# We *already built* the light water breeder?

## Main Question or Discussion Point

In another thread http://en.wikipedia.org/wiki/Shippingport_Atomic_Power_Station was mentioned. I went to the link, thinking "okay, I remember this one, the first commercial reactor exclusively for electricity generation, not weapon needs. Mostly only historical interesting..."

I was reading the article at leisure when it struck me. It was a U-233/thorium reactor. It was a light water reactor. It was a breeder. It succeeded in demonstrating that breeding was achieved.

WAIT A SECOND. The light water thorium breeder. The holy grail of unlocking full energy reserves of thorium and possibly all uranium too (as opposed to only using U-235, a much more scarce resource). We already had it, and it *worked*. In *1982*.

Why our today's reactors aren't light water thorium breeders, then?

Why Indians are torturing themselves trying to build a heavy water breeder, which is more expensive and has some additional problems such as tritium generation?

Why did US and other nations spent lots of effort trying to develop fast breeders, which have significant problems too?

Last edited:

## Answers and Replies

Related Nuclear Engineering News on Phys.org
Doug Huffman
Gold Member
[ ... ] Why our today's reactors aren't light water thorium breeders, then? [ ... ]
It is a long hard step from a one off demonstration to economically practical, as my friend Rod Adams explains.
Unfortunately, the program leaders were not focused on factors that make new innovations successful in the market. The following weaknesses prevented commercial success.

There was little effort to promote the technology. Knowledge of the program is rare even within the nuclear industry. There is little chance of an unknown idea – particularly one with as much potential impact as a light water breeder reactor – becoming a new technical standard.
• The core engineers did not pay enough attention to production difficulties. The assembly of the core modules required a great deal of manual labor including 2,000 precise measurements for each module. This effort implies a high production cost even if raw materials are used more efficiently.
• There was no effort to develop other uranium-thorium reactors in an effort to help spread the fixed cost of fuel material production.
• The program was viewed as Admiral Rickover’s pet project.

It is a long hard step from a one off demonstration to economically practical, as my friend Rod Adams explains.
But... "the reactor reached criticality on December 2, 1957" (wiki). It was one of the very first power reactors. The "hard step" from experiment to commercially viable power stations wasn't yet made for _any_ reactor type.

From what I read here http://www.inl.gov/technicalpublications/Documents/2664750.pdf [Broken]
the reactor's core and fuel weren't radically different from what is in use today. Ceramic oxide pellets as fuel. Zirconium cladding.

I just don't see an obvious reason why this particular type of reactor (thorium LWR breeder) wasn't developed further, but uranium NON-breeders were. The basics of their designs are reasonably similar. I know this place has people very knowledgeable of US nuclear power history, I hope someone knows the reason "why".

Last edited by a moderator:
Doug Huffman
Gold Member
Shippingport is where my seniors got their experience before the resources that I enjoyed existed.

About why, explanations are ignored, denied and disputed. "Similar" is a Sorites heap paradox; 2000 exceptional hand measurements per module is hardly similar. I suspect that our disputant knows as little of fuel physics as do I, but I had a physicist at my elbow while I tested his product.

Shippingport originally went online in '57 as a highly enriched uranium LWR. The thorium breeder re-work came online 20 years later.
http://en.wikipedia.org/wiki/Shippingport_Atomic_Power_Station#Cores
Ah... that probably explains it: by the time successful thorium breeding was confirmed, uranium LWR designs were already built and running commercially.

Ah... that probably explains it: by the time successful thorium breeding was confirmed, uranium LWR designs were already built and running commercially.
The key for commercial power plants is the cost of producing power while maintaining license requirements. What makes you think thorium is any better at this goal? UO2 works, has a long (relatively) well understood behavior (easier to license), its fuel cost are relatively cheap with well established global supply chains and there is no risk of running out of it in the short term. If you enrich it sufficiently you can get relatively high burnup and run long batch cycles with no need to reprocess.

Thorium has some interesting benefits (better material properties), large global reserves and the possibility of thermal breeding. However, these don't necessarily translate into cheaper power or higher profits. Some how people have gotten the impression that thorium is some sort of superfuel, but I don't see where the major gains as supposed to come from.

Doug Huffman
Gold Member
Thorium is done no favors being conflated with liquid alkaline metal coolant/moderator/vehicle.

