The reason that breakeven can only be exceeded for a short period of time is extremely complicated. I can give a bit of a brief explanation and some papers that explain it in more detail, but understanding what is going on is going to involve a lot of work.
In early tokamaks, the confinement times of energy and particles were seen to be much lower than what was anticipated from theory and also much lower than what is needed for fusion. The reason for this is attributed to anomalous transport of particles and energy across the magnetic field lines (which are intended to prevent this) due to various instability modes in the plasma.
Also, early tokamaks could not be heated to anywhere near the temperatures required for fusion since they used exclusively inductive current drive to heat the plasma, which is ineffective beyond ~1 keV. Auxiliary heating methods such as wave heating with electron cyclotron, ion cyclotron or lower hybrid waves and neutral beam injection were then adopted in order to further heat the plasma beyond the ohmic limit. Initially these were observed to further reduce the confinement time from the ohmic limit. Later the ASDEX tokamak observed an increase in the confinment time back to ohmic levels with auxiliary neutral beam heating. This regime of operation is referred to as H-mode. It required the auxiliary heating power to exceed a certain minimum threshold. http://prola.aps.org/abstract/PRL/v49/i19/p1408_1
Additionally, it has been roughly observed that the confinement times in tokamaks scale with the plasma parameters such as plasma current, toroidal magnetic field, plasma volume, etc. ITER was designed using a database of other tokamaks, their parameters and their performance, in order to scale the machine to the desired performance http://www.iop.org/EJ/abstract/0029-5515/32/2/I11. Superconducting magnetic systems will allow ITER to have long pulse durations like Tore Supra (> 6 minutes) with better plasma performance than any of the big tokamaks now (JT-60, JET, DIII-D, T-15).
There is certainly the risk of proliferation from fusion technology. IMO proliferation will just become a bigger and bigger concern even without fusion since as technology progresses it will become easier and easier and accessible to more countries.
ITER will NOT generate power. It's not even close.
This isn't one of ITER's goals. There are other projects that are designed to address these issues. IFMIF is designed to address the environmental, safety and economic concerns of fusion power http://www.frascati.enea.it/ifmif/. Sometime after ITER and IFMIF there would be DEMO which would first replicate ITER's performance and the preliminary track would then be to produce 1 GW of electric power with DEMO. From there it would be PROTO, a prototype reactor. Concurrent to ITER there are several projects planned such as IGNITOR http://www.frascati.enea.it/ignitor/ and FIRE http://fire.pppl.gov/
Fusion plasmas today already put out more energy then we put in, but we can't turn that energy into electricity yet.
Do you have a source for this? I know of only two tokamaks that have performed D-T fusion, JET (16.1 MW, Q = 0.6) and TFTR (10.7 MW, Q = 0.27). JT-60 has achieved plasma performance which corresponds with Q = 1.25, however JT-60 is not designed to handle tritium fuel and hence has never performