Talk:Liquid fluoride thorium reactor

I propose merging the separate Disadvantages and Design challenges sections into one section called "Disadvantages and design challenges", since the difference between them is very ambiquitous. Most disadvantages seems to have an engineering (design) solution to counter or mitigate them (and such design challenges can also be seen as disadvantages). Keeping it separate only causes editors to make duplicate entries. — Preceding unsigned comment added by ShotmanMaslo (talk • contribs) 20:27, 28 April 2012 (UTC)

Article is now over 91k in length. Please summarize into bullet points. Also remember all contents for Wikipedia must be verifiable, and no original research. That also applies to objections. Speculative, uncited objections cannot be included. Same applies to uncited explanation. MegaHasher (talk) 04:35, 3 May 2012 (UTC)

Some of the "Design Challenges" could also be seen as "Development Issues". Let's see what kind of headings work best. "Disadvantages" do not work that well because a later edit could make an item not a disadvantage. MegaHasher (talk) 23:37, 3 May 2012 (UTC)

I noticed that a lot of the 'disadvantages' points had counterpoints toward the end to soften them. But the 'advantages' tended to not have such counterpoints. The whole article seemed to me to be heavy on optimism, especially for a technology that hasn't been implemented in decades. 75.101.23.13 (talk) 02:48, 23 June 2013 (UTC)

I have moved the program cancellation section to the MSR article where there is more extensive materials to support this section of text. Of course, the MSRE was not cancelled but ran to a successfully conclusion. It was actually the follow-on effort that was canceled. MegaHasher (talk) 03:37, 3 May 2012 (UTC)

We have two sections dealing with the historical ORNL one fluid and two fluid designs. If we want to include texts about modern one fluid or two fluid designs it might be best to use their own headings, in order not to confuse the modern with the old. MegaHasher (talk) 23:47, 3 May 2012 (UTC)

One important subject is the choice of subject heading in this article could lead to point of view biases, which is prohibited by Wikipedia. I would encourage editors to group texts into subject headings of bounded size, and to minimize bias. MegaHasher (talk) 18:30, 4 May 2012 (UTC)

Many of the References appear several times. There is method to do this more elegant - having the references only, with several links back. This will help to keep the list a little shorter.--Ulrich67 (talk) 23:24, 4 May 2012 (UTC)

Many of the references are from Thorium Energy Alliance conferences. As much as I would like to attend one maybe next year the best references are US government funded nuclear reports. You can quote them more extensively, and they lead you to more useful references. MegaHasher (talk) 04:16, 5 May 2012 (UTC)

There are a lot of low quality references, that do not meet the wikipedia quality standards. It does not help to give many of these low quality sources - this tends to make a hidden link list from the reference list. The slides or videos of conference talks and blogs are often highly biased and thus rather bad references especially for things like pros and cons.--Ulrich67 (talk) 16:54, 5 June 2012 (UTC)

Somehow what appears to be TEAC Conference minutes appeared as a section in this article. This section need to be re-organized, or reverted all together. Much of the content is unrelated to Small Modular Reactors. We can't have anything that starts with "Kirk Sorensen said...". Maybe best to just revert. MegaHasher (talk) 04:55, 5 May 2012 (UTC)

+1 This section about costs should be removed. There is no good ref to this, and other Refs. to costs more tend to say that costs of the reactor are similar to LWR. Cost estimates are always difficult, especially with a new technology and often tends to be biased. Even with conventional reactor costs sometimes increase by a factor of 2 and more.--Ulrich67 (talk) 15:10, 25 May 2012 (UTC)

Comparing Brayton cycle versus Rankine cycle in web search had a mixed result. Some studies showed Brayton as more efficient, while other ones had Rankine as more efficient. MegaHasher (talk) 20:44, 12 May 2012 (UTC)

Answer to MegaHasher:

The steam rankine cycle referred in the ORNL MSBR studies had an efficiency of about 44%.

Brayton cycles can be more efficient, but only when using closed gas, multiple reheat, intercoolers etc. The University of Berkeley work assumes such closed gas, multiple reheat cycles. At the ORNL MSBR operating temperature the efficiency is around 46%, but the temperature is higher than a steam cycle. At elevated temperatures >900 degrees Celsius, efficiency is over 55%.

Open Brayton cycles, such as today's natural gas burning gas turbines, are not more efficient unless extremely high temperatures can be attained, and even then are only slightly more efficient than supercritical steam cycles. However the advantage with a gas turbine burning natural gas is that the waste heat gas is very hot so can be used to power another, bottoming cycle, usually steam. This combined cycle allows up to 60% efficiency in the latest designs. — Preceding unsigned comment added by Siphon06 (talk • contribs) 13:31, 16 May 2012 (UTC)

For a single fluid reactor the MSRE was only 8 mega-watts. That is already very small. The sentence about "breeding breakeven" refers to the breeding ratio, and could it be a duplicated point about breeding ratio? I am pretty sure it is false to suggest only a two fluid reactor can have flexibility in reactor size. MegaHasher (talk) 21:02, 18 May 2012 (UTC)

These two points are related, but not exactly the same. With the 1 fluid design the breeding ratio drops significantly if it gets smaller. The breeding break-even already needs quite a large unit (like 500 kg of fissile material - maybe 200 MW). For the 2-fluid design the breeding ratio does not drop much with smaller size. The MSRE (38 kg fissile material, low power density) is more like the inner core of a 2 fluid design, not a 1 fluid LFTR. So it is true that the single fluid design is less flexible in size, if one wants significant breeding - even if less than break-even. Some 2-fluid designs have difficulties with large units, but here using several smaller units is an option.--Ulrich67 (talk) 12:31, 19 May 2012 (UTC)

The US Small modular reactor program defined a small reactor as less than 300 MWe. The Fuji MSR was a 155 MWe design. Your example of a 200 MW design would be a small reactor as well. MegaHasher (talk) 18:08, 20 May 2012 (UTC)