The key for commercial power plants is the cost of producing power while maintaining license requirements. What makes you think thorium is any better at this goal?
Nothing makes me think that way. You jumped to a conclusion that I do.

I know that the motivation behind breeders is not lower cost, but more complete usage of our finite fissionable resources. I said: "The holy grail of unlocking full energy reserves of thorium and possibly all uranium too (as opposed to only using U-235, a much more scarce resource)."

Nothing makes me think that way. You jumped to a conclusion that I do.

I know that the motivation behind breeders is not lower cost, but more complete usage of our finite fissionable resources. I said: "The holy grail of unlocking full energy reserves of thorium and possibly all uranium too (as opposed to only using U-235, a much more scarce resource)."
Fair enough! That was my bad. I'm so used to people thinking thorium is magic that I interpreted your interest that way.

mheslep
Gold Member
The key for commercial power plants is the cost of producing power while maintaining license requirements. What makes you think thorium is any better at this goal? UO2 works, has a long (relatively) well understood behavior (easier to license), its fuel cost are relatively cheap with well established global supply chains and there is no risk of running out of it in the short term. If you enrich it sufficiently you can get relatively high burnup and run long batch cycles with no need to reprocess.

Thorium has some interesting benefits (better material properties), large global reserves and the possibility of thermal breeding. However, these don't necessarily translate into cheaper power or higher profits. Some how people have gotten the impression that thorium is some sort of superfuel, but I don't see where the major gains as supposed to come from.
I'd like to comment, if I may.

Thorium has some advantages. In particular, a switch to thorium i) eliminates the long lived actinides leaving only fission products in the spent fuel stream, thereby reducing the storage time of harmful radiotoxicity from tens of thousands to a couple hundred years, ii) eliminates the enrichment fuel cycle aside from the need for breeding seeds; thus, regardless of the cost of the uranium enriched fuel cycle, thorium as a fuel must be less expensive, and iii) the high burnup afforded by thorium (170 GWd/t Th at Fort St Vrain) as compared to the typical LWR (35 or 40 GWd/t U) means fuel service intervals might be extended and the waste stream mass greatly reduced.

The suggestion for uranium fuel high burnup via higher enrichment ("enrich it sufficiently") runs counter to the goal of minimizing proliferation potential. Yes highly enriched uranium reduces the hazard of the waste stream, but has its own proliferation issues. If operating within existing regulations is a goal, 5% enrichment is the legal limit in the US. As long as U-238 is in the fuel stream of LWRs Pu will also appear in the waste stream.

The enriched fuel cycle is not entirely "global". Visibly there are several countries that buy and sell enriched fuel to each other, but that list is highly restricted. The like of Yemen, Nigeria, Burma, Iran are not allowed in, nor should they be IMO for security reasons. A future where the developing world starts adding nuclear plants instead of coal plants is made more likely without enriched fuel.

mheslep
Gold Member
I list below the IAEA given "benefits" of thorium. "Challenges" are also given in the document.
http://www-pub.iaea.org/mtcd/publications/pdf/te_1450_web.pdf

1. Thorium is 3 to 4 times more abundant than uranium, widely distributed in nature as aneasily exploitable resource in many countries and has not been exploited commercially so far. Thorium fuels, therefore, complement uranium fuels and ensure long term sustainability of nuclear power

2. Thorium fuel cycle is an attractive way to produce long term nuclear energy with low radiotoxicity waste. In addition, the transition to thorium could be done through the incineration of weapons grade plutonium (WPu) or civilian plutonium.

3. The absorption cross-section for thermal neutrons of 232Th (7.4 barns) is nearly three times that of 238U (2.7 barns). Hence, a higher conversion (to 233U) is possible with 232Th than with 238U (to 239Pu). Thus, thorium is a better ‘fertile’ material than 238U in thermal reactors but thorium is inferior to depleted uranium as a ‘fertile’ material in fast reactor.

4. For the ‘fissile’ 233U nuclei, the number of neutrons liberated per neutron absorbed (represented as η) is greater than 2.0 over a wide range of thermal neutron spectrum, unlike 235U and 239Pu. Thus, contrary to 238U–239Pu cycle in which breeding can be obtained only with fast neutron spectra, the 232Th–233U fuel cycle can operate with fast, epithermal or thermal spectra.

