Integral Fast Reactor: Why Did Funding Stop?

[vanesch]

Interesting breakdown page 15: 200,000 cubic meters concrete, 10,000 tons steel per reactor. Some time ago Vanesch and I estimated 70,000 cubic meters concrete, 29,000 tons steel for the EPR, in a comparison of materials costs between wind and nuclear for equivalent power (wind requires a lot more steel than nuclear). We only estimated the reactor plant, so I can see how we were light on concrete, but I don't see how we could have been heavy on the steel for the pressure vessel and containment building.
https://www.physicsforums.com/showpost.php?p=1729108&postcount=242

I think that in fact not all concrete has rebar steel in it. I have to say that I'm more puzzled by the large amount of concrete here: where do they put it ?? I wonder if they didn't confuse cubic meters and tonnes...

BTW, there is a funny typo in that document elsewhere where they say that the AP-1000 (a PWR) can run on enriched or natural uranium

 

[Astronuc]

I think that in fact not all concrete has rebar steel in it. I have to say that I'm more puzzled by the large amount of concrete here: where do they put it ?? I wonder if they didn't confuse cubic meters and tonnes...

The concrete supports and fills the volume between the rebar. The EPR has a double containment. The external structure will have a higher density of rebar, and internal structures less so. The external structures much survive impact loads, while internal structures provide support primarily, and both must be designed for appropriate seismic loads.

BTW, there is a funny typo in that document elsewhere where they say that the AP-1000 (a PWR) can run on enriched or natural uranium

Presumably, the AP-1000 fueled with natural U would be a PHWR. That's feasible.


The actual cost of the plant will fluctuate with the costs of materials and labor.

The economics of Russia and China are different from the US, Canada and Europe.


One thing to consider in an IFR plant is the reprocessing facility, which is not found at a conventional NPP. Back in the early days of the commercial nuclear industry, it was invisioned that typical LWRs would operate on annual cycles with reloads of about 1/3 of the core (some used reloads of 1/4 core) - so an entire core's worth of fuel would be used every 3 to 4 years - and the fuel would be reprocessed so that MOX fuel would be used. Alternatively, the fuel could be reprocessed and the MOX fuel used in a fast reactor, and possibly the Pu produced in the breeder would be used in MOX fuel in LWRs. Well all that changed - reprocessing was abandoned and all that spent fuel began accumulating.

To reduce usage, plants have gone to longer cycles 18-24 mo, and many now reload about 45-50% of the core. Burnups are on the order of 4-5% FIMA (40-50 GWd/tU) on a batch average basis, with local burnups pushing 6-6.5% FIMA (~60-65 GWd/tU).

 

[mheslep]

The South African reactors projects in play with either Areva or Westinghouse are dead for now.

JOHANNESBURG -- South Africa's state power company, which has been forced to ration electricity to mines and smelters, Friday shelved plans to build the country's second nuclear power station, saying it can't afford to make the investment...
The utility had been expected to make a decision by the end of the year between two 1,650-megawatt reactors proposed by a consortium led by French nuclear-engineering giant Areva SA and three 1,140-megawatt reactors to be built by a group led by Toshiba Corp.'s Westinghouse.

http://online.wsj.com/articl/SB122868998183686411.html

and the Flamanville EPR project announced its price going up 20%, delayed a year.
http://www.reuters.com/article/rbssIndustryMaterialsUtilitiesNews/idUSB29884120081202

 

[Astronuc]

South Africa's state power company, which has been forced to ration electricity to mines and smelters,

Interesting. This evening, I was talking with a former classmate, who is involved with independent power supply systems. We discussed the potential in Africa, and he mentioned that various mines and plants in S. Africa are royal pissed at ESKOM. Apparently the companies are looking a building their own independent power supplies.

Both EPRs, TVO-3 and now Flamanville, are delayed. NPPs are big capital projects, and the new ones are supposed to take about 5 yrs (60 months) to build.

