Passive Reactors: Theory & Toshiba 4S | Nuclear News

[theCandyman]

Could someone tell me and explain explain the theory behind passive reactors? I just read about the Toshiba 4S in Nuclear News when I got back to school (one of the only nuclear related perodicals in the library). Right now I take it that instead of producing more heat at higher tempetures, it slows down the fission process. Is that it?

 

[Morbius]

Could someone tell me and explain explain the theory behind passive reactors? I just read about the Toshiba 4S in Nuclear News when I got back to school (one of the only nuclear related perodicals in the library). Right now I take it that instead of producing more heat at higher tempetures, it slows down the fission process. Is that it?

Candyman,

I believe you are inquiring about a "passively safe" reactor - or one with "passive shutdown".

It means a reactor that, in case of an accident, will safely shutdown and cool itself without
the need of an "engineered system" to work. Emergency cooling pumps, emergency
shutdown control rods... any system that is supposed to work to handle an emergency;
theoretically has a finite probability of failing. In a "passively safe" or "inherently safe"
reactor - the safety of the reactor doesn't depend on some bit of machinery "working".

All reactors have inherent feedback mechanisms that tend to shutdown the reactor
should it start to runaway or get too hot. In a water moderated reactor, like a typical
power reactor; as the temperature of the reactor and its coolant goes up - the water
becomes less dense at the higher temperature. Therefore, it is less of a moderator,
and this tends to decrease reactor power - which is just what you want to happen if
the reactor starts getting too hot.

There's another mechanism, more complex; called Doppler broadening of absorption
resonances that also works to shutdown the reactor as it gets hotter.

All reactors have these feedback mechanisms, but in a passively safe reactor those
inherent feedbacks are designed to be so strong that even if the control rods fail - they
are strong enough to shutdown the reactor.

There's the additional problem of getting rid of the decay heat after the reactor has
shutdown. Most have emergency cooling pumps - but a pump could always fail. In
a passively safe reactor is designed so that natural convection cooling will suffice.

So the safety of a passively safe or inherently safe reactor doesn't depend on a pump
working, or a control rod dropping when requested. No - the reactor is kept safe by
the laws of physics which don't fail.

Dr. Gregory Greenman
Physicist

 

[Clausius2]

All reactors have inherent feedback mechanisms that tend to shutdown the reactor
should it start to runaway or get too hot. In a water moderated reactor, like a typical
power reactor; as the temperature of the reactor and its coolant goes up - the water
becomes less dense at the higher temperature. Therefore, it is less of a moderator,
and this tends to decrease reactor power - which is just what you want to happen if
the reactor starts getting too hot.

Sometime I heard Chernobyl reactor wasn't an inherently safe reactor. Is it true?. Then, why it wasn't?

 

[hitssquad]

Sometime I heard Chernobyl reactor wasn't an inherently safe reactor. Is it true?. Then, why it wasn't?

RBMK reactors feature a positive void coefficient instead of a negative void coefficient. When the coolant is voided in a normal reactor, the reaction is poisoned (slowed down or stopped). When the coolant in an RBMK is voided the reaction speeds up.

RBMK reactors are also unstable at low operating power. The fuel ([EDIT: fuel should be control) rods have to be withdrawn farther and farther to maintain constant power levels, but with the control rods ultimately so far out the reactor power can suddenly spike upward. For this reason, RBMK reactors are supposed to never be run at low power levels.

en.wikipedia.org/wiki/Chernobyl_accident

 

[theCandyman]

The fuel rods have to be withdrawn farther and farther to maintain constant power levels, but with the control rods ultimately so far out the reactor power can suddenly spike upward.

How does the power spike with the fuel rods withdrawn?

Edit: Control rods withdrawn, not fuel rods.

 

[hitssquad]

How does the power spike with the fuel rods withdrawn?

Typographical error.

 

[Clausius2]

Thanks, hittsquad.

But now, why the hell russia employed these kind of reactors? Did they have any advantage like being more cheaper or so? I remember they employed CO2 gas as coolant (it is possible I'm wrong here).

1) am I wrong?
2) if I don't, why using a gas as coolant instead of water?.

 

[Morbius]

How does the power spike with the fuel rods withdrawn?

Candyman,

Another indication of the poor design of the RBMK reactor.

The control rods in the RBMK had fueled "followers". The axial center section of the
control rod contained a neutron absorber. However, the axial ends of the control rod
contained fissile fuel.

