Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by GiraffeNecktie on Tuesday March 27, 2012 @08:52AM
(#3761183)
I'm not sure why we're still pursuing the incrediblly difficult and expensive route of fusion when thorium is simple, safe, inexpensive and incredibly abundant.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by torstesu on Tuesday March 27, 2012 @09:47AM
(#3761445)
I completely agree, especially since Norway has large reserves of the stuff. One challenge though, is that uranium is also abundantly available and still much cheaper to mine than thorium. Another challenge is that it is not the fuel which drives the cost, as in conventional power plants; it is the capital cost concerned with building the reactors themselves. Therefore, a thorium-fueled reactor has to be considerably cheaper to build than a uranium-fueled reactor and it is easy to pose the question if this will ever be true. There has to be some serious subsidizing to incentivize commercial development, adaption and investment.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by drucken on Tuesday March 27, 2012 @10:16AM
(#3761589)
1. Thorium is a natural by-product, in fact it is in the waste slurry, from rare earth mineral mining.
2. Tiny amounts are required relative to uranium for electricity production. We have already mined in the West more than we will need for a hundred years, if we had production Molten Salt Reactors today.
3. It is far more electrically efficient than uranium.
So, there is no comparison purely based on which is the better fuel for fission. Thorium wins by multiple factors if not magnitude across many aspects, including fission waste prodcuts.
However, that said, the trust cost of fission is, as you mentioned, the upfront cost of the nuclear reactors. Since this quasi-government industry has seen little innovation in practice for 70+ years, conventional nuclear power plants have remained enormous capital sinks and risky endeavours.
There is now some serious research going into MSRs (including thorium) and other nuclear tech. But it is mainly being led by China, excluding sidetracking efforts like the Travelling Wave Reactors of Bill Gates et. al, who need as much energy as they can get.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by DanBC on Tuesday March 27, 2012 @10:15AM
(#3761575)
People are scared of "Nuclear" - mention uranium and many people will panic.
There's the problem of allowing some countries to have a Uranium energy programme which gives them access to the beginnings of nuclear weapons technology.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by tocomment on Tuesday March 27, 2012 @08:34AM
(#3761126)
Fusion idea that I've always had. Why can't we scale up a something like a car engine's piston and cylinder to be 1000's of times bigger and explode hydrogen bombs in it to move the piston up and down?
I know it sounds kind of absurd to think about but it seems like it would work to me. Can anyone explain why it wouldn't work? Is it just a cost issue?
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by tgflynn on Tuesday March 27, 2012 @09:40AM
(#3761408)
If I understand what you're suggesting I think it addresses only the aspect of how to convert the energy released by fusion reactions into useful work (ie. mechanical or electrical energy). This is not actually the difficult part of building a practical fusion reactor.
Even if your design could work you need a way to produce pure fusion explosions. There is currently no sufficiently energy efficient way of doing this. Thermonuclear bombs require a fission explosion to trigger them (in fact I believe most produce more energy from fission than from fusion).
The laser based inertial confinement schemes actually are designed to produce tiny pure fusion explosions. If such a system could be built that produced more fusion energy than the driving lasers consume there would be many ways to convert that energy into electrical energy (photovoltaics, steam generators/turbines, etc.). I suspect an internal combustion engine would not be the most practical of these methods from an engineering point of view.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by tgflynn on Tuesday March 27, 2012 @10:45AM
(#3761757)
Just to be clear, now that I have a better understanding of your proposal, I didn't mean that a device in which the majority of the energy comes from fusion isn't feasible, just not typical.
Quantitatively I don't know how large you can make the fusion/fission energy ratio but there will always be some fission with the radioactive decay product problems inherent to that.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by tocomment on Tuesday March 27, 2012 @10:07AM
(#3761537)
I don't seem to be allowed to reply to the comment below:
> How does your design address that ? Are you saying the fusion reactions would be triggered by the mechanical compression of the pistons in the cylinder ?
I'm saying fusion reactions are triggered by the detonation of a hydrogen bomb (fission triggers fusion inside the bomb).