 

[RobertW]

Good point. If you were to use solar electricity to produce hydrogen by electrolysis (for use in cars, say) an average supply works just fine.


The capital cost of the IFR must be huge on a per watt basis. A 1 GW conventional nuclear plant will run about $4 billion. which is $4 per Watt. An IFR would be at least double, maybe triple that so $8 - $12 billion not including development costs. While an IFR if very efficient, it does use fuel and has a significant operating cost. Accepting that my figures for cost may be out by a factor of 6, that puts solar at today's prices about $12/watt. So it I appears to me that solar would be competitive. A big advantage of solar would be the speed at which it could be implemented.


AM

The estimated cost for a one-GW(e) IFR, if they are mass produced, is $1.5 to $2.0 billion. IFRs do not require huge cooling water reservoirs or cooling towers. A modest water supply would be required if steam is to be the operating fluid. If supercritical gas is used as the operating fluid, then water would only be required for sanitation and pyroprocessing purposes. An IFR is fueled only once - when it is built. An IFR creates only 1,700 POUNDS of waste per year that will be "safe in 300 to 400 years as opposed to 17,000 TONS of waste per year for a PWR or LWR that will be safe in 10,000 to 200,000 years. The LWR and PWR must be refuled every three to five years - a very costly maintenance item; no refueling required for the IFR. Fuel for the LWR and PWR must be trucked to the reactor and spent fuel trucked from the reactor. IFR fuel is reprocessed on site - a major safety consideration.
The IFR is 99.5% efficient; the PWR and LWR are typically 3% to 5% efficient.
You can calculate, manipulate, cogitate, and rationalize all you want, there currently is no cheaper, safer, more efficient way to reliably meet all energy needs of the U.S. on a 24/7 basis than the IFR. Suggest reading the whole thread.

 

[vanesch]

The concrete supports and fills the volume between the rebar. The EPR has a double containment. The external structure will have a higher density of rebar, and internal structures less so. The external structures much survive impact loads, while internal structures provide support primarily, and both must be designed for appropriate seismic loads.

Yes, but the calculation mheslep and I did was a geometry calculation of the primary grade kind: the volume of the double containment building walls, simplified as cylinders, with wall thickness of 1.3 meters (times two, for the two buildings).

I will retake it here (I even think we made a mistake back then):

cylinder of 55 meters high, 48 meters diameter:

Surface of wall: pi x 48 m x 55 m = 8289 m^2

Surface of bottom = surface of roof: pi * (48m/2)^2 = 1808 m^2

Total surface = 8289 m^2 + 2 x 1808 m^2 = 11907 m^2

That, times the double thickness of 2.6 m gives us a total wall volume of:
30957 m^3

So 31 000 m^3.

How do they come at 200 000 m^3 of concrete, which is almost 7 times more ??

 

[vanesch]

Presumably, the AP-1000 fueled with natural U would be a PHWR. That's feasible.

I think it was more a kind of typo, as they said that the CANDU needs to work on enriched U...

 

[gmax137]

Yes, but the calculation mheslep and I did was a geometry calculation of the primary grade kind: the volume of the double containment building walls, simplified as cylinders, with wall thickness of 1.3 meters (times two, for the two buildings).

I will retake it here (I even think we made a mistake back then):

cylinder of 55 meters high, 48 meters diameter:

Surface of wall: pi x 48 m x 55 m = 8289 m^2

Surface of bottom = surface of roof: pi * (48m/2)^2 = 1808 m^2

Total surface = 8289 m^2 + 2 x 1808 m^2 = 11907 m^2

That, times the double thickness of 2.6 m gives us a total wall volume of:
30957 m^3

So 31 000 m^3.

How do they come at 200 000 m^3 of concrete, which is almost 7 times more ??