5. Thorium dioxide is chemically more stable and has higher radiation resistance than uranium dioxide. The fission product release rate for ThO2–based fuels are one order of magnitude lower than that of UO2. ThO2 has favourable thermophysical properties because of the higher thermal conductivity and lower co-efficient of thermal expansion compared to UO2. Thus, ThO2–based fuels are expected to have better in–pile performance than that of UO2 and UO2–based mixed oxide.

6. ThO2 is relatively inert and does not oxidize unlike UO2, which oxidizes easily to U3O8 and UO3. Hence, long term interim storage and permanent disposal in repository of spent ThO2–based fuel are simpler without the problem of oxidation.

7. Th–based fuels and fuel cycles have intrinsic proliferation-resistance due to the formation of 232U via (n,2n) reactions with 232Th, 233Pa and 233U. The half-life of 232U is only 73.6 years and the daughter products have very short half-life and some like 212Bi and 208Tl emit strong gamma radiations. From the same consideration, 232U could be utilized as an attractive carrier of highly enriched uranium (HEU) and weapons grade plutonium (WPu) to avoid their proliferation for non-peaceful purpose.

8. For incineration of WPu or civilian Pu in ‘once-through’ cycle, (Th, Pu)O2 fuel is more attractive, as compared to (U, Pu)O2, since plutonium is not bred in the former and the 232U formed after the ‘once-through’ cycle in the spent fuel ensures proliferationresistance.

9. In 232Th–233U fuel cycle, much lesser quantity of plutonium and long-lived Minor Actinides (MA: Np, Am and Cm) are formed as compared to the 238U–239Pu fuel cycle, thereby minimizing the radiotoxicity associated in spent fuel. However, in the back end of 232Th–233U fuel cycle, there are other radionuclides such as 231Pa, 229Th and 230U, which may have long term radiological impact.

Yes highly enriched uranium reduces the hazard of the waste stream
I assume this means "... by increasing burnup, and thus decreasing amount of waste".

Well, this is "almost untrue". Granted, the weight of generated spent fuel is less. But the amount of actual waste (fission products + actinides) will be about the same.

IOW: 1 ton of spent fuel at 100 GWd/t burnup has about the same amount of fission products as 2 tons of spent fuel at 50 GWd/t. After reprocessing and, say, vitrification you'll have about the same amount of glass to bury.

mheslep
Gold Member
I don't mean fission products, which have relatively short half-lives, but the actinides which derive from U238. Reduce the ratio of U238 via enrichment, and one reduces actinides.
https://www.hzdr.de/db/PicOri?pOid=30404

I'm not disagreeing that thorium can do some cool things. I'm suggesting that doing cool things doesn't directly translate into cheaper power. Longer cycle times I'd wager is the most attractive aspect of thorium economically.

Are utilities incentivized to reduce waste? Does reducing production of actinides affect cost? Not really waste disposal is a very small component of the total cost. Greater chemical stability? Well maybe a little bit of a source term reduction, but it doesn't fundamentally change the game. Greater abundance of thorium doesn't decrease cost because there isn't the same supply chain and infrastructure in place (yet). Proliferation concerns don't really matter in countries that already have nuclear weapons or already posses other routes to nuclear weapons.

The benefits have to be balanced against the cost of licensing something different. And training everyone to work with 'different'. Investments in nuclear are already extremely expensive (capital) and high risk (financially speaking). Adding first of a kind (or generation) technology adds to those troubles.

I was under the impression that LWRs cant get a high enough burnup to really take advantage of thorium. Thorium requires a higher neutron economy due to some parasitic effects and that combined with actinide buildup would limit available burnup from the fuel,

mheslep
Gold Member
I was under the impression that LWRs cant get a high enough burnup to really take advantage of thorium. Thorium requires a higher neutron economy due to some parasitic effects and that combined with actinide buildup would limit available burnup from the fuel,
IAEA:
"4. For the ‘fissile’ 233U nuclei [bred from Th232], the number of neutrons liberated per neutron absorbed (represented as η) is greater than 2.0 over a wide range of thermal neutron spectrum, unlike 235U and 239Pu."
High burnup with Th (mixed with HEU) has been demonstrated (170 GWd/t)

In a LWR?

