contains estimates which contradict the actual costs coming in. From the Summary:

Cost of Construction. Historically, construction costs for nuclear facilities have been roughly double initial estimates. NRC’s revised licensing process for nuclear power plants is expected to reduce midconstruction modifications, which were blamed for many cost overruns in the past. Moreover, vendors argue that advanced reactors will have lower construction costs because they have fewer parts than older reactors. As a result, CBO’s base-case assumption for construction costs is about 25 percent lower than the historical average—a figure that reflects recent experience in the construction of advanced reactors in Japan. If those factors turned out not to reduce construction costs in the United States, nuclear capacity would probably be an unattractive investment even with EPAct incentives, unless substantial carbon dioxide charges were imposed.

So why is nuclear construction so much cheaper in Japan?

Details:
http://www.cbo.gov/ftpdocs/91xx/doc9133/Chapter2.5.1.shtml

CBO’s base-case assumptions include overnight costs of about $2.4 million for each megawatt of capacity for new nuclear plants and innovative coal plants but lower costs for conventional coal, conventional natural gas, and innovative natural gas technologies. For nuclear and innovative coal and natural gas technologies, the assumptions are intended to represent plants built over the next decade but do not incorporate the first-of-a-kind costs that are assumed to be covered by federal research and development programs. The estimate for nuclear plants, taken from the EIA’s most recent analysis, is roughly 10 percent above the estimate of overnight costs used in MIT’s study,which was published in 2003,

(That is the MIT study mentioned up thread.)

before construction costs for most types of power plants surged. CBO also calculated construction costs for each technology using alternative assumptions designed to capture plausible variations in those costs. For nuclear and innovative coal technologies, CBO considered construction costs ranging from about $1.2 million per megawatt of capacity to roughly $4.8 million per megawatt of capacity. The breadth of that range reflects the uncertainty associated with the cost of building new nuclear plants in the United States and is wide enough to capture plausible further increases in construction costs, which could affect conventional fossil-fuel plants as well.

Followed up by a caveat on the cost overages in the past:

CBO’s assumption about the cost of building new nuclear power plants in the United States is particularly uncertain because of the industry’s history of construction cost overruns. For the 75 nuclear power plants built in the United States between 1966 and 1986, the average actual cost of construction exceeded the initial estimates by over 200 percent (see Table 2-1). Although no new nuclear power plants were proposed after the partial core meltdown at Three Mile Island in 1979, utilities attempted to complete more than 40 nuclear power projects already under way. For those plants, construction cost overruns exceeded 250 percent.3 (An average of 12 years elapsed between the start of construction and the point at which the plants began commercial operation. The overruns in overnight costs did not include additional financing costs that were attributable to post-accident construction delays.)4

 

[mheslep]

Some of the biggest capital expense is for the reactor pressure vessel and steam generators, which are BIG forgings, and the reactor coolant pumps and high pressure piping.

That would explain some of it. The price of steel has gone up ~3x since 2000, but %40 of that was a spike in just this last month - April to May.
http://www.dot.state.oh.us/construction/oca/Steel_Index_for_PN525.htm

Of course that's going to impact any new construction, just nuclear more than most.

 

[vanesch]

As I said above wind has to be coupled with other forms of power such as pump storage (26GW in the US so far) and perhaps hydrogen from electrolysis and then to gas turbines*. With a widely distributed wind system I don't believe we're looking at that much backup required to have always available power.

*Or power plant scale fuel cells, looks like FCs will arrive there much sooner than vehicles.

My point is: the proof of the pudding is the eating. There are many non-nuclear power countries, or countries which have a high desire to get rid of their nuclear (Germany comes to mind). Their research in alternatives is not recent, but runs already for a few decades. Why haven't we seen these wonderful systems then ? Why are they, when the light threatens to go out, building brown coal plants ?

I'm manifestly pro-nuclear, but not because of some vested interest, but rather because I'm honestly convinced that in the coming 3 or 4 decades at least, they are the only large-scale alternative to coal and gas. I'm not against alternatives, I'm not against wind, at all. I would indeed prefer that wind and solar could provide us with abundant energy. A priori, I like that also more than nuclear. But when you sit down, and crunch a bit some numbers, the vision changes.

Pumping stations are great buffers, but they increase the price of the installation seriously, they need some specific geography, and moreover their capacity is limited. You cannot live on a pumping station full load for a week (or it has to be a very big one). In fact, if you want to provide an entire country with "pumping station capacity" for a week (which is what you need when you rely on wind), you have to imagine that you are going to provide your country with 100% hydro, except for the river input.

If the point is that "in an extended region, the wind always blows *somewhere*", then you need to install the FULL capacity of an entire country several times over: if the wind only blows in the north-east, then there has to be enough wind capacity in the north-east to provide the entire country with electricity. If the next day, it only blows in the south-west, then again, you need to have the full capacity of an entire country a second time in the south-west.

I don't believe that any fuel cell station that has a significant fraction of the capacity of a country will come online in the next few decades.

If it were that simple, it would already have been done on some scale. I'm all for experimenting - it would really be nice if this rosy picture could come true one day, but you cannot yet count on this as a large-scale policy, to drop proven techniques from the list.

Again, it was not the motivation that was missing in some European countries. But they ended up building coal plants again.

It is rather this observation which makes me pro-nuclear. Not vice versa.

 

[mheslep]

My point is: the proof of the pudding is the eating. There are many non-nuclear power countries, or countries which have a high desire to get rid of their nuclear (Germany comes to mind). Their research in alternatives is not recent, but runs already for a few decades. Why haven't we seen these wonderful systems then ? Why are they, when the light threatens to go out, building brown coal plants ?

The proof is in the recent evidence. Wind power has not been cost effective until the last few years with the creation of 30% capacity 1.5MW turbines. That is reflected in the large jumps in installed wind in just the last 2-3 years. The only power source growing anywhere close to as fast as wind is CCGT. The early, subsidized wind installations in the EU were done on hype.

I'm manifestly pro-nuclear, but not because of some vested interest, but rather because I'm honestly convinced that in the coming 3 or 4 decades at least, they are the only large-scale alternative to coal and gas. I'm not against alternatives, I'm not against wind, at all. I would indeed prefer that wind and solar could provide us with abundant energy. A priori, I like that also more than nuclear. But when you sit down, and crunch a bit some numbers, the vision changes.

Pumping stations are great buffers, but they increase the price of the installation seriously,

$1/watt installed here in 2000. All you need is the space and elevation. Still, agree, that has to be added in some backup percentage to the cost of wind (or other renewable) installation.

they need some specific geography, and moreover their capacity is limited. You cannot live on a pumping station full load for a week (or it has to be a very big one). In fact, if you want to provide an entire country with "pumping station capacity" for a week (which is what you need when you rely on wind), you have to imagine that you are going to provide your country with 100% hydro, except for the river input.

If the point is that "in an extended region, the wind always blows *somewhere*", then you need to install the FULL capacity of an entire country several times over: if the wind only blows in the north-east, then there has to be enough wind capacity in the north-east to provide the entire country with electricity. If the next day, it only blows in the south-west, then again, you need to have the full capacity of an entire country a second time in the south-west.

I don't think this is a correct analysis. Assuming for a moment transmission is unlimited, then if the wind is not blowing in 10% of a country, then you only need 10% backup, installed anywhere, and power can flow to the outage area. Then with realistic transmission limitations that backup percentage has to increase, but not I think to multiples of nationwide full capacity.

I don't believe that any fuel cell station that has a significant fraction of the capacity of a country will come online in the next few decades.

Solid Oxide FCs, $0.80/Watt, 49% efficient. 2006
A Practical Fuel-Cell Power Plant
http://www.technologyreview.com/read_article.aspx?id=17644&ch=energy&a=f
10MW demo plants scheduled to come online 2012, construction starts next year:
http://www.netl.doe.gov/publications/proceedings/07/SECA_Workshop/pdf/SECA%20Overview%20-%20Wayne%20Surdoval%2C%20NETL.pdf, slide 8.
As far as I can see the only hold up on these GE systems is the fuel supply, which at the moment is envisioned to be hydrocarbons - gasified coal - keeping that ~clean is a slowdown. If the fuel is electrolyzed clean H then it could start now.

In general I'm also pro nuclear, but I don't see a good crunching of the numbers on the nuclear side for an industry that plans to scale up by 3 or 4x. On looking at the numbers so far, I see nuclear not properly accounting for the risks of proliferation, and more recently its off by some large multipliers in construction costs, at least in the US. At these cost levels ( $6-8/Watt installed) even solar looks reasonable.

 

[vanesch]

The proof is in the recent evidence. Wind power has not been cost effective until the last few years with the creation of 30% capacity 1.5MW turbines. That is reflected in the large jumps in installed wind in just the last 2-3 years. The only power source growing anywhere close to as fast as wind is CCGT. The early, subsidized wind installations in the EU were done on hype.

We'll see. I had a look at a recent project in my own native country, which is a rather modest 300 MW installed / 100 MW estimated average delivery offshore wind farm to be constructed with 60 5 MW units:
http://www.c-power.be/applet_mernu_en/index01_en.htm
It is a to be realized project as of yet, so I would take it to be state of the art.

Price tag: 8 Euros per average delivered W (~ $13,- per average delivered watt): the project is projected to cost 800 million Euro. We don't have any cost overruns yet, as it is still in the project phase.

And one has to say that this farm is going to be placed on a very good, windy place.

You should compare "installed nuclear" with "average wind" because for nuclear, there is a utility factor of over 90%, while for wind, 30% is very optimistic and can only be reached on "good windy places".

If I compare that to an EPR unit, even assuming 50% cost overrun, 3.3 BEuro x 1.5 = 5 B Euro for a 1.6 GW unit, I find 3 Euro per installed (and delivered) Watt, and this adds flexibility to the grid.

That said, I'm all in favor for this kind of constructions. But I refuse to phase out a technology that has proven its utility for a technology that didn't prove - in a real world example, and not a projected scaleup - its capacity to do its thing.

As you correctly point out, it is also ridiculous to think that the world will switch to nuclear. My hope is that renewables and nuclear will both grind away on fossil fuel, and they will meet somewhere when all fossil fuel, or a large part of it, has been replaced by either. The relative fractions will then be settled by the relative capacity of both. I would dream to see 60 or 70 % renewable, but I think that it will stop somewhere around 20 or 30%.

 

[vanesch]

I don't think this is a correct analysis. Assuming for a moment transmission is unlimited, then if the wind is not blowing in 10% of a country, then you only need 10% backup, installed anywhere, and power can flow to the outage area. Then with realistic transmission limitations that backup percentage has to increase, but not I think to multiples of nationwide full capacity.

I was more thinking of a good, blocked anti-cyclone that gives you nice weather and no wind except for a morning and evening breeze for 4 or 5 days over a large portion of a medium-sized country...

 

[Phy6explorer]

What about finding a way to make the waste lose energy by making them emit radiation in the form of particles or electromagnetic waves,that is, do the process of radioactive decay by ourselves?

 

[malawi_glenn]

you mean speed up the process? The radioactive waste is undergoing that prodcess already...

Transmutation is working in that kinda way.

 

[Phy6explorer]

Yes, I am talking about speeding up artificial transmutation to keep pace with the amount of nuclear waste coming in. Natural transmutation is okay, but it takes lot of time and a lot of work to make deep geological repositories and stuff.

 

[malawi_glenn]

Yes, artificial transmutation is under investigation. By bombarding the waste with protons and neutrons, one is hoping for to realease their energy and taking them to a "safer" state. (i.e a shorter lived state)

http://www.neutron.kth.se/courses/transmutation/Chapter1.pdf
http://en.wikipedia.org/wiki/Nuclear_transmutation

 

[Phy6explorer]

I have already checked out the wikipedia link.The first link is very very useful.Thanks a lot!

 

[malawi_glenn]

I have already checked out the wikipedia link.The first link is very very useful.Thanks a lot!

oh, that's great!

It is from a university in the neighbouring town to mine ;)

 

[vanesch]

Yes, I am talking about speeding up artificial transmutation to keep pace with the amount of nuclear waste coming in. Natural transmutation is okay, but it takes lot of time and a lot of work to make deep geological repositories and stuff.

