Energy From Thorium Discussion Forum

It is currently Aug 18, 2018 1:54 am

All times are UTC - 6 hours [ DST ]

Post new topic Reply to topic  [ 26 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Sep 14, 2012 3:08 pm 
User avatar

Joined: Nov 30, 2006 3:30 pm
Posts: 3731
Location: Alabama
Last year several of us in the thorium community had an opportunity to review early versions of a series of reports that were being drafted by the National Nuclear Lab (NNL) in the UK for their Department of Energy and Climate Change (DECC). These reports were a major part of our discussions with DECC personnel when Kirk Dorius and I visited the UK last September.

DECC: Advanced Reactors and Fuel Cycle Reports

The first of these is a discussion of 42 metrics used in the analyses of advanced reactors. I reviewed these and in my opinion thought that the list could be consolidated, in that many of them seemed to be slightly different shades of the same idea.

The second paper scores each of the advanced reactor types against the 42 metrics and creates numerical rankings for each of the reactor types. Without going into too many details, the scoring for the MSR option improved considerably between the earlier draft and the one released yesterday. I think the reviewers made a case for a more favorable consideration of the MSR. You can see those scores that changed because they are in a dark blue color rather than the black of the original text.

The third paper is a review of the thorium fuel cycle, with a heavy emphasis on solid-fueled reactors. It is basically an augmented version of the infamous NNL "white paper" written by Kevin Hesketh. I wrote a series of comments on this paper that do not appear to have led to any alteration in it.

Drafts of these three papers were the ones I reviewed in the fall of 2011, along with other reviewers. We submitted comments and suggestions back to the DECC, who presumably submitted them back to NNL.

Two newer studies were also released yesterday.

The fourth study is an addendum to the first paper, and it appears to be a response to the issue of metrics that are duplicates and overlaps. The 42 metrics are reduced to 33 metrics grouped under seven headings: Generating cost; Inherent Proliferation Resistance and Physical Protection (PRPP); Safety; Strategic; Deployability; Sustainability and Waste. Weighting factors for each metric are also assigned and rationalized. Weightings were something I had suggested in my review in order to help emphasize certain metrics of greater importance. The original NNL analysis had flat weightings (every category had the same value in the final score).

The fifth study is an addendum to the reactor scoring, using the newer set of metrics along with a set of weightings. There was no global scoring done here, but MSR did quite well in many of the categories.

These studies have been underway for quite some time and it is a bit of relief to me that they are finally now public. We can review and examine them, and help shine a light on sections where MSR may not have been given a fair shake (and there are quite a few). But I have been very impressed with the people I have worked with in DECC and their efforts to make sure this was done fairly and that outside opinion was considered. Otherwise, this would have represented only what the NNL knew (or didn't know) about the reactor concepts.

PostPosted: Sep 14, 2012 4:17 pm 
User avatar

Joined: Dec 24, 2011 5:25 am
Posts: 3
Location: Cypress, Tx
I have been wandering about this: "Larger systems such as SFR, GFR and MSR may not be as straightforward to dismantle and would be disadvantaged.", this was in one of the reports of NNL. Without I have been so much immersed on MSRs, it seems that this may be right. If somebody have any online information about this point please share, I'd like to read.

PostPosted: Sep 15, 2012 8:50 am 

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
jfernand57 wrote:
I have been wandering about this: "Larger systems such as SFR, GFR and MSR may not be as straightforward to dismantle and would be disadvantaged.", this was in one of the reports of NNL. Without I have been so much immersed on MSRs, it seems that this may be right. If somebody have any online information about this point please share, I'd like to read.

My own analysis of decommissioning costs suggests the opposite: decommissioning costs have huge economy of scale. Dismantling a 500 MWe reactor costs almost as much as dismantling a 1000 MWe reactor. See for example this IAEA report: ... 783942.pdf

The two HWRs decommissioning costs are directly comparable as they are both HWRs in Canada. The 880 MWe reactor cost only 20% more to decommission as the 400 MWe reactor.

PostPosted: Sep 15, 2012 7:40 pm 
User avatar

Joined: Dec 29, 2011 10:14 am
Posts: 217
And leave it to the news media to create division: ... overstated

Reading the comments is a mixture of amusement and bafflement. Who is feeding this kind of "miss-information" to the public? Inquiring minds want to know... who exactly are these big bad nuclear companies plotting their next big heist using thorium? Are they the same ones with the black helicopters from Area 51?

PostPosted: Sep 16, 2012 10:33 am 
User avatar

Joined: Sep 10, 2008 7:40 pm
Posts: 299
I sent this email message to David MacKay at UK DECC.

David MacKay,

I have read the report Comparison of thorium and uranium fuel cycles, ... ycles.pdfl and have the following comments.

In general the report is clear and accurate.

Section 4.3 Economics deals just with the use of thorium in LWRs, which is only of marginal economic value.

Section 4.5 Proliferation states "Attempts to lower the fissile content of uranium by adding U-238 are considered to offer only weak protection, as the U-233 could be separated relatively easily in a centrifuge cascade in the same way that U-235 is separated from U-238 in the standard uranium fuel cycle."

However, the contaminating, gamma-emitting U-232 makes maintenance of U-233/U-238 separation centrifuge arrays very much more difficult, so the words "relatively easily" are quite incorrect. Actually the single-fluid DMSR (denatured molten salt reactor) is the most proliferation-resistant design of all nuclear reactors.

Section 4.6 Reprocessing refers to reprocessing of solid fuel. The THOREX process would not be used in LFTR (liquid fluoride thorium reactor). The fluoride volatility process would be used in a LFTR. The DMSR or any one-fluid molten salt reactor would not remove the U-233 from the fuel salt at all.

Section 6.2 VHTR (Very High Temperature Reactor) discusses TRISO fuel. High temperature reactors, including thorium molten salt reactors, have advantages of more efficient electric/thermal power conversion and applicability to industrial processes such as water dissociation for hydrogen production. UC Berkeley (with others) has designed a PB-AHTR (pebble bed advanced high temperature reactor) that is molten salt cooled. The TRISO pebbles can also be fabricated with thorium, as in the case of the German THTR-300.

Section 6.4 MSR (Molten Salt Reactor) states the technological readiness is (indeed) low. However the challenges of fuel processing for U-233 separation and fission product removal are largely bypassed in single-fluid MSRs such as DMSR.

Section 8 Discussion states "Innovative thorium fuelled reactors will not be a viable alternative for at least 20 to 30 years.." is true only if countries like UK and USA continues cumbersome regulatory and development processes. I personally advocate a 5-year pilot-plant development program with LFTR production in 10 years. Rickover built the first US commercial power reactor in 39 months. China is building a prototype PB-AHTR by 2015 and an MSR by 2017. The timescale for MSR development depends on the perceived criticality of the world climate crisis, the need for energy to support economic growth, and the need to achieve nations' energy security to avoid natural resource contention and causes for war.

I'm going to China next month to see what's happening there.

Robert Hargraves

PostPosted: Sep 17, 2012 3:07 am 

Joined: Sep 15, 2011 7:58 pm
Posts: 186
robert.hargraves wrote:
Section 4.5 Proliferation states "Attempts to lower the fissile content of uranium by adding U-238 are considered to offer only weak protection, as the U-233 could be separated relatively easily in a centrifuge cascade in the same way that U-235 is separated from U-238 in the standard uranium fuel cycle."
I must have missed a memo. If that constitutes only weak protection, exactly what is strong protection? I was under the impression that obtaining the fissile material is the hard part, and the rest is comparatively child's play. I was also under the impression that it's not terribly difficult to obtain raw uranium ore, and that the difficult part is to centrifuge it. How is this anything but intellectual dishonesty, ignorance, or a gross accidental error? Perhaps it makes more sense in context?

PostPosted: Sep 17, 2012 9:09 am 
User avatar

Joined: Nov 30, 2006 3:30 pm
Posts: 3731
Location: Alabama
Here are some of my comments on the metrics where MSR scored particularly low in the NNL assessment:

Metric 9: Sabotage resistance assessment
Highest ranking: VHTR, V HIGH (4)
MSR ranking: LOW (1)

The notes accompanying this metric state "this analysis concentrates on assessing the sabotage resistance of the reactors."

Much like the discussion of spent fuel characteristics, the physical aspects of a LFTR argue strongly against any attempt at sabotage. The entire primary loop of a LFTR would be inside a stout containment structure that is not designed to be opened—quite unlike a solid-fueled reactor whose containment is designed to be opened and whose reactor vessel much be opened for refueling operations. The entire primary loop is also protected by high temperatures and strong radiation fields. Any attempt to sabotage the reactor would have to contend with the formidable challenges of the reactor design itself. Attempting to reroute fluid fuel lines or to "tap" fuel would simply be impossible in the temperatures and radiation fields of the reactor.

Perhaps the goals of the saboteurs is not fissile material removal but instead the release of radioactive material to terrorize the public. Here again they would be thwarted by the physical aspects of the reactors. The reactor operates at low pressure, so there is no pressure term to release through a compromised pressure vessel. The fuel is fluid and freezes at temperatures below 400C, so even an intentional breach of the reactor vessel, depending on its size, could range from a situation where the frozen salt "plugs" the breach to a situation where the liquid salt runs down the side of the reactor, drips into the catch basin, and drains into the drain tank, as it would under a standard shutdown.

Gas-cooled reactors and water-cooled reactors have to contend with a loss of pressure, fluoride reactors do not.

Metal-cooled reactors do not have pressure concerns but must contend with the chemical energy release potential of liquid-sodium and metal fuel exposed to air or water. A catastrophic release of chemical energy is possible, leading to the mobilization and release of radioactivity assuming that the containment has been sabotaged.

The fluoride reactor has a chemically-stable fuel form that will not react vigorously with air and water. Thus catastrophic release is not possible.

A more accurate representation of the "spent fuel characteristics" would be to ask whether the fissile material in the reactor is suitable for diversion to clandestine purposes, and by this metric the uranium-233 fuel in a LFTR, strongly contaminated by uranium-232, is a very poor candidate for diversion. Indeed, there are 70,000 nuclear weapons in the world and none are based on uranium-233 or thorium. All of them are based on uranium-235 and plutonium-239. The LFTR, with its high temperatures, strong radiation fields, and inherently unattractive fuel for diversion, deserves a much higher score on this metric.

Metric 10: Reliability assessment
Highest ranking: VHTR, V HIGH (4)
MSR ranking: MED (2)

Reliability is a difficult metric to assess for any advanced reactor concept in the absence of significant operational experience, nevertheless, the arguments put forward as to why the VHTR should have a V HIGH reliability assessment also apply to the LFTR. LFTR also uses a closed-cycle gas turbine power conversion system, that unlike the VHTR does not depend on flowing the working fluid gas of the power conversion system through the core and pressurizing it to high pressures. This gives the LFTR configuration safety and simplicity, as well as the ability to employ gas-turbine reactor design techniques like multiple stages of reheat that improve both system efficiency and system net work (work per unit mass of gas flow).

Reliability should actually be higher in a LFTR than a VHTR because the LFTR will not have to shut down for refueling of the reactor. The LFTR is continuously being refueled by the uranium-233 generated in its blanket and extracted through fluorination and then reduced to UF4 in the core salt. A VHTR must shut down for solid fuel removal, addition, and reshuffling. With the expected shutdown times typical of all other solid-fueled reactors, it is doubtful that a capacity factor higher than 90% can be achieved. In order to improve the time between refueling, longer core lives will have to be pursued for VHTR. This implies higher enrichment levels, higher development costs, and a longer development schedule to assess the irradiation performance of a high-enrichment long-lived fuel.

LFTRs on the other hand have a fuel form that is impervious to radiation and they maintain their reactivity through the continuous addition of new fissile material generated in their blankets. They effortlessly reject the most significant of the fission product poisons, xenon-135, through the natural off-gassing of fission product gases in the primary salt pump. This effect was conclusively demonstrated in the successful five-year operation of the Molten-Salt Reactor Experiment from 1965 to 1969. Despite being simply an experiment, that reactor showed very high reliability in operational modes that are more similar to a LFTR than previous gas-cooled reactors have been to a VHTR.

Metric 11: Radiological exposure assessment
Highest ranking: VHTR, V HIGH (4)
MSR ranking: MED (2)

The MSR technology is being penalized inaccurately and unfairly on this metric due to its ability to remove fission product gases during operation. Rather than being a disadvantage, this is a great advantage in reliability, safety, load-following control, and radiological release. Solid-fueled reactors depend on a continuously degrading fuel form to contain fission product gases. In an accident scenario this barrier fails very quickly, as we saw at Three Mile Island and Fukushima. It is doubtful that a sober consideration of safety would rank the solid fuel form as an effective barrier against radiological exposure.

A LFTR, on the other hand, continuously removes and addresses its fission product gas inventory. Unlike solid-fueled reactors, the offgas system is designed to handle the entire inventory, not simply the small amount from a few failed elements. We see the significance of this in accident scenarios.

Radiological exposures are minimized in the LFTR because unlike solid-fueled reactors there is never the need to shut down, depressurize the reactor, remove, reshuffle, and replace solid fuel. The issue simply doesn't exist. The entire core fuel inventory is homogeneous in composition. Reshuffling is meaningless in a fluid-fueled reactor. The reactor is continuously refueling itself through the operation of the blanket and the fluorination/reduction processing system. There is no need to add fuel or remove fuel. Since refueling of a solid-fueled reactor is the primary pathway to radiological exposure of workers, this is a very important feature.

It is puzzling why a solid fueled reactor like a VHTR would receive a V HIGH ranking on this metric when it has the need for regular refuelings but a reactor that does not have this requirement like a LFTR receives a MED ranking.

Metric 12: Safety
Highest ranking: VHTR, V HIGH (4)
MSR ranking: MED (2)

The VHTR is ranked V HIGH (4) for "Very robust fuel form" while MSR is ranked MED (2) for "Safety approach remains to be developed". The safety case of any reactor is based around the likelihood that operating personnel or the public will be harmed by the operation of the reactor or by an accident scenario. MSRs and LFTRs in particular have a safety case that is vastly better than any solid fueled reactor and certainly much better than the VHTR that received a V HIGH rating on this topic.

There are several bases for this argument:
1. Absence of stored energy sources (pressure, chemical reactivity, etc.)
2. No excess reactivity
3. Strongly-negative temperature coefficients of reactivity
4. Totally passive decay heat removal
5. Fuel form that does not undergo structural or chemical changes
6. Chemically stable environment for dangerous fission products (strontium, cesium) and continuous removal of volatile fission products (xenon, krypton).

The same fission rate generates the same amount of thermal power and the same production rate of volatile fission products (xenon and krypton) in an MSR as in any other reactor. But MSRs and LFTRs are better on this metric because they need less thermal power to generate a given amount of electrical power due to their higher operating temperature and superior thermal efficiency. Even setting that aside, in a solid-fueled reactor volatiles are inside the fuel, where they are cracking the fuel, causing the fuel to swell, corrupting its reactivity, reducing its conversion ratio and sustainability, and compromising its load-following capabilities. This report appears to consider that a virtue, enhancing the safety profile of the reactor. My opinion, backed by reality, is that it is not a virtue to have xenon and krypton trapped in solid fuel, rather it is a great drawback and should not be counted as a safety advantage of a solid-fueled reactor.

By contrast, the volatiles in a LFTR are not in the fuel. They are continuously removed and addressed by the off-gas system, which was demonstrated successfully to run for over four years in the Molten-Salt Reactor Experiment. In a LFTR, the xenon doesn't corrupt the reactivity, doesn't reduce conversion efficiency, and doesn't inhibit load-following capabilities. Xenon and krypton don't cause swelling and cracking issues in a liquid-fuel form—they come right out of solution.

Even more importantly, in fluoride chemistry, the very significant radioisotopes of cesium are no longer volatile at all! Rather they are incredibly stable and non-volatile fluorides. So fluoride-based MSRs have LESS volatile inventory than a solid-fueled reactor AND they address that volatile inventory continuously as a course of normal practice rather than assuming that the solid-fuel is the desirable location for fission product gases.

The real issues for safety have to do with decay heat management and reactivity control. These have been the bases for the real nuclear accidents that have taken place in the world. Three Mile Island-2, Windscale, and Fukushima were all related to decay heat management, and Chernobyl and SL-1 were accidents related to reactivity control.

Compounding both of these issues is the operational state of the reactor, specifically whether or not the reactor operates at high pressure. High pressure compounds the management of decay heat removal because emergency core cooling systems (ECCS) in water-cooled reactors must operate at a variety of different pressurization conditions in the reactor, from fully pressurized to fully depressurized to every state in between. Gas-cooled reactors, which are the general category that includes the VHTR that received such a high score in this metric, operate at very high pressures because of their low heat-transfer effectiveness. This essentially condemns them to operate at low core power density and low economic effectiveness. Gas reactor advocates try to promote this deleterious economic condition as a virtue by pointing out the slow response time of a reactor in the case of a transient and the large thermal mass of the system, but all of these are bought at a price.

Decay heat removal in nearly all MSRs and certainly in LFTR is benefitted tremendously by the low operating pressures of the reactor. There is no "driving" term seeking to move radioactivity out of the reactor. No depressurization issue to contend with like in a light-water reactor or a gas-cooled reactor.

Furthermore, if primary heat removal is lost, only fluid-fueled reactors like MSRs and LFTRs can "downshift" into a completely different core cooling configuration.

It is absurd that a high-pressure light-water reactor could have a HIGH rating on this metric while MSR has MED. MSRs and LFTRs in particular address all of the safety issues of nuclear reactors by removing the issues from the design. In addition, they do not have a higher volatile inventory, rather they have a lower one. On this all-important metric, it is clear that both the MSR should be assigned a safety rating of V HIGH (4).

Metric 15: Low uncertainties on dominant phenomena assessment
Highest ranking: SFR, VHTR, Small LWR, LWR once-through, LWR recycle, HIGH (3)
MSR ranking: LOW (1)

The description of this metric stated: "Low uncertainties on dominant phenomena refers to the uncertainties affecting the engineering parameters controlling safety at the plant." The MSR was scored LOW on the assessment despite the fact that the physical phenomena that control the MSR are not poorly understood or immature as the notes assert, but rather are simple and well-understood. A great deal of this has to do with the fact that so many of the safety-related phenomena present in other reactors, like high-pressure in water-cooled and gas-cooled reactor, or chemical reactivity in metal-cooled reactors, simply are not present in the MSR design.

The elegance of the governing principles of the MSR family of reactor designs, particularly expressed in the LFTR configuration being developed by Flibe Energy, have very low uncertainties on the dominant phenomena. The long-term effects of fuel circulation, fission product accumulation, fission product gas removal, redox balance in the salt, reactivity response, and corrosion were all identified and addressed in the Molten-Salt Reactor Experiment and its follow-on research. While some aspects of these realizations present challenges, the uncertainties around them are low, even if they are poorly appreciated by the nuclear community as a whole.

Metric 17: Integral experiment scalability assessment
Highest ranking: Small LWR, LWR once-through, LWR recycle, V HIGH (4)
MSR ranking: LOW (1)

Throughout the Molten-Salt Reactor Program at Oak Ridge, scaled experiments were able to model the fluid behavior of the reactor core. This was aided tremendously by the fluid nature of the core and the ability to match similarity parameters (Reynolds, Nusselt, etc.) in order to create scaled assessments. These scaled results were then favorably matched to real testing in the Molten-Salt Reactor Experiment and the reactor operators did not encounter surprises.

Recent work by Dr. Per Peterson and his students at the University of California-Berkeley point to a future of scaled experiments using simulants for flibe salt (LiF-BeF2) that can match even more non-dimensional parameters, and that lead to very favorable scaled tests on the fluid dynamics of the reactor. Since the core of the reactor is fluid, homogeneity and uniform burnup is enforced, leading to the elimination of hot-spots and the ability to develop full-scale systems with great confidence.

Bardet and Peterson, "Options for Scaled Experiments for High-Temperature Liquid Salt and Helium Fluid Mechanics and Convective Heat Transfer," Nuclear Technology, Vol. 163, September 2008.

Metric 18: Source term assessment
Highest ranking: VHTR, V HIGH (4)
MSR ranking: LOW (1)

The source term is absolutely minimized in a LFTR due to several factors. Fission product gases are continuously removed and treated. Thus there are no fission gases to contend with in an accidental release, and furthermore, there are no driving forces (pressure, chemical reactivity) that would drive a release in the first place from the reactor. Overall inventory is absolutely minimized through continuous breeding of new fuel and a thermal spectrum. Contrast this with the VHTR, which has to carry large amounts of excess reactivity due to its poor breeding characteristics. Fission gases are trapped in the fuel, leading to swelling, cracking, and reactivity distortions during high power operation. To contend with this effect the entire core has to operate at a low core power density, compromising economics and increasing costs since the entire reactor operates under high pressure.

I do not understand how a VHTR could merit the V HIGH metric in this category while the MSR gets ranked LOW. If the contention is that the fission product gas removal constitutes a deficiency for the MSR design, I would contend vigorously with that point of view. Fission product gas removal should strengthen its ranking on this metric, not reduce it. MSR deserves a higher ranking on this metric due to its minimal inventory, continuous fission product gas removal, and the ability to stabilize reactivity through thorium breeding. The VHTR can do none of these things.

Haubenreich and Engel, "Experience with the Molten-Salt Reactor Experiment," Nuclear Applications and Technology, Vol. 8, pg 118 (1970).

Metric 21: Effective hold-up assessment
Highest ranking: VHTR, V HIGH (4)
MSR ranking: MED (2)

The explanatory notes state: "This metric refers to in-built mechanisms for retaining volatile fission products". Once again, the MSR is penalized for what should be considered one of its greatest advantages, the ability to remove and continually process fission product gases. By addressing these fission product gases continuously and effectively, it should have the highest ranking on this metric, not the VHTR, which has to contend with the buildup of fission product gases and their attendant damage and cracking of the fuel form.

Furthermore, cesium is one of the most important volatile fission products that must be considered in a release scenario. In oxide chemistry, cesium is volatile. In fluoride chemistry, cesium is incredibly stable and is bound in the fuel form. For these reasons, it is the MSR technology that should be given a higher ranking on this category. It has fewer volatiles that solid-oxide fueled reactors and it continuously addresses the behavior of noble fission gases in a way that does not compromise the reactor's performance.

Metric 23: Production costs assessment
Highest ranking: all except Small LWR, MED (2)
MSR ranking: LOW (1)

Production costs is defined in this study as the operating and maintenance (O&M) costs. There is reason to assume that these costs will be lower in an MSR than in a solid-fueled reactor. There is no need to fabricate fuel, nor to reshuffle fuel bundles. Fuel composition is continuously homogenized by the pumping action of the core. The reactor operates with essentially no excess reactivity unlike a solid-fueled reactor, and xenon is effortlessly removed from the fuel salt. Thus reactor operators do not have to worry about spatial xenon transients or chemical shim of the reactor fluid, which are significant operational challenges in current light-water reactors. These advantages are diminished somewhat by the need to monitor the off-gas system, the fluorination and reduction columns, and the production rate of new fissile in the blanket. Therefore, a much higher ranking is appropriate.

Metric 26: R&D cost assessment
Highest ranking: all LWR, V HIGH (3)
MSR ranking: LOW (1)

R&D costs will likely be high for the MSR, thus we have no objection to the current rank, but we do challenge the notion that the VHTR should have a HIGH rank in this category. It should be LOW like the other advanced concepts due to the expense of high-temperature fuel qualification.

Metric 30: Timescales to deployment assessment
Highest ranking: LWR once-through, LWR recycle, V HIGH (4)
MSR ranking: LOW (1)

The timescale to deployment is an extremely subjective metric which is difficult to quantify. There are no explanatory notes associated with this metric in the study, thus we do not have a basis to challenge the current ranking.

Metric 31: Technology Readiness Level
Highest ranking: LWR once-through, LWR recycle, V HIGH (3)
MSR ranking: LOW (1)

This metric appears to be an overlap with the "R&D cost assessment" metric, and we recognize that the technology readiness level of the MSR is lower than the other concepts. We hasten to point out that an accessory metric that should be considered is the cost of improving the TRL level. By analogy, if TRL is position, then the "velocity" of a technology on the TRL scale, with respect to the investment required (the partial derivative of TRL with respect to investment) should be carefully considered as well. Many reactor concepts have higher TRLs but will require a great deal of financial investment to advance, particularly the VHTR with its high-temperature solid-fuel form. The MSR in comparison has a simplified fuel form that does not require expensive qualification procedures, rather it only needs exposure in a high-flux materials testing reactor.

Metric 32: Flexibility in location assessment
Highest ranking: VHTR, ADTR, HPM, Small LWR, HIGH (3)
MSR ranking: MED (2)

The LFTR can have exceptional flexibility in location due to its use of a gas-turbine power conversion system and air-cooling, thus freeing it from the need to be based near bodies of water that can supply cooling water for steam turbine condensers. It merits a very high ranking in this category.

Metric 36: Associated fuel cycle
Highest ranking: SCWR, all LWR, HIGH (3)
MSR ranking: LOW (1)

Inside a liquid fluoride thorium reactor, or LFTR (pronounced LIF-ter), the processes of the thorium fuel cycle are manifested in the equipment of the reactor. Thorium in the outer region of the reactor, called the “blanket”, absorb about half of the neutrons produced in fission. As uranium-233 is formed in the thorium blanket, it is removed by fluorination from uranium tetrafluoride (UF4) which is its stable chemical form in the salt, to uranium hexafluoride (UF6) which is a gas and will come out of solution from the blanket salt. Thorium exists only as a tetrafluoride (ThF4) and has no gaseous hexafluoride state, so uranium is removed while thorium is left behind in this straightforward chemical process. Uranium fluorination is done every day on tonnage scale as part of today’s preparation of uranium fuel for enrichment and is a well-understood chemical process.

Once the uranium hexafluoride gas is removed, it is reduced back from UF6 to UF4 in the presence of the “core salt” of the LFTR using hydrogen gas. Thus the core salt is constantly refueled by new uranium-233 produced in the blanket and the uranium-233 consumed in fission is replaced. In a similar manner the blanket is continually creating new uranium-233 from thorium using the neutrons from fission in the core salt. Thorium tetrafluoride is fed into the blanket to make up for the consumption of thorium in the blanket.

After reduction in the core salt, hydrogen fluoride (HF) gas is produced and recovered from the reduction column. It is then electrolyzed to produce fluorine (F2) gas for the fluorinator and hydrogen (H2) gas for the reduction column. HF electrolysis units are common industrial equipment and allow the HF to be recycled indefinitely.

Periodically or continuously, depending on the economics, the core salt is fluorinated and distilled to remove fission products from the core salt. Some of these, like molybdenum-99, are very valuable and will be segregated for their own purposes. The regenerated core salt is then returned to the core. The chemical form of the salts render them impervious to radiation damage, allowing them to function as a medium for nuclear reactions essentially forever.

Metric 37: Proliferation resistance assessment
Highest ranking: VHTR, V HIGH (4)
MSR ranking: MED (2)

Proliferation resistance is a highly subjective metric. To some merely the use of fissile materials implies low proliferation resistance. There are approximately 70,000 nuclear weapons in the world. None are based on the use of thorium and uranium-233. There is a good reason for this. It is difficult to create uranium-233 without forming uranium-232. Uranium-232 has a half-life of 69 years and follows the same decay chain as thorium-232. The thorium-232 decay chain includes thallium-208 and bismuth-212, both of which emit strong gamma radiation.

U-232 as a contaminant in U-233 can play a role in nonproliferation of nuclear weapons and still allow U-233 to be used in the nuclear fuel cycle. A recent work by Dr. Ralph Moir, commissioned by Lawrence Livermore Laboratory, describes the two features of U-232, its gamma radiation and its heat release, both of which causes problems for use in nuclear weapons. The gamma radiation at low levels causes health problems and even death for people nearby after prolonged exposures. At high levels it causes degradation of the high explosives of the weapons after a sufficiently long exposure. The heat causes problems in weapons design, for example, thermal degradation of high explosives. These properties of U-232 are quantified in his note. The role of U-232 as a potential nuclear weapon contaminant is similar to that of Pu238 in that both give off troublesome heat, U-232 much more than Pu-238, but the strong radiation from U-232 makes its role unique as Pu-238 has little radiation associated with it.

Dr. Moir's examinations of gamma doses and heat generation indicated that the high explosives used in weapons would be degraded sufficiently to render them unsuitable for stockpile use by a nation-state.

Consider then the theft of fissile materials, and where and how they might be stolen from the LFTR. The logical place for a theft to occur would be in the chemical processing systems that move either protactinium or uranium out of the blanket salt. We consider first the protactinium isolation case, since the proliferation argument is often levied against protactinium isolation without careful consideration in the fluoride reactor. Assume that the LFTR has the ability to isolate protactinium chemically from its blanket and can allow it to decay into very pure uranium-233 with little uranium-232 contamination. How might a determined thief attempt to steal it?

A great deal depends on the design of the chemical processing system, which it is safe to assume has been built according to a layout that was accepted by national and international guidelines to preclude such events. The design favored by Flibe Energy when considering the merits of protactinium isolation involves a lithium fluoride-beryllium fluoride salt wherein is dissolved both uranium-233 tetrafluoride and protactinium tetrafluoride that has not yet decayed. The lithium used in this region is enriched in lithium-6 so that it strongly absorbs neutrons. The use of enriched lithium fluoride can be configured so that accidental criticality of the uranium and decaying protactinium is averted, even if the reactor shuts down and all of the protactinium decays to uranium.

The gamma dose to a worker at 1 m from 5 kg of U-233 with 2.4% U-232 would be 100 rem/hr, 1 year after chemical separation. The dose would be 4 rem after 65 hours of exposure, the limit for annual exposure to workers and a fatal dose of 300 rem would occur in 300 hours of exposure after separation. The high explosive is predicted to degrade owing to ionizing radiation after a little over 0.5 year. The heat rate is 77 W just after separation and climbs to over 600 W ten years later.

The high explosive HMX commonly used in nuclear explosives can withstand up to 100 Mr of radiation. The effects of this radiation dose are gas evolution, crumbling and other undesirable effects. Various levels of U232/U233 on gamma dose rate from a sphere of U233 of 5 kg reflected by beryllium that would be just critical are considered. At U232/U233 = 0.024 the dose rate at 1 m is 100 rem/h after 1 year from separation. We have normalized the dose rate of Fig. 2 to 100 rem/h at 1 year and plotted the result in Fig. 3.

High explosive can tolerate about 100 Mr before degradation. 1600 hours to accumulate the tolerable dose for 1 year after separation of U232. At nine years the dose rate is 2.9 times that at 1 y. The time to degrade or shelf life would be 550 hours (Fig. 4).

LLNL-TR-438648, "U232 Nonproliferation Features", Ralph Moir, June 2010

Metric 38: Ease of construction
Highest ranking: VHTR, HPM, Small LWR, HIGH (3)
MSR ranking: LOW (1)

Ease of construction was ranked "high" for the VHTR because of "modular construction". Perhaps there is some reference case for a VHTR that I have not been exposed to, but construction costs for a reactor whose central element is a high-pressure, high-temperature reactor vessel would not strike me immediately as having high merit.

Modular construction is also a principle that applies to the LFTR concept. LFTR operates at low pressure not high like the VHTR. Thus its reactor vessel can be much thinner-walled than the VHTR, making it easier to fabricate through bending, joining, and welding processes.

The simplified approach to decay heat removal, utilizing the freeze plug and drain tank concept, also reduces construction challenges when compared to the multiple reactor-grade emergency core cooling systems required in a high-pressure reactor. This is particularly evident when comparing with a reactor whose high-pressure coolant can undergo a phase change in the event of depressurization—a water-cooled reactor.

The salts are chemically stable and their processing techniques (fluorination, reduction, electrolysis, and distillation) are also mechanically simple and radiation-hard processes. There is no need to fabricate expensive fuel elements such as are required in solid-fueled reactors, nor is there a need to remove fuel before it has been fully consumed such as in solid-fueled reactors.

The complicated systems that adjust reactivity in water-cooled reactor (boron injection and mechanical control rods and drives) are also not present in the LFTR, since excess reactivity can be kept to an absolute minimum by the ability to add and remove fuel during operation.

The physical footprint of the nuclear plant can also be minimized relative to other reactors by the use of a high-core-power-density reactor with dense heat transfer fluid in a close-fitting containment driving compact power conversion equipment. All of these factors argue for a dramatically reduced cost of construction of a LFTR relative to solid-fueled reactor concepts.

Metric 41: Decommissioning costs
Highest ranking: HPM, V HIGH (4)
MSR ranking: MED (2)

Decommissioning aspects in a LFTR are greatly simplified over a comparable solid-fueled reactor, primarily because of the ease of recycling the primary and secondary fluids that make up the fuel and blanket of the reactor. Fluoride salts are impervious to radiation damage due to the ionic nature of their chemical bonding. This is a strong contrast to the covalently-bonded solid fuels, where neutron and gamma bombardment cause lattice dislocations, swelling, and cracking. Fluid fuels give up fission product gases effortlessly during operation; solid fuels retain fission product gases which enhance swelling and cracking. The fluoride salt combination used in the fuel of the LFTR is LiF-BeF2-UF4, along with the fission product fluorides that will accumulate. When it is time for decommissioning, the UF4 can be easily recovered from the fuel salt by fluorination to UF6 gas. The valuable LiF-BeF2 carrier salt can be removed from the fission product fluorides by high-temperature distillation, a technique actually demonstrated in May 1969 with 12 liters of fuel salt during the operation of the Molten-Salt Reactor Experiment. The LiF-BeF2 recovered from distillation and the UF6 recovered from fluorination can then be recombined into LiF-BeF2-UF4 fuel salt and recycled into the next generation of LFTR. The remaining fission product fluorides can be further separated and distilled to remove valuable non-radioactive components like neodymium and potentially valuable radioactive components like strontium and cesium. Finally, the very low value fluorides remaining can be immobilized in iron phosphate glass and sent to a burial site.

The blanket salt of the reactor (LiF-BeF2-ThF4) should have a low fission product inventory, particularly if protactinium isolation has been employed in reactor operation. If the fission product inventory is sufficiently low, then a simple fluorination step to remove any residual UF4 should be all that is necessary to prepare the blanket salt for recycling to the next LFTR reactor. If there is a more significant fission product inventory then additional purification steps will be necessary. Thorium is not extracted through fluorination, nor is it removed by distillation, so there is a strong incentive to keep fission product inventory at a minimum rather than to have to deal with the challenge of trying to separate lanthanide trifluoride fission products from thorium tetrafluoride salts. This challenge was considered by the Molten-Salt Reactor Program during their investigation of the "one-fluid" reactor concept and was found to be quite challenging; for this reason we have been strongly incentivized to investigate the "two-fluid" reactor concept that is the basis for LFTR.

The interior surfaces of the reactor structures (reactor vessel, primary heat exchangers, chemical processing equipment) will be contaminated with fission products and will be highly radioactive. "Flushing" interior surfaces with "clean" LiF-BeF2 salt was shown to remove much of this contamination by the MSRE, but it does not completely remove fission products. The techniques necessary to separate metals and graphite from fission products are still an area of active consideration, but right now the leading contender for the graphite appears to be oxidation to CO2 and venting to the atmosphere. The full recycling of metallic structures, primarily Hastelloy-N, is the goal of the LFTR program, and other metallic structures that make up the chemical processing systems will also have to be investigated.

Leaving UF4 in the fuel salt for long periods of time (decades) is not a good idea. As the salt cools and freezes, radiolysis of UF4 leads to the formation of F2 gas that will fluorinate some of the UF4 to UF6, which then becomes mobile. Despite the warnings of MSRE chemical engineers, this is exactly the course of (in)action that was taken after the shutdown of the MSRE, leading to a costly and expensive decontamination and remediation program in the 1990s and 2000s. All of this can be avoided simply by fluorinating the UF4 out of the fuel salt within several years of shutdown, as was recommended by the chemical engineers who worked on the reactor.

Compared to the other reactor options in the NNL study, the MSR was ranked "medium" due to "bulky reactor structure and graphite disposal" yet the VHTR, which also uses graphite as a moderator was ranked "high". The VHTR runs at high pressure and needs a large, thick pressure vessel to hold in the high-pressure gases used for reactor coolant. The LFTR, along with other MSRs, uses fluoride salts at low pressure to transport heat. Since fluoride salts are thousands of times more effective at heat transport per unit volume, "bulky reactor structure" is a poor assessment of the MSR technology. This appellation would be better suited for the VHTR with its high-pressure reactor vessel. VHTR spent fuel is incredibly robust and resistant to processing or separation of valuable fuel or fission products. It is very likely it will be disposed of waste in large volumes, since the graphite moderator that encloses and encapsulates the TRISO uranium-oxide fuel particles in the VHTR fuel will also likely be sent to a disposal facility. Fuel and blanket salts used in a LFTR can be processed in a manner described previously and actual waste generation can be minimized by tremendous factors relative to the VHTR.

Even more inexplicably, the Hyperion Power Module (HPM) was ranked "very high" on this assessment due to its "disposable reactor module". One is left to wonder what that means, since the HPM has uranium nitride fuel that will contain large amounts of unburned uranium and high-quality plutonium that has formed in a fast-neutron spectrum, along with fission products. It will also contain lead-bismuth eutectic (LBE) coolant. Neither of these materials will be in a form suitable for long-term disposal.

ORNL-4577, "Low-Pressure Distillation of a Portion of the Fuel Carrier Salt from the MSRE", August 1971.

PostPosted: Sep 17, 2012 3:47 pm 
User avatar

Joined: Nov 30, 2006 3:30 pm
Posts: 3731
Location: Alabama
The next three posts contain the NNL study's consideration of the fuel cycle, with my comments in blue.

PostPosted: Sep 17, 2012 3:47 pm 
User avatar

Joined: Nov 30, 2006 3:30 pm
Posts: 3731
Location: Alabama
1. Introduction

The UK National Nuclear Laboratory (NNL) has been contracted by the Department for Energy and Climate Change (DECC) to review and assess the relevance to the UK of the advanced reactor systems currently being developed internationally. Part of the task specification relates to comparison of the thorium and uranium fuel cycles. Worldwide, there has for a long time been a sustained interest in the thorium fuel cycle and presently there are several major research initiatives which are either focused specifically on the thorium fuel cycle or on systems which use thorium as the fertile seed instead of U-238. In the UK, the thorium fuel cycle isn’t currently regarded as a mainstream option and it is important for DECC to understand why this is the case and whether there is a valid argument for adopting a different position.

I would ask the author to "stick to the facts" and to avoid prejudicial language like "not a mainstream option". Decision makers are interested in what might be done with thorium far more than in what has been done with thorium.

All commercial power reactors in operation today are reliant on the uranium-plutonium (U-Pu) fuel cycle, in which U-235 is the principal fissile nuclide providing the fission neutrons needed to maintain criticality and power output. Most of the commercial reactor fleet uses Low Enriched Uranium (LEU), containing typically < 5 weight percent of U-235, though some reactor types (CANDU being the most prevalent) use natural uranium at 0.71 weight %. In all these reactors the U-238 (which constitutes the bulk of the fuel mass) undergoes fertile neutron captures to produce Pu-239, which is fissile and which increases the useful energy that can be extracted from the fuel. The Pu-239, along with the higher plutonium isotopes produced by neutron captures, is partly consumed as it is produced and some remains in the spent fuel. The remaining plutonium can be recycled in various forms to increase the energy extracted and in a fast reactor breeding cycle the extracted plutonium is sufficient to meet the reactor’s fuel requirements without any further inputs of uranium.

It should be noted that the uranium-plutonium cycle, utilized in a thermal-spectrum reactor like those commonly employed, does not produce sufficient neutrons per fissile absorption to sustain the consumption of uranium through successive recycles. Thorium has that potential, which is why it is of surpassing interest to future nuclear planners.

The thorium fuel cycle is an alternative to uranium-plutonium. Thorium is widespread in the Earth’s crust and is known to occur in economically accessible deposits in many locations. Natural thorium is made up entirely of the isotope Th-232, which is fertile, but not fissile. When irradiated by neutrons, Th-232 is converted to U-233, which is fissile. In principle, if there are sufficient spare neutrons from a uranium-fuelled reactor available, useful quantities of U-233 can be produced by irradiating Th-232. The U-233 can then either fission in situ in the fuel, increasing its useful energy output, or be separated and recycled into new fuel. This is the basis of the thorium fuel cycle.

The thorium fuel cycle is more basic than this. The previous paragraph is already an attempt to constrain careful consideration of the thorium fuel cycle to solid fuels, a criticism that I have for many other sections in this report, and which is not remedied in the small section of fluid-fueled reactors where it should be. Thorium can be the basis of a sustainable cycle of nuclear energy production in a thermal-spectrum reactor due to the neutron production of U-233 in thermal-spectrum fission. This is its singular advantage above all others and this is the basis of the thorium fuel cycle.

NNL has recently produced a position paper on thorium [1] which attempts to take a balanced view of the relative advantages and disadvantages of the thorium fuel cycle. This report develops the arguments further by considering the possible relevance of the thorium fuel cycle in the context of potential applications in the UK, comparison being with the uranium-plutonium fuel cycle.

Conversations with the author of this position paper (Kevin Hesketh) reveal that he did not consider fluid-fueled reactors in his development of the position paper, and this intentional oversight substantially diminishes its value as a guide for decision makers. Mr. Hesketh can hardly claim to have made a "balanced view" when he neglects to consider the safest and most compelling approach to thorium utilization.

Section 2 gives a brief history of the development of the thorium fuel cycle. Section 3 describes the salient features of the uranium-plutonium fuel cycle, which forms a reference point for the rest of the report. Section 4 describes the thorium fuel cycle and explains why it is of interest in the international research community. Section 5 discusses the potential role of thorium-plutonium fuel as an option for plutonium management in the UK. Section 6 discusses the potential role of thorium in the nine advanced reactor systems considered in a report written for the first phase of this study [2]. Finally, Sections 7 and 8 discuss the key points that need to be considered in determining the best strategies that the UK might adopt if it decides to recycle its plutonium stockpile.

2. Thorium history

The potential benefits of the thorium fuel cycle have led to a number of historic R&D projects world-wide: The first attempt to demonstrate the thorium fuel cycle at large scale was in the Shippingport PWR in the late 1950s [3]. This operated for a while with a high enriched uranium (HEU) driver fuel and thorium fertile targets. The ultimate objective was to reprocess the thorium targets and recycle the U-233 into new driver fuel assemblies which would in turn provide the neutrons for the next generation of thorium targets. The reasons why the demonstration programme ended are likely to have been a combination of technological and economic issues. Shippingport was eventually converted to LEU oxide fuel which was ultimately adopted in the commercial fuel cycle.

This is incorrect. Shippingport did not operate with a thorium core until 1978, and the fuel was a mixture of uranium-233 dioxide and thorium dioxide. It did not contain HEU as a driver fuel. Post irradiation examination of the fuel elements in 1982 reveals that more uranium-233 had been generated during operation of the reactor over a five-year period than had been consumed by fission, demonstrating that thorium breeding was possible in a thermal-spectrum reactor.

Shippingport had been operating on a uranium-plutonium fuel cycle since it was started in 1957. The thorium experiment was the last fuel that was ever loaded into the core.

A good article on the thorium-uranium-233 experiment in Shippingport can be found here: ... ystem.html

The Indian Point-1 reactor in New York operated with an HEU/thoria core in the early 1960s, and some of the uranium-233 generated in the thoria (thorium dioxide) fuel elements was removed and is still stored at the Oak Ridge National Laboratory in Tennessee.

Further R&D on thorium fuels was carried out in the USA and Germany as part of the early High Temperature Reactor (HTR) programmes of those countries which started in the early 1960s and continued to the mid-1980s. It was recognised that that HTRs are especially suited to thorium fuels, because HTR fuels are capable of very high burnups, which is an essential requirement if U-233 is to be utilised in-situ in a once-through fuel cycle. Moreover, some of the HTR fuel cycle schemes being considered at that time involved reprocessing and U-233 recycle, which offered the possibility of high conversion ratios and low fuel requirements.

A long standing R&D programme currently led by LightBridge [4] is developing a two-part fuel assembly for PWRs in which a central LEU driver sub-assembly provides the seed neutrons to breed U-233 in an outer thorium sub-assembly. This seed-blanket concept uses a once-through fuel cycle in which the seed sub-assemblies are replaced more frequently than the blanket sub-assembly. This allows the U-233 in the blanket more time to build up and for it to be fissioned more completely. Lightbridge have been working closely with Russian researchers and the seed sub-assembly uses LEU metal fuel elements based on submarine reactor technology.

In the past two decades, there has been a large amount of interest in Accelerator Driven Systems (ADS) using thorium fuels. These use a sub-critical reactor core which maintains a steady fission power with an external source of neutrons generated by a spallation source driven by a beam of high energy protons from an accelerator. This was an old idea that was revived by Nobel laureate Carlo Rubbia in the Energy Amplifier [5] and has recently been taken up in projects such as the Accelerator Driven Thorium Reactor (ADTR) and the Accelerator Driven Sub-critical Reactor (ADSR) [6,7] proposed by Aker solutions and a consortium of universities. ADS are capable of burning any type of fuel and choosing thorium potentially provides low radiotoxicity, fuel diversity and proliferation resistance. It is noted here that ADS have to some extent been promoted on the grounds that they are different to normal critical reactors and that the choice of thorium helps add an additional element of product differentiation.

More recently, there have been several small companies involved in promoting thorium fuels, mostly with links to Norway (eg Thor Energy), which has large thorium reserves. In this case, the main driver is to establish a market for thorium that Norway could subsequently exploit.

The European Union Framework Programme (FP) has sponsored several projects related to thorium fuel and a new proposal for FP-7 called THORIZON is currently being developed by NRG and AREVA; NNL are also partners and have participated in Thorium projects, including irradiations in previous FP-5 and FP-6 projects. AREVA’s interest centres on the use of thorium fuel in PWRs with reprocessing and recycle of the U-233. They claim studies showing a benefit of up to 40% in reduced uranium demand if plutonium, reprocessed uranium and U-233 are recycled [8]. AREVA feel this is worthwhile justification for a modest level of R&D spend on thorium in PWRs. AREVA make the point that without recycle of U-233, the uranium demand benefits are too small to justify the necessary investment. THORIZON will therefore be looking at thorium reprocessing technologies and will also be investigating molten salt reactors.

AREVA is correct to think this way. Without recycle of bred U-233, there is little reason to investigate thorium as a fuel. The particularly compelling advantage of the liquid-fluoride thorium reactor (LFTR, a type of molten-salt reactor) is that this recycling is dramatically simplified.

Acting largely independent of international developments on thorium, India has maintained a sustained interest in thorium fuels for many decades. To date, this has mainly been focused on India’s Heavy Water Reactors (HWR). Future plans will involve the breeding of U-233 in thorium blanket assemblies in India’s planned fast reactor fleet. The U-233 will then be separated and manufactured into U-233/Th fuel assemblies to be irradiated in the planned Advanced Heavy Water Reactors (AHWR). These reactors could meet about two-thirds of their long term U-233 requirement from breeding in the thorium matrix, with the balance being provided by the fast reactor breeder blankets.

India’s situation is special. The main justification for thorium is that India has large thorium reserves, but no reserves of uranium. India has been isolated from the broader international nuclear R&D community because of not having signed the Non-Proliferation Treaty (NPT). India’s nuclear industry does not operate on the same commercial footing as most other countries, so that justification of the thorium fuel cycle does not have to be made on the same basis. Finally, there is a strong element of India wanting to demonstrate its technical prowess and the thorium fuel cycle provides a powerful vehicle for this purpose.

3. Uranium-plutonium fuel cycle

3.1. General principles

The uranium-plutonium fuel cycle is the only one that has been used in commercial reactors, despite there having been an early interest in the thorium/U-233 fuel cycle. In the uranium-plutonium fuel cycle the primary fissioning nuclide is U-235. U-235 is the only naturally occurring fissile nuclide and was necessarily therefore the starting point for both military and civil nuclear programmes. In graphite-moderated systems (such as MAGNOX) and heavy water moderated systems (such as CANDU), the 0.71 weight % abundance of uranium is sufficient to achieve criticality. In other systems, such as light water reactors (LWRs), criticality is only achievable with low-enriched uranium (LEU). In both natural and LEU systems, U-235 accounts for about 60% of the fission events in the nuclear fuels over their irradiation lifetimes, with the balance coming principally from Pu-239 and lesser contributions from Pu-241 and U-238. Plutonium is generated in uranium fuel by neutron capture events in U-238, which after two beta decays results in Pu-239. This is the fertile capture mechanism whereby the fertile nuclide U-238, which is reluctant to fission, is converted via a neutron capture event into a fissile nuclide which fissions readily.

Fort St. Vrain was a commercial reactor that utilized thorium-uranium fuels. The opening statement of this paragraph is therefore incorrect. Perhaps in a paragraph titled "General Principles" the author ought to stick to the physical facts (cross-sections, capture-to-fission ratios, etc.) rather than talking about how reactors have developed.

Fertile conversion is a key element of the uranium-plutonium fuel cycle. The production of Pu-239 in this manner contributes to about 30% of the fission events in the nuclear fuel over its lifetime, increasing the effective energy output over that achievable with U-235 only. This contributes to reducing fuel costs, since the fuel throughput is decreased proportionally and at the same time uranium ore requirements are reduced by the same amount. The U-238 fertile capture mechanism also has a key role in reactor safety because many of the neutron captures in U-238 occur in the resonant energy range from 6 eV upwards. The resonances are very sharply defined peaks in the neutron capture cross-section that are broadened by the thermal motion of the atoms. When the fuel temperature increases, this Doppler broadening of the resonances increases neutron captures because the population of neutrons with kinetic energies matching the resonances is increased. This is a fast acting negative feedback effect that is essential to ensure safe operation of all reactors.

The author should note that today's reactors are incapable of burning all of the uranium present (both U-238 and U-235) due to the inherent properties of uranium-235 and plutonium-239 fuel in a thermal spectrum. Therefore, the implementation of today's uranium-plutonium reactors is effectively confined to the role of "burning" the small fraction of uranium that is naturally fissile (U-235) and leaving the overwhelming majority of the energy content of the fuel unutilized. This can be very different in thorium and thermal-spectrum reactors.

Another benefit of fertile conversion is that not all of the plutonium produced in the fuel is fissioned before the fuel is discharged. The residual plutonium at discharge (typically about 1% of the heavy metal mass) can potentially be recovered in reprocessing and either recycled in a thermal reactor (as is the case today in France) or in a fast reactor. Thermal reactor plutonium recycle as Mixed Oxide (MOX) fuel gives approximately a 15% increase in the energy recovered from the original uranium ore. On the other hand, plutonium recycle in a fast reactor gives the possibility of much higher energy recovery in a breeding cycle, with the possibility of a fully self-sustained fuel cycle with minimal uranium ore input required. In principle, it is possible to attain about a 50 to 100-fold improvement in the energy extracted from uranium ore. This is the justification for fast reactors, which would allow a country to be strategically independent of the uranium market. However, the practical difficulties of realising this theoretical gain have not yet been overcome in any country. Moreover, such a large gain could only be realised over a large number of recycle steps. Since each cycle of irradiation, cooling and recycle last about 10 years at the minimum, the timescales involved extend to around one hundred years. Whether such timescales are actually relevant and meaningful in practice is questionable and in any practically relevant scenario, the recoverable energy is likely to be much lower. This is a practical limitation that is usually glossed over in the literature and in strategic analyses.

Even with MOX fabrication and further irradiation the performance benefits of uranium in thermal-spectrum reactors are modest. Only by going to fast-spectrum reactors can significant improvements in uranium fuel utilization be realized. With thorium fuels these benefits can be achieved in thermal-spectrum reactors.

The fertile conversion of U-238 to Pu-239 in uranium fuel is the first step in a chain of neutron capture events that leads to higher isotopes of plutonium and to the production of the minor actinides (principally neptunium, americium and curium). This has important implications for this report in two respects:

Firstly, the accumulation with burnup of higher plutonium isotopes (especially Pu-240) is seen as beneficial for reducing the potential proliferation risk, because in high burnup fuels the proportion of Pu-240 makes the plutonium unattractive for weapons applications. Although all plutonium is formally regarded for safeguards purposes as being weapons usable, there is undeniably a vast difference in attractiveness between plutonium in low burnup fuels and fuels discharged at high burnups from modern LWRs; plutonium with Pu-240 < 6% is classified as “weapons plutonium” . This can be regarded as a beneficial characteristic of high burnup uranium-plutonium fuel cycles. This point is noted here because, as will be seen later, it contrasts strongly with the thorium fuel cycle.

It is very strange that the author would flag the poor performance of plutonium-239 fuel in a thermal spectrum as any sort of "advantage" of uranium relative to thorium. The formation of plutonium-240 is actually one of the great deficiencies of the uranium-plutonium fuel cycle in the thermal spectrum. If plutonium-239 was more likely to fission than absorb in the thermal spectrum, then sustained conversion of uranium-238 to plutonium-239 in the thermal spectrum might be possible. But plutonium-239 performs poorly, fissioning only 2 out of 3 times it absorbs a thermal neutron. The other 1 out of 3 times it absorbs the neutron, forming plutonium-240 which is not fissile in the thermal spectrum. By contrast, uranium-233 will fission 9 out of 10 times it absorbs a thermal neutron.

If weapons plutonium production is desired, all that needs to be done is to limit the exposure duration, and the small amount of plutonium formed will be of high isotopic quality. There is no inherent difficulty with this approach, which has historically been undertaken by every major power that has attempted to weaponize plutonium.

Refer to later discussion of U-232 as a proliferation deterrent for thorium fuel cycle similar to the Pu-240 is for uranium cycle.

Secondly, though the total minor actinide content of LWR fuel is only of the order of 0.1 weight %, the minor actinides contribute significantly to radiotoxicity, heat production and neutron output in spent fuel or vitrified high level waste (VHLW) from reprocessing. The presence of U-238 in the fresh fuel makes it impossible to avoid significant production of minor actinides. Again, this is a point which is strongly contrasting in the thorium fuel cycle.

Yes, in a "pure" U-233/thorium fuel cycle the production of minor actinides can be drastically reduced due to the absence of uranium-238 from the fuel.

3.2. Resource availability

Total identified resources of uranium ore have been estimated by OECD-NEA [9, 10] to be sufficient to meet 100 years of supply at 2008 rates of consumption. While uranium availability poses a strategic risk, this will most likely materialise as an escalation of uranium prices that will have only a limited impact on total generating costs; uranium ore makes up only a small percentage of the overall nuclear generating cost. If world nuclear capacity remains static or grows slowly, uranium price escalation is unlikely to have be a major limitation. Pressure on uranium ore prices is likely to be most severe in a scenario with rapid growth of world nuclear capacity. Estimates of economic uranium reserves are strongly linked to market prices – an increase in market price greatly increases the reserves which are economically viable. Therefore, even in high growth scenarios, uranium availability is not likely to be limiting and utilities are unlikely to view alternatives to uranium as a strategic priority for some considerable time yet. Also, reprocessed uranium and plutonium recycle are available to help mitigate this risk if required.

If carbon-free nuclear energy is desired to address the effects of climate change and to improve global living standards, then a huge expansion of nuclear energy will be necessary. If this energy continues to be generated by the inefficient use of uranium in thermal-spectrum uranium-plutonium reactors then uranium supplies WILL be an issue, as will be issues regarding the disposition of spent nuclear fuel. Reprocessing into MOX followed by additional irradiation will have a modest effect on this conclusion but not significant. Only a "game-changing" decision to move into efficient thorium reactors or fast-spectrum uranium reactors will alter this conclusion, and only thorium can be used efficiently in the thermal spectrum; uranium cannot.

3.3. Economics

The total generating cost of a nuclear power plant is dominated by the capital cost (typically ~60%), followed by operating and maintenance (~20%) and then the fuel cost (~15%), as illustrated in Reference [11]. The back-end fuel cost and the decommissioning provision cost accounts for the remainder. The fuel cost is comprised of the cost of buying uranium ore on the world market, the cost of converting and enriching the uranium and finally the cost of fuel fabrication. The uranium ore cost is variable, being determined by market prices, but at the present long term contract prices, it equates to about one third of the fuel cost. Therefore, the cost to a utility of uranium ore represents only a small component of overall generating costs (typically ~5%) and overall generating costs are relatively insensitive to escalations in uranium ore prices. Again, reprocessed uranium and plutonium recycle are available to help mitigate this risk if required.

Fuel costs are a small factor in today's nuclear generation costs—that will not necessarily be the case if a huge expansion of nuclear energy is considered. Furthermore, even from the limited perspective of fuel costs, thorium used in a LFTR can eliminate most of the cost steps involved with uranium fuel. It is not mined per se, but is recovered as an undesired byproduct of rare-earth mining. It needs no enrichment since it has only one natural isotope. In a LFTR it is fluorinated and introduced into the blanket, where it is irradiated and converts to uranium-233. The U-233 is removed by fluorination and introduced to the core salt by reduction. Fuel is not removed from the reactor, only fission products.

At present, although there are pressures on utilities from uranium market trends, these are insufficient at present to force them to seriously look at alternatives.

3.4. Radiotoxicity

The radiotoxicity of spent uranium fuel is dominated for the first 500 years by fission products. After this time the fission products have mostly decayed and the radiotoxicity becomes dominated principally by transuranic elements, particularly plutonium. This persists until approximately 100,000 years, when the long-lived fission products such as I-129 become the dominant contributors. The radiotoxicity is an important measure of the hazard potential in the geological repository. The period between 500 years and 10,000 years is usually considered to be a key factor in repository performance, since this is when waste packages are likely to lose their integrity and radionuclide transport out of the repository is most significant.

The fact that using the "pure" U-233/thorium fuel cycle reduces the production of transuranic actinides is incredibly important and worthy of note in this section.

Reducing radiotoxicity is currently not regarded by utilities as a concern in reactor operations. Radiotoxicity has been cited in justification arguments for new build in the UK and are likely to be used by utilities in justifying future fuel cycle and operational strategies. Nevertheless, the practical impact of radiotoxicity calculations has to date been relatively low and there is no prospect of it becoming important enough to warrant major changes of strategy by an utility. Plutonium recycle as MOX can be shown to give a modest reduction in overall radiotoxicity.

Radiotoxicity may not be a concern to utilities but it is a central concern to governments that have to site and build geological repositories. If the intended audience for this report is the government then this important fact should be noted along with the potential importance of thorium in helping to achieve this goal.

3.5. Proliferation risk

Spent uranium fuel from LWRs contains just over 1 weight % of plutonium. This is considered to represent the main proliferation risk associated with the uranium fuel cycle. In a once-through fuel cycle, the plutonium remains relatively inaccessible in the spent fuel. In contrast, in a reprocessing fuel cycle, separated plutonium oxide is produced which needs to be subject to stringent physical protection. Measures to increase the inherent proliferation resistance of the reprocessing fuel, such as avoiding the separation of pure plutonium oxide are considered desirable in designing new reactors and associated fuel cycle facilities. However, reducing proliferation risk is not a factor in strategic decision making for utilities and is unlikely to become so in the foreseeable future. Therefore, there currently is no incentive for utilities to seek
alternatives to U-Pu fuel.

Plutonium is always accessible in spent nuclear fuel by chemical processing. Whether that separation has taken place (in a reprocessing-based cycle) or has not taken place (in a once-through thorium cycle) is immaterial.

The situation with regards to proliferation may not be a significant concern for utilities but if we were to contemplate a global expansion of nuclear energy then the associated fuel cycle techniques of uranium enrichment and plutonium reprocessing may expand as well. These techniques find application both in the fuel cycle and in weapons production. Since a LFTR that has already been started by fissile material (U-233) does not require additional enrichment for its sustenance, this is a significant non-proliferation consideration.

3.6. Reprocessing

The purpose of reprocessing spent uranium fuel is to separate it into a pure uranium stream for recycle, a pure plutonium stream for recycle and a vitrified high level waste stream which contains all the fission products and transuranics other than plutonium for disposal. The PUREX process is used for this purpose. It involves mechanically shearing the spent fuel and dissolving the fuel pellets in nitric acid. Tri Butyl Phosphate (TBP) in kerosene is used as the extractant to separate out the fission products and to separate the uranium from the plutonium. The PUREX process is well established, with commercial plants operational in France, Japan, Russian Federation and the UK. The recycled uranium can be re-enriched for re-use as Reprocessed Uranium (REP U) fuel. The recycled plutonium can be reused as UO2/PuO2 mixed oxide (MOX) fuel.

Only a small proportion of utilities have their fuel reprocessed, the vast majority preferring to opt for a once-through strategy with direct disposal of spent fuel, which utilities regard as a less expensive option. Many utilities which already have reprocessing contracts in place have been seeking to reduce their commitments in favour of direct disposal. Although this attitude may change in the future, it is unlikely to occur in the near future and there is not likely to be any pressure from utilities to invest in new reprocessing plants.

These decisions and pressures are based in large part on the stance taken by governments with regards to the disposal of various materials from spent nuclear fuel. The situation in this respect is already changing in the United States with the cancellation of the Yucca Mountain waste repository. It is almost certain that with a significant expansion of nuclear energy worldwide that there will be far greater pressures on governments to address the long-term disposition of waste in a sustainable manner, and that these pressures will be communicated from governments to utilities. It is also likely that this will lead to a change in the way utilities assess the relative values of fuel cycles, and could strongly favor a thorium fuel cycle that absolutely minimizes the production of transuranic waste, as is possible in a LFTR.

PostPosted: Sep 17, 2012 3:48 pm 
User avatar

Joined: Nov 30, 2006 3:30 pm
Posts: 3731
Location: Alabama
4. Thorium fuel cycle

This section discusses the potential advantages of the thorium fuel cycle and comments on the validity of the claims made.

4.1. General principles

In the thorium fuel cycle, the starting point is naturally occurring thorium, which consists of just the single isotope Th-232. This is a fertile isotope, analogous to U-238, and on capturing a neutron it is transformed via two beta decays to U-233, which is fissile. There are two strategies for making use of the U-233 produced in this way: The once-through strategy involves the fissioning of as much of the U-233 as possible in situ in the thorium matrix and not to attempt to recover any from spent fuel. The spent fuel eventually undergoes geological disposal. The recycle strategy involves the reprocessing of the thorium fuel to recover the U-233 for fabrication into U-233/Th fuels. Only the latter option is capable of delivering the full benefits of the thorium fuel cycle, but technological difficulties with thorium fuel reprocessing and fabricating U-233 fuels have posed obstacles to its implementation to date.

I would state this differently. The starting point of the thorium fuel cycle is uranium-233, the fissile derivative of thorium, formed from neutron capture. Uranium-233 serves as a type of nuclear "catalyst" for the sustained consumption of thorium in a thermal-spectrum reactor. This is possible because of the superior qualities of uranium-233 in a thermal-spectrum beyond all other fissile materials.

Any reasonable discussion of the thorium fuel cycle should begin with a clear delineation between the use of thorium in solid fuels and its use in liquid form. Then the relative benefits and challenges of each should also be clearly delineated. Blanket discussions of benefits and challenges of thorium, without the express context of the fuel form are not informative or productive. A balanced discussion should also discuss relative benefits and challenges separately, lest immediate juxtapositioning of the two be inadvertently interpreted as marginalizing those benefits. Here, the author immediately delves into a discussion of the possible paths that a solid-fueled reactor might take and highlights difficulties of use of thorium in solid fuels. This is a level of detail that should follow the careful explanation of the inherent benefit of the thorium-uranium-233 approach, which is not delineated here.

In the thorium fuel cycle Th-232 is analogous to U-238 in the U-Pu fuel cycle, because it is the fertile component. Some of the neutron captures in Th-232 are resonance events, so that the Doppler broadening role of U-238 is very closely replicated. U-233 is analogous to Pu-239 in the uranium fuel cycle, because these are the fissile nuclides produced from fertile neutron captures. Unlike the uranium fuel cycle, in the thorium fuel cycle there is no naturally occurring fissile isotope analogous to U-235 to provide the first neutrons and the thorium fuel cycle relies on the supply of neutrons from another source to produce the fissile material, which in practice means fission neutrons from U-235 or Pu-239. This is a crucial difference between the thorium and uranium fuel cycles that in large part explains why the uranium fuel cycle became established first and why the thorium fuel cycle has failed to make any inroads: the uranium fuel cycle necessarily had to be established first and once the infrastructure was in place it was difficult to justify further investment for thorium.

The author seems to persist in ignoring the central advantage of thorium and its fissile derivative uranium-233. Uranium-233 produces enough neutrons per absorption of a thermal neutron to enable the continued consumption of thorium. Plutonium-239 does not. This is the summum bonum of the thorium story, and it cannot be emphasized enough. Further, smaller advantages include the fact that thorium-232 is roughly three times more absorptive of thermal neutrons than uranium-238, meaning that the optical depth of a blanket region to neutrons can be reduced in a thorium fuel cycle relative to a uranium fuel cycle, but since most uranium reactors make little attempt to breed new fissile fuel, this is an advantage of thorium that is not particularly well-appreciated.

The presence or absence of a fissile isotope of thorium is unnecessary to emphasize. The fissile inventory for a thorium reactor can be obtained from U-233, U-235, Pu-239, or any combination thereof. If we had absolutely no fissile on Earth at all but U-235, we would simply enrich uranium to a high level (to remove as much U-238 as possible) and then use that as the starting material, as was done at the commercial Ft. St. Vrain reactor. But we are awash in unused fissile material, including hundreds of tonnes of HEU in the United States and over 100 tonnes of plutonium in the United Kingdom. Thorium reactors can productively use these materials; LFTRs can use them inexpensively and sustainably. Once LFTRs are started on fissile material they can thereafter sustain energy generation on only a feed of inexpensive natural thorium.

The real reason that thorium was not developed is that thorium and uranium-233 were not suitable for weapons and were not developed extensively during the Manhattan Project. Uranium and plutonium were suitable for weapons and the difficult and expensive techniques of uranium enrichment and plutonium separation were developed during the war. After the war, it was easier to design uranium-plutonium reactors around accepted and understood enrichment, fabrication, and separation techniques than it was to "start over" with thorium.

4.2. Resource availability

Thorium represents an alternative resource to uranium and has a higher abundance and a different geographic distribution to uranium, as illustrated in Table 1, which shows the distribution of thorium resources according to the World Nuclear Association [12]. It may therefore be a strategic benefit in the event of high uranium prices. Nevertheless, there remains the need for large efforts regarding full Thorium mine prospecting and development of a Thorium purification process at the front end of the fuel cycle. Furthermore, it should also be noted that the external radiation dose is much higher for Thorium than Uranium leading up to the purification stages because of the decay to Thallium-208. Nevertheless, mining of open pit monazite deposits (presently the main source of thorium) is easier than that of most uranium bearing ores, and management of thorium mine tailings is also simpler than in the case of uranium mainly because of the much shorter half live of “Thoron” (= 220Rn : 55 sec) than of Radon (222Rn : 8 days, daughter of 226Ra, 1600 years).

There is no need for "large efforts" to prospect for thorium mines. None whatsoever. Far more thorium than will ever be needed by the entire planet is already being mined in the effort to obtain rare-earth minerals. Many thousands of metric tonnes of already-separated thorium exist in the United States and France. In this respect thorium has a staggering advantage over uranium as a fuel.

It should also be noted, however, on the basis that neutrons from U-235 or plutonium are needed to convert fertile Th-232 to fissile U-233, the thorium fuel cycle would not be completely independent of uranium until a fully self-sustaining thorium cycle is eventually established.

Given the excess fissile inventories of the US and the UK a sustainable thorium utilization program could be established that would be utterly independent of further uranium mining.

The thorium fuel cycle is in principle capable of achieving higher conversion ratios in thermal reactors than uranium fuel, which is advantageous for resource availability. The conversion ratio is the number of fissile atoms generated by fertile captures divided by the number of fissile atoms consumed in the process. A conversion ratio of 1.0 implies that a fuel cycle is capable of operating a self-sustained cycle. Thermal reactors with uranium fuel typically have conversion ratios in the region of 0.6 and although this can be increased, it is not feasible to reach 1.0. Thorium fuel, however, in a thermal reactor, can achieve conversion ratios much closer to 1.0 and this was one of the main attractions. The basis of these statements comes from the nuclear cross-sections of U-233 in comparison with U-235 and Pu-239. The so-called eta parameter, which is the number of neutrons per fission multiplied by the ratio of the fission and absorption cross-sections, in the thermal neutron energy range, is highest for U-233 [13].

I commend the author in finally discussing the central advantages of the thorium fuel cycle, here on page 14 of a 30 page report. These important facts should appear in one of the very first paragraphs. Further exposition would explain to the reader the profound implications of these properties—that thorium can be nearly completely consumed in a liquid-fluoride thorium reactor, and that this will unlock staggering amounts of energy from the world's rich supply of thorium. I would also ask the author to point out to the reader that this remarkable property of thorium is possible in the thermal spectrum, which makes thorium unique amongst nuclear fuels.

4.3. Economics

While economic benefits are theoretically achievable by using thorium fuels, in current market conditions the position is marginal and insufficient to justify major investment by utilities:

"the position in solid-fueled reactors is marginal" should be the proper form of that statement. The statement is not justified when one considers liquid-fluoride thorium reactors, particularly with uncertainties about spent nuclear fuel disposal and the siting of geological waste repositories.

In a once-through thorium cycle, thorium will displace only a fraction of the uranium fuel, the latter being necessary to provide the neutrons to convert the fertile Th-232 to fissile U-233. Moreover, the uranium fuel remaining will need to have a higher U-235 enrichment to compensate for neutron captures in Th-232, so that any savings in uranium ore and enrichment costs are likely to be marginal. On the other hand, the thorium fuel will require new fuel production facilities, with a substantial investment. Any marginal reductions in uranium ore and enrichment costs are unlikely to justify the necessary investment.

This statement is true for solid-fueled reactors but not for liquid-fueled reactors. The author should make clear the fuel form context of any statements regarding thorium. Blanket statements about thorium fuel cycles are meaningless without this context. Liquid-fluoride thorium reactors could be started on U-233 from the US, HEU from the US, and even plutonium from the UK. The cost of making fluid fuel formulations is trivial compared with the costs of fabricating, testing, and qualifying solid fuel forms.

In a reprocessing cycle, in which the U-233 is recycled, the uranium ore and enrichment savings are likely to be improved still further and could even be eliminated altogether in the long term if a breeding cycle could be established. However, to reach this position will require major investment in thorium reprocessing and fuel fabrication plants, with significant technical and investment risk which appears not to be merited by current or reasonably foreseeable market conditions.

Again, delineation between the performance of thorium in solid and liquid fuels is in order. For the liquid-fluoride thorium reactor, the processing steps (fluorination, reduction, and distillation) are straightforward. No fuel fabrication is required. The future market condition that could justify this investment is associated with a substantial expansion of nuclear power in order to reduce a nation's reliance on carbon-emitting energy sources.

It cannot be ruled out that the thorium fuel cycle may become competitive in a future market environment of restricted uranium ore availability and thus very high uranium prices. This is not considered very likely for the foreseeable future, given that economically recoverable uranium reserves are thought to be very price dependent and therefore if uranium prices were to increase, then more uranium would be available to the market. Under these circumstances, fast reactors would also become more viable. Therefore, NNL’s view is that thorium will only be an economic option for the very long term future.

Uranium ore costs will not drive towards thorium, because thorium is not advantageous in a solid-fueled reactor. Waste concerns will be the impetus to move towards a thorium fuel cycle in a liquid-fluoride thorium reactor. If Mr. Hesketh presumes to speak for all of NNL, then he is limiting himself to a consideration of only solid-fueled reactors by saying that thorium will be an economic option only for the very long term future.

4.4. Radiotoxicity

The thorium fuel cycle generates only trace quantities of plutonium and higher actinides, which can reduce the long term radiotoxicity of spent nuclear fuel. Figure 1 shows a typical result for thorium systems for a scenario which was analysed for this study. The green curve shows the radiotoxicity in Sieverts per tonne of initial Heavy Metal (tHM) for a Light Water Reactor (LWR) fuelled with Th-Pu as a function of cooling time after discharge. For comparison, the blue curve shows the radiotoxicity profile for UO2 fuel irradiated to the same discharge burnup. Between 100 and 100,000 years, the thorium fuel cycle shows a modest reduction. The thorium case, is however, higher after 100,000 years, a result which is typical of many studies, due to the in-growth of daughter nuclides from the thorium decay chain. There is virtually no difference in the cooling time needed to reach the radiotoxicity of uranium ore in the uranium fuel cycle, which is often used as a reference point. The key point to note is that the comparison between the radiotoxicities of the thorium and uranium cycles depends on the decay time being considered.