Need a reference. MegaHasher (talk) 02:37, 22 May 2012 (UTC)

Wash 1097 (section 5.3.2, currently Ref. no. 27) gives a minimum practical size of about 1000 MW(e) (or a little less) - this is a design with about 1500 kg fissile material. Without removal Pa the power per amount of fissile material would be smaller and thus a lower power rating (more like 200-500 MW(e)) for the same size of core. This is not a sharp limit in size - it just gets increasingly difficult getting smaller. So the 1 fluid LFTR is limited to relatively large size. Very small breeding reactors just need a blanket to reduce neutron leakage.--Ulrich67 (talk) 15:04, 25 May 2012 (UTC)

The Fuji Designs have a relatively low power at something like 150-250 MW(e) (depending on the Ref. used, but it still uses quite a large amount of fissile material, like 1950 kg for the 250 MW(e) version, and even this just at break even. It is always possible to limit the power of a large reactor so a small reactor should be better defined as one with a small amount of fissile material, not from the rated power. In conclusion this means that the LFTR is actually less flexible in size than the LWR. So the point "LFTRs scale well" under advantages should better be replaced with some extra informations to the sections on the 1-fluid (only large units) vs. 2-fluid (preferred small).--Ulrich67 (talk) 18:12, 1 June 2012 (UTC)

I am not sure if the reduced breeding ratio quoted in Wash 1097 could be attributed to switching from 2 fluid to 1 fluid. The change from a 2000 MWe design to a 1000 MWe design could have done it too. This bullet item is probably a combination of breeding ratio and reactor size, but these items were already mentioned, which could be a form of double counting. MegaHasher (talk) 22:41, 1 June 2012 (UTC)

Reactor size still is an issue: If one does not want the possibility for small sizes as a positive feature of the 2 fluid (and 1,5 fluid) design, we should note it for the 1 fluid design as a design limit.--Ulrich67 (talk) 19:50, 4 June 2012 (UTC)

One of the smallest one-fluid design I can find is the FUJI-U3: ^[1] 200 MWe, breeding ratio 1.01, 30 year graphite life. MegaHasher (talk) 06:49, 5 June 2012 (UTC)

Though small in power this design still uses 1,5 t of fissile material. So it's something like a large core at low power density, needed to get good graphite lifetime and to avoid Pa removal. The ORNL 2-fluid design was 315 kg U233 and 250 MW(e) at a relative high power density. At a reduced power, to circumvent Pa removal and to improve core lifetime, this would be more like 50 MW. Already here they needed the complicated interleaving of core and blanket to make it that large.--Ulrich67 (talk) 20:27, 5 June 2012 (UTC)

The section Advantages describes some of the properties - not all are actually positive (the rather slow breeding from thorium compared to U238->Pu (about 10 times faster) is more a problem, than positive). Similar other points have both positive and negative sides to it or are just requirements to make the LFTR safe. Most of these points appear in some talks - but it is questionable if this belong in this article: the talks are given to point out the advantages, so they are highly biased sources (and they are far from sufficient quality to qualify as a reliable source). There just is no need make the article biased - this article is not about selling something, unlike some of the talks and videos.

It should be better to have a sections about things like fuel (could be mainly a link to thorium fuel cycle), safety-systems and waste. That have both the positive an negative parts. --Ulrich67 (talk) 15:53, 25 May 2012 (UTC)

There are some good stuff in the deleted material, but it needs to be summarized in a non-biased form suitable for Wikipedia. A generic subject heading like "Advantages" is generally inferior to a specific subject heading such as "Plant cost estimates", or "Energy unit cost comparison". MegaHasher (talk) 05:26, 1 June 2012 (UTC)

I agree that some information on costs may be useful. However it is rather difficult, to find a good source on this. The deleted part on costs was rather useless: highly biased and not at all a reliable source. Cost estimates are very difficult, even mature technique like PWRs. For the LFTR there are so many open questions (e.g. 1 vs. 2 fluid, type of reprocessing, safety requirements) that it is just to early for a serious estimate. The cost estimates for the fast breeders were also very much lower. So at best one would get something like ... claimed low costs, though it's to early for a serious estimate.--Ulrich67 (talk) 21:03, 3 June 2012 (UTC)

Under Advantages there is the point transparent coolant. Is this point really relevant ? This should only apply to the secondary salt (without fuel), and only in the hot liquid state. Water used in PWR is transparent too - so it's not really an advantage, just on par. As the list and the whole article is already on the long side, I suggest removing this small paragraph. --Ulrich67 (talk) 23:32, 9 July 2012 (UTC)

The section under advantages- safety about the temperature coefficient of reactivity seem a little to simplistic and optimistic. Newer, more detailed calculations (e.g. [ http://arxiv.org/abs/nucl-ex/0506004] and ) show that the temperature coefficient is not negative in all cases - especially the ORNL-MSBR design often used as a basis likely does not have a negative temperature coefficient. Its only the coefficient for the salt itself that is negative - but due to the graphite the overall coefficient is positive (or at least not sure to be negative). So this design would not be intrinsically safe and self regulating, though not necessarily unsafe. The simple way to describe the influence of salt expansion is also not correct: In an under-moderated reactor (like the MSBR) removing salt from the core increases reactivity. The simple picture of less fuel salt means less reactivity is true for a small reactor where neutron leakage is important. In an under-moderated large core, less salt could mean better moderation and thus more reactivity. Thus the void coefficient can be positive too - a serious safety concern. So a large negative temperature coefficient is more a restriction to observe, but not a general advantage of a LFTR. At best one could say that it is likely possible meet this requirement, but this comes at a price, like reduced breeding. So this section needs editing - possibly moved outside advantages as a separate section (e.g. reactivity coefficients). The fist thing would be to correct the part on expansion of the salt.--Ulrich67 (talk) 15:39, 17 July 2012 (UTC)