I know it can in a HWR. Or a high temp has reactor. (Ft at vrain was where the 170gwd/ton came from, that's an htgr, not a LWR)

Last edited:
mheslep
Gold Member
...
Are utilities incentivized to reduce waste? Does reducing production of actinides affect cost? Not really waste disposal is a very small component of the total cost.
Apparently you are considering only O&M money? Backup to new construction and staffing, and consider the spent waste pool construction for a given tonnage, long term waste storage for a given tonnage. Thus, the relevant question is, can the utility save money in new nuclear construction. Consider the cost of the construction of new enrichment plants at $4B each. Greater abundance of thorium doesn't decrease cost because there isn't the same supply chain and infrastructure in place (yet). A bit of a strawman: nothing is effective until it is built, and note that thorium has been used in commercial reactors three or four times so far. Proliferation concerns don't really matter in countries that already have nuclear weapons or already posses other routes to nuclear weapons. Yes. Proliferation though is a concern by definition about countries that do not have weapons. The A.Q. Khan episode was all about his delivery of enrichment technology to rogue states, and about which he had long experience and education. Take away the enrichment industry, and the likelihood of such events must decrease. Investments in nuclear are already extremely expensive (capital) and high risk (financially speaking). Adding first of a kind (or generation) technology adds to those troubles. Yes, which explains in part why no new nuclear projects have been completed in the US last 30 years. The five new reactors now under construction will just barely manage to maintain the collective output of the US nuclear fleet. The reality is that the public's unease, justified or not, about waste and proliferation and accidents drive costs (relative to say India,$1500/KW), made that way by regulation and legal battles. Thorium can help with the first two.

mheslep
Gold Member
In a LWR?

I know it can in a HWR. Or a high temp has reactor. (Ft at vrain was where the 170gwd/ton came from, that's an htgr, not a LWR)
Sorry. For LWRs you may be correct.
The European Framework Program has supported a number of relevant research activities into thorium fuel use in LWRs. Three distinct trial irradiations have been performed on thorium-plutonium fuels, including a test pin loaded in the Obrigheim PWR over 2002-06 during which it achieved about 38 GWd/t burnup...
A small amount of thorium-plutonium fuel was irradiated in the 60 MWe Lingen BWR in Germany in the early 1970s. The fuel contained 2.6 % of high fissile-grade plutonium (86% Pu-239) and the fuel achieved about 20 GWd/t burnup.
Maybe only gas, heavy water, and molten salt can obtain high burnup.

Astronuc
Staff Emeritus
Heavy water (CANDU) plants generally have low burnups (~ 10 to 12 GWd/tU) because they use natural U. Burnup could be increased with enrichment and residence time. HW could be used in an LWR, but HWR is expensive.

As I recall, FSV fuel was highly enriched, and IIRC, dispersed in a graphite matrix, so they could achieve high burnup.

Modern PWRs are getting peak assembly burnups of about 50 - 54 GWd/tU, with peak rods ~59-60 GWd/tU. BWRs are somewhat lower, although one German plant has irradiated assemblies to 67 GWd/tU and slightly higher.

From the folks at ANT International.
The highest burnups in BWRs are in the US, Germany, Spain and Switzerland. The batch averages range between 43 and 57 GWD/MT, peak assemblies between 51 and 68 GWD/MT and the peak rods between 55 and 73 GWD/MT. The burnup levels in BWRs are catching up to those of the PWRs in the US, probably because both reactor types are reaching the current NRC burnup limitation.

The highest peak rod burnups achieved in LTAs have been in GWD/MT: 63 (Spain), 65 (US), 72 (Japan), 73 (Switzerland), 75 (Germany).
http://www.epri.com/abstracts/Pages/ProductAbstract.aspx?ProductId=000000000001016624

Some interesting history at FSV - http://www.fsvfolks.org/FSVHistory_2.html [Broken]

I used to work for a company that did work at FSV, but unfortunately, I joined just before FSV was shutdown.