Figure 1 is intended to illustrate the typical behaviour observed with thorium systems. It should be stressed that the results are system and scenario specific and other thorium scenarios (especially those involving recycle of U-233) can be envisaged which would give larger reductions in radiotoxicity than in this particular example. However, care is required because such results are often quoted for an equilibrium fuel cycle in which U-233 is fully established in a self-sustained system. To get to an equilibrium condition will require the use of fuels containing U-235 or plutonium which will contribute higher radiotoxicities. In realistic scenarios, with the evolution of the scenario modelled explicitly in time, the overall radiotoxicity is usually significantly higher than the equilibrium case. The overall conclusion is that while trace production of minor actinides in the thorium fuel cycle is without question advantageous for radiotoxicity, there is insufficient potential benefit to utilities to encourage the necessary investment.

This is an absurd example to choose to represent the behavior of thorium with respect to radiotoxicity. The author has zoomed in on one of many possible scenarios for thorium utilization. He has chosen a solid-fueled reactor with limited irradiation time and a plutonium-thorium fuel mixture to demonstrate the performance of thorium fuel. The results will show little difference because in each case plutonium will dominate long-term radiotoxicity. I really have to ask, is the author attempting to do a good job describing the potential performance of thorium fuels, or is he carefully choosing cases that will show little distinction to justify his thesis that thorium shows little promise?

A better example would be the radiotoxicity of the distillation still bottoms of a liquid-fluoride thorium reactor over the course of sustained irradiation of the liquid fuel. In that scenario, the reader would notice that the still bottoms of the distillation still exclude actinides, including thorium and uranium-233, which remain in the reactor as they are fully consumed. The fission products would decay rapidly, with strontium-90 and cesium-137 making up the bulk of the radiotoxicity after approximately 30 years post-exposure. Longer-term radiotoxicity (>300 years) would be vastly less than the solid uranium case or the Th-Pu case because plutonium would not be present in the waste stream.

Mr. Hesketh should refrain from deciding what utilities might or might not do under the changing conditions of future scenarios, particularly those where much larger deployments of nuclear energy might be desired. Radiotoxicity reduction is currently an advantage for the governments that must dispose of nuclear waste, but as they assign higher importance to that function it is likely that that emphasis will be passed on to utilities.

4.5. Proliferation risk

In the thorium fuel cycle, the absence of plutonium is often claimed to reduce the risk of nuclear weapons proliferation, though Reference [1] questions whether is this is completely valid, given that there were a number of U-233 nuclear tests (the “Teapot tests”) in the US in the 1950s. U-233 is in many respects very well suited for weapons use, because it has a low critical mass, a low spontaneous neutron source and low heat output. It has been stated [eg Wikipedia entry on U-233] that because U-233 has a higher spontaneous neutron source than Pu-239, then this makes it more of a technical challenge. However, this is erroneous, because even in weapons grade plutonium the main neutron source is from Pu-240. A further consideration is that the U-233 produced in thorium fuel is isotopically very pure, with only trace quantities of U-232 and U-234 produced. Although the U-232 presents problems with radiological protection during fuel fabrication, the fissile quality does not degrade with irradiation. Therefore, if it is accepted that U-233 is weapons useable, this remains the case at all burnups and there is no degradation in weapons attractiveness with burnup, unlike the U-Pu cycle.

No, the claim for the thorium fuel cycle is that the presence of U-232 makes weapons fabrication from U-233 significantly less attractive than the use of HEU or plutonium. The Teapot test, as far as is known publically, was the only attempt to detonate an explosive from U-233, and the core was both plutonium and U-233. The shot has a yield below anticipation, and the fact remains that no operational nuclear weapon has ever been fabricated from thorium or uranium-233, for good reasons. Of the tens of thousands of warheads in the world's arsenals, none are based on U-233 or the thorium-fuel cycle.

"Equilibrium" uranium in the core fluid of a LFTR would consist of a mixture of five different isotopes of uranium, with mass numbers from 232 to 236. The bulk of the uranium would be 233, but a substantial amount would be 234 and a radiologically significant amount would be 232.

The presence of trace amounts of U-232 is beneficial in that it provides a significant gamma dose field that would complicate weapons fabrication and this has been claimed to make U-233 proliferation resistant. However, there are mitigating strategies and the U-232 dose rate cannot be regarded as a completely effective barrier to proliferation. As such, U-233 should be considered weapons usable in the same way as HEU and plutonium. This is also the position taken by the IAEA, which under the Convention on the Physical Protection of Nuclear Materials [14] categorises U-233 in the same way as plutonium. Under the IAEA classification, 2 kg or more of U-233 or plutonium are designated as Category I Nuclear Material and as such are subject to appropriate controls. By way of comparison, the mass of U-235 for Category I material is 5 kg. Attempts to lower the fissile content of uranium by adding U-238 are considered to offer only weak protection, as the U-233 could be separated relatively easily in a centrifuge cascade in the same way that U-235 is separated from U-238 in the standard uranium fuel cycle.

There is no claim that thorium is a "completely effective barrier to proliferation." The claim is that uranium-233 is a far less attractive route towards weapons production that the enrichment of natural uranium or the production of plutonium from uranium irradiation. This claim is borne out by experience in weapons programs across the world.

Unlike plutonium, uranium-233 in liquid fluoride form could be instantaneously denatured by mixture with uranium-238 salts in the event that the security of a facility was compromised by hostile forces. There is no natural form of plutonium with which isotopic dilution might be executed like there is for uranium, and "just-in-time" denaturing of solid fuel is impossible.

Separating uranium isotopes by centrifuge is utterly impractical for the isotopic mixture that would be found in the equilibrium uranium core salt in a LFTR. Five isotopes of uranium are present and are only separated by a single atomic mass each. The centrifuge cascade would be hopelessly contaminated by the introduction of this uranium mixture and the radiation exposure (mostly from U-232) would be unacceptable for personnel.

The overall conclusion is that while there may be some justification for the thorium fuel cycle posing a reduced proliferation risk, the justification is not very strong and, as noted in Section 3.5, is not a major factor for utilities. Regardless of the details, those safeguards and security measures in place for the U-Pu cycle will have to remain in place for the thorium fuel cycle and there is no overall benefit.

On the contrary, real experience bears out the claim that the use of thorium reduces the risk of proliferation. Use of thorium produces an isotope that renders the material highly unsuitable for weapons, the uranium itself can be diluted in liquid form, and there is no national experience base with which to confidently design and build a uranium-233 weapon like there is with HEU or plutonium.

4.6. Reprocessing

The purpose of reprocessing of thorium fuels is to separate the U-233 from the bulk Th-232 and the fission products. The U-233 and the Th-232 are then purified to leave only trace quantities of other radioactive materials suitable for recycle. The THOREX process has been developed for this purpose and is similar to the PUREX process used for separating uranium and plutonium. The THOREX process starts by shearing fuel assemblies and uses a mix of nitric and hydrofluoric acid to dissolve the nuclear fuel pellets. Tri Butyl Phosphate (TBP) in kerosene is used as the extractant to separate out the fission products, to separate the uranium from the thorium and to purify them for recycle. The process is analogous to the PUREX process for reprocessing uranium/plutonium fuel, but there are a number of difficulties that are best illustrated by comparing THOREX and PUREX:

Once again, the author is presupposing that we are discussing thorium in solid-oxide form. In liquid-fluoride form it is straightforward to remove uranium-233 from the thorium tetrafluoride blanket fluid (LiF-BeF2-ThF4) through fluorination. UF4 becomes UF6 which is gaseous and bubbles out of the blanket salt. By continuous removal of fissile material from the blanket, fission product buildup in the blanket is minimized and the simplified strategy of fluorination is appropriate.

The difficulties of THOREX relative to PUREX are unimportant for further discussion if thorium-oxide fuel is not the technique for thorium utilization under consideration, as would be the case if the liquid-fluoride thorium reactor was the baseline thorium reactor being considered for future deployment.

1. PUREX uses nitric acid for dissolution, which is sufficient to dissolve uranium/plutonium fuel. THOREX requires hydrofluoric acid in addition to nitric, because thorium is not completely dissolved in nitric acid on its own. Hydrofluoric acid is much more reactive than nitric acid towards structural metals and requires special alloys for reaction vessels, pipework, valves, pumps and sensors. Corrosion of the reprocessing plant components will need careful control to ensure the operational lifetime is not compromised.

Another argument against using solid thorium oxide fuel.

2. The PUREX process takes advantage of the chemistry of uranium and plutonium, which are easily separated from one another. In the THOREX process, thorium is characterised by relatively poor extraction and this will complicate the design of the reprocessing plant, with possibly a cost penalty.

Thorium-uranium chemical separation is advantageous in fluoride media and processing is drastically simplified. No dissolution is necessary when using fluoride forms of uranium and thorium.

3. Waste streams from a THOREX plant will be different from those from a PUREX plant, because of the different reagents used and work will be required to establish if they can be managed using existing methods. The THOREX process, for example, is expected to generate 50-70 % more glass than PUREX [8].

4. The THOREX process has not been demonstrated beyond laboratory scale, which represents a technical risk when scaling up to commercial throughputs. Considerable R&D spend will be required to demonstrate the process at commercial scale and with a minimum timescale of 15-20 years before consideration of commercial scale facility could be considered. The THOREX process will be reviewed as part of the THORIZON FP7 proposal.

5. The U-233 product will contain trace quantities of U-232 which has a very energetic gamma emitter as part of its decay chain. A short while after production of the U-233, the surface gamma dose will build-up to very significant levels and if there is a requirement to store the material, a shielded and remote access storage facility will be needed.

Again, none of these points are significant against the thorium fuel cycle if it is considered in the liquid-fluoride thorium reactor. The trace quantities of U-232 are not an issue when solid fuel fabrication is unnecessary, as the fluoride mixture can be constituted remotely.

4.7. Recycle

Recycling U-233 presents some difficult challenges in fuel fabrication because of the daughter products from U-232. U-232 builds up to part per million (ppm) levels in the U-233, compared with parts per billion concentrations in reprocessed uranium fuels. U-232 has a half-life of 68.9 years and its decay chain includes daughters with very energetic gamma emissions, especially Tl-208. When the U-233 is chemically separated from the thorium fuel, the daughter products from U-232 are partitioned with the VHLW and the gamma activity of the U-232 is initially very low. However, the U-232 decay daughters re-establish themselves quite quickly, reaching equilibrium after 2 years, at which point the U-233 has a very high gamma field. The activity of U-232 becomes significant at parts per billion (ppb) levels, so that the ppm concentrations in U-233 are very serious, demanding substantial shielding and remote fabrication methods. This is a significant technological barrier to full recycle of U-233 and poses a technical risk.

Since we're talking about the specific techniques of how solid thorium fuels are used, let me hasten to point out that the U-233 is not "recycled" in the core salt of a liquid-fluoride reactor in the usual sense, but is retained in the core salt until it is fissioned. When it is necessary to process the core salt, the uranium can be removed by fluorination and the remaining LiF-BeF2 carrier salt can be purified by distillation. The extracted UF6 is then recombined with purified LiF-BeF2 to reconstitute the core salt. Neither of these techniques is affected by the presence of U-232 in the salt, and no equivalent to solid-fuel fabrication is necessary. This is another significant advantage of the use of thorium in fluoride reactors.

4.8. Technological readiness

NNL has assessed the Technology Readiness Levels (TRLs) of the thorium fuel cycle. For all of the system options more work is needed at the fundamental level to establish the basic knowledge and understanding. Thorium reprocessing and waste management are poorly understood. The thorium fuel cycle cannot be considered to be mature in any area. Much of the fundamental knowledge requirements and experimental measurements at laboratory scale have a high degree of commonality for the different systems. The relative immaturity of the thorium fuel cycle is reflected in its inclusion in the European Framework Programme.

I vigorously disagree. There is no fundamental breakthrough needed to formulate, utilize, or process thorium fuel in a fluoride reactor. All of the fundamental feasibility assessments were done during the Molten-Salt Reactor Program (MSRP) at the Oak Ridge National Laboratory from 1957 to 1974. What remains now is the engineering design and development of the various systems. Using the TRL scale, I would assess most MSR technologies at TRL 4 to 6. Thorium processing in fluoride form is well understood due to the efforts of the MSRP. I cannot speak to the degree of technology readiness of solid thorium fuels but it is of little concern to LFTR development efforts.

PostPosted: Sep 17, 2012 3:48 pm 
User avatar

Joined: Nov 30, 2006 3:30 pm
Posts: 3731
Location: Alabama
5. Plutonium-thorium fuel

Plutonium-thorium fuel is a technical option that potentially could be of interest in the UK. This would consist of a mix of plutonium and thorium oxides (PuO2-ThO) analogous to conventional PuO2-UO2 mixed oxide (MOX) fuel. Although there is only laboratory scale experience of manufacturing and irradiation Pu/Th fuel, its physical properties are very close to PuO2-UO2 MOX fuel and its in-reactor behaviour would be expected to be comparable.

In the UK, the application would be as an option for the disposition of separated plutonium instead of PuO2-UO2 MOX. There are some potential advantages in this approach:

Firstly, ThO matrix has been cited as potentially being more stable than a UO2 matrix in geological disposal, with reduced leaching of plutonium. If this can be confirmed, this would fit with a strategy of irradiation of the PuO2-ThO fuel followed by spent fuel storage and eventual disposal.

Secondly, the use of ThO as the matrix implies that there is no production of new Pu-239, as is the case in conventional MOX fuel. Production of new Pu-239 in conventional MOX to some extent offsets the destruction by fission of the Pu-239 that was present in the fresh fuel. While there is still net destruction of plutonium in conventional MOX, the discharge inventory remains quite high at typically about two-thirds the initial inventory.

In PuO2-ThO fuel the lack of a Pu-239 fertile production source causes the discharge plutonium inventory to be lower (Pu-239 is reduced to about one-third its initial inventory and total plutonium to about half). The fissile quality of the plutonium at discharge is therefore exceptionally low, well below the level at which it would realistically be attractive for weapons use. The radiotoxicity of PuO2-ThO fuel is also lower than that of the equivalent MOX fuel, though the difference is fairly marginal.

These potential advantages need to be balanced against the disadvantages, which are dominated by the relative immaturity of PuO2-ThO fuel technology. R&D work would be needed to better determine the fuel thermo-physical properties and establish fuel fabrication methods, carry out irradiation testing (starting out with small scale irradiation trials in a research reactor and progressing eventually to commercial scale tests in a power reactor). In addition, R&D would be needed to demonstrate the impact on core nuclear design behaviour and to better understand the behaviour of PuO2-ThO fuel during irradiation, in store and in the repository environment. This R&D would still be required even though PuO2-ThO fuel is only a relatively small step removed from conventional MOX fuel and would involve a significant R&D spend. The timescales required are quite protracted, as any R&D programme involving irradiation testing is necessarily a long process. Realistically, it is difficult to envisage such an R&D programme being completed in less than 10-15 years even with significant investment. This defines the minimum feasible timescale for such a strategy, but if it was accepted that there is no immediate urgency for plutonium disposition in less than, say 15-20 years, such a programme might be regarded as feasible.