The picture of the liquid FiLBe only shows, that the pure carrier salt is transparent. The picture under Uranium-233 shows, that at least the solid form with uranium is dark green. Melting likely will not make it transparent. Still there is no Ref. for the optical Inspection in the hot state: Even if transparent this would be hard to impossible - so I don't see any advantage from this. --Ulrich67 (talk) 17:48, 8 September 2012 (UTC)

I tend to agree that "Other Uses" should have its own heading instead of appearing under "Advantages". Again the TEAC3 reference is not that great because they are self published presentations. Article submitters should submit sections with good heading classification, in summary form, and specifically referenced. We as editors don't have an obligation to fix every borderline additions. In Wikipedia a section deletion needs no justification other than that section not meeting the published standards of Wikipedia. Also it is not useful for article writers to re-submit deleted section completely unmodified. At least use the talk page to find out what is wrong about it. MegaHasher (talk) 23:28, 1 June 2012 (UTC)

Currently there is a section "Ease of reprocessing". This should better called just reprocessing as reprocessing is not per se easy. If it would be so easy as sometimes claimed, it would likely be used for solid fuel too. --Ulrich67 (talk) 20:46, 5 June 2012 (UTC)

The reference given for the "very small" reprocessing plant still gives a size of the reprocessing units that is about 1/4 to 1/2 the size of the core itself. This is for the 2 fluid design - which makes reprocessing much easier than the more practical 1 fluid (or 1,5 fluid) design. In addition this is assuming a rather small throughput of 15 ft³ a day, of about 50 days cycle time for the core. So my conclusion from this rather old ref. would be more it needs a rather large reprocessing plant to reach 10 days cycle - possibly larger and more expensive than the rest of the hole reactor.--Ulrich67 (talk) 20:02, 7 June 2012 (UTC)

One Problem here is, that small or vary small is rather vague - it's the question small compared to what. With the reprocessing part it also very much depends on the frequency of reprocessing. Going from 10 days (needed for Pa removal) in the older designs to something like 100 days, reduces the needed throughput by a factor of 10. There are also several different suggested methods - and some early laboratory tests of some parts. So it may be to early to get a good estimate of the size. --Ulrich67 (talk) 12:12, 10 June 2012 (UTC)

Hi everybody. I'm new here and no native speaker. Translating LFTR into Polish. I find In solid-fuel reactors, it remains in the fuel and interferes with reactor control. under the Easy to control heading inconsistent with the next sentence. May I suggest to begin the next sentence like "Continuous LFTR salt processing in which Xenon-135 is removed, combined with the low reactivity..." etc? Pliftr (talk) 00:05, 9 June 2012 (UTC)

The problem is in the formulation with "this feature,..." - probably the sentence before that was added later and causes the confusion. I would suggest removing the sentence "In solid-fuel reactors, it remains in the fuel and interferes with reactor control" - there is not much new information here - most is said just 2 sentences before.

This article is not yet in a quality and stability really suitable for a direct translation. As you can see it is still frequently changing and contains some disputed parts and active discussion going on.--91.3.57.52 (talk) 09:03, 9 June 2012 (UTC)

Alternatively one could also delete the last sentence of the point, as it only repeats other points already mentioned. --Ulrich67 (talk) 12:25, 10 June 2012 (UTC)

Yes, major editing requires thorough cleanup. It lacked probably also in "Low mobility of radioactivity" where "MSFR" remained.Pliftr (talk) 16:15, 10 June 2012 (UTC)

This section contains a lot of wishful thinking: There is very little reuse of fission products in conventional reprocessing and the liquid salt form does not offer so much advantages. Most of the refs given are just for the properties of the materials. The 2 refs at the end really about use from fission products are rather low/no quality. Just a few weak points: Technetium-99 is actually a type of more problematic long lived waste - not valuable, though there is a very limited use to it. For separation the amount of actinides in the waste does not make a big difference, most of it is already removed in conventional reprocessing. The tricky part is to get the last few ppm out. So this section needs a cleanup or maybe removed altogether, as this is only a small side effect, and quite speculative.--Ulrich67 (talk) 20:33, 10 June 2012 (UTC)

There is only one low Quality source (company Website, giving this as possible future plan). Some sentences are even copies from this site and thus would likely be URV. Even as a Speculation this does not make much sense: The LFTR produces less (about half compared to other reactors) Xe-136. Neodymium is expensive because of the difficult and waste producing separation - so separation from the radioactive waste is extremely expensive (likely more difficult and expensive than the PUREX process). Where the noble metals end up is not clear - its only clear that they will not end up in the salt. Already the nickel filter is speculation. So reclaiming these elements is likely challenging. Getting these elements from LWR Spend fuel is likely simpler. Bismuth 213 is just a decay product of U-233. Technetium99m can not be used directly from the reactor, but needs the decay from Mo-99 to get reasonable purity and a little more time. So much of the current text is of no real use and I don't see a reason keeping it.--Ulrich67 (talk) 20:48, 22 October 2012 (UTC)

Hi ShotmanMaslo, I'm new, not a native speaker and translating the LTFR Wiki article to Polish. New also to Wiki. Please tell if I should better use "talk" at the article page or may I ask questions to someone like you? I'd appreciate every advice.

Well, since I am here already see my question:

I find "energy density" term in the "Economy and efficiency" section. It reads "The energy density is millions of times higher than any fossil fuel..." In my opinion such strong claim deserves a link to the Wiki Energy density article. Sadly thorium has only a minor remark there. Try to look also at the Polish version of the article and you will see that more detailed information appears in the table here (I added short English translation)

Fuzja deuter-tryt (fusion) 337 000 000

Rozszczepienie uranu (100% U-235) (U fission) 88 250 000 1 500 000 000 (MJ/L)

Naturalny uran (99,3% U-238, 0,7% U-235) w reaktorze powielającym (fast breeding) 24 000 000

Uran wzbogacony (3,5% U-235) w reaktorze jądrowym (Enriched U in reactor fission) 3 456 000

Naturalny uran (0,7% U-235) w reaktorze jądrowym (natural U in ractor fission) 443 000

It makes good comparison I think. But thorium is absent here as well.