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by tgflynn on Tuesday March 27, 2012 @10:17AM
(#3761593)
OK, so then what you're proposing isn't a pure fusion reactor but a hybrid fusion-fission reactor. That's probably going to combine many of the disadvantages of fusion and fission. You get both the highly radioactive and long-life decay products of fission, still need relatively rare fissile fuel and you get the containment problems of fusion.
Moreover if you need a fission explosion trigger you need a critical mass of fissile material for each detonation so there's a limit to how far you can scale down the process.
I think it's highly unlikely this could be practical.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by tgflynn on Tuesday March 27, 2012 @10:04AM
(#3761527)
I mainly suggest my design because it seems so difficult to trigger fusion without fission using current technology.
How does your design address that ? Are you saying the fusion reactions would be triggered by the mechanical compression of the pistons in the cylinder ?
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by bradleyland on Tuesday March 27, 2012 @10:33AM
(#3761677)
Lots of great answers here, but only a couple really get down to brass tacks.
TGFlynn answers the question of why we can't use fusion bombs: "Thermonuclear bombs require a fission explosion to trigger them (in fact I believe most produce more energy from fission than from fusion)." So in short, even setting off "fusion" bombs produces lots of bad byproducts.
The other part of your question is, "why not a big piston and cylinder?" Car engines are heat engines as well. They convert heat in to kinetic energy (motion). The problem is that they're relatively inefficient when compared to the typical means of heat/motion conversion used in a power plant. They're also far more complex and require more maintenance than other, simpler designs.
The design of a car engine has advantages that suit it well to use in an automobile. Namely that it's power output is easily variable. You don't need peak horsepower when sitting at a stoplight, so you throttle the engine back and generate far less power. This requirement is almost nonexistent in power plants. There is some variability, but you're dealing with average power consumption from thousands of households. The rate of increase/decrease is much, much slower than you find in an automobile.
Because of this, power plants use heat engines that might require a lot of energy to get started, but are more efficient once operating. Steam generators coupled with turbines are most common. The heat is used to boil water, which creates steam, which is used to drive some type of turbine. Turbines are wildly efficient if they're designed properly. The primary inefficiency in a steam/turbine power plant is startup and shutdown, but plant operators try to minimize this impact by employing operation strategies that distribute load in a way that can scale up and down minimally.
So, why not use a large piston in cylinder design? Because it's less efficient, more complex, requires more maintenance, and requires more materials to construct.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by arethuza on Tuesday March 27, 2012 @09:25AM
(#3761329)
One idea that was actually investigated was to explode atomic bombs underground and use the heat released to produce steam that could be user to generate power. The Project Gnome test used a relatively small bomb (3.1kt) which created a 20m wide cavity - which people eventually entered (!):
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by mseebach on Tuesday March 27, 2012 @09:25AM
(#3761330)
I think that combustion engines are mainly good for their scalability in terms of output: You can make a combustion engine go from cold to pulling a tractor trailer up a mountain in seconds. For sustained loads (electricity generation), the steam pressure and turbines setup is much more efficient.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by bradleyland on Tuesday March 27, 2012 @11:26AM
(#3761993)
There's a greater lesson in all this discussion. Great ideas can be stifled by the limitations of knowledge. That is to say, if you "know" that an idea cannot work, you may not explore it. However, if you know nothing about the principles that limit a field, you can't even begin to postulate solutions, because you can't see that an idea is as absurd as putting the cart in front of the horse.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by alphaBetaGamma on Tuesday March 27, 2012 @09:27AM
(#3761340)
Nuclear fusion needs high temperature (so that the hydrogen nuclei have enough energy to get close together). "High" means ~10 million kelvin, several orders of magnitude higher than the fusion/evaporation point of any material.
So your giant piston will melt and evaporate long long long before you get fusion.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by tocomment on Tuesday March 27, 2012 @09:29AM
(#3761349)
I'm talking a really giant piston and cylinder so that the piston and cylinder walls are far enough away from the blast that they don't experience destructive levels of heat. (You'd explode the bomb in the middle of the cylinder (during freefall))
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by bradleyland on Tuesday March 27, 2012 @11:08AM
(#3761865)
The efficiency of a piston-in-sleve design is reduced, drastically, as you introduce free space in to the combustion chamber.