Therefore, when the control rods are inserted into the reactor - the very first part of the
control rod insertion involves the fueled follower entering the core. So during the very
first part of the control rod insertion, the control rods are actually adding fuel or moving
fuel to a more reactive part of the core. It is only after this initial reactivity insertion,
that the control poison is inserted into the reactive parts of the core.

It is the intial insertion of fuel, and its increase in reactivity that causes the power spike.

Dr. Gregory Greenman
Physicist

 

[hitssquad]

Why the RBMK design was chosen for Chernobyl

The reactors were intended to be frequently refuelable so that weapons grade plutonium could be manufactured while electric power was simultaneously produced. The RBMK design was chosen as the most-expedient way to achieve that dual-goal.

Water is not just a coolant candidate. It is also a neutron moderator. In the RBMK design water is used, but in other graphite-moderated designs such as the one at http://www.nukeworker.com/nuke_facilities/North_America/usa/NRC_Facilities/Region_4/fort_st_vrain/index.shtml in Platteville, Colorado (now decommissioned), CO2 was used.

 

[Morbius]

But now, why the hell russia employed these kind of reactors? Did they have any advantage like being more cheaper or so? I remember they employed CO2 gas as coolant (it is possible I'm wrong here).

Clausius2,

The RBMK were dual purpose reactors. They were intended to produce power as well as
produce plutonium for the Soviet nuclear weapons program.

The RBMK is actually a scaled up version of one of the U.S.S.R.'s production reactor
designs. In fact, that's part of the problem with it. When they scaled it up - they didn't
redo the nuclear design for the larger size.

Imagine you had a smaller production reactor - and you built a larger version by stacking
multiple copies of the small reactor in a 2 x 2 x 2 block. [ Imagine stacking 8 child's
building blocks].

The new reactor would have less neutron leakage than the 8 small reactors. At any of
the internal interfaces were one block touches another - one reactor leaks neutrons into
its neighbor - and by symmetry - vis-a-versa. Therefore, at these symmetry planes, there
is no net neutron leakage - whereas there would be if the small reactor block was free-standing.

The RBMK designers didn't account for the reduced leakage in the nuclear design of the
RBMK - they kept the nuclear design the same as their old tried and true production
reactor.

So safety lost out to the interest in producing material for Soviet nuclear weapons.

Dr. Gregory Greenman
Physicist

 

[theCandyman]

Typographical error.

I noted that in my post as well, then.

The control rods in the RBMK had fueled "followers". The axial center section of the
control rod contained a neutron absorber. However, the axial ends of the control rod
contained fissile fuel.

What is the reasoning behind that, to "jumpstart" the reactor?

In fact, that's part of the problem with it. When they scaled it up - they didn't
redo the nuclear design for the larger size.

The RBMK designers didn't account for the reduced leakage in the nuclear design of the
RBMK - they kept the nuclear design the same as their old tried and true production
reactor.

That seems like a silly mistake, if they had thought about it more, would they have noticed that problem?

 

[Morbius]

What is the reasoning behind that, to "jumpstart" the reactor?

Candyman,

I think they were trying to flatten the power distribution by putting more fuel near the
top and bottom of the reactor where the closer proximity of the reactor boundary leads
to increased neutron loss due to leakage - or something like that.

Whatever they were trying to do it was a DUMB, DUMB, DUMB idea!

Any benefit of flatting power or increasing efficiency is not worth the compromise of
the safety system.

That seems like a silly mistake, if they had thought about it more, would they have noticed that problem?

They knew about the problem.

I remember one documentary on Chernobyl that featured the late Nobel Laureate Hans Bethe
saying that U.S. scientists had told the Soviets of the problems with their reactors - and
that they needed to have containment buildings. Bethe said the response received from
the Soviets was "Oh no; our reactors are safe".

They were just so hell-bent on getting more weapons material that safety went by the
boards.

Dr. Gregory Greenman
Physicist

 

[Astronuc]

Fort St. Vrain was a high-temperature, gas-cooled reactor (HTGR or HTGCR) which used He gas for cooling not CO₂. IIRC, CO₂ would have reacted with the graphite to produce CO. FSV was a one-of-a-kind reactor, designed by Gulf General Atomic (now General Atomic). There were significant issue with hydraulic instability, and steam instrusions into the He primary circuit. There were plans for 8 other units of a more advanced design, but those died after TMI. But there were numerous technical problems with the design.

Passive reactors are designed to shutdown and remove residual (decay) heat after shutdown without need of active heat removal systems or operator intervention. Natural convection one of the mechanisms used.