Last edited by a moderator:
Apparently you are considering only O&M money? Backup to new construction and staffing, and consider the spent waste pool construction for a given tonnage, long term waste storage for a given tonnage. Thus, the relevant question is, can the utility save money in new nuclear construction. Consider the cost of the construction of new enrichment plants at $4B each. Lets look at the NEI website. Long term waste storage cost 0.1 kWhe to the nuclear waste fund (not paid based on tonnage, radioactivity or isotopic composition). I agree that you would save on the spent fuel pool, but this is already relatively cheap expense, so its effect should be relatively small. A bit of a strawman: nothing is effective until it is built, and note that thorium has been used in commercial reactors three or four times so far. This isn't a strawman, its acknowledging that the uranium industry leverages infrastructure that is already built . Its a comparison between the marginal cost of each new piece of infrastructure. Its quite possible that thorium would be cheaper if it had the same level of investment but we cannot reinvest that money that has already been spent on uranium. Its gone and all we can do is figure out the best way forward based on our current situation. Yes. Proliferation though is a concern by definition about countries that do not have weapons. The A.Q. Khan episode was all about his delivery of enrichment technology to rogue states, and about which he had long experience and education. Take away the enrichment industry, and the likelihood of such events must decrease. Personally, I disagree. Countries like US, France, UK, China, Russia, India, Pakistan, Isreal already have nuclear weapons and the capability to build more. What difference does it make if they use a PUREX cycle now? Then there are non-nuclear weapons states like Canada and Japan (and quite a few others) which have the infrastructure to make weapons but haven't. What difference does it make if you have 3 ways to make plutonium instead of 2? Yes proliferation matters for countries like Iran, which is why we should be selling them all the low enriched uranium they want near cost (so they don't develop their own ability to enrich). You can also 'lease' them fuel and take it home with you too. Certainly proliferation resistance designs help in these cases but they are not the only option. Yes, which explains in part why no new nuclear projects have been completed in the US last 30 years. The five new reactors now under construction will just barely manage to maintain the collective output of the US nuclear fleet. The reality is that the public's unease, justified or not, about waste and proliferation and accidents drive costs (relative to say India,$1500/KW), made that way by regulation and legal battles. Thorium can help with the first two.
There is more at play than simply public unease. There has been numerous failed projects, changing energy markets (cost of coal and natural gas) and reduced demand. To further complicate matters plants are now expected to have a 60+ life time but no one can accurately predict market conditions 5 years from now. This adds a great deal of uncertainly to the present value calculation.

Astronuc
Staff Emeritus
New plants may not be cost/price competitive because of the cost of material, e.g., high grade steel and concrete, which have accelerated in price 2x, 3x, 5x, since 2000. In the late 1990s and early 2000s, we were talking about between $1-2 billion for a new Gen3+ plant, but with the acceleration in cost of materials, the cost of a new unit went quickly to$5 billion then to about $7 billion, per unit. Here is one example of the cost of materials - http://steelbenchmarker.com/files/history.pdf - and the cost/price is volatile. There are plenty of Gen 3+ designs that have been approved, so that money has already been spent. Now it's a matter of finding customers. A few years ago, I found a study by one state's transportation department that showed the cost of concrete had increased by a factor of 5, while steel had gone up by a factor of about 3. Those factors are subject to change, and the steelbenchmarker file provides a limited example of price volatility. The customers (utilities) are faced with abundant fossil fuel supplies (coal, oil and natural gas). The capital cost for a gas-fired plant and the construction time is a lot less than for a nuclear unit, so they are more attractive. If one builds a combined cycle plant, then that's even better. Of course, if natural gas prices spike, then natural gas-fired plants are less attractive or even yield a loss. And to add to the mix, wind power is heavily subsidized and utilities are forced to take the wind, when it's generating, and have to reduce generation of other units, which cuts into revenues of baseload plants, e.g., nuclear. Last edited: mfb I thought that cost of labor and very long delays due to paperwork are dominating the cost... Astronuc Staff Emeritus Science Advisor I thought that cost of labor and very long delays due to paperwork are dominating the cost... That was the case for some of the later Gen 3 plants built in the late 70's and 80's. There were increases in labor costs, litigation costs, redesign costs (particularly after the fire at Browns Ferry and the accident at TMI), and just general delays. As a result, over 100 nuclear plants were cancelled in the US, and several like WPPSS, Bellefonte, Zimmer, that were in various stages of construction were stopped. Shoreham (on Long Island) actually operated before it was shutdown and decommissioned. It probably cost over$6 billion when all was said and done.