Sunday, March 20, 2011

How the LFTR would have survived the Japanese Earthquake/Tsunami

Future nuclear safety tests should include capacity to survive the events which lead to the Fukushima Dai-ichi nuclear plant crisis. The real survival test would require the same flawed backup generator system that was destroyed by the tsunami that struck the plant. Of course, we are not going to talk about every conceivable tsunami. For example, it is highly likely that the island of Oahu will rupture someday, dropping a large part of it into the Pacific Ocean, creating a huge tsunami. That would be a megatsunami. Since megatsunami can be up to 1000 meters high, we need not worry about a coastal reactor surviving all tsunami. No one will be around in the vicinity of a 1000 meter megatsunami to worry about a subsequent nuclear accident.

This leads us to the question of acceptable nuclear risks, a fit topic for another post. At any rate we are looking, right now, at how well the Liquid Fluoride Thorium Reactor (LFTR) or any otherMolten Salt Reactor (MSR) would have survive the natural disaster that overwhelmed the Fukushima Dai-ichi nuclear plant.

First it should be noted that Molten Salt Reactors do not require water cooling at all. Hence the loss of the emergency cooling generator system would have not been a serious problem by itself. Lets explore an accident scenarios that is at worst remote possibility, a key pipe ruptures triggering a loss of coolant accident. Say the entire content of the core coolant system - a liquid salt mixture - drains on to the floor of the reactor chamber. and forms a puddle. Since the nuclear fuel is dissolved in the coolant salts it will be deposited into the puddle. Now the interesting thing about a puddle is that its geometry is not at all conducive to a chain reaction, so the loss of coolant in turn triggers a withdrawal of nuclear fuel from the core, which in turn triggers a termination of the chain reaction, so the reactor automatically stops functioning.

Now the puddle, even though we can expect it to be short lived, might be a problem because radioactive gaseous fission products, dissolved in the fuel salt, are likely to bubble out along with volatile fission products.

Our puddle will not stay on the floor - it quickly drains into a pipe leading into a set of emergency coolant tanks, which are intended to hold the fuel mixture until the reactor can be repaired and restarted. The geometry of the tanks would be intended to prevent a chain reaction from occurring, but the fuel salt would include some radioactive fission products capable of generating the sort of post reactor shutdown heat that created so many problems in the Japanese reactors. Is there any way to insure that the liquid salt in the emergency coolant tanks does not start boiling and releasing a lot of nasty stuff? There turnout to be several passive solutions to this problem.

One: Draw air over a simple heat exchange system designed to dissipate some of the heat in the emergency tanks. Not too much heat, since we want the coolant liquid to remain hot and well, liquid. The heated air can be directed to a chimney, through which it flows into the atmosphere. The system thus requires no power, no controls and no operators. It works automatically, relying on the laws of nature to function.

Two: Rely on a thermal sink, most likely a molten or solid salt, such as the salt that carries the fuel. The inner emergency coolant tanks could be surrounded by an outer thermal sink tank. Or the coolant tank could be shaped like a donut, with inner and outer thermal sink tanks removing heat. A large enough thermal sink would probably be sufficient to dissipate heat without requiring any further heat transfer system. There is a cost for a large thermal sink salt tank system. The Integral Fast Reactor (IFR), for example, relies on this thermal inertia to prevent the reactor from overheating during an emergency shutdown, but this approach is likely to add considerably to IFR costs. However, a thermal salt vault approach combining energy storage with the safety function of decay hear dispassion, would probably would add minimally to reactor cost, while offering a source of reserve electricity.