The benefits of this approach (PuO2-ThO2) do not seem to justify its research and development costs.

6. Thorium in advanced reactors
This section considers how thorium fuels fit with the nine advanced reactor systems considered in the earlier report produced in this study [2]. These systems are:

1. Sodium Fast Reactor (SFR).
2. Gas Fast Reactor (GFR)
3. Lead Fast Reactor (LFR)
4. Very High Temperature Reactor (VHTR)
5. Super Critical Water Reactor (SCWR)
6. Molten Salt Reactor (MSR)
7. Accelerator Driven Thorium Reactor (ADTR)
8. Hyperion Power Module (HPM)
9. Small modular Light Water Reactor (LWR)

The first three systems are all lumped together, while the remaining systems are considered independently.

6.1. SFR, GFR & LFR
These are the three fast reactor systems being developed as part of the Generation IV international collaboration. All three systems could in principle operate with the thorium in place of the conventional U-Pu cycle, although there is currently no work specific to thorium in the Generation IV programme.

6.1.1. Uranium-plutonium fuel cycle
The base assumption in GIF is that all three systems will be U-Pu fuelled. The fuel will be plutonium in a natural or depleted uranium diluent and the plutonium will be recycled in a fully sustainable fuel cycle independent of uranium ore requirements. In principle, a fully self-sustaining U-Pu cycle could extract up to a factor of 100 times more useful energy from each kg of uranium ore than the once-through LWR cycle i.e. using/converting all of the uranium (U235 and U238) compared with only 0.71% U235 in a once-through fuel cycle. The U-Pu fuel cycle is capable of achieving conversion ratios well over 1.0, which is an essential requirement for a breeding cycle to be established. It is particularly important for the breeding ratio to exceed 1.0 by a comfortable margin in a scenario in which the number of fast reactors is increasing rapidly, because this margin determines how quickly new reactors can be phased in. The initial deployment of fast reactors on a self-sustaining fuel cycle will be limited by the availability of stocks of fissile material generated by breeding. In the case of the U-Pu fuel cycle the initial deployment will be limited by the availability of plutonium.

The U-Pu fuel cycle is well understood, much of it being based on existing thermal reactor recycle technology. However, there are exceptions, the principal one being the strong drive to adopt alternatives to PUREX reprocessing to avoid separating pure plutonium. Other notable exceptions are that the characteristics of fast reactor fuel are different to those of thermal reactor fuel and therefore the specifications of the recycle plant will need to be modified to reflect the differences. Fast reactor recycle has been demonstrated only at sub-commercial scale and further development will be needed for full commercial readiness. Therefore the U-Pu fuel cycle for the GIF fast reactors can be regarded as being well understood, but in need of further development for commercial readiness.

6.1.2. Thorium fuel cycle
GIF is not planning to investigate the thorium fuel cycle for the Generation IV fast reactors, although thorium does represent a possible alternative.

A thorium fuel cycle in a fast reactor is compatible with a self-sustained breeding fuel cycle, potentially extracting up to 100 times more energy from each kg of Th-232 compared with 1 kg of uranium ore in the once-through thermal reactor cycle i.e. based on comparisons with a once-through LWR cycle, using only the U235. However, the same comment applies that complete energy conversion is only achievable on very long timescales.

Although the thorium fuel cycle is capable of achieving a conversion ratio greater than 1.0 in a fast reactor fuel cycle, the breeding ratio for the thorium fuel cycle in a fast neutron system is smaller than that of the U-Pu cycle. Depending on the specific scenario for deployment of fast reactors, this is potentially a disadvantage for the thorium fuel cycle that might slow the rate at which new reactors can be deployed. In the case of the thorium fuel cycle the limiting factor is the availability of U-233. A smaller breeding ratio for the thorium fuel cycle will limit the initial deployment of fast reactors and the time needed for the fast reactor fleet to expand will be longer. The doubling time for initial deployment is very sensitive to the breeding ratio and relatively small changes can have a large impact, so this is potentially a limiting factor in scenarios where rapid deployment of fast reactors is required. This loss of responsiveness in the deployment of fast reactors and the extra complexity of reprocessing thorium fuel and recycling U-233 are factors which are likely to have discouraged the Generation IV project from pursuing the thorium option for SFR, GFR and LFR. For fast reactors, the thorium fuel cycle offers no advantage in terms of sustainability because the U-Pu cycle is already fully self-sustainable. For these reasons, there is reduced incentive for thorium in fast reactors, although there would be a modest benefit in terms of reduced radiotoxicity.

Another consideration is whether GIF would take the view that it is better to avoid the production of pure U-233, perhaps by diluting the U-233 with U-238 during reprocessing. Such a strategy would be analogous to avoiding the production of pure plutonium in the U-Pu cycle, but the presence of U-238 would lead to the production of plutonium and other transuranics that would to some extent negate the benefits of the thorium cycle. The thorium fuel cycle is much less technologically mature than the U-Pu fuel cycle. As noted in Sections 4.6 and 4.7, the THOREX process has not been demonstrated beyond laboratory scale and fabrication of U-233 fuels will have to account for the strong gamma dose fields.

Since the central reason for considering a fast-reactor system is to sustainably consume uranium fuel, I wonder why anyone is even considering using the thorium fuel cycle in such a machine.

6.2. VHTR
The Very High Temperature Reactor (VHTR) in Generation IV is envisaged as operating with a once-through fuel cycle.

The fuel is in the form of 1 micro-metre diameter microspheres, containing the fuel kernel encapsulated in a TRISO fuel microsphere. The TRISO particle consists of a pyrolytic graphite layer, a silicon carbide or zirconium carbide layer (SiC/ZrC) and a second pyrolytic graphite layer. The fuel microsphere is embedded in a graphite matrix. Recycling VHTR fuel involves separating the microspheres from the bulk graphite and then mechanically cracking the hard SiC/ZrC shell. These are difficult steps that would complicate the recycle of the fuel.

VHTR is in principle capable of using a wide range of fuel kernels, including thorium and minor actinide fuels, but the base assumption in GIF is that it will use uranium.

6.2.1. Uranium-plutonium fuel cycle
The technology base for TRISO fuel manufacture and fuel performance was established from the 1960s to the mid-1980s in the various HTR prototypes that operated then, some of which used uranium fuel kernels. It is likely that further development work will be needed to meet the requirements of the VHTR programme, so the technology cannot be considered fully mature.

VHTR is potentially well suited for plutonium disposition, because the high burnup of the fuel kernels allows a larger proportion of the initially loaded plutonium to be destroyed. Moreover, the isotopic quality of VHTR plutonium fuel at discharge is very low. Deep Burn VHTR fuel is specifically designed to maximise the burnup and minimise the isotopic quality of plutonium fuel to such an extent that it is rendered almost unusable for weapons purposes.

6.2.2. Thorium fuel cycle
Some of the HTR prototypes operated from the 1960s to the mid-1980s used thorium fuel kernels and therefore the Technology Readiness Level of thorium can be regarded as comparable to that of uranium fuel.

There are several reasons why VHTR is especially well suited to thorium fuels:

1. The thorium fuel cycle is capable of achieving higher conversion ratios in a thermal reactor than the U-Pu fuel cycle and since VHTR has a thermal neutron spectrum, it is well suited to thorium. The result is a system with higher fertile conversion and therefore less dependence on external fissile materials.

2. It is important that the fuel discharge burnup should be as high as possible to ensure that the U-233 is fissioned efficiently in a once-through fuel cycle. VHTR fuels will have the required high burnup capability, by using higher enrichments.

3. Thorium oxide is thought to be potentially a more stable matrix than UO2 for geological disposal of spent fuel and this would be an advantage for a once-through fuel cycle.

These features were the reason why many of the early HTR projects (DRAGON, Peach Bottom, Jülich HTR and Fort St Vrain) all used thorium fuel kernels. These would have been used to breed U-233 that would subsequently undergo fission in the core. The conversion of Th-232 to U-233 requires a source of neutrons, which would have been provided by the U-235 or Pu-239 fissions. This demands that the fresh fuel kernels should contain a mix of U-Th or Pu-Th.

A potential limitation of thorium fuels in HTRs is the need to have relatively high initial enrichments of U-235 in the uranium driver fuel needed to drive the initial conversion of Th-232 to U-233. Some fuel designs developed to date use U-235 enriched to 20.0 weight percent, which is the upper limit for LEU, leaving no margin for design flexibility.

The Technology Readiness Level of uranium and thorium fuel cycles for VHTR fuels can be regarded as being comparable.

6.3. SCWR
The Super Critical Water Reactor (SCWR) is one of the least well developed of the Generation IV concepts, for which one of the main research requirements is to develop fuel and primary circuit materials that can withstand the extremely challenging core conditions of the super-critical water moderator/coolant. The SCWR design is at a very early conceptual stage and no specific consideration appears to have been made towards thorium fuels. However, there is no doubt that thorium fuels could be used in SCWR, but the precise role of SCWR is insufficiently developed to comment further.

6.4. MSR
The Molten Salt Reactor (MSR) being developed by Generation IV is, like earlier molten salt cores, specifically designed for the thorium fuel cycle. For this reason, the discussion here is limited to the thorium fuel cycle.

There are several aspects of the MSR design that are particularly suited to the thorium fuel cycle:

1. MSR has a thermal neutron spectrum in which the thorium cycle can achieve a higher conversion ratio than the uranium/plutonium cycle.

The conversion ratio can reach above 1.0. This is the key feature that allows an MSR to unlock the staggeringly large energy resources of thorium. This point cannot be over-emphasized.

2. MSR avoids some of the loss of conversion efficiency that occurs due to neutron capture events in Pa-233. The conversion of Th-232 to U-233 proceeds via two intermediates Th-233 and Pa-233 which undergo beta decay. Pa-233 has a relatively long half-life of 27 days and a significant fraction of it is removed by neutron captures which reduces U-233 production. The nuclear fuel in MSR is unique in that it circulates through the entire primary circuit and spends only a fraction of its time in the active core. This reduces the time-averaged neutron flux that the Pa-233 sees and significantly reduces the proportion of Pa-233 atoms that are lost to neutron captures.

The behavior of Pa-233 (including its decay and absorption cross-sections) is important to understanding the thorium fuel cycle and should have been discussed in detail in the "General Principles" section of the thorium fuel cycle rather than waiting to discuss it in the particular case of the MSR. Nevertheless, the ability of an MSR design to retain blanket fluid out-of-core or to employ protactinium separation on blanket fluid is a tremendous advantage when considering how to implement the thorium fuel cycle.

3. MSR continually reprocesses the nuclear fuel as it re-circulates in the primary circuit, removing fission products as they are generated. The U-233 produced by fertile captures on Th-232 is recycled simply by being left in the primary circuit. MSR therefore completely avoids the difficulties in conventional reactors with fabricating U-233 fuels (which have high gamma activities from U-232 daughters).

The author appears to be referring to a "one-fluid" fluoride MSR, where fertile ThF4 and fissile UF4 are both in the same salt. In the LFTR under development by Flibe Energy, fertile ThF4 and fissile UF4 are kept separate from one another. This is referred to as a "two-fluid" fluoride reactor.

4. Since the nuclear fuel is a molten salt, there are no fuel mechanical performance issues to consider. There is no distinction in this respect between different fuels and therefore no barrier to the adoption of thorium as there is in conventional reactors.

This is very significant.

The Technology Readiness Level of the MSR fuel cycle should be regarded as low, because it has never been demonstrated as a whole and experience to date has been limited to small scale laboratory experiments. To date, there has been a very low level of commitment to MSR within GIF.

Define "low". The TRL level of the various components of a fluoride MSR range between 4 to 6. This statement discounts the successful extended demonstration of 15,000 hours of operation of the MSR experiments at ORNL.

6.5. ADTR
The Accelerator Driven Thorium Reactor (ADTR) [6] is a sub-critical neutron multiplying system in which the external neutron source needed to support steady state operation is provided by a high power proton beam impinging on a spallation source. Most sub-critical concepts, including ADTR and the Energy Amplifier are designed around the thorium fuel cycle. For this reason, the discussion here is limited to the thorium fuel cycle.

ADTR, along with other sub-critical systems, is claimed to be safer than conventional reactors on the grounds that it is sub-critical and therefore less vulnerable to reactivity insertion accidents. ADTR also claims to be more internationalisable because of its high proliferation resistance. Finally, it is claimed that ADSR will be more economic than LWRs once uranium ore prices begin to rise in response to demand exceeding supply.

If thorium is used in solid form in an ADSR, as most designs I have seen anticipate, then there is little reason to believe than an ADSR will have attractive fuel cycle performance. Rather, it will be very similar to one of the metal-cooled fast reactors operating with solid thorium fuel.

All these claims are undemonstrated and they can all be disputed. For example, the main threat to the safety of conventional reactors is from decay heat production, not reactivity insertion events and ADTR will be no different in this respect.

This is correct and poorly appreciated by ADSR advocates.

ADTR presents many technological risks, one of which is associated with the fuel cycle. To meet its stated objectives, ADTR requires the implementation of full recycle of U-233. This brings with it the technological risk for the THOREX process noted in Section 4.6, as well as the fuel manufacture challenges noted in Section 4.7. Establishing the complete thorium fuel cycle will require very considerable investment.

This assumes that ADSR will use solid thorium fuel, as appears likely. Some ADSR designs have considered liquid fluoride or chloride fuel.

There is no doubt that thorium fuels do offer advantages of increased strategic independence, reduced radiotoxicity and possibly lower proliferation risk. However, there may be other considerations at play. Just as sub-critical systems are being differentiated on the grounds that they are safer than critical reactors, there is also an element of differentiating them on the grounds that the fuel is not uranium or plutonium.

6.6. HPM
The Hyperion Power Module (HPM) is an autonomous small power reactor with a capacity of 25 MWe. It is a liquid metal reactor that uses uranium nitride fuel and lead-bismuth coolant. It is designed for passive cooling, passive safety and has a long core life.

Hyperion makes no mention of a thorium fuelled option. As a very small power unit, it has only very limited relevance to the UK.

HPM is not specifically intended for plutonium recycle, but it is likely to be flexible enough to accommodate it if required, with additional investment needed to fabricate plutonium fuel.

6.7. Small modular water reactors
Small modular water reactor designs are based on existing Light Water Reactor (LWR) technology, but scaled down to benefit from increased applicability of passive safety. Small modular LWRs might use conventional UO2 fuels or UO2-PuO2 MOX fuels or thorium-based fuels.