Would you share my opinion that thorium entry should be added to both articles? What are the corresponding MJ per kilogram and MJ per litre values?

Regards Maciek 77.252.246.129 (talk) 16:13, 15 June 2012 (UTC) — Preceding unsigned comment added by 77.252.246.129 (talk) 16:09, 15 June 2012 (UTC)

Indented lineHi.

You can use the article talk page I think. If you find a good reference for thorium energy density, then it should be added, also into the english LFTR article. I have not found specific figures, but it would probably be comparable to U-238 in fast breeder reactors (24 000 000 MJ/L).

I'm deep in my translation and making some occassional remarks on the run. Can anybody help and find the needed values for me?

Pliftr (talk) 18:03, 15 June 2012 (UTC)

Is there really less activated waste with the LFTR ? The first question here is less than what ? The obvious choice here would be the normal LWR. To me this not at all obvious that the LFTR causes less activation, though it is hard to compare because there is no definite design, but a variety. Usually the LFTR has quite a lot of graphite in the moderator (e.g. about 300 m³ every 4 years in the ORNL MSBR design), then there is the reactor vessel, the whole primary circuit and much of the reprocessing. Recycling the graphite moderator is likely not possible, since the LFTR needs a high quality graphite. So my estimate would be that the graphite alone would be more activated material than with normal LWR operation. Anyway the amount of activated material is more an economic issue - less safety relevant. --Ulrich67 (talk) 16:55, 25 June 2012 (UTC)

Dr. David Leblanc has figures on parasitic captures in LWRs and CANDU, compared to DMSR. They are 22%, 12% and 5% respectively. Graphite absorbs very few neutrons compared to the fuel claddings used in LWRs and CANDUs, and the undermoderated region results in reduced neutron leakage. With less metal inside the core, and no metal cladding that must be replaced with new fuel every year or two, the amount of activated waste is much reduced.

http://thoriumenergyalliance.com/downloads/TEAC4%20presentations/LeBlanc_TEAC4.pdf

Siphon06 (talk) —Preceding undated comment added 13:54, 2 July 2012 (UTC)

This still is only a comparison of the neutron loss. The biggest neutron loss in the LWR is conversion H to D, which is not activated waste. The CANDU has quite a lot of internal structure (pressure tubes) - so it is not obvious that in the LWR, the neutron loss other than to H is more than 5%. It can be still possible that one type of LFTR can create less activated waste, but this is still unclear how to compare - by volume, by activity, by radio-toxicity or how else. In any way this would need a Ref. and preferably a clearer formulation. --Ulrich67 (talk) 19:41, 2 July 2012 (UTC)

The CANDU has 12% parasitic captures and almost no H (in stead has D2O), whereas the DMSR has 5%. Therefore the CANDU has much more neutron capture than the DMSR, even though graphite absorbs more than heavy water. Which is proof that there is less activation in the LFTR. — Preceding unsigned comment added by 83.138.2.141 (talk) 12:17, 6 September 2012 (UTC)

So we have a low quality reference that shows that a CANDU reactor has more parasitic neutron loss than a proposed LFTR. This still is only a indirect hint that makes the claim likely. With the standard LWR the question is still very open: The Shippingport LWR was able to reach breeding in the Th-U233 cycle: this indicates a very low loss of neutrons less than about 12%, including the loss to H2O). I see this as an indication that there is not much loss to the structural material. So the claim is far from obvious and needs a much better reference.--Ulrich67 (talk) 16:43, 6 September 2012 (UTC)

This section needs some cleanup: The argument of more efficient fuel use is repeated several times, with sightly different aspects. There also seems to be NPOV issue, giving too much emphasis to some rather unimportant points (e.g. once the price for the Th is negligible low, the details don't matter any more). The ref. given to show that there are enough natural resources is rather old, and for Bismuth specified in many reprocessing schemes it even states that that there will be a shortage (even by about 2000) - for the whole point it is questionable if we need this point here at all. It is more a reply to earlier text that questioned the availability of Li or Be. The points of efficiency in fuel use and LFTRs are cleaner assumes 100% efficient reprocessing. As far as I found reprocessing is significant less efficient. The last point (from waste to resource) has only a very week source (filbe web page) for the main point - the other refs. are just for more or less obvious details. Finally there is the general problem with the whole "advantages" section: it is not clear what to compare. There are different suggested LFTR designs and different current conventional reactors and fuel cycle options. So it may be better to put the whole economy part outside, and formulate it in a NPOV way with positive and negative aspects in one section, even if it is difficult as most of the available sources are primary sources and thus likely biased.--Ulrich67 (talk) 20:24, 1 August 2012 (UTC)

Yes, POV is an issue and I have restored User:nneonneo's Advert tag. Also agree that, per NPOV, we need to have discussion of positive and negative aspects integrated together. Johnfos (talk) 00:15, 2 August 2012 (UTC)

What all these wiki thorium articles are lacking is the observation that the cost of the fuel is hardly ever the major cost factor in the operating costs of a nuclear power plant; fuel cost is 10%-15% of the operating costs for a Uranium-based plant . Contrast with 50% for coal-based and 90% for a gas-based plant . Even if Thorium fuel were absolutely cost free, it won't make that much of a difference in operating costs... 86.127.138.234 (talk) 13:30, 29 January 2015 (UTC)

I notice in the above comments that the issue of POV or non-neutrality has come up quite a bit with this article and I would agree that the article is not balanced, in that it advocates the use of the LFTR.