Three: One proactive LFTR safety approach would be to remove some or even most of the fission products from the reactor. Removal of radioactive gases would be very desirable for a number of reasons. For example. as Uri Gat, and H.L. Dodds pointed out,
The source term, which is the inventory of radioisotopes in the reactor available for dispersion to the environment, contributes two-fold to an accident. The source term is the measure of the radiation which needs to be contained from reaching any sensitive location or target. The energy contained in the source term also provides the driving force for the dispersion of the source term as it is also a measure of the after heat, or the energy, to damage a reactor in the event of heat-removal failure or loss-of-coolant accident (LOCA). For an MSR, as for any fluid fuel reactor, on-line fuel processing can be applied. The on-line processing, at the least, removes the gaseous and volatile part of the source term. This part is the most likely to be dispersed when there is a breach of containment. Fuel processing also reduces the inventory of longer and long-lived isotopes as their accumulation is time dependent. The MSRs processing can be adjusted to have a small source term. The safety advantages of this small source term are many fold: The driving force for dispersion is reduced; the gaseous and volatile components, which are the most likely to disperse, are essentially all but eliminated; the long half-life isotopes (elements) are reduced such that the long-term effect of even the most unlikely accident is not severe; and, the short-lived isotopes require a proportionately short-term protection time till they decay. Thus, even a hypothetical severe accident is ameliorated a priori.

A properly designed processing facility quickly removes the separated radioisotopes from the purview of the reactor. This makes them totally unavailable to the reactor source term even under the most extreme hypothesized circumstances.
In addition to removing fission product gases, and volatile fission products, removal of nobel metals would be highly desirable from an operational point of view. Gat and Dodds state,
The source term, which is the inventory of radioisotopes in the reactor available for dispersion to the environment, contributes two-fold to an accident. The source term is the measure of the radiation which needs to be contained from reaching any sensitive location or target. The energy contained in the source term also provides the driving force for the dispersion of the source term as it is also a measure of the after heat, or the energy, to damage a reactor in the event of heat-removal failure or loss-of-coolant accident (LOCA). For an MSR, as for any fluid fuel reactor, on-line fuel processing can be applied. The on-line processing, at the least, removes the gaseous and volatile part of the source term. This part is the most likely to be dispersed when there is a breach of containment. Fuel processing also reduces the inventory of longer and long-lived isotopes as their accumulation is time dependent. The MSRs processing can be adjusted to have a small source term. The safety advantages of this small source term are many fold: The driving force for dispersion is reduced; the gaseous and volatile components, which are the most likely to disperse, are essentially all but eliminated; the long half-life isotopes (elements) are reduced such that the long-term effect of even the most unlikely accident is not severe; and, the short-lived isotopes require a proportionately short-term protection time till they decay. Thus, even a hypothetical severe accident is ameliorated a priori.

A properly designed processing facility quickly removes the separated radioisotopes from the purview of the reactor. This makes them totally unavailable to the reactor source term even under the most extreme hypothesized circumstances.
Removing all radioisotopes from a Molten Salt Reactor removes the protection that those isotopes afford. As long as the salt contains radioactive fission products, it will be far too dangerous to handle for nefarious purposes, such as the eternal bogeyman of nuclear proliferation. Salt processing can be conducted by automatic equipment inside the reactor core hot cell. The heat and radiation inside the hot cell would prevent anyone having near real-time access to a MSR.

One way of managing a reactor situation that is likely to lead to an accident, is to design a built in failure point, analogous to an electrical fuse or other weak link, which will fail before anything else. One such deliberate failure point in the MSR is the freeze valve; if a LFTR or other MSR begins to overheat, the freeze valve is designed to melt as Gat and Dodds explained,
The MSR can utilize freeze valves in critical locations or where desired. Freeze valves can be ordinary sections of pipe which are exposed to a cooling stream of environmental gas to the extent that it creates a frozen plug that blocks the flow and acts as a valve. Where such a valve has a safety function, as in draining the fuel to the storage tanks, it is prudent to design it such that the required flow is
gravity-driven. The frozen valve itself can be designed such that when the salt rises above a certain predetermined temperature the heat overrides the cooling, melts the frozen plug and opens the valve. Such an arrangement is passive, inherent and non-tamperable (PINT-safe).