I don't know the details, but the source says 200,000 m3 concrete for a two unit plant. Your calc of the containment concrete neglects any internal walls & floors within the containment; perhaps more importantly it neglects the auxiliary building, the tubine building, the intake structures and or cooling towers, the switchyard, the maintenence and admin buildings, parking lots, security structures etc etc...

 

[Andrew Mason]

I think it was more a kind of typo, as they said that the CANDU needs to work on enriched U...

Not a typo. This Candu is designed to use lightly enriched uranium.

AM

 

[mheslep]

The estimated cost for a one-GW(e) IFR, if they are mass produced, is $1.5 to $2.0 billion. ...

Source?

 

[mheslep]

Yes, but the calculation mheslep and I did was a geometry calculation of the primary grade kind: the volume of the double containment building walls, simplified as cylinders, with wall thickness of 1.3 meters (times two, for the two buildings).

I will retake it here (I even think we made a mistake back then):

cylinder of 55 meters high, 48 meters diameter:

Surface of wall: pi x 48 m x 55 m = 8289 m^2

Surface of bottom = surface of roof: pi * (48m/2)^2 = 1808 m^2

Total surface = 8289 m^2 + 2 x 1808 m^2 = 11907 m^2

That, times the double thickness of 2.6 m gives us a total wall volume of:
30957 m^3

So 31 000 m^3.

How do they come at 200 000 m^3 of concrete, which is almost 7 times more ??

I assumed balance of plant might explain that - large areas of foundation and slab, though that's all relatively low strength concrete and doesn't require much steel reinforcement or rigorous inspection and regulation.

 

[vanesch]

I don't know the details, but the source says 200,000 m3 concrete for a two unit plant. Your calc of the containment concrete neglects any internal walls & floors within the containment; perhaps more importantly it neglects the auxiliary building, the tubine building, the intake structures and or cooling towers, the switchyard, the maintenence and admin buildings, parking lots, security structures etc etc...

Well, what makes the confinement building so important is the wall thickness. A cooling tower, although it is often bigger than the reactor building, uses in fact much less concrete. The largest cooling tower in the world (for a coal fired plant in Germany) is 200m high and 100 m diameter, with a wall thickness of about 0.2 m. If we make the approximation of a cylinder, then this corresponds to 200m x 3.14 x 100 m x 0.2 m = 12 560 m^3. And that's the largest one that exists. So the largest cooling tower in the world uses about 3 times less concrete than the EPR confinement building.

 

[mheslep]

I don't know the details, but the source says 200,000 m3 concrete for a two unit plant. ...

The BrucePower source says 400,000 cubic meters of concrete for a two unit plant; I optimistically called it 200k for one.

 

[mheslep]

This Flamanville construction picture gives some idea of the scale of the containment structure at its base - massive.
http://newsimg.bbc.co.uk/media/images/44516000/jpg/_44516043_nuclearflamanvilleafpg203.jpg

 

[RobertW]

Source?

D.O.E. - about 14 years estimated the cost of a sodium-cooled 1GW(e) IFR to be $985 million. Add cost of pyroprocessing cell and inflation = about $1.5 to $2 billion. Assumes IFRs are built in quantity like tract houses. Under this assumption, some of the reactor structure can be pre-cast and trucked to the construction site. Google the "STAR" and "STAR-LM" reactors for more cost info on this type of reactor construction.

 

[mheslep]

D.O.E. - about 14 years estimated the cost of a sodium-cooled 1GW(e) IFR to be $985 million. ...

An estimate that precise must be linkable somewhere at DOE? Googling at DOE gives me nothing.

 

[RobertW]

An estimate that precise must be linkable somewhere at DOE? Googling at DOE gives me nothing.