What is under study is the transmutation of minor actinides into fission products (in other words, use them as fuel). This is accomplished in a fast neutron spectrum, which can be obtained in a fast or breeder reactor, or in an ADS (accelerator driven system).

However, there are two caveats to this. Although it makes for great publicity to "reduce the radiotoxicity of the waste from 10000 years to 400 years", there are 2 points:

1) it is absolutely not sure that this can be fully done, with less risk than the extremely small risk these things represent in the future. You need to include extra transportations, extra activities, extra handling today, to avoid a small potential risk in the future and it is not sure you will win. It is also far from sure that you can do this with ALL of the actinides - you might reduce them but are you going to fission 99.9 % of them ?

2) Most repository analyses show that the actinides don't migrate well. So, the tiny dose that one might impose upon a hypothetical future generation by a repository is usually NOT due at all to the actinides.

So the question is: what do we objectively win by transmuting actinides ? And what does it cost us ?

I'm far from convinced that this is a win situation.

 

[vanesch]

Its probably because its almost impossible to produce u-233 without u-232 contamination and u-232 has a very nasty gamma daughter in its decay chain.
http://www.princeton.edu/~globsec/publications/pdf/9_1kang.pdf

Mmm, if I look at the numbers in that paper, and not at the textual arguments, I find U-233 by far the easiest (everything is relative) way to make a bomb.

 

[mheslep]

We'll see. I had a look at a recent project in my own native country, which is a rather modest 300 MW installed / 100 MW estimated average delivery offshore wind farm to be constructed with 60 5 MW units:
http://www.c-power.be/applet_mernu_en/index01_en.htm
It is a to be realized project as of yet, so I would take it to be state of the art.

Price tag: 8 Euros per average delivered W (~ $13,- per average delivered watt): the project is projected to cost 800 million Euro. We don't have any cost overruns yet, as it is still in the project phase.

Curious that a project is going ahead now w/ 5MW units - I had read that the sweet spot in turbine $/W was currently at 1.5MW. Those 5MW units are truly gigantic. The offshore component is apparent in the C-Power project as well. The 142 km of high voltage submarine cable is nearly the same cost as the turbines in Phase I.

And one has to say that this farm is going to be placed on a very good, windy place.

You should compare "installed nuclear" with "average wind" because for nuclear, there is a utility factor of over 90%, while for wind, 30% is very optimistic and can only be reached on "good windy places".

If I compare that to an EPR unit, even assuming 50% cost overrun, 3.3 BEuro x 1.5 = 5 B Euro for a 1.6 GW unit, I find 3 Euro per installed (and delivered) Watt, and this adds flexibility to the grid.

Well I'm suspicious that recent world price spike in materials, esp. steel, is making cost comparisons unreliable unless they are exact contemporaries. I suspect this imminent C-power wind project is reflecting these price hikes and thus only an imminent nuclear project (ala the $14B/2GW Florida plan) would be truly comparable.

 

[vanesch]

Well I'm suspicious that recent world price spike in materials, esp. steel, is making cost comparisons unreliable unless they are exact contemporaries. I suspect this imminent C-power wind project is reflecting these price hikes and thus only an imminent nuclear project (ala the $14B/2GW Florida plan) would be truly comparable.

Well, the Finnish EPR has just started construction, so that is the closest I can come up with (3.3 B Euro and estimated 50% cost overrun).

Actually, a related question is: how does the material bill (steel and concrete) compare between wind and nuclear ? I would think nuclear consumes more material, but is it gigantically more so ?

For the EPR reactor, the pressure vessel contains about 500 tons of steel, and I don't have the numbers for the steam generators, but compared to the N4 French plants, the steam generators are of the same order of magnitude as the pressure vessel (about 20% less, and there are 4 of them), so an estimate would be 2000 tons of steel for an EPR of 1.6 GW electric.

The double containment building, about 55 m high, 2 x 1.3 m thick, and diameter 48 m, must contain about:
3.14 x 48 x 55 x 2.6 = 22 000 m^3 of concrete for the wall, and
2 x 3.14 x 48^2 x 2.6 = 37 000 m^3 of concrete for roof and bottom, so total of 60 000 m^3 concrete.

