Nuclear Power and Reactor Scale.

[ohwilleke]

Most commercial nuclear fission reactors are large, on the 1000MW order of magnitude.

Some of the early experimental reactors were small, and the reactors in nuclear submarines are relatively small. I understand that the Navy even has one very small (crew of 7 people, 400 tons v. 7,000-9,000 for nuclear attack submarines) nuclear submarine that also moves very slowly 4 knots, which does searchs of the ocean floor (e.g. for wreckage). See here: http://www.chinfo.navy.mil/navpalib/factfile/ships/ship-nr1.html A couple of commercial ships of a freighter size were once nuclear, but have been converted back to conventional fuel.

I've also heard about self-contained small nuclear reactors that they're talking about trying in an Alaskan village (on a pebble bed concept, IIRC). See http://www.adn.com/front/story/4214182p-4226215c.html

What are the practical limits that impact the scale of a nuclear fission reactor? For example, what prevents someone from developing a nuclear powered airplane?

 

[hitssquad]

Know Nukes on small and airborne reactors

I've also heard about self-contained small nuclear reactors that they're talking about trying in an Alaskan village

The Toshiba 4s.
http://www.google.com/search?q=toshiba+4s+sodium

(on a pebble bed concept, IIRC)

The Toshiba 4S is a sodium-cooled fast-neutron reactor that has virtually no moving parts and that for years can opertate without refueling. The pebble bed is a gas-cooled slow-neutron reactor with lots of moving parts and that needs to be refueled constantly and continuously.

What are the practical limits that impact the scale of a nuclear fission reactor?

One frequently-cited factor is that heat engines tend to be more efficient when they are larger.

For example, what prevents someone from developing a nuclear powered airplane?

Graham Cowen weighed in on this in the recent Better liquid oxide coolant thread on the Know Nukes list:
http://groups.yahoo.com/group/Know_Nukes/message/12046

--
> >No they wouldn't. Multiple separate rad shields
> >on separate reactors would be needed, and each
> >would have to be for real: decimetres of dense hydrogen
> >plus decimetres of dense heavy metal, all blocking every
> >ray path. Eight-pi shielding, or 12-pi, counting
> >the multiplicity of reactors. A nuclear aircraft
> >would be a thousand-tonne aircraft, maybe 2,000.
> I misread the original post. You expect to fly a reactor!

Zero or two or four, I think; not just one.
One did fly ~50 years ago, once, as a passenger.
The shield weighed 60,000 pounds IIRC;
that's not for real.

I think nuclear aircraft may fly if some worthwhile job
is found that only aircraft of 2,000 tonnes or more can do,
because airport and flight path neighbours will be more
comfortable with them than with conventionally powered
aircraft that big, i.e. with ~1,000 tonnes of kerosene or lH2,
50 to 100 terajoules of potential BLEVE, on board.
Wouldn't you be?
--

In short, nuclear aircraft would certainly be safer than chemical aircraft, but they might have to be very large.

 

[Astronuc]

Right now the largest practical light water reactor (LWR) in commercial operation is the French N4 series. Civaux-1 is one of four of the new N4 nuclear reactor series. The other three being: Chooz B-1, Chooz B-2 and Civaux-2, each with a capacity of 1516 MWe (gross), 1450 MWe (net), with a thermal capacity of approximately 4270MWth. The core consists of 205 assemblies with an active fuel length of 4.27 m (14 ft).

Some background - http://www.memagazine.org/backissues/aug98/features/reactor/reactor.html

and

http://www.worldenergy.org/wec-geis/publications/default/tech_papers/17th_congress/3_2_03.asp

There is also the EPR (European Pressurized Reactor) Project. The EPR is a four-loop reactor designed for a thermal output of 4250 MWth and an electrical output capacity of 1500 MWe. Its core comprises 241 fuel assemblies, each containing 264 fuel rods and 81 control rods.

Apparently the Russians are planning a VVER-1500 (1500 MWe) design as well.

Nuclear powered aircraft would seem to be impractical from a thrust to weight ratio. The major impediment is the shielding.

 

[Morbius]

What are the practical limits that impact the scale of a nuclear fission reactor? For example, what prevents someone from developing a nuclear powered airplane?

ohwilleke,

The reactor core can be quite small, physically - but still put out a lot of
power. The limit to how much power you get out of a reactor has more
to do with how effectively you can cool it, rather than any limit as to
how much the core can generate.

The core of the research reactor at M.I.T. is about 15 inches in diameter
and produces 5 Mwt:

http://web.mit.edu/nrl/www/reactor/reactor.htm

http://web.mit.edu/nrl/www/reactor/core_description.htm

The core of the now shutdown University of Michigan reactor was a
2-foot cube from which they extracted 2 Mwt. However, that core was
essentially the core of a submarine reactor - and the 2 Mwt limit was
only because that's how much heat the cooling system for the University
of Michigan's reactor could extract. Put a cooling system with more
capacity on it - and you can power a submarine.

http://www-ners.engin.umich.edu/research/index.shtml

Although you could devise a reactor that could output the power of
a few jet engines - the shielding necessary would make the plane too
heavy. So a reactor-powered plane isn't practical.