I obtained the following by Googling "STAR-LM REACTOR:
Supercritical CO2Brayton Cycle Control Strategy for Autonomous Liquid Metal-Cooled Reactors
http://www.osti.gov/bridge/servlets/purl/840371-mR3VlE/native/840371.pdf

STAR-LM Concept
http://www.ne.anl.gov/research/ardt/hlmr/index.html

POWER OPTIMIZATION IN THE STAR-LM MODULAR NATURAL CONVECTION REACTOR SYSTEM
http://www.ipd.anl.gov/anlpubs/2002/02/42316.pdf

Supercritical Steam Cycle for Lead Cooled Nuclear Systems
http://nuklear-server.ka.fzk.de/OFMS/Web%2FMain%2FPublications%2F2005%2FGLOBAL2005%2FP_C.Boehm_GLOBAL2005.pdf

Heavy Metal – Cooled Reactors: Pros and Cons
http://nucleartimes.jrc.nl/Doc/Global03final2.pdf

These will get you started. There are at least 100 papers and articles on the Secure, transportable, autonomous reactor (STAR). A STAR-LM is a liquid metal cooled fast reactor. If one were to triple or quadruple the size of of the STAR-LM and integrate a pyroprocessing cell with the reactor, one would have an IFR. There are numerous references at the end of the papers that can lead you to cost info done by someone. I don't have time to search for specific cost estimates. But you can be certain that the construction costs will be a fraction of the costs you are estimating for PWR, LWR, or PBMR reactors and the fuel costs will be about 2% or 3% of these older designs. The liquid metal ones are 99.5% efficient if the fuel is reprocessed in a co-located pyroprocessing cell. Also, the design life of the IFR can be extended to 60 years.

 

[Astronuc]

But you can be certain that the construction costs will be a fraction of the costs you are estimating for PWR, LWR, or PBMR reactors and the fuel costs will be about 2% or 3% of these older designs. The liquid metal ones are 99.5% efficient if the fuel is reprocessed in a co-located pyroprocessing cell. Also, the design life of the IFR can be extended to 60 years.

Um - no! Based on industry and personal experience.

 

[mheslep]

I obtained the following by Googling "STAR-LM REACTOR:
Supercritical CO2Brayton Cycle Control Strategy for Autonomous Liquid Metal-Cooled Reactors
http://www.osti.gov/bridge/servlets/purl/840371-mR3VlE/native/840371.pdf

STAR-LM Concept
http://www.ne.anl.gov/research/ardt/hlmr/index.html

POWER OPTIMIZATION IN THE STAR-LM MODULAR NATURAL CONVECTION REACTOR SYSTEM
http://www.ipd.anl.gov/anlpubs/2002/02/42316.pdf

Supercritical Steam Cycle for Lead Cooled Nuclear Systems
http://nuklear-server.ka.fzk.de/OFMS/Web%2FMain%2FPublications%2F2005%2FGLOBAL2005%2FP_C.Boehm_GLOBAL2005.pdf

Heavy Metal – Cooled Reactors: Pros and Cons
http://nucleartimes.jrc.nl/Doc/Global03final2.pdf

These will get you started. There are at least 100 papers and articles on the Secure, transportable, autonomous reactor (STAR). A STAR-LM is a liquid metal cooled fast reactor. If one were to triple or quadruple the size of of the STAR-LM and integrate a pyroprocessing cell with the reactor, one would have an IFR. There are numerous references at the end of the papers that can lead you to cost info done by someone. I don't have time to search for specific cost estimates.

Neither do I, and all of the links above appear to be architecture oriented and not informative regards cost; the reference papers are not are generally not publicly available. Therefore,

But you can be certain that the construction costs will be a fraction of the costs you are estimating for PWR, LWR, or PBMR reactors and the fuel costs will be about 2% or 3% of these older designs. ...

I am not certain of any cost information presented here.

 

[RobertW]

Um - no! Based on industry and personal experience.

Why not?

 

[RobertW]

Neither do I, and all of the links above appear to be architecture oriented and not informative regards cost; the reference papers are not are generally not publicly available. Therefore,
I am not certain of any cost information presented here.