In terms of the layout, the recent GA reviewer has said that the long-list format doesn't work very well. If you look at WP:LAYOUT, then you will see that bullet points are only suitable for lists of short items, unlike in the article at current. Instead, the sections should be divided into groups and each point presented in a paragraph. Moreover an integrated discussion is to be preferred to WP:Pro and con lists.

As the GA reviewer said, per WP:Verifiability, every controversial statement needs a reference. This is pretty much every statement, unless it is patently obvious. This means that almost all of the article requires a citation. There are far too few references provided in the article at the moment. Unsourced material may be challenged and removed at any time.

Unfair as it is, that the LTFR was cancelled for its uselessness in bomb-making, nevertheless a good deal of the LTFR advocacy slights the FBR solution, and this Wikipedia article seems to echo that. In particular, the widespread horror of "long-lived transuranic" products is a persistent feature of fanatically anti-nuclear propaganda, and I see no reason to endorse the idea at all. U-233, while not quite a transuranic, is certainly not naturally-occurring, and is longer lived than Pu-239. The fast neutron breeder Integral Fast Reactor, cancelled in 1994 because it was designed to generate Plutonium, was as meltdown-proof and as self-sustaining as the LFTR. It also consumed, with its fast neutrons, its own transuranic products. The plutonium produced was just as useless for bomb-making as the LFTR's uranium.^[1]cite web|website=http://www.pbs.org/wgbh/pages/frontline/shows/reaction/interviews/till.html}}</ref> DaveyHume (talk) 19:49, 10 May 2015 (UTC)

WP:Linkrot is an issue for many of the references. Citations require details that allow others to verify the statement. Given than weblinks sometimes break, this means providing things like the page's title, date, author, work and publisher. If you use a template like {{cite web}} for web links, then the fields should provide a guide as to what it is necessary to include (not all the fields, but at least some).

A smaller number of high quality external links should be used, avoiding advocacy links.

I have added some tags which reflect these comments and hopefully should provide guidance. Johnfos (talk) 00:18, 2 August 2012 (UTC)

A agree with changing to a more integrated discussion - I already started, by moving the part on reprocessing outside. However this part needs a major cleanup or rewrite, not just more references. For good reasons here much less information is available, and finding a neutral point position likely causes some discussion. Other topics that may deserve a separate section may be waste, economy, and safety - with pros and cons. So we may not need the section advantages any more, and the Design challenges could be reduced to something like research to do.

Just small changes and additional Refs. for the tagged point will not solve the structural problem of article. For now the references with just weblinks is not good, but still one of the smaller problems - this could be fixed when a reasonable stable version is there. A bigger problem is that some of the sources are of low quality, especially the slides an videos from conference talks. --Ulrich67 (talk) 17:23, 3 August 2012 (UTC)

Agree that changing to a more integrated discussion is the main thing, and thank you for making a start on this. Also agree that some of the sources are of low quality, especially the slides an videos from conference talks, and these need to be upgraded or removed. Johnfos (talk) 21:31, 5 August 2012 (UTC)

The references given for the point slow heatup show data on the specific heat capacity, not much more. However the specific heat capacity is only one point, and the difference is not that big, comparing to a classical LWR reactor: The fuel salt itself may have a slightly higher heat capacity, but there is much more moderator than fuel in both cases - as the LFTR tends to be more compact, expect it to have the smaller capacity. Anyway, the difference and advantage of the LFTR is more in the larger allowed rise in temperature: In a water cooled reactor only limited temperature excursions are allowed before the pressure rises to much. The salt temperature cold rise several 100 K, before immediate catastrophic occurs. The amount of heat that can be absorbed is specific heat capacity * mass * temperature rise. The LFTR wins because of the last factor. So at best this point is unclear and missed a ref. for something else than details. --Ulrich67 (talk) 21:53, 26 September 2012 (UTC)

There are still some unreliable sources: Slides from talks are usually self published work. In addition they may give a very simplified often exaggerated view with the talk itself missing. In this way videos may be a little better, but still this is not a good source. I don't think we should rely on such sources. At least we should keep them marked as less reliable. The videos are also candidates for future dead links. --Ulrich67 (talk) 22:23, 15 October 2012 (UTC)

I wrote a lot of the initial article, and I have some suggestions for sources. The best references are probably the U.S. government reports. The basic references for the MSRE, the last real laboratory research on LFTRs, are "The Use of Thorium in Nuclear Power", WASH-1097 (1969) and H.E. McCoy, and B. McNabb, Intergranular Cracking of INOR-8 in the MSRE, ORNL-4829 (1972), Technical Report, Springfield, VA, USA. I believe both are on the web at http://www.energyfromthorium.com/pdf and they're also available from U.S. National Technical Information Service, U.S. Dept. of Commerce. The 1969 report, WASH-1097, discusses the MSRE, and seems to indicate that it was killed because of corrosion. ORNL-4829 summarizes the successful corrosion research completed in 1972. Most of the "unreliable" web sites are discussing implications of these two reports. Ralph Moir has published a number of more recent, peer-reviewed papers, and made them available at http://www.ralphmoir.com/fission/. One of his most interesting papers is: Moir, Ralph W. "Cost of Electricity from Molten Salt Reactors (MSR)." Nuclear Technology, 2002: 138 93-5; He and Edward Teller also published a survey article about LFTR advantages. One of the best ways to cope with LFTR wastes is to convert them to iron-phosphate glasses, which are less soluble in ground water than borated glasses like Pyrex. A good reference for that is: Darryl D. Siemer, "Molten Salt Breeder Reactor Waste Management", Nuclear Technology, volume 185, no 1, January 2014 p100-108. (This is more laboratory work.) Ray Van De Walker 06:24, 11 January 2014 (UTC)