Dr. Gregory Greenman
Physicist

 

[villiami]

I'm not sure about this, but I've heard some pacemakers where powered by small amounts of plutonium

 

[Astronuc]

Plutonium-238 (non-fissile isotope) has been used to power scientific equipment in spacecraft and implanted heart pacemakers.

See also Nuclear Batteries

 

[Morbius]

I'm not sure about this, but I've heard some pacemakers where powered by small amounts of plutonium

villiami,

Those aren't nuclear reactors. They are called RTGs for
Radioisotope Thermal Generators.

Whereas nuclear reactors derive their energy from the fissioning of
fissile materials like U-235 and Pu-239; RTGs get their energy from the
heat produced by radioactive decay of some radioisotope, most notably
Pu-238, as Astronuc states.

In an RTG, you let the radioisotope generate heat due to its radioactive
decay, and you then convert the heat to electricity - a thermocouple is
one way of doing that.

The requirements for the "fuels" of reactors and RTG are different. The
reactor requires a fuel like U-235 and/or Pu-239 which is fissile. The
fuel doesn't have to be radioactive - in fact U-235 has a pretty low level
of radioactivity.

For the RTG, being fissile is immaterial - you want something that is
quite radioactive so it produces a lot of heat. That's Pu-238.

So nuclear reactors and RTGs are two very, very different animals. The
RTGs are what power pacemakers.

Dr. Gregory Greenman
Physicist

 

[cwbiii]

Nasa did some experimentation with using a reactor as a rocket engine... to my understanding the tests were successful as to whether it would work or not. I don't think it was practical for actual use because of the radiation it spewed out in the process... and what we send up tends to come back down again, sometimes unexpectantly. It could probably be used in deep space for power generation for an ion rocket engine... but it would need to be assembled there and kept a long way from
the rest of the science package or crew.

 

[theCandyman]

When did they test this? And how would you use a reactor as a rocket engine in the first place?

 

[hitssquad]

Nuclear rocketry

When did they test this?

http://www.nas.nasa.gov/About/Educa...sons/contributed/thomas/Adv.prop/advprop.html

--
Nuclear rockets have, in one way or another, been studied for the past fifty years.
--

And how would you use a reactor as a rocket engine in the first place?

You use the reactor-generated heat directly, or you use electricity generated by the reactor, or you use both of those things to launch propellant out of a rocket nozzle. For example, if the propellant is ionized hydrogen, you might use electrical energy to power electromagnets arrayed so as to accelerate the hydrogen ions. Some nuclear-rocket designs that use this mode of operation also use heat from the reactor to heat up the hydrogen propellant, and hence these designs use both heat and electromagnetics to provide energy to the propellant.

The above link has more detail on nuclear rockets. Here is another one:
http://www.lascruces.com/~mrpbar/rocket.html

There are may nuclear propulsion designs. One I am interested in is the vapor core (also known as gas core) reactor with MHD generator. Vapor (or gas) core means the fission fuel is normally in a vapor state instead of a solid or liquid state. This technology in a nuclear rocket might be very efficient since it allows the reactor to operate at a high temperature and because heat engine efficienvy tends to rise with rising temperature (per law[/url]). Here is a link to a general description of a vapor core reactor with MHD generator:
http://www.inspi.ufl.edu/research/gcr/

 

[Morbius]

When did they test this? And how would you use a reactor as a rocket engine in the first place?

Candyman,

Do a Google search on "NERVA" - Nuclear Engine for Rocket Vehicle Applications.

Here's some of what I found:

http://grin.hq.nasa.gov/ABSTRACTS/GPN-2002-000144.html

[This one has some good drawings of the NERVA engine when you click
on the "Image Information" section. ]

http://www.aemann.pwp.blueyonder.co.uk/ spacecraft /nerva/reactor.html

http://www.fas.org/nuke/space/c04rover.htm

Dr. Gregory Greenman
Physicist

 

[CrazedMathematician]

There is also MITEE, which is a rather recent design study on nuclear thermal propulsion. It can be found at this site:

http://www.newworlds.com/mitee.html

The weight of there main reactor design using U-235 is about 70 kg. What's more surprising is the whole engine weighs just 140 kg, puts out 14000 N of thrust, and has a specific impulse of 1000 seconds (over twice that of liquid oxygen/liquid hydrogen). They also have some other reactor designs using more exotic nuclear materials like U-233 and Am-242m, which shrink the reactor down to 40 kg and 25 kg, respectively.