Now how does that compare to about 1000 5 MW wind turbines (equal average power),
http://www.c-power.be/applet_mernu_en/index01_en.htm
each 120 m high (the tower), with wings 63 meter long each (total height 184 m, 2/3 of the Eiffel tower) ?

Do you do with so much less for one windmill than 2 tons of steel and 60 m^3 of concrete (half a m^3 per meter height) ?

It is my opinion - I can be wrong - that the material investment in steel and concrete of one EPR unit is comparable to the amount of steel and concrete needed for these 1000 turbines.

If that's the case, then material cost is going to have a null-effect on comparisons.

 

[baywax]

What is the attraction about CANDU reactors? As a Canadian I should know but I don't! Are they cheaper, produce less waste or what?

 

[vanesch]

What is the attraction about CANDU reactors? As a Canadian I should know but I don't! Are they cheaper, produce less waste or what?

CANDU reactors are heavy-water moderated and cooled natural uranium reactors.

The main advantage is that they can use natural uranium. As such, they are comparable to the Gen I graphite reactors, but they are a lot safer. They can be loaded and unloaded continuously (that can be seen as an advantage, or a disadvantage, because as such, they are ideal tools to make weapon grade plutonium, or do the thorium - U-233 conversion).

The disadvantage is that they need heavy water, which is expensive to produce (by distillation of ammoniak, I think), and there's another disadvantage connected to the heavy water: they produce loads of tritium (well, that too can be considered an advantage).

Another disadvantage is that they can only moderately "use up" the U-235 in the natural uranium (which can be compensated for by using slightly enriched U, so that the burn-up can be much higher, but then this undoes the main advantage of a CANDU).

 

[mheslep]

What is the attraction about CANDU reactors? As a Canadian I should know but I don't! Are they cheaper, produce less waste or what?

No enrichment required, they work (can) on natural uranium. Either CANDU or NRX reactors (Im not sure which, perhaps both) can then be used to make Pu and give one a path to a bomb and bypass a technologically difficult enrichment program.

 

[baywax]

CANDU reactors are heavy-water moderated and cooled natural uranium reactors.

The main advantage is that they can use natural uranium. As such, they are comparable to the Gen I graphite reactors, but they are a lot safer. They can be loaded and unloaded continuously (that can be seen as an advantage, or a disadvantage, because as such, they are ideal tools to make weapon grade plutonium, or do the thorium - U-233 conversion).

The disadvantage is that they need heavy water, which is expensive to produce (by distillation of ammoniak, I think), and there's another disadvantage connected to the heavy water: they produce loads of tritium (well, that too can be considered an advantage).

Another disadvantage is that they can only moderately "use up" the U-235 in the natural uranium (which can be compensated for by using slightly enriched U, so that the burn-up can be much higher, but then this undoes the main advantage of a CANDU).

Thank you. Theres another reactor in Ontario that was shut down for some reason to do with safety but then a global outcry because it was one of the only reactors that could produce medical grade imaging or treatment grade material. Then the watchdog that shut it down was fired by the Conservative Govt... (like Republicans) and the plant has been re-started... safety issues aside. (?)

 

[baywax]

No enrichment required, they work (can) on natural uranium. Either CANDU or NRX reactors (Im not sure which, perhaps both) can then be used to make Pu and give one a path to a bomb and bypass a technologically difficult enrichment program.

I think we've sold CANDUs to many different countries. Oops.

 

[mheslep]

...If the point is that "in an extended region, the wind always blows *somewhere*", then you need to install the FULL capacity of an entire country several times over:...

I don't believe that any fuel cell station that has a significant fraction of the capacity of a country will come online in the next few decades...

Here's the basic idea at an NREL test facility.
http://www.nrel.gov/hydrogen/proj_wind_hydrogen_animation.html
Wind runs straight to the grid when needed, when its peaking over load it generates H via electrolysis, is pressurized and stored. Fixed plant storage is not that problematic, unlike vehicle storage. Then when needed the H2 drives either and H2 ICE (in low production now) generators or fuel cells.