Furthermore, the properly sized external cooling of the freeze valve cooling drive, such as an electric driven fan, will cease with any failure of the power and release the valve to melt and perform its safety function. This mode of operation is again PINT-safe.
Once the freeze melts, a MSR will simulate a total loss of coolant accident, with fuel/coolant salts dumped into a tank or tanks that are designed with a criticality inhibiting geometry. In his paper 2006 paper, Molten-Salt-Reactor Technology Gaps (Proceedings of ICAPP ‘06, Reno, NV USA, June 4–8, 2006, Paper 6295), MIT nuclear scientist Charles Forsberg stated,
Under emergency conditions, the liquid fuel is drained to passively cooled critically safe dump tanks. By the use of freeze valves (cooled sections of piping) and other techniques, this safety system can be passively initiated upon overheating of the coolant salt. MSRs operate at steady-state conditions, with no change in the nuclear reactivity of the fuel as a function of time. Last, the option exists to remove fission products online and then solidify those radionuclides into a stable waste form. This minimizes the radioactive inventory (accident source term) in the reactor core and potential accident consequences.
In addition to the very useful freeze plug, the capacity of molten salt to freeze at a still relatively high temperature is directly responsible for another MSR, its automatic leak control. As hot salt leaks from reactor piping, it begins to cool on contact with hot cooler air, and as it cools, it freezes, blocking further escape of coolant salt. Reportedly this mechanism is very effective in stopping leaks if they occur.

At this point I have established the case which I have sought to prove, the tsunami that destroyed the back up generators of the Fukushima Dai-ichi nuclear plant, would have left the safety systems of the LFTR in tact.

This is not the only MSR advantage. According to a group pf French nuclear scientist from the University of Grenoble, the MSR does not simply offer a high probability of safety, it offers an
excellent level of deterministic safety,
That is, safety depending only on the laws of nature and thus safety that is beyond doubt. The MSR is uniquely stable. It can be designed to safely operate without any human intervention, until such time as repairs or parts replacement is required. Thus, MSRs do not require on site operators, and indeed the stability and load following ability of the MSR are such that operators would have quite literally nothing to do. The absence of human operators would probably add to MSR safety, rather than inhibit it. that is safety that depends on the laws of nature and thus safety that is beyond doubt.

It is probably true that the safety systems of the AP-1000 and the ESBWR would have survived the Dai-ichi tsunami. But compared to the simple safety features of MSRs, the safety systems of even advanced LWRs are complex and expensive. Molten salt nuclear technology offers many potential cost saving advantages, and if all of them are employed, MSR costs could be substantially lower than the costs of LWRs. Part of the MSR advantage is higher safety at lower cost.

13 comments:

Mark Roy said...

Hubris.

Charles Barton said...

Really, Mark Roy. You make that judgement without the slightest hint of its basis. Perhaps you believe that you don't have to identify a problem, before you announce that a mistake has been made. Now that is hubris.

Greg Barton said...

I think Mark's post is ironically self referential.

Robert Steinhaus said...

There is an earthquake intensity number that will destroy any energy production system, including Thorium LFTRs. The troubled Japanese reactors sustained a 500% over design event in the recent earthquake – tsunami. Charles, I am fully prepared to lose this argument (as I would prefer not to give aid and comfort to anti-Thorium anti-Nukes) but I would like to suggest that, until a full end to end review of a full complete modern LFTR design is made, we will not be able to say that there are NO safety vulnerabilities left in the reactor design. Safety vulnerabilities frequently lurk in the design detail . . . rare strange little places in the design where rare events can expose safety vulnerability.
ORNL designers went to some trouble to keep water out of their molten salt reactors. In the case of the MSRE experiment, ORNL spent several days baking out all of the plumbing in the reactor and backfilling with non-reactive gas to eliminate all water and even water vapor from all of the plumbing and pipes that would carry hot molten salt. One of the stronger intrinsic safety features of a LFTR is that it normally operates at low pressures as molten salt is pumped around the reactor at near one atmosphere. The vast majority of a LFTR reactor is totally enclosed inside of pipes that are easily strong enough to stand up to low pressure, but there are typically some small entry points (in the area of the pump bowl of the molten salt cooling pump and in the support chemical plant where hot salt can be accessed directly) that are often open. If a tsunami were to wash up over a modern LFTR and flood the reactor room with water, the water could find any open entries in the closed plumbing and piping of the reactor. Significant amounts of water could get inside of pipes intended to normally only hold low pressures, and there is the possibility of safety problems from high pressure steam. There are also potential concerns for what might occur if the earthquake were strong enough to break the molten salt drain tank and allow molten salt to cover the bottom of the reactor building and collect in small confined spaces like perhaps the well of a sump pump designed to pump out any water that might enter the facility. In the case of a strong earthquake, even for designs as intrinsically safe and robust as LFTR, there is need of a careful and comprehensive safety revue to expose all potential safety vulnerabilities.