I too have wondered if a reactor could some how be miniaturized, small enough to power something like a car, boat, or small airplane. Maybe using a closed cycle with a small gas turbine for power generation? I had the idea looking over this site:

http://www.microjeteng.com/shaft.html

It's a miniature turbo shaft engine. You would need a lot of radiation shielding though, wouldn't you? :grumpy:

The real question is, how far can you miniaturize it? Like, could you make one the size of a desktop computer power supply, or a laptop battery? :tongue2:

 

[Astronuc]

Shielding is certainly an issue with any fissile system.

Miniturization will be limited by critical mass at whatever enrichment one chooses. The pits in nuclear weapons are about as small as it gets. Am242 would allow a slightly small CM than Pu239. The key issues in a mini-reactor for vehicle transportation are:

Control of the mini-reactor.
Heat transfer.
Shielding.
Disposition of fission products.

Could you make one the size of a desktop computer power supply, or a laptop battery?

No. The diameter of CMs of pure fissile materials are on the order of cm's, and then one must add shielding.

 

[enigma]

What are the practical limits that impact the scale of a nuclear fission reactor? For example, what prevents someone from developing a nuclear powered airplane?

What if it crashes?

What if it's hijacked?

Nuclear reactors and subs are able to be reasonably isolated and kept secure from the general populace. Aircraft are not.

 

[Morbius]

What if it crashes?

What if it's hijacked?

Nuclear reactors and subs are able to be reasonably isolated and kept secure from the general populace. Aircraft are not.

ohwilleke/enigma,

Nuclear propulsion would not be used in aircraft - nuclear propulsion is
for SPACECRAFT!

I believe that falls in the "reasonably isolated" category - even more so
than subs.

As far as crashing - NOBODY intends a nuclear propulsion system to be
used at takeoff. The nuclear powered spacecraft would be lifted to
Earth orbit via regular chemical rockets.

Since the reactor won't have been started yet - the fuel will be no more
radioactive than it was when first dug out of the ground. Therefore,
a crash due to failure of the chemical rockets during boost phase would
have minimal consequences.

After the rocket is in orbit - and the nuclear engines are fired up - the
rocket is not coming back down - it already has orbital velocity - and
unless one were to use the nuclear engine as a retro-rocket - and nobody
would ever do that - the craft is not coming back to Earth.

Dr. Gregory Greenman
Physicist

 

[theCandyman]

I was wondering about that. Thanks for asking and clearing that up enigma and Morbius.

 

[enigma]

ohwilleke/enigma,

Nuclear propulsion would not be used in aircraft - nuclear propulsion is
for SPACECRAFT!

Well, his original question was:

what prevents someone from developing a nuclear powered airplane?

and I was responding to that question with a few real-life reasons why it would be a patently bad idea.

 

[ohwilleke]

So does this follow because there are economies of scale in shielding? Or is it simply that anything short of a boat or submarine can't allocate a large enough proportion of vehicle weight to propulsion and shielding to make it work?

 

[Morbius]

So does this follow because there are economies of scale in shielding? Or is it simply that anything short of a boat or submarine can't allocate a large enough proportion of vehicle weight to propulsion and shielding to make it work?

ohwilleke,

That's pretty much it. You have to have a vehicle big enough so the penalty
that you pay in shielding is a small fraction of the vehicle.

If you want to power an airplane - weight is a premium - and a reactor
doesn't make much sense.

However, if you are going to power an aircraft carrier - then the shielding
is a very, very small part of the vehicle - and reactors are good powerplants
for carriers.

Even though Nimitz-class carriers are among the largest warships afloat ;
they can out run a lot of the smaller non-nuclear powered vessels.

Dr. Gregory Greenman
Physicist

 

[turbomotive]

Safety issues aside, could you take a steam railway locomotive, put a reactor core in the tender and feed coolant gases into the firebox? Is this still too small? Or is the idea just daft?

 

[Morbius]

Safety issues aside, could you take a steam railway locomotive, put a reactor core in the tender and feed coolant gases into the firebox? Is this still too small? Or is the idea just daft?

turbomotive,

Reactor cores can be very small. Many universities have nuclear reactors
that are quite small. The core of the now shutdown Ford Nuclear Reactor
at the University of Michigan was a 2 foot cube. The FNR core was
similar to a Polaris submarine reactor - U of M even got their fuel
from the U.S. Navy.

The core of the still operating M.I.T. reactor is about as big as a couple
coffee cans stacked one on top of the other.

http://web.mit.edu/nrl/www/reactor/core_description.htm

However, the problem is the shielding. The whole reactor at M.I.T is about
20 feet across, practically all of which is shielding. [ If you look at the
link below - the picture labelled "E" is a picture of the experimental
hall - the reactor is the green cylinder:

http://web.mit.edu/nrl/www/photos.htm

So the requirements for a shield is what sinks using the reactor to
power something like a train. However, a larger vehicle like a
submarine can use a reactor, and our newest aircraft carriers like the
U.S.S. Ronald Reagan:

http://www.reagan.navy.mil/fact.htm

use 2 nuclear reactors for propulsion.