Keep in mind that converting heat to motion can be easily expressed a math problem. You have a specific amount of heat that is output by the nuclear reaction. The units that are used to measure heat can be directly converted to the units that measure motion energy and electromagnetic energy.
So, let's assume that you have built a massive piston engine that has a target thermal efficiency of other piston engines in common use. A highly efficient diesel ship engine (some of the most efficient made) will reach 50% thermal efficiency. Most car engines are around 35%.
Because of the intense heat of the fusion reaction, you must make some design compromises. Namely, the massive distances required to protect your materials. Your materials must be far enough away that the heat reaching them is below their thermal design limit. This means that all the heat energy between the temperature of the actual thermonuclear reaction and your material design limits must be dissipated somehow. In your design, you propose to use an air-gap, which means the heat will dissipate in to the air, which will be discharged during the exhaust cycle. There's a big loss in efficiency.
You'll also have the problem of heat soak. As you repetitively detonate thermonuclear bombs in your combustion chamber, some of the energy will transfer in to the materials that make up your engine. You'll have to carry this wasted heat energy away some how. In cars, this is a water cooling jacket and radiator. In power plants, this heat energy is normally used to boil water in what is called a "combined cycle" generator. You could apply the same principle here. But let's look at the piston-in-sleve side of things, because that's your primary design idea. The combined cycle side of the equation is irrelevant, because thermonuclear reactions generate heat that is easily converted to steam without the intermediary piston engine mechanicals.
Inefficiency scales with size, so it doesn't matter that we're generating insane amounts of energy, we're still facing losses here due to design compromises. The losses due to the massive air gaps required cannot be overstated. The compression ratio in a car engine has a lot to do with its efficiency. A car engine that loses even 20% of it's compression can fail to sustain operation. That is, the energy created by the fuel-air explosion isn't sufficient to overcome the parasitic forces present in the engine.
So let's make a wild guess and say that your engine is one-tenth (optimistic) as efficient as an optimally efficient diesel engine; around 5% thermal efficiency. Ouch.
So what is the thermal efficiency of a thermonuclear power plant that simply uses the heat energy from the reaction to generate steam? Current designs are limited to around 30% because of materials limits. In a typical nuclear reactor, the rods are actually inside the pressure vessel, which must be actuated to regulate the reaction, as well as accessible for servicing.
Keep in mind that I have a very limited understanding of these principles, but much of it is hands-on. I used to be a small-engine mechanic. I worked for a retired Army power plant engineer, who taught me more than I'll ever need to know that a gigantic piston-in-sleve design is a monumental waste of time.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by dbcooper on Tuesday March 27, 2012 @09:25AM
(#3761326)
The coulomb barrier (electromagnetic force derived) for hydrogen nuclei is not overcome until a temperature of about 120 million kelvin. Obviously, no metallic materials can operate at this temperature.
Controlled magnetic fields are required to contain a plasma above the coulomb limit.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by aqme28 on Tuesday March 27, 2012 @09:36AM
(#3761388)
Despite their name, hydrogen bombs are primarily uranium fission bombs[1].
What you've described is just another kind of fissile nuclear reactor, though surely far less stable and likely far less efficient.
The other problem is that materials weaken and swell (as in gain actual volume) when bombarded with radiation. It would be pretty difficult to make a piston that changes shape throughout its life.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by huxley on Tuesday March 27, 2012 @09:26AM
(#3761335)
Are there actually any materials which could withstand repeated contained hydrogen bomb explosions in the same manner as a piston? No point having a 1000x sized piston you only use 2 or 3 times.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by GiraffeNecktie on Tuesday March 27, 2012 @08:49AM
(#3761174)
Sounds great except for the absurdly impractical parts (i.e. exploding hydrogen bombs to move a giant piston). Where would you propose expelling the radioactive gases after the first stroke of the engine?
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by rbarooah on Tuesday March 27, 2012 @09:17AM
(#3761285)
My guess would be that adequate sealing for such a unit would be very hard to achieve, and that mechanical wear and tear would make it hard to maintain without access to the hot zones.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by aqme28 on Tuesday March 27, 2012 @09:46AM
(#3761431)
This article talks about tokamaks like they're the only magnetic confinement technology out there. Things like stellarators and dipoles have been gaining a lot of ground lately and could one day overtake tokamaks as the standard magnetic confinement tech.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by DanWaterworth on Tuesday March 27, 2012 @08:36AM
(#3761132)
I thought that, but I now think it means in a single run, using just half a gram of hydrogen, the peak output power is 500MW, it would be helpful if it also mentioned how much usable energy was generated too.
