Monday, December 28, 2009

Updating ORNL MSR Design and Cost Studies

In order to understand the importance of the LFTR, we must return to the original research reports and other working documents prepared at Oak Ridge National Laboratory between 1950 when ORNL research on the Molten Salt Reactor concept began at ORNL and 1980 when ORNL scientists who had MSR expertize, wrote their last proposal. ORNL researchers in the 1960's and 1970's argued that MSR development would not cost a lot and commercial MSR was cost competitive with then current LWE designs. Since the 1970's some of the development problems for the MSR have been independently solved. Much of the most challenging MSR technology is also used in other high temperature reactors, and in fusion reactors. So ongoing research and development on these technologies has contributed to the development of the technology proposed for the MSR. At the same time proposed design of the Thorium Breeding Molten Salt Reactor (the LFTR), has increasingly diverged from current Light Water Reactor (LWR) designs. Canadian physicist David LeBlanc has proposed a radically simplified of the already simple MSR/LFTR core design. LeBlanc's redesigned core could be assembled in less than a day, in a LFTR factory. LeBlanc also suggests the use of
less expensive iron alloys including the common stainless steels 304 and 316 [which] have also shown promise at somewhat lower operating temperatures.
The use of multiple cost saving strategies also holds promise for lower LFTR costs. These would include the factory construction of small LFTRs, which would be transported by truck, rail or barge to the power plant location, the recycling of old coal fired electrical generation facilities as LFTR locations. The use of underground housing, rather than massive containment structures. The manufacture of large numbers of reactors on a factory assembly line will increase the speed of progression on the manufacturing learning curve, leading to even lower prices.

During the last 40 years the cost of light water reactors has risen dramatically. Rigorous NRC certification standards have dramatically increased the cost of parts design. Nuclear critics point to $50,000 being spent to design as $5 nuclear part. Mass production means that each part can be used on hundreds and even thousands of reactors, lowering nuclear costs. Recent developments in LWR design point to reactor simplification as an important step for lowering nuclear costs. There is little doubt that adopting LFTR would dramatically simplify nuclear designs.

In the absence of a serious design effort by a contemporary nuclear design team, it is difficult to estimate exactly how much LFTRs would cost. ORNL MSBR design studies, although old, do hold clues to nuclear costs. ORNL-TM-1851 (SUMMARY OF THE OBJECTIVES, THE DESIGN, AND A PROGRAM OF DEVELOPMENT OF MOLTEN-SALT BREEDER REACTORS) is a good starting point for looking at LFTR cost estimates. TM-1851 estimated that a LFTR type MSBR could be developed in 8 years at a cost of 125,000,000 1967 dollars.

This estimate was made before ORNL researchers had their three experience of with the Molten Salt Reactor Experiment (MSRE). The MSRE pointed to some real but hardly insurmountable technological problems to be overcome before a LFTR type MSBR could be commercially viable. Thus later ORNL design studies, and R & D plans were more realistic about development costs and time frames.

The early ORNL MSBR designs were fairly complex. TN-1851 developed its cost estimates from analogies to the costs of the light water reactors of the time. But in fact later developments in both technologies diverged.

Recent thinking about LFTR design has moved on the 1960's and 1970's designs, and more recent actually point to lower, and potentially much lower LFTR costs, than could be obtained by simply replicating old ORNL designs. Thus it is quite possible that old ORNL estimates for MSBR costs, when adjusted for inflation could be actually higher than future LFTR costs. TM-1851 estimated that power could be generated by MSBRs for as little as 2.6 mills per kWh, of about 1.7 cents per kWh, inflation adjusted 2009 costs. This would be most encouraging, if we could rely on this figure.

In another post, I pointed to yet other ORNL studies which also point to similar conclusions about LFTE costs. Finally I point in the same post to the development and manufacturing costs for the Airbus 380 aircraft. Airbus invested €11 billion plus that in the development of the A380. At a cost of $327 million the A380 would be if anything more complex and more expensive than the modular LFTR. Thus we have a reasonable hope that LFTR costs would come in at under $2 per watt of generation capacity, and $1 per watt or even less is not beyond the realm of possibility. We need more research to get a better understanding of LFTR cost estimate, but my preliminary studies suggest that the LFTR could represent a dramatic breakthrough in lowering the cost of post carbon energy.


Frank Kandrnal said...

I think your figures are little too pessimistic. Like Alvin Weinberg believed, I too believe it can be done for less expenditure.
Present day example how inexpensive nuclear electricity can be just look at the latest UAE deal with South Korean Consortium.
Four unit reactor contract was just awarded to South Korean Consortium. Despite having one of the best conditions for solar power, UAE decided to use nuclear power instead because of much better economics and power availability for industries they are planning. When all considered, this is the best decision how to invest national financial resources and get things done.
By the way, the electricity production cost of these 4 reactors will be about 1.6cents/kwh when 40 billion is spread over 60 years with 90% power plant availability factor. I calculate $20 billion for construction cost and $20 billion for operation for 60 year period, no interest on finances is counted into calculation. This is what I call economics! It makes alternate solar/wind generated electricity cost look very sick in comparison.
Alvin Weinberg knew very well that molten salt reactor technology is simpler hence would cost much less than light water reactor.
I was very impressed by various hardware designs for 1000mwe power plant design ORNL designers came up with and the projected cost for various components. Those costs were quite realistic for those days and the designs were made so it could be easily manufactured.
There is no question that today those components can be still made cheaper than the components for pressurized light water reactors.

Charles Barton said...

Frank, My suspicion is that LFTR costs could well be under a dollar a watt, but i think such a low cost estimate would be dismissed as a fantasy,

Jim Bowery said...

A minimum-size LFTR demonstrator would provide an upper-bounds on the cost per installed watt.

Doesn't that seem far more likely to actually happen in the near term?

What is the smallest LFTR demonstrator that could be built and how much would it cost to build it?

Alex P said...

" Frank, My suspicion is that LFTR costs could well be under a dollar a watt "

Do you mean a full thorium breeder (with a quite complex on-site reprocessing system) or a simply converter LFTR (with a minimal fuel reprocessing) ?

Alex P said...

I personally think (and hope for) that the first prototypes at a R&D scale will be of a power of less than 100 MW thermal (including if possible a fast chloride reactor working with a thorium blanket and LWR transuranics "waste" as "seed" to effectivly burn them), in order to develop future commercial size reactors of less than 1000 MW thermal (or 300-500 MW electric)

Jim Bowery said...

I agree that the demonstrator will have to include utilization of waste as seed. It seems to me that the most important thing about going with the lower scale is that it provides industrial learning curve via mass production. Mass production industrial learning curve on this kind of thing promises to be so enormous that the diseconomies of thermal scale may not matter in real deployment.

I see no reason that the R&D prototype cannot be the same scale as the manufacturing prototype, with both being low enough power that rural municipalities could afford them at mass production prices.


Blog Archive

Some neat videos

Nuclear Advocacy Webring
Ring Owner: Nuclear is Our Future Site: Nuclear is Our Future
Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet
Get Your Free Web Ring
Dr. Joe Bonometti speaking on thorium/LFTR technology at Georgia Tech David LeBlanc on LFTR/MSR technology Robert Hargraves on AIM High