Here is the URL - p. 83. This is where I found the cost estimates. The document was published in 2002.
http://gif.inel.gov/roadmap/pdfs/gen_iv_roadmap.pdf
Quite honestly, you cannot be certain about any of the info you find here. The only way you can be certain what it will cost to build an IFR is to build one - any other approach IS uncertain.

 

[mheslep]

...Heavy Metal – Cooled Reactors: Pros and Cons
http://nucleartimes.jrc.nl/Doc/Global03final2.pdf

In this source, some capital estimates are given (in reference to another source table I): $661.5/kW for a liquid metal reactor (SVBR), however it also cites in comparison $749.8/kW as the capital costs for a same size traditional PWR (VVER) and we know that is ridiculously low for a total cost, with Olkiluoto at $3,000/kW and the AP1000's in Florida quoted at $6,000/kW.
http://www2.tbo.com/content/2008/jan/15/bz-nuclear-costs-explode/
http://www.economist.com/business/displaystory.cfm?story_id=12724850

 

[RobertW]

Um - no! Based on industry and personal experience.

You say no, and these folk say yes - who is right?

http://skirsch.com/politics/globalwarming/ifrBerkeley.htm

An Introduction to Argonne National Laboratory's INTEGRAL FAST REACTOR (IFR) PROGRAM

 

[mheslep]

RW you stated

...But you can be certain that the construction costs will be a fraction of the costs you are estimating for PWR, LWR, or PBMR reactors and the fuel costs will be about 2% or 3% of these older designs. The liquid metal ones are 99.5% efficient if the fuel is reprocessed in a co-located pyroprocessing cell. Also, the design life of the IFR can be extended to 60 years.

- and Astronuc replied no, based on personal experience. In response you give this Argonne article:

You say no, and these folk say yes - who is right?

http://skirsch.com/politics/globalwarming/ifrBerkeley.htm

An Introduction to Argonne National Laboratory's INTEGRAL FAST REACTOR (IFR) PROGRAM

and though it suggests IFRs would be more economic than existing reactors, it says very little to none at all to support the assertions you made above about 'fractional' and '2 or 3%' fuel costs.

 

[Astronuc]

Why not?

Because if one sizes and IFR to 1 GWe, it's going to take about the same amount of material (and labor) as a 1 GWe LWR. One still needs a containment structure - and one is going to throw in steam generators and a turbine/generator set - and cooling systems because one still uses a steam (Rankine) cycle.

On top of that, one will include a pyroprocessing center and fuel fabrication site if one is planning to reuse the spent fuel. That is not a trivial matter, especially if one is recycling metric ton levels. What has been done at INL is not anywhere near a commercial scale.

Also bear in mind that EBR-II was a small core/unit (~62 MWt/20 MWe) and FFTF was 400 MWt. Scaling to a ~3000 MWt IFR (~1 GWe) is not trivial.

I've seen one estimate that FFTF would cost $2-4 billion to build today, and that's probably an underestimate. FPL did an estimate earlier this year that put a twin EPR unit plant at about $12-14 billion. I think it was reported in the WSJ.

It’s the Economics, Stupid: Nuclear Power’s Bogeyman
http://blogs.wsj.com/environmentalc...the-economics-stupid-nuclear-powers-bogeyman/

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. Part of the cost escalation is bad luck. 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.

Fast reactors require specialty steels and those are alloys are quite expensive, not to mention that they have not been fabricated in the high tonnage quantities for large plants.

And given the materials problems (failures) I've seen (and been involved with) in the nuclear industry over the last 20+ years, I imagine there's still a lot of R&D to do on advanced reactor concepts.

Any IFR plant will have to get into the pipeline with all the other LWRs already ahead, so they are not going to pop up very quickly - if at all.

 

[mheslep]

...It’s the Economics, Stupid: Nuclear Power’s Bogeyman
http://blogs.wsj.com/environmentalc...the-economics-stupid-nuclear-powers-bogeyman/...