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by mrsebastian on Tuesday March 27, 2012 @08:48AM
(#3761168)
I don't think anyone knows how much 'usable' energy ITER will generate (plus, ITER won't be a power plant -- the heat produced will actually be thrown away -- so I guess we won't ever know.)
Re:500MW from half a gram of hydrogen: The hunt for fusion p...
by pieter on Tuesday March 27, 2012 @08:36AM
(#3761131)
ITER won't produce any stable power at all, so they most likely mean something like "A single run of ITER will consume about half a gram of hydrogen while having a peak output of 500MW (for a few seconds).
2. Tiny amounts are required relative to uranium for electricity production. We have already mined in the West more than we will need for a hundred years, if we had production Molten Salt Reactors today.
3. It is far more electrically efficient than uranium.
So, there is no comparison purely based on which is the better fuel for fission. Thorium wins by multiple factors if not magnitude across many aspects, including fission waste prodcuts.
However, that said, the trust cost of fission is, as you mentioned, the upfront cost of the nuclear reactors. Since this quasi-government industry has seen little innovation in practice for 70+ years, conventional nuclear power plants have remained enormous capital sinks and risky endeavours.
There is now some serious research going into MSRs (including thorium) and other nuclear tech. But it is mainly being led by China, excluding sidetracking efforts like the Travelling Wave Reactors of Bill Gates et. al, who need as much energy as they can get.
There's the problem of allowing some countries to have a Uranium energy programme which gives them access to the beginnings of nuclear weapons technology.
I know it sounds kind of absurd to think about but it seems like it would work to me. Can anyone explain why it wouldn't work? Is it just a cost issue?
Even if your design could work you need a way to produce pure fusion explosions. There is currently no sufficiently energy efficient way of doing this. Thermonuclear bombs require a fission explosion to trigger them (in fact I believe most produce more energy from fission than from fusion).
The laser based inertial confinement schemes actually are designed to produce tiny pure fusion explosions. If such a system could be built that produced more fusion energy than the driving lasers consume there would be many ways to convert that energy into electrical energy (photovoltaics, steam generators/turbines, etc.). I suspect an internal combustion engine would not be the most practical of these methods from an engineering point of view.
If this is true, then that puts a kink in my design. I assumed the vast majority of energy output was from fusion. (why else would they build them?)
I mainly suggest my design because it seems so difficult to trigger fusion without fission using current technology.
Quantitatively I don't know how large you can make the fusion/fission energy ratio but there will always be some fission with the radioactive decay product problems inherent to that.
I'm saying fusion reactions are triggered by the detonation of a hydrogen bomb (fission triggers fusion inside the bomb).
Moreover if you need a fission explosion trigger you need a critical mass of fissile material for each detonation so there's a limit to how far you can scale down the process.
I think it's highly unlikely this could be practical.
How does your design address that ? Are you saying the fusion reactions would be triggered by the mechanical compression of the pistons in the cylinder ?
TGFlynn answers the question of why we can't use fusion bombs: "Thermonuclear bombs require a fission explosion to trigger them (in fact I believe most produce more energy from fission than from fusion)." So in short, even setting off "fusion" bombs produces lots of bad byproducts.
The other part of your question is, "why not a big piston and cylinder?" Car engines are heat engines as well. They convert heat in to kinetic energy (motion). The problem is that they're relatively inefficient when compared to the typical means of heat/motion conversion used in a power plant. They're also far more complex and require more maintenance than other, simpler designs.
The design of a car engine has advantages that suit it well to use in an automobile. Namely that it's power output is easily variable. You don't need peak horsepower when sitting at a stoplight, so you throttle the engine back and generate far less power. This requirement is almost nonexistent in power plants. There is some variability, but you're dealing with average power consumption from thousands of households. The rate of increase/decrease is much, much slower than you find in an automobile.