Charles Barton said...

Robert, I did not say that there are no safety problems with the LFTR. There are safety choices, and choices mean that mistakes are possible. But What I argue here is that the LFTR would have survived the Dai-ichi accident. This is true of some other reactor designs.

seamus said...

Nice writeup. The linked PDFs are interesting too.

Sione said...

Charles

What Robert is pointing out is that you do not know whether a LFTR reactor would have survived. Potentially it might have, but potentially the present jap reactors might have been made robust enough to survive intact and safe. Turns out they were vulnerable to circumstances the designers and operators failed to consider (presumably because like all right-thinking "scientists" they said the probabilities of those circumstances occurring were too low to worry about).

The details of specific design are what counts. Until the real design is actually prepared and properly reviewed and tested, it is not possible to know that a LFTR would have done any better than those wrecked jap disasater piles. All that can be said is that according to your opinion, LFTR should be superior in this regard.

Still, next time a design team prepares a reactor design it is likely they'll incorporate some lessons from this present situation (or so one might hope). Perhaps they'll find that the attributes of the LFTR process makes it easier to design a safer plant. I hope so.

I'm sympathetic to your claims and opinions regarding LFTR but claims and opinions are not the equivalent to knowledge.

Sione

Charles Barton said...
This comment has been removed by the author.
Charles Barton said...

Robert and Sione, I have referenced safety features which have been included in all MSR designs and safety features that are simply essential characteristics of all Molten Salt Reactors. i have not relied on my own judgement, but on the judgement of scientists and engineers who have studied the safety features of molten salt reactors. These are undisputed judgements, since no one has ever produced a serious paper disputing those statements. I would thus regard them as well attested facts, unless or until someone demonstrates that they is contradictory evidence.

donb said...

Robert Steinhaus brought up a number of legitimate concerns. I agree that we need to make nuclear reactors safe.

However, we also need to keep perspective. If our standard is to make reactors "as safe as possible", we will never build any. If we never build any, other much more dangerous forms of energy will be used in their place. We have "the perfect" preventing "the better" from being built.

No only do we need to make nuclear reactors safe, we also need to consider what will happen if we do not build those reactors. We will not be moving ahead in safety if (say) a coal-fired power plant gets built because we are unable to build the perfectly safe nuclear power plant. Instead, we need to build the nuclear power plant that is safer than the coal power plant, sooner rather than later. And we need to continue making reasonably safe nuclear power even safer.

Charles Barton said...

donb, i am writing a post to address the issues which Robert raised. I regard Robert's comments as non responsive to my posts, since Robert alters my assumptions, and focuses on issues which I indicated I would not address.

Frank Kandrnal said...

In my opinion the LFTR reactor has far better safety features than LWR. The big difference is in fuel inventory. LWR must have more fuel than LFTR to operate. In addition LWR must have enough fuel to run for several years, therefore, over critical mass is present. For this reason there is always greater risk for accident and radioactive release.
LFTR does not need excess fuel in the core, it is added as it is burned away so there is always far less fissile material than in LWR. When fission products are removed on the go in order to improve neutron economy there is then very little radioactive material to be spread around when all reactor is somehow destroyed by extreme earthquake or bunker busting bomb

Jim Baerg said...

Kirk Sorenson has suggested putting LFTRs in submarines places well off shore. Since tsumanis only become destructive as the water gets shallow, this suggests a way to make LFTRs immune to damage from tsunamis & earthquakes.

Followers

Blog Archive

Some neat videos

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