6.7.1. Uranium-plutonium fuel cycle
The U-Pu fuel cycle in small modular LWRs is identical to that currently deployed in current LWRs and can therefore be considered to be fully technologically mature. Small modular LWRs could be well suited for plutonium disposition in the UK, with their capacities potentially matching better the requirements for siting at Sellafield, co-located with a MOX fabrication facility. Some small modular LWR cores are designed with long life cores for which MOX fuel is well suited.

6.7.2. Thorium fuel cycle
As with conventional LWRs, thorium fuels are a potential option that would have the major benefit of reducing dependence on uranium ore. There are two approaches that might be used, one based on a once-through cycle and one based on recycle of the U-233:

The Lightbridge fuel assembly discussed in Section 2 [4] is one example of a once-through thorium fuel cycle option that could be used without modification in small modular LWRs. The Lightbridge fuel design should be regarded as having a low Technology Readiness Level at present, because it has innovative design features that have only been demonstrated at small scale. Other options can be envisaged in which current LWR assembly mechanical designs are used without modification, with either a heterogeneous or homogeneous distribution of thorium in the fuel rods. This latter option could be regarded as having a higher Technology Readiness Level, though there may still remain issues related to thorium fuel manufacture and fuel performance that remain to be demonstrated. The benefits of a once-through thorium fuel cycle are a modest reduction in uranium ore requirements and a modest reduction in radiotoxicity. Full recycle of U-233 is another potential option for small modular LWRs. This option is already being considered by AREVA [8] for large LWRs and would give more substantial reductions in uranium ore requirements and radiotoxicity that once-through approaches.

However, as noted earlier, recycle of U-233 requires the THOREX process and remote fuel fabrication methods, both of which have not been developed, which puts this option at a low Technology Readiness Level.

7. Plutonium recycle strategies
By the time the UK’s MAGNOX and THORP reprocessing plants cease operation, it is projected that the UK will have more than 100 tonnes of separated plutonium in store. One option for the eventual disposition of this plutonium would be to recycle it in future reactors. This section explains some of the issues that need to be considered if the UK was to adopt such a strategy and it is hoped that this will guide any future assessment of advanced reactor systems for the UK. In general, there are three plutonium recycle strategies available:

1. Multiple recycle via MOX fuel in thermal reactors such as LWRs; the intention being to maximise depletion of the fissile material, with the minor actinides treated as waste.

2. Single recycle through existing reactors, such as LWRs, followed by reprocessing and burning in a fast reactor; this strategy incinerates some of the minor actinides too.

3. Optimised number of recycles in existing reactors such as LWRs, and as such reduce the number of fast burner reactors required; this strategy takes advantage of existing facilities.

Option 1 is very much a theoretical option as there are technical considerations that limit the number of recycles to a maximum of two. Therefore, that leaves Options 2 and 3 for consideration for any potential full recycle in a closed, sustainable sense. This is in line with other options being considered internationally, including in France and in Generation IV for example. Nevertheless, each of these recycling options carries its own associated risk (technical, economic etc) and limitations. Furthermore, the future options can also be seen to fit into three temporal phases in the management of the plutonium:

1. Gradual introduction of the recycling of MOX fuel up to an industrial scale in existing reactors e.g. LWRs. During this time, the stockpile of Pu tends to increase. This is the world-wide position as of today.

2. Ongoing but irregular expansion of MOX recycling in which more countries consider MOX fuel, develop the technology and additional reactors are licensed for MOX. This is up to around 2030.

3. Introduction of advanced reactor systems (thermal and fast) alongside technologies specifically designed for MOX fuel.

The reactor technology and the fuel type chosen by the UK will specifically determine the assumed plutonium loadings in terms of number of tonnes of fuel that can be taken to be loaded each year e.g. current plants are likely to be able to accommodate approximately 30% MOX core fraction versus 50% or even 100% for future LWRs. The new reactor technologies discussed earlier will have a range of plutonium loadings and proportion of plutonium destroyed, but other than LWRs, none have yet been proven to be able to accommodate plutonium or indeed thorium fuels.

Regardless of the chosen option for the management of the plutonium on these timescales (single or multiple recycle, use of fast or only thermal reactors etc), according to the OECD-Nuclear Energy Agency (NEA) , there are five high-level issues that need to be considered in assessing the technical options:

1. Plutonium management strategies should be consistent in maintaining high standards of safety.

2. Plutonium management strategies should preferably maintain flexibility in the fuel cycle, such that future options are not foreclosed.

3. Plutonium management strategies should be consistent with maintaining satisfactory standards of security and safeguards against proliferation.

4. The quantities and forms of radioactive wastes arising from each technical option are very important considerations.

5. A clear requirement is that the overall fuel cycle should remain economically competitive, though the economics should not be assessed in isolation, but rather as part of a Life Cycle Analysis (LCA) that accounts for all externalities.

Any future UK proposed fuel cycle and Plutonium management options should at least be cognisant and make reference to these key issues and strategies. In particular recognising that the economics are not the only consideration in evaluating plutonium management options e.g. environmental impact, radiotoxicity, proliferation resistance etc. As such, it is important to consider the limitations of any plutonium or Thorium scenario (as outlined below) as well as the need to complete a Life Cycle Analysis (LCA) of the issues, other than simply economics.

Furthermore, before the UK can truly determine the most appropriate technology for either long term sustainability and moreover, plutonium management it is vital that the UK understands whether the mission goal is reduction or destruction of the plutonium stocks as quickly as possible, or construction of an integrated fuel cycle and use of the plutonium as a potential valuable resource in the future, e.g., in fast reactors. Associated with this is the determination of the timescales that the UK wishes to address the chosen driver from these three goals. The timescales will in turn dictate which technology is at the UK’s disposal at that time e.g. if the UK wishes to reduce its separated plutonium stockpile in the next 20 years, the only possible options available will be MOX (or a related derivative) in light water reactors. A decision by the UK is therefore required on this before the most appropriate technical choice for a reactor re-use option can be made.

8. Discussion
Thorium fuel cycle R&D has a long history dating back to the very beginning of the nuclear industry. Though there are potential advantages, with the exception of India, it has failed to become established in commercial reactors for the reasons that have been explained in this report. Even in India, utilisation of thorium fuels still remains at relatively small scale. In recent years the thorium fuel cycle has been very heavily promoted by many research groups and technical companies such as Lightbridge and Thor Energy, which are dedicated to promoting it.

The only reason put forward in this report for the exclusion of thorium from today's reactors was the lack of a fissile isotope of thorium, which is patently incorrect.

While the thorium fuel cycle has some benefits compared with the uranium-plutonium fuel cycle, these have yet to be demonstrated or substantiated, particularly in a commercial or regulatory environment.

The overwhelming advantage of thorium relative to uranium is the ability to breed in the thermal spectrum. This was demonstrated in the last core of the Shippingport reactor from 1977 to 1982, and the technology to economically utilize thorium was advanced during the Molten-Salt Reactor Program. These benefits HAVE been demonstrated and substantiated and the author is incorrect to state that they have not.

This is very relevant to the UK, especially at the present time in view of plans to start a new build programme in the UK based on LWRs. It could be argued that the main priority for the UK is to ensure the momentum that the new build programme currently has built up is maintained, in order that the new build plants will be available in good time to meet the projected shortfalls of low carbon electrical capacity. This only permits existing reactor designs with the uranium-plutonium fuel cycle. Innovative thorium fuelled reactors will not be a viable alternative for at least 20 to 30 years and definitely cannot meet the new build timescales. A limited role for thorium fuels in new build LWRs might be possible at a later date, with perhaps a partial transition to thorium-U233 fuels later in their lifetimes and any major shift towards the thorium fuel cycle would only be realistic in a follow-on programme of reactor construction.

Is the new build program focused on "momentum" or on the successful and safe use of nuclear energy? These may be very different outcomes as the UK considers the use of thorium in liquid-fluoride thorium reactors. If the development schedule for the new-build of LWRs is slowed by cost overruns and safety concerns then the ultimate outcome of the new-build will be far less than is currently planned, in a manner very similar to what happened to the build-out plan that ultimately resulted in the Sizewell B LWR. The role for thorium in the UK's nuclear future is not as a solid-oxide fuel in LWRs but rather as a fluoride fuel in LFTRs.

Thorium fuelled reactors have already been advocated as being inherently safer than LWRs [15], but the basis of these claims is not sufficiently substantiated and will not be for many years, if at all. Suggesting that the UK should consider thorium reactors as a safer alternative to LWRs is not a viable option at this time as the UK energy shortfall and demand is on much shorter timescales than Thorium fuelled reactors could respond to. Furthermore, since the energy market is driven by private investment and with none of the utility companies investing or currently developing either Thorium fuels or Thorium fuelled reactor concepts, it is clear that there is little appetite or belief in the safety or performance claims.

Liquid-fluoride reactors can achieve greater levels of safety for lower costs than LWRs because they remove the basic sources of accidents in a nuclear design. They operate at low pressure, with chemically-stable fluids, and without fluids that undergo phase changes. They can passively drain in the event of a loss-of-cooling into a configuration designed to preclude criticality and maximize the rejection of decay heat to the environment. These features have been demonstrated on a proof-of-concept reactor at Oak Ridge in the 1960s.

Mr. Hesketh appears to have clear knowledge of the intention of utilities into the foreseeable future, an enviable trait. Furthermore he is able to deduce from the plans of utilities that wise planners have examined the thorium fuel cycle in close detail and found it wanting, which he was not able to do in this report.

The only area where thorium fuel might be of interest to the UK is possibility of using thorium-plutonium fuels in new build LWRs as a means of dispositioning the UK’s plutonium stocks. As discussed, this might offer technical advantages over uranium-plutonium (MOX) fuels, though this remains to be demonstrated. The value of using thorium fuel for plutonium disposition would need to be assessed against the high level issues identified in Section 6 concerning the importance of maintaining high standards of safety, security and protection against proliferation, as well as meeting other essential strategic goals related to maintaining flexibility in the fuel cycle, optimising waste arisings and economic competitiveness. It is important that the UK should be very clear as to what the overall objectives should be and the timescales for achieving these objectives.

Mr. Hesketh is kind enough to tell us that there is no merit to any other approach to thorium but the thorium-plutonium MOX technique. He completely ignores the merits of MSRs, which have tremendous advantages in the thorium fuel cycle that he did not disprove in this paper.

Overall, the conclusion is reached that the thorium fuel cycle at best has only limited relevance to the UK as an alternative plutonium disposition strategy and as a possible strategic option in the very long term for any follow-up reactor construction programme after LWR new build. Suggestions that thorium fuelled reactors may be able to achieve superior safety performance to new build LWRs will take many years to substantiate and are not likely to be helpful to meeting the UK’s strategic priorities. Nevertheless, it is important to recognise that world-wide there remains interest in thorium fuel cycles and this is not likely to diminish in the near future. It is may therefore be judicious for the UK to maintain a low level of engagement in thorium fuel cycle R&D by involvement in international collaborative research activities. This will enable the UK to keep up with developments, comment from a position of knowledge and to some extent influence the direction of research. Participation will also ensure that the UK is more ready to respond if changes in technology or market forces bring the thorium fuel cycle more to the fore.

This is Mr. Hesketh's conclusion. It is an incorrect one. I urge you to examine all of the data he chose to ignore, particularly how MSR technology makes safe and economical use of thorium possible.

PostPosted: Sep 17, 2012 4:54 pm 

Joined: Jan 16, 2012 7:15 am
Posts: 94
With respect to the comments on Metric 10 - Reliability: You argue that an MSR avoids down time because it doesn't need to stop operations to refuel. This is a good point, but it is not relevant to reliability. It is relevant to Availability, which is not the same thing.

The US Army has an acronym for the integrated requirements of Reliability, Availability, and Maintainability (RAM) . The uppermost requirement is a high degree of Availability. Reliable systems provide the simplest route to high availability. (However, you can have an unreliable system that provides good availability if you can fix it quickly. Not that I recommend this approach. 8) )

The NNL report uses reliability, rather than availability, as a measure of merit. This is their mistake. Your argument is valid, but outside their limited definition. You must first convince them that Availability is a more relevant measure, then your argument makes sense.

PostPosted: Sep 17, 2012 6:24 pm 

Joined: Sep 15, 2011 7:58 pm
Posts: 186
Joshua Maurice wrote:
robert.hargraves wrote:
Section 4.5 Proliferation states "Attempts to lower the fissile content of uranium by adding U-238 are considered to offer only weak protection, as the U-233 could be separated relatively easily in a centrifuge cascade in the same way that U-235 is separated from U-238 in the standard uranium fuel cycle."
I must have missed a memo. If that constitutes only weak protection, exactly what is strong protection? I was under the impression that obtaining the fissile material is the hard part, and the rest is comparatively child's play. I was also under the impression that it's not terribly difficult to obtain raw uranium ore, and that the difficult part is to centrifuge it. How is this anything but intellectual dishonesty, ignorance, or a gross accidental error? Perhaps it makes more sense in context?

Again, I'm still curious how this is anything other than an outright lie.

PostPosted: Sep 17, 2012 7:56 pm 
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1947
Location: Montreal
Joshua Maurice wrote:
Joshua Maurice wrote:
robert.hargraves wrote:
Section 4.5 Proliferation states "Attempts to lower the fissile content of uranium by adding U-238 are considered to offer only weak protection, as the U-233 could be separated relatively easily in a centrifuge cascade in the same way that U-235 is separated from U-238 in the standard uranium fuel cycle."
I must have missed a memo. If that constitutes only weak protection, exactly what is strong protection? I was under the impression that obtaining the fissile material is the hard part, and the rest is comparatively child's play. I was also under the impression that it's not terribly difficult to obtain raw uranium ore, and that the difficult part is to centrifuge it. How is this anything but intellectual dishonesty, ignorance, or a gross accidental error? Perhaps it makes more sense in context?

Again, I'm still curious how this is anything other than an outright lie.

I would tend to agree with you: After all, that is the justification for ORNL's project to denature its stock of U233 - by denaturing with U238.

I suspect however, that the NNL may be referring to a particular degree of denaturing, which is not equivalent to natural uranium (i.e. 0.71% fissile).
Since they fail to make that clear, its only speculation on my part.
But its clear that denaturing to only, say, 20% fissile, makes centrifuging to HEU relatively easy....

PostPosted: Sep 18, 2012 2:59 am 

Joined: Sep 15, 2011 7:58 pm
Posts: 186
Ah jaro, thanks. That makes a lot of sense. Would be nice if they clarified that though, lol.

Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 26 posts ]  Go to page 1, 2  Next

All times are UTC - 6 hours [ DST ]

Who is online

Users browsing this forum: No registered users and 1 guest

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group