Because of this, power plants use heat engines that might require a lot of energy to get started, but are more efficient once operating. Steam generators coupled with turbines are most common. The heat is used to boil water, which creates steam, which is used to drive some type of turbine. Turbines are wildly efficient if they're designed properly. The primary inefficiency in a steam/turbine power plant is startup and shutdown, but plant operators try to minimize this impact by employing operation strategies that distribute load in a way that can scale up and down minimally.
So, why not use a large piston in cylinder design? Because it's less efficient, more complex, requires more maintenance, and requires more materials to construct.
http://en.wikipedia.org/wiki/Project_Gnome
http://www.wipp.energy.gov/science/ug_lab/gnome/gnome.htm
So your giant piston will melt and evaporate long long long before you get fusion.
I don't know what the Mach stem effect of a contained cylinder would be, but the materials might also need to deal with that.
Keep in mind that converting heat to motion can be easily expressed a math problem. You have a specific amount of heat that is output by the nuclear reaction. The units that are used to measure heat can be directly converted to the units that measure motion energy and electromagnetic energy.
So, let's assume that you have built a massive piston engine that has a target thermal efficiency of other piston engines in common use. A highly efficient diesel ship engine (some of the most efficient made) will reach 50% thermal efficiency. Most car engines are around 35%.
Because of the intense heat of the fusion reaction, you must make some design compromises. Namely, the massive distances required to protect your materials. Your materials must be far enough away that the heat reaching them is below their thermal design limit. This means that all the heat energy between the temperature of the actual thermonuclear reaction and your material design limits must be dissipated somehow. In your design, you propose to use an air-gap, which means the heat will dissipate in to the air, which will be discharged during the exhaust cycle. There's a big loss in efficiency.
You'll also have the problem of heat soak. As you repetitively detonate thermonuclear bombs in your combustion chamber, some of the energy will transfer in to the materials that make up your engine. You'll have to carry this wasted heat energy away some how. In cars, this is a water cooling jacket and radiator. In power plants, this heat energy is normally used to boil water in what is called a "combined cycle" generator. You could apply the same principle here. But let's look at the piston-in-sleve side of things, because that's your primary design idea. The combined cycle side of the equation is irrelevant, because thermonuclear reactions generate heat that is easily converted to steam without the intermediary piston engine mechanicals.
Inefficiency scales with size, so it doesn't matter that we're generating insane amounts of energy, we're still facing losses here due to design compromises. The losses due to the massive air gaps required cannot be overstated. The compression ratio in a car engine has a lot to do with its efficiency. A car engine that loses even 20% of it's compression can fail to sustain operation. That is, the energy created by the fuel-air explosion isn't sufficient to overcome the parasitic forces present in the engine.
So let's make a wild guess and say that your engine is one-tenth (optimistic) as efficient as an optimally efficient diesel engine; around 5% thermal efficiency. Ouch.
So what is the thermal efficiency of a thermonuclear power plant that simply uses the heat energy from the reaction to generate steam? Current designs are limited to around 30% because of materials limits. In a typical nuclear reactor, the rods are actually inside the pressure vessel, which must be actuated to regulate the reaction, as well as accessible for servicing.
Keep in mind that I have a very limited understanding of these principles, but much of it is hands-on. I used to be a small-engine mechanic. I worked for a retired Army power plant engineer, who taught me more than I'll ever need to know that a gigantic piston-in-sleve design is a monumental waste of time.
Controlled magnetic fields are required to contain a plasma above the coulomb limit.
What you've described is just another kind of fissile nuclear reactor, though surely far less stable and likely far less efficient.
The other problem is that materials weaken and swell (as in gain actual volume) when bombarded with radiation. It would be pretty difficult to make a piston that changes shape throughout its life.
[1] http://en.wikipedia.org/wiki/Teller%E2%80%93Ulam_design
Perhaps we're part of the problem? We should discuss radical energy ideas, teach others what wouldn't work, and how to change them, not poo poo them.
http://en.wikipedia.org/wiki/Wendelstein_7-X http://en.wikipedia.org/wiki/Levitated_dipole
The current record is, like, 10 seconds and no one has broken even on input/output power...
Edit: Ah, actually the current record is less than 1 second: http://en.wikipedia.org/wiki/ITER#Background