Under Advantages there is one Point "Excellent heat transfer". Here the ref. ORNL-TM-2006-12 (currently number 51) does not support the claim that the salt gives better heat transfer than water. It gives the opposite: water being better than molten salt (e.g. lower figure of merit), even for pure Filbe, and more for other salts listed. The comparison is difficult anyway, as water and the salt can hardly work at the same temperature, and the salt may allow for a higher temperature difference. However if the temperature is lowered, the viscosity of the salt increases quite fast and makes heat transfer more difficult at temperature much below 700°C. The direct comparison of thermal conductivity and volumetric heat capacity is somewhat misleading - viscosity and density are also important factors, this time favoring water. Also the addition of ThF to the salt is expected to have a negative influence (higher viscosity, higher density, less thermal conductivity and lower volumetric heat capacity).--Ulrich67 (talk) 21:50, 3 December 2012 (UTC)

I have finished my Polish version of the LFTR article in the end and uploaded it to the Polish Wiki as "Reaktor torowy na ciekłych fluorkach". This is my first Wiki article. It has been reviewed and is now available for the Polish readers. Please advise if I may add interwiki links

en:Liquid fluoride thorium reactor to the Polish version and

pl:Reaktor torowy na ciekłych fluorkach to the original article?

Pliftr (talk) 18:12, 30 May 2013 (UTC)

I just read through this article, and I wanted to post some feedback. It seems to have a large base of contributors, which is good. However, for how long the article is, it could certainly be more concise. Specifically, each disadvantage listed has one or more counter-arguments listed for how that disadvantage could be overcome. It is really quite tedious to read both subsections of 'Advantages' and then get into the 'Disadvantages' only to be reading a third section of advantages. It certainly does not seem to be written even-handedly. 64.114.134.52 (talk) 22:02, 7 June 2013 (UTC)

That's basically because a few of the "disadvantages" actually belong in the advantages section. When someone with a lay education in the nuclear field hears the word "nuclear" or "uranium", they begin to worry that this could lead to things like nuclear proliferation. However, in this case, it is rather difficult to use this technology for nuclear weapons (continuing with the example), so much so that it effectively can not be. Which is why it is an advantage (and not a disadvantage as a layman may think it is). Former nuclear engineer → — al-Shimoni (talk) 01:20, 2 July 2013 (UTC)

Using long lists of advantages and disadvantages is problematic since things are usually not that simple, that one has clear disadvantages / advantages. Especially for points that are problematic and thus disadvantages at first, there are special designs needed to more or less deal with the possible weaknesses - in some cases causing new problems. Some topics like proliferation are just not that clear - there are both pros and cons, with different sources giving more weight to one side or the other. So these controversial parts should better have a separate section outside the pro and con list.--Ulrich67 (talk) 20:55, 26 July 2013 (UTC)

There is this point under Economic advantages. There is a Ref. given - however the contend of this section is not supported by the two Refs. - more of the opposite: corrosion is still an open question and more research and tests are needed. So I Suggest removing this point all together - most of the corrosion problem is already handled better further down under disadvantages (with the same Refs.). The numbers given for the replacement graphite are also of rather limited value. They are rather old cost estimates of the new material only - the bigger issue with graphite replacement is the waste and downtime and provisions needed to do the replacement.--Ulrich67 (talk) 11:41, 22 June 2013 (UTC)

The MSRE had no problems with corrosion from the salt. This is documented in WASH-1097 (1969), the MSRE's main research report. It had problems with a tellurium fission product slowly dissolving the grain boundaries in the nickel alloy, "Hastelloy-N." The corrosion was slow because there wasn't much tellurium in the salt. The tellurium corrosion was researched and fixed after WASH-1097 was published, which I think may be causing the confusion. The research showed that a bit of niobium in the hastelloy fixes it. This is discussed in a later research report, McCoy, H.E., and B. McNabb. "Intergranular Cracking of INOR-8 in the MSRE," ORNL-4829 (1972). Technical Report, Springfield, VA, USA, or http://www.energyfromthorium.com/pdf/ORNL-4829.pdf: An authoritative publisher is: U.S. National Technical Information Service, U.S. Dept. of Commerce.

The summery (p. 170-173) of ORNL-4829 confirms that there was some corrosion in MSRE. Weather this corrosion is serious or just superficial is hard to say: it's just a few mils deep most of the time, but cracks in a brittle material are serious even if small. At least ORNL took it rather serious and did quite a lot of follow up. Still they say even more test are required to be sure. There was some improvement after wash-1097, but far more research is needed and this is far from the level of confidence that exists in water based reactors. So adding niobium may help, but it is not sure the corrosion problem is solved.--Ulrich67 (talk) 19:15, 12 January 2014 (UTC)

This point under advantages is a little off topic for the LFTR: It may be possible to use some SNF components as a fuel in a MSR, but not in LFTR as described in this article. Also this is only a still rather vague suggestion with may open points. As the LFTR as a thermal reactor with only marginal negative reactivity coefficients adding extra problematic elements is hard (it may work in small non breeding MSR with better safety) and the buildup of MA is no better (likely even worse because of slightly harder spectrum) than in a LWR. Also the LFTR has very few spare neutrons, and adding waste with MA tends to reduce breeding even further. The real option for transmutation of MA is using a fast MSR , possibly even accelerator driven - this is rather different from a LFTR. So if at all (because this are just plans with may open points) this point should go to the more general MSR article.--Ulrich67 (talk) 08:05, 27 July 2013 (UTC)

moved the section here, in case someone wants to have something in the MSR article:

"*Destruction of existing long lived wastes. ^{[failed verification – see discussion]}LFTRs can use existing transuranic wastes for their initial fissile startup charge better than any solid fueled reactor for various technical and physical reasons.^[1] Because the fuel is a liquid homogeneous solution, it is always perfectly mixed, impervious to radiation damage and can accept any composition of plutonium, neptunium, americium and curium up to the solubility limit. Solid fueled reactors, such as solid fueled fast reactors, while theoretically outperforming the LFTR in burning of these higher actinides, can only accept limited amounts of these higher actinides (neptunium, americium and curium are often called minor actinides). This is because the fuel is not perfectly mixed, as it is confined in solid fuel elements, and also because the coolant void coefficient (coolant overheating) can become positive for too high levels of minor actinides.^[2] In addition, manufacturing solid fuels with high amounts of americium and curium is also difficult due to decay heat generation and helium production rates.^[1] As a result, solid fueled reactors usually only use reprocessed plutonium but do not use the americium and curium, which constitute a large portion of the radiotoxicity of the long lived waste.^[3]"

This whole article makes no reference to an estimate of the costs per kwh. Surely this is the most important point of any discussion of alternative energy sources!? Let's make that happen! ;-) 129.199.82.143 (talk) 10:50, 5 August 2013 (UTC)

The main developement is rather old (late 1960's). At that time there was a cost estimate at cost comparable to a conventional nuclear power supply. Due to may still open questions this is a rather crude estimate. In addition regulations have changed since then, and chemical reprocessing turned out to be much more expensive for solid fuel - so the old numbers are rather outdated too. It may be worth having a separate section on costs - though the conclusion is more ore less that the final costs are unknown and estimates are difficult so early in development. The difficulty is to find a current unbiased source.--Ulrich67 (talk) 10:01, 10 August 2013 (UTC)

Ralph Moir wrote a published, peer-reviewed article estimating LFTR costs. It was $0.038/Kwh, vs. $0.041 for dirty-style coal plants of the same era. This was based on comprehensive cost estimates done at Oak Ridge National Labs, which designed a full-sized LFTR power plant. This means that potentially they're cheaper than anything else out there. The article is well-known to LFTR aficionados, who often quote the conclusions without attribution. See: Moir, Ralph W. "Cost of Electricity from Molten Salt Reactors (MSR)." Nuclear Technology, 2002: 138 93-5, or see http://ralphmoir.com/fission, accessed 2014-1-10. Ray Van De Walker 05:14, 11 January 2014 (UTC)

The article of Ralph Moir only adjusted the old numbers for inflation. Doing this old, crude estimates don't get better this way - they are just outdated. The French programs suggest that fast reprocessing as planed in the MSBR is likely not feasible - thus too expensive to be competitive.--Ulrich67 (talk) 21:46, 7 March 2014 (UTC)

I agree with you on the prohibitive cost of MSBR-style processing. Do you have a reference for the French cost estimate/feasibility? markmassie (talk) 18:28, 8 March 2014 (UTC)

One ref. ist currently Ref 43 in the article. Slightly more details this in this article from the same group . There is no detailed cost estimate (it's just to early to give serious numbers), just mentioning that fast reprocessing may not be feasible and assuming a 6 mouth cycle as realistic. The best I could find as a cost estimate were some estimates on pyro-processing metallic fuel for a fast reactor - somewhat similar, but not that much. These numbers would also make a 10 day cycle to expensive and something like 6 months realistic.--Ulrich67 (talk) 19:44, 8 March 2014 (UTC)

Under Advantages there is this point. However for new nuclear designs passive emergency cooling is more or less a requirement. On first sight the high temperatures may help designing a passive cooling - however, the only ref. given for this section essentially only gives crude first ideas and at the end says that more research is needed to validate and understand these things. The need to limit cooling (to prevent uncontrolled salt freezing) and the requirements not to use water and keep oxygen out can actually make passive decay heat cooling more difficult than in a water based system. At least ORNL used a quite intricate active cooling system for the salt storage system of the MSER. So far the given Ref. is more supporting the lack of a proven concept for passive cooling - this is definitely not an advantage. So unless someone finds a good Ref. on this - this point should be deleted or even moved in revised form to difficulties.--Ulrich67 (talk) 18:07, 30 December 2013 (UTC)

The advantage is that the fluid decay products can be drained from the heat-conserving reactor to a scram tank, which can passively emit large amounts of heat. For example, reactors are generally large squat cylinders with minimal surface area. A large LFTR's scram tank can be an array of pipes or hollow plates with much greater surface area, and they can be submerged in meltable salts or water to diffuse the heat. Further, the MSRE had a clever passive method to self-scram, by blocking the drain with frozen salt. If the cooling failed, the reactor would melt the plug and self-scram. Solid-fueled reactors of whatever type can't do these things. References... WASH-1097, the official report, refers to the MSRE's drain tank, but does not discuss its geometry and cooling characteristics. Ray Van De Walker 05:34, 11 January 2014 (UTC)

Wash 1097 - does not mention passive cooling as this was not a requirement at that time. Some other refs mention passively cooled tanks to be planed, but don't give any details. The freeze valve is a different point, and just a fist step (not needed in other reactors) in decay heat cooling. The Ref. we have in this section suggests that development of passive cooling is still open. This may not be a big deal, but still its an area that needs development - so it's to early to say how good passive cooling works. In comparison, new LWR designs usually include a well developed passive decay heat cooling.--Ulrich67 (talk) 21:28, 7 March 2014 (UTC)

Rather more important than passive cooling per se is that the fuel is continuously cleaned of fission products, the main source of decay heating in solid fuel reactors. If you don't create the heat in the first place, then cooling it is not as high a priority. SkoreKeep (talk) 18:24, 26 August 2015 (UTC)

Removing the fission products does not change much: Usually chemical separation is not done very often - like 0.2-2 times a year, compared to a 3-5 year excange cycle in conventional reactors. So the main difference is from decay product with a half live of some 0.5-5 years - this is only a rather small part of the decay heat problem. Critical is usually the heat in the first weeks, and here the relativly long livetime of Pa-233 from the Th bredding cycly makes a real difference (about twice the decay heat from day 3-7).Ulrich67 (talk) 20:14, 30 August 2015 (UTC)