 

[mheslep]

Well, the Finnish EPR has just started construction, so that is the closest I can come up with (3.3 B Euro and estimated 50% cost overrun).

Yes those are the costs I see, however Olkiluoto started in 2005, they've long since contracted for all their materials.

Actually, a related question is: how does the material bill (steel and concrete) compare between wind and nuclear ? I would think nuclear consumes more material, but is it gigantically more so ?

For the EPR reactor, the pressure vessel contains about 500 tons of steel, and I don't have the numbers for the steam generators, but compared to the N4 French plants, the steam generators are of the same order of magnitude as the pressure vessel (about 20% less, and there are 4 of them), so an estimate would be 2000 tons of steel for an EPR of 1.6 GW electric.

The double containment building, about 55 m high, 2 x 1.3 m thick, and diameter 48 m, must contain about:
3.14 x 48 x 55 x 2.6 = 22 000 m^3 of concrete for the wall, and
2 x 3.14 x 48^2 x 2.6 = 37 000 m^3 of concrete for roof and bottom, so total of 60 000 m^3 concrete.

Plus rebar steel. Strong reinforced concrete about 5% steel by cross section: 60000m^3*0.05=3500m^3; 3500m^3 * 7900kg/m^3 * 907kg/ton = 26000tons rebar steel.

Is the cooling tower significant? Schmehausen, for example: average diameter ~120M, height 180M, 0.1M thick average(?)= 3.14*120*180*0.1 = 6800 m^3 walls
base = 3.15*60^2* 0.2(?) = 2300m^3, total 9000m^3 for the tower. Say 1000tons rebar steel. Not so much compared to that containment bldg. Call it 70,000M^3 concrete, 29000 tons steel, total project.

Now how does that compare to about 1000 5 MW wind turbines (equal average power),
http://www.c-power.be/applet_mernu_en/index01_en.htm
each 120 m high (the tower), with wings 63 meter long each (total height 184 m, 2/3 of the Eiffel tower) ?

The blades are fiber glass composite now. The tower is almost always steel, though concrete has some inroads. Weight? 1http://www.google.com/url?sa=t&ct=res&cd=8&url=http%3A%2F%2Fwww.mecal.nl%2Ffiles%2Falgemeen%2Fewec2003-ATS_paper.pdf&ei=QgwqSMf-Apys8ASW0PHGCw&usg=AFQjCNGXC0r4dfJh_7uSZ72bJ6y2ShyGYQ&sig2=sUhooGLwJ0sBCJ2WugGG4w" , so let's say 700tons/tower. So 700,000 tons for a 5GW 1000 tower field. The concrete I think is small in comparison.

Do you do with so much less for one windmill than 2 tons of steel and 60 m^3 of concrete (half a m^3 per meter height) ?

It is my opinion - I can be wrong - that the material investment in steel and concrete of one EPR unit is comparable to the amount of steel and concrete needed for these 1000 turbines.

If that's the case, then material cost is going to have a null-effect on comparisons.

Apparently Wind requires a lot of steel; I doubt very much concrete.

 

[vanesch]

Plus rebar steel. Strong reinforced concrete about 5% steel by cross section: 60000m^3*0.05=3500m^3; 3500m^3 * 7900kg/m^3 * 907kg/ton = 26000tons rebar steel.

Hey, that's funny, there's 10 times more steel in the building than in the actual reactor part.
One should make then out of fibre glass...

Is the cooling tower significant? Schmehausen, for example: average diameter ~120M, height 180M, 0.1M thick average(?)= 3.14*120*180*0.1 = 6800 m^3 walls
base = 3.15*60^2* 0.2(?) = 2300m^3, total 9000m^3 for the tower. Say 1000tons rebar steel. Not so much compared to that containment bldg. Call it 70,000M^3 concrete, 29000 tons steel, total project.

Right, which gives us an equivalent budget of 29 tons of steel and 70 m^3 of concrete per turbine, or essentially 70 m^3 of reenforced concrete.

The blades are fiber glass composite now. The tower is almost always steel, though concrete has some inroads. Weight? 1http://www.google.com/url?sa=t&ct=res&cd=8&url=http%3A%2F%2Fwww.mecal.nl%2Ffiles%2Falgemeen%2Fewec2003-ATS_paper.pdf&ei=QgwqSMf-Apys8ASW0PHGCw&usg=AFQjCNGXC0r4dfJh_7uSZ72bJ6y2ShyGYQ&sig2=sUhooGLwJ0sBCJ2WugGG4w" , so let's say 700tons/tower. So 700,000 tons for a 5GW 1000 tower field. The concrete I think is small in comparison.

Apparently Wind requires a lot of steel; I doubt very much concrete.

For these offshore applications, I'm pretty sure that the base on the seafloor requires quite some concrete, but I don't know how it compares to 70 m^3 which would make a base plate of 10 m x 10 m x 70 cm.

 

[baywax]

Here's the basic idea at an NREL test facility.
http://www.nrel.gov/hydrogen/proj_wind_hydrogen_animation.html
Wind runs straight to the grid when needed, when its peaking over load it generates H via electrolysis, is pressurized and stored. Fixed plant storage is not that problematic, unlike vehicle storage. Then when needed the H2 drives either and H2 ICE (in low production now) generators or fuel cells.

Another source of presently wasted H2 is at electrolysis plants around the world. They just burn off the excess H2 instead of storing it. What does it take to collect the H2 rather than burn it off? Probably the same amount of equipment. There must be many other chemical processes going on in industry today that discharge hydrogen rather than store it for re-sale. This represents work that has been done anyway so the efficiency goes up 100%... and you get two or more outcomes for the price of one.

 

[vanesch]

Here's the basic idea at an NREL test facility.
http://www.nrel.gov/hydrogen/proj_wind_hydrogen_animation.html
Wind runs straight to the grid when needed, when its peaking over load it generates H via electrolysis, is pressurized and stored. Fixed plant storage is not that problematic, unlike vehicle storage. Then when needed the H2 drives either and H2 ICE (in low production now) generators or fuel cells.

Sure, all this is nice. This test facility (that's why it is a test facility) is on the 10 KW scale. IN the US we need to go to a 10 million times bigger scale. This is not going to happen in the next few decades, that's the point.
Also, I wonder what the efficiency is of "generated electricity" - "generated hydrogen" - "re-generated electricity". I wonder if you get overall over 30% (especially if the hydrogen is used in a combustion engine), which means that you need 3 times the capacity to account for the variability.

Again, I'm not against this, on the contrary. But these are experiments on a scale where nuclear power was in the 40ies. It took at least 4 decades before this became a major player in the world energy provision.

 

[Astronuc]

No enrichment required, they work (can) on natural uranium. Either CANDU or NRX reactors (Im not sure which, perhaps both) can then be used to make Pu and give one a path to a bomb and bypass a technologically difficult enrichment program.

With respect to CANDU fuel, AECL has been offering slightly enriched fuel (CANFLEX) for some time.

http://www.aecl.ca/Commercial/Services/Expertise/CANDU-Fuel.htm

NRX was a research reactor and not appropriate for power reactor, although certainly one could breed fissile isotopes. It is now being decommissioned.

Production of fissile materials requires processing of the converted material, which requires chemical processing.

 

[vanesch]

NRX was a research reactor and not appropriate for power reactor, although certainly one could breed fissile isotopes. It is now being decommissioned.

Production of fissile materials requires processing of the converted material, which requires chemical processing.

Well, there's a link of course. When you look at the Wiki entry (I don't know if it is correct), India got his nuclear weapons (in 1974 already) from a CANDU-style reactor:

http://en.wikipedia.org/wiki/Nuclear_proliferation

It is widely believed that the nuclear programs of India and Pakistan used CANDU reactors to produce fissionable materials for their weapons; however, this is not accurate. Both Canada (by supplying the 40 MW research reactor) and the United States (by supplying 21 tons of heavy water) supplied India with the technology necessary to create a nuclear weapons program, dubbed CIRUS (Canada-India Reactor, United States). Canada sold India the reactor on the condition that the reactor and any by-products would be "employed for peaceful purposes only.". Similarly, the U.S. sold New Delhi heavy water for use in the reactor "only... in connection with research into and the use of atomic energy for peaceful purposes". India, in violation of these agreements, used the Canadian-supplied reactor and American-supplied heavy water to produce plutonium for their first nuclear explosion, Smiling Buddha.[16] The Indian government controversially justified this, however, by claiming that Smiling Buddha was a "peaceful nuclear explosion."