Water-cooled, graphite moderated reactors are inherently unsafe because the cooling water is actually an absorber of neutrons, which are primarily moderated by graphite. The reduction in density, particularly when the water goes to steam results in an increase in reactivity - so if a system is critical, positive reactivity ($\Delta k$ > 0) causes the system to go supercritical. The greater the $\Delta k$, the greater the power multiplication, and some technical limits - such as fuel melting or system pressure - may be exceeded - as was the case at Chernobyl.

 

[Morbius]

Water-cooled, graphite moderated reactors are inherently unsafe because the cooling water is actually an absorber of neutrons, which are primarily moderated by graphite. The reduction in density, particularly when the water goes to steam results in an increase in reactivity - so if a system is critical, positive reactivity ($\Delta k$ > 0) causes the system to go supercritical.

Astronuc,

You can't make that categorical statement about water-cooled, graphite moderated reactors.

The problem that leads to the instability that you detail above is due to the reactor being
"over-moderated". That is the reactor has more moderator than optimal - thus any decrease
in moderator - such as by the heating and/or voiding of the water; results in a positive
reactivity insertion.

However, just because a reactor has both a graphite moderator and is water cooled
doesn't mean, of necessity, that it is over-moderated. It is quite possible to design
a water cooled graphite reactor that is under-moderatored and stable. One just needs
to have less graphite - so that the cooling water is needed to provide moderation just as
in a water cooled / water moderated reactor like a typical power reactor.

The RBMK was over-moderated and unstable - but that is not inherent in the reactor
design type. It was a result of poor execution of the design.

Dr. Gregory Greenman
Physicist

 

[hitssquad]

Fort St. Vrain was a high-temperature, gas-cooled reactor (HTGR or HTGCR) which used He gas for cooling not CO₂.

OK. Thanks. Sorry about reporting it was CO₂.

 

[Astronuc]

NBD. Possibly you were thinking of the Magnox or AGR plants in the UK. Both types are gas-cooled, using CO₂.

http://en.wikipedia.org/wiki/Magnox

http://en.wikipedia.org/wiki/Advanced_gas-cooled_reactor

The AGRs use graphite though, so it would appear to be compatible with grahphite up to 640°C.

 

[theCandyman]

Recently, a couple of AREVA employees come to talk about job opportunities with the company. They gave a slide show and one of the reactors had a container for the core to melt into, I just wanted to know if it was still possible for a rector designed today to have a meltdown or is the container just for the public to feel safe?

 

[Astronuc]

That's a tough question to ask, because a designer must assume the worst possible scenario and then attempt to prevent the possibility. Most likely, passive LWR designs would not permit a core melt, and most current designs will likely not permit a core melt either.

One has to consider what type of conditions would lead to a core melt situation. First, there must be some type of 'disruptive' accident that 1) total disrupts the core geometry (loss of coolable geometry) - e.g. a loss of coolant accident (LOCA) or reactivity insertion accident (RIA), and then 2) complete loss of cooling capability, e.g. loss of main pumps, auxillary pumps and sump pumps.

The worst time would be at end of full power operation after some long period. At this point the decay heat would be maximum - about 0.6% of full rate power.

Taking a step back - one has to ask if the reactor can be scramed, i.e. shutdown during the accident. If not, then the power density in the core may be slightly higher than decay heat, and possibly, there might be a region that maintains criticality.

Besides the ceramic fuel, the bulk of the core structure is some zirconium alloy, with some stainsteel from control elements, and perhaps some Inconel from other structural components. If the temperature gets high enough, the Zr alloy may react with steam and form ZrO₂ and H₂ - this was one problem at TMI-2.

The melting point of Zr-alloys is about 1860°C, but it gets soft well before that, and would oxidize to ZrO₂ anyway. The oxidation of Zr also generates heat as well.

The UO₂ melts at about 2840°C, but it would also tend to oxidize to higher order oxides such as U₄O₉ and U₃O₈ with oxidation occurring along grain boundaries and so the fuel pellets would eventually crumble. Of couse, I am leaving out the thermochemical behavior of the various fission products.

The problem of accident analysis involves trying to determine how much core and fuel material accumulates in an undesirable, i.e. uncoolable, geometry at the worst possible location, which usually infers the bottom head of the pressure vessel, and this means the core/fuel material has to get below the core support plate. At TMI-2, the core partially disintegrated and accumulated to one side, and the core baffle, core support plate, and pressure vessel wall (IIRC) was somewhat eroded (partial melt). Had the pressure vessel wall ruptured, then the contents would have poured out, but it would not have necessarily melted.

At Chernobyl, there was some melting of the core structure, but that was primarily a graphite structure which had actually caught fire.