That seems like a pretty odd conclusion to come to, when the removal frequency is 30 times that of the water reactors - by your numbers). My understandig was the the chemical processing was planned to be done continuously, if for no other reason than to remove the neutron poisons. The fuel reshuffling that occurs in water reactors also doesn't get rid of all of the FPs, only about a quarter of them (if 1/4 of the fuel is new). That would make it about 120:1 (best case), 10:1 (worst case), which must make things easier on the storage tanks; presumably it would have had considerable impact on Fukushima if such had been possible. Yes, the short lived isotopes are the ones which cause most of the heating, so that would reduce the advantage. And then, of course, that is just that reactor design. But I'm probably not as up on this as you are, so take me with a grain. SkoreKeep (talk) 21:28, 30 August 2015 (UTC)

There are many different designs for the LFTR. The early ones assumed something like 10-20 days to remove the FP and even Pa. Later ones assume more like 100-200 days for cleaning, with only removing a part of them. So the avarage time most FP stay in the reactor can range from 100 days to many years. FP removed as gas will stay in the gas handling system, likely for a long time. In a conventional reactor FP stay on average in the reactor for something like 2-3 years (about half the fuel life). So there is not a big difference in decay heat in the reactor, it depends on details that are still open. The inital MSBR design would have less FP in the reactor, but it turned out that so much chemical prcessing is likely not feasable.Ulrich67 (talk) 21:03, 8 September 2015 (UTC)

Thanks for the explanation. I appreciate it. SkoreKeep (talk) 21:49, 9 September 2015 (UTC)

The fragment "four times greater abundance in the earth's crust than" appears in the article. This is ambiguous. If it is five times as abundant as, then let's say so. If it is four times as abundant as, then let's say that.

The fragment "20x smaller than" also appears. What the hell is that supposed to mean? 1/20 as large as?

Jack Waugh (talk) 22:27, 23 April 2014 (UTC)

I don't see the ambiguity. The sentence could be rewritten like "thorium is four times more abundant than uranium in the earth's crust", but the meaning is still the same. — Preceding unsigned comment added by 105.210.21.243 (talk) 13:41, 30 April 2015 (UTC)

This article is a few years old. Has anyone created a working Thorium reactor since then? Larsumms (talk) 07:56, 7 May 2015 (UTC) Larry Summers

Hello fellow Wikipedians,

I have just modified 2 external links on Liquid fluoride thorium reactor. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Corrected formatting/usage for http://home.engineering.iastate.edu/~pjscott/Sorensen_Google_LFTR.pdf
• Corrected formatting/usage for http://home.earthlink.net/~bhoglund/mSR_Adventure.html

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—^{cyberbot II}_{Talk to my owner:Online} 06:07, 1 April 2016 (UTC)

The lede currently has nothing about the status of development, proof-of-concepts, nor power production projections, so a bit hard for the casual reader to get a "read" on the status of this technology on the "theoretical paper" to "lab experimentation" to "proof-of-concept implementations" to "small-scale production of the nn MW class" to "production reactors" anticipated in 2023.

Obviously, any of that could only be said if info is verifiably sourced. But it does seems the lead could use a good summary of where this tech is on the long-term development path, if such sources exist, even if it is only still theoretical and early lab stage stuff. Cheers. N2e (talk) 20:09, 8 January 2017 (UTC)

Hello fellow Wikipedians,

I have just modified 2 external links on Liquid fluoride thorium reactor. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20140808054526/http://www.inl.gov/technicalpublications/Documents/4502650.pdf to http://www.inl.gov/technicalpublications/Documents/4502650.pdf
• Added archive https://web.archive.org/web/20160310143100/http://www.epri.com/abstracts/Pages/ProductAbstract.aspx?ProductId=000000003002005460 to http://www.epri.com/abstracts/Pages/ProductAbstract.aspx?ProductId=000000003002005460

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 01:09, 24 December 2017 (UTC)

You may read the criticism on the forum. These were the most criticised claims on Disadvantages:

1. Highly questionable economics
2. Reaching break even breeding is questionable

Someone may want to check those and correct possible errors. —Nikolas Ojala (talk) 18:32, 27 July 2018 (UTC)

Hi Nikolas.

to Point 1.

Highly questionable economics: there is no doubt that there is no nuclear reactor who was ever economical successful. Can it be that a Atomic Reactor will be economical successful in the future ? No, there is no Insurance Company who can ever cover the risk of an nuclear reactor due to the enormous cost if a nuclear reactor had an accident eg. Chernobyl up to 2016 700 billion Dollar or Fukushima 260 billion Dollar and this are not the biggest possible accident (thermonuclear explosion) and the cost running at least for the next 100 to 1000 Years. So the government need to take the risk of a nuclear reactor, with out Government Sponsoring there is no nuclear reactor hence it is per definition Impossible that any nuclear reactor ever can be economic successful... Eliot128 (talk) 23:59, 17 January 2025 (UTC)

Only as an addition... The yearly insurance cost would be between 5 to 25 billion Dollar for a nuclear power station. Eliot128 (talk) 00:16, 18 January 2025 (UTC)

Hello Guys.

I am wondering that this article never mentioned the chemistry behind it ? Thorium(IV) fluoride (ThF4) is a Fluoride of Thorium maid like ThO2 + 4HF -> ThF4 +2 H2O so if there is any hydrogen in the near the ThF4 will also deca (chem.) into Th X and HF what is fluoride acid. This Reactor Can't be practically build due to the extreme corrosive property of HF, the high temperature and the radioactive stress to the existing HF proof materials. The LFTR will literally dissolve itself into an ugly pile of black goo.

Any Idea how we can mentioned this with out leading the article ad absurdum ? Eliot128 (talk) 20:15, 17 January 2025 (UTC)

The Chinese thorium reactor is now fully operational…e.g. see https://engineerine.com/worlds-first-thorium-reactor/ 50.53.9.147 (talk) 21:44, 27 April 2025 (UTC)