Dr. Gregory Greenman
Physicist

 

[candu600]

In talking about nuclear powered aircraft there are issues of technical feasibility, economic feasibility and political correctness. The nuclear aircraft does not score highly on the last two, but if a large aircraft (one million pounds these days folks) fell in my neighbourhood I would rather take my chances on some radiation exposure than trying to outrun 300,000 pounds of flaming gasoline. The technical feasibility of nuclear powered aircraft was demonstrated pretty conclusively http://www.aviation-history.com/articles/nuke-american.htm back in the 50’s. It would be more feasible today. The reactor components are only a small fraction of the weight. The issue is shielding. One would envision a canard aircraft with the reactor and shield in the rear, a cargo section, possibly borated hydrocarbon fuel for takeoff and landing (go rounds) and a set of optimized shields - LiH, Boron, Gadolinium?. The fuel load of a one million pound hydrocarbon aircraft is about 300,000 pounds, so that is the weight budget one would have to play with.

 

[Vanadium 50]

To do a proper comparison, one needs to know a few numbers that are kept very secret - the weight, cost and power output of a naval reactor, and the cost of a commercial airplane engine.

A guess is that a reactor produces 50 kW/kg of weight. So your 300,000 pounds works out to more power than the airplane needs. However, I would estimate that the cost is an order of magnitude more - around a billion dollars rather than one hundred million. You'd certainly need at least two of them, so what's the payback period? 30y? 50y? 100y?

Also, while a gasoline fireball is not a pleasant place to be, 12 hours later, it's perfectly safe. The same cannot be said of a reactor dropped from 40,000 feet.

 

[joelupchurch]

The Soviets worked a lot on nuclear powered bombers. They never got one to work, but they did have a bomber fly with a working reactor on it.

http://www.aviation-history.com/articles/nuke-bombers.htm"

The smallest reactor I know of was the Snap-10a. All the other Snap reactors were RTGs.

http://en.wikipedia.org/wiki/SNAP-10A"

Here is a picture of the reactor core being assembled.

http://www.etec.energy.gov/History/Major-Operations/MajorOpsImages/S8DR_Core_assembly_31Jul64.jpg

http://www.etec.energy.gov/History/Major-Operations/SNAP-Overview.html"

 

[candu600]

To do a proper comparison, one needs to know a few numbers that are kept very secret - the weight, cost and power output of a naval reactor, and the cost of a commercial airplane engine.

A guess is that a reactor produces 50 kW/kg of weight. So your 300,000 pounds works out to more power than the airplane needs. However, I would estimate that the cost is an order of magnitude more - around a billion dollars rather than one hundred million. You'd certainly need at least two of them, so what's the payback period? 30y? 50y? 100y?

Also, while a gasoline fireball is not a pleasant place to be, 12 hours later, it's perfectly safe. The same cannot be said of a reactor dropped from 40,000 feet.

I think what Vanadium is saying is that my numbers are conservative. From searching around on the net I find that a 747-200 can carry 349,000 pounds of fuel and an A380 can carry 545,000 pounds, so I was a little low on those numbers. Two R.R. Trent 900s weigh 28,000 ( 14k each). So if we were “converting” an A380 to nuclear propulsion (forgetting for the moment that you need to completely redesign the aircraft to be nuclear) we would keep two of the Trents and say 50,000 pounds of fuel for them. So our nuclear package would need to provide 140,000 pounds of thrust and weigh less than (545,000 + 28,000 - 50,000) which is 523,000 pounds. We don’t need 140,000 pounds of thrust for cruise but we probably do need it for the lose-one-engine-on-takeoff scenario. Anyway I found a reference to a nuclear marine unit with 140,000 SHP but it’s at 12lbs/SHP which is about three times too heavy. Interestingly a 1977 study by Lockheed resulted in a design for 1.3 to 1.5 million lb. aircraft with a 543,000 lb. nuclear propulsion system. This was a canard aircraft with a 400,00. to 600,000 lb. payload and included provision for ‘emergency’ range on regular jet fuel. I think somewhat smaller nuclear aircraft are feasible, but I think it is largely a shielding issue. As to falling from 40,000 feet (my nuke maybe only cruises at 30) I think that would be what is called a “design basis accident”. Falling into a nice swamp would probably be just fine. Falling into water you have to worry about criticality issues and falling onto a big slab of granite might be a challenge. I am reminded of George Carlins’s remark “You know those indestructible black boxes, why don’t they make the whole aircraft out of that stuff?” What you would do is utilize a good chunk of the shield material to protect the core. The trick is to get the heat out but keep the fission products in.