 

[mheslep]

Sure, all this is nice. This test facility (that's why it is a test facility) is on the 10 KW scale. IN the US we need to go to a 10 million times bigger scale. This is not going to happen in the next few decades, that's the point.
Also, I wonder what the efficiency is of "generated electricity" - "generated hydrogen" - "re-generated electricity". I wonder if you get overall over 30% (especially if the hydrogen is used in a combustion engine), which means that you need 3 times the capacity to account for the variability.

Yes there are some significant losses. Electrolysis requires 1.4 electric joules to make 1 joule of H2 (71%), and the fuel cell is 50% unless the FC heat is reused in a combined cycle, absent that about 35% total as you guessed. Is this a problem? It depends on the outage ratio (not the turbine average capacity). If it is 1 week a year as posited some threads ago, then the wind system needs to be plus rated (1/51)/.3 = 6.5% to supply the needed storage energy over it's on time, so producing the back up H2 is no problem. A more complicated problem is that the FC or H2 ICE has to have the same power capability as the downed percentage of wind. That also gets back to the transmission and weather scenario guess work which I don't enough about.

Again, I'm not against this, on the contrary. But these are experiments on a scale where nuclear power was in the 40ies. It took at least 4 decades before this became a major player in the world energy provision.

A big part of that time must be credited to safety concerns, political games, major development of reactor technology, complexities of plant operation, and the development of very evolved government regulatory bodies (e.g. NRC, necessary IMO); the nuclear history doesn't make the argument that any new technology must take 40 years to roll out. The internet? 5-10 years. Cell telephones? 5-10 years. I don't believe nuclear development parallels must necessarily be drawn to Wind; solar perhaps needs a couple more generations (ala reactors) but not wind.

 

[vanesch]

Yes there are some significant losses. Electrolysis requires 1.4 electric joules to make 1 joule of H2 (71%), and the fuel cell is 50% unless the FC heat is reused in a combined cycle, absent that about 35% total as you guessed. Is this a problem? It depends on the outage ratio (not the turbine average capacity). If it is 1 week a year as posited some threads ago, then the wind system needs to be plus rated (1/51)/.3 = 6.5% to supply the needed storage energy over it's on time, so producing the back up H2 is no problem. A more complicated problem is that the FC or H2 ICE has to have the same power capability as the downed percentage of wind. That also gets back to the transmission and weather scenario guess work which I don't enough about.

The Belgian project I referred to earlier (and is placed on one of the better spots in the world) http://www.c-power.be/applet_mernu_en/welcome/presentatie2/presentatie2.html
tells me that about 46% of the time, the unit is below half of its installed power, and 20% of the time below 1/5 of its installed power (which means it is below its average of 1/3 of installed power - so at that point, one needs an intervention from the backup - 4% of the time, it is totally dead).
The problem is that this simulation doesn't give us a distribution of the consecutive times when this happens, but as I said, typical anti-cyclone situations take 4-5 days.

 

[mheslep]

...Right, which gives us an equivalent budget of 29 tons of steel and 70 m^3 of concrete per turbine, or essentially 70 m^3 of reenforced concrete.

For these offshore applications, I'm pretty sure that the base on the seafloor requires quite some concrete, but I don't know how it compares to 70 m^3 which would make a base plate of 10 m x 10 m x 70 cm.

Hmm apparently at least three foundation techniques in use/planned. One for onshore and two for off shore. Looks like a 6.8M^3 concrete base for the onshore. Offshore: phase one, zero concrete, 'monopole' towers are just pile driven (more steel); phase two uses flared prefab concrete bases w/ excavation and then the base is filled, probably <10M^3 for the base.
http://www.c-power.be/applet_mernu_en/index01_en.htm

In sum the structural support materials cost for wind is going to be all in the steel, concrete relatively nil.