Thanks for this link. I saw the article when it came out, but missed the very good CBO article referenced therein.
http://www.cbo.gov/ftpdocs/91xx/doc9133/Chapter2.5.1.shtml#1090614

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

The worst case period before 3 Mi Island was 1974 to 1975, with 14 plants underway. Industry average estimate: $1,263/kW, actual $4,817/kW (281%)

 

[Astronuc]

Here's a better link to the CBO report

Nuclear Power's Role in Generating Electricity
http://www.cbo.gov/doc.cfm?index=9133

One can download a pdf as opposed to the html. If one knows the 4-digit index number, one can use the doc.cfm?index to find the download page for a report.

Besides the increased cost of steel and concrete (which cost a little more for NPPs because they must be nuclear grade and certified to higher quality standards than non-nuclear facilities), there is a shortage of qualified craftsmen who can work at NPP's.


The cost overruns on the plants being construct during the late 70's and the early 80's reflect the redesigns and modifications that were imposed as a result of the fire at Browns Ferry 1 (March, 1975) and the TMI-2 failure (1979). The NSSS vendors, AE and construction contractors had to fix design deficiencies that had contributed to both accidents. And then there were big screw ups at several of the plants on top of that. And for some plants, intervenors caused delays which ran up the legal bills and interest payments. On the other hand, the intervenors wouldn't have had much of case if the industry hadn't been so sloppy.

The industry is a lot better than it used to be, and some utilities/AE's are much better than others. Still, I imagine that in some new plants, there will be costly screw ups.

 

[RobertW]

RW you stated - and Astronuc replied no, based on personal experience. In response you give this Argonne article:

and though it suggests IFRs would be more economic than existing reactors, it says very little to none at all to support the assertions you made above about 'fractional' and '2 or 3%' fuel costs.

Let's be clear about the cost of nuclear fuel - the real costs of an open fuel cycle. The real costs of an open nuclear fuel cycle include the mining and processing of uranium, the conversion of uranium into reactor fuel, the transportation costs of the fuel to and spent fuel from reactor sites, the reprocessing costs if the fuel is MOXed, the encapsulation of unuseable nuclear waste, and the storage of unuseable nuclear waste for 10,000 years or more. Now, how many of these costs could be eliminated or nearly eliminated by using an IFR with a closed fuel cycle? I can't prove my estimate is correct and you can't prove it is wrong - neither of us have sufficient emperical data. However, the IFR closed fuel cycle, from a simple economics point of view, should greatly reduce the real costs of nuclear fuel.

 

[RobertW]

Because if one sizes and IFR to 1 GWe, it's going to take about the same amount of material (and labor) as a 1 GWe LWR. One still needs a containment structure - and one is going to throw in steam generators and a turbine/generator set - and cooling systems because one still uses a steam (Rankine) cycle.

On top of that, one will include a pyroprocessing center and fuel fabrication site if one is planning to reuse the spent fuel. That is not a trivial matter, especially if one is recycling metric ton levels. What has been done at INL is not anywhere near a commercial scale.

Also bear in mind that EBR-II was a small core/unit (~62 MWt/20 MWe) and FFTF was 400 MWt. Scaling to a ~3000 MWt IFR (~1 GWe) is not trivial.

I've seen one estimate that FFTF would cost $2-4 billion to build today, and that's probably an underestimate. FPL did an estimate earlier this year that put a twin EPR unit plant at about $12-14 billion. I think it was reported in the WSJ.

It’s the Economics, Stupid: Nuclear Power’s Bogeyman
http://blogs.wsj.com/environmentalc...the-economics-stupid-nuclear-powers-bogeyman/
Fast reactors require specialty steels and those are alloys are quite expensive, not to mention that they have not been fabricated in the high tonnage quantities for large plants.

And given the materials problems (failures) I've seen (and been involved with) in the nuclear industry over the last 20+ years, I imagine there's still a lot of R&D to do on advanced reactor concepts.

Any IFR plant will have to get into the pipeline with all the other LWRs already ahead, so they are not going to pop up very quickly - if at all.

I enjoyed the "It’s the Economics, Stupid: Nuclear Power’s Bogeyman" article, but I believe the author's views to be myopic. There is more to consider than the cost increase of nuclear reactors. What about the cost of national security, the economic future of the U.S., and the alternative uses of nuclear energy. Boone Pickens estimated that the U.S. would pay $700 billion for foreign oil if oil stayed at $150 per barrel. Let's assume that oil will settle at about $50 per barrel for the next year or two, and then start to increase again - oil consumption is increasing exponentially and there is a corresponding decrease in supply - the price will go up. If oil is at $50 per barrel, that means that the U.S. will pay about $230 billion per year for foreign oil - using Pickens' figures. The cost of oil is passed on to consumers; this cost has the same effect as a tax on consumers and is a serious drag on our economy. Now assume that 12 one-GW(e) reactors (Brayton Cycle IFRs of course) are built in the shale fields of the west-central U.S. It is estimated that there are one trillion barrels of oil locked in shale in this portion of the U.S. Let's assume further that the electricity produced by these reactors is used to heat the oil shale in the manner described in the Shell in situ conversion process - see: http://www.shell.us/home/content/usa/aboutshell/shell_businesses/upstream/locations_projects/onshore/mahogany/mrp_technology.html

The heat produced by these 12 reactors could yield 10 to 12 million barrels of shale oil per day at a cost of about $20 to $25 per barrel if the government provided the reactors. If shale oil is extracted at a cost of $25 per barrel and oil is selling on the open market at $50 per barrel, that would save the U.S. about $115 billion per year. If the reactors each cost $20 billion, they would pay for themselves in about six years (the shale has to be heated for 3 to 4 years before it produces oil and natural gas). The savings would continue for as many years as the U.S. imports this amount of oil. If the cost of oil goes up, the amount of savings will increase. Nuclear power cannot replace gas in cars, but nuclear power can help put gasoline in cars.

I believe the rapid rise in reactor costs is due to an increase in the perceived risk associated with building nuclear power plants in the U.S. Some of the increased risk is prompted by the general public's fear of nuclear power and their associated reactions to nuclear power plants. Our concern about the cost of nuclear energy should be balanced against the consequences of not developing more nuclear power. In the future, I believe it will not be the cost of labor that determines which country is dominant in the world. Rather, the country that will dominate in the world will be the country that can furnish all of its energy needs at the lowest possible cost.

 

[Andrew Mason]

Let's be clear about the cost of nuclear fuel - the real costs of an open fuel cycle. The real costs of an open nuclear fuel cycle include the mining and processing of uranium, the conversion of uranium into reactor fuel, the transportation costs of the fuel to and spent fuel from reactor sites, the reprocessing costs if the fuel is MOXed, the encapsulation of unuseable nuclear waste, and the storage of unuseable nuclear waste for 10,000 years or more. Now, how many of these costs could be eliminated or nearly eliminated by using an IFR with a closed fuel cycle? I can't prove my estimate is correct and you can't prove it is wrong - neither of us have sufficient emperical data. However, the IFR closed fuel cycle, from a simple economics point of view, should greatly reduce the real costs of nuclear fuel.

These are all very valid points. However, the economic advantages of the IFR will be realized only when competing with plants using once-through fuel. If the world were to eventually replace its nuclear plants with IFRs, the demand for new U fuel would drop dramatically. This would result in the price of U going down to a level that would make all but the richest deposits economic to mine.

A few hundred miles north of where I live is the world's richest uranium deposit at MacArthur River, Saskatchewan. It is 24% U. Every load of ore that is hauled to the mill is worth about half a million dollars. This single deposit could supply the world with Uranium to generate all of the world's electricity for several hundred years if all of the world's electricity were produced by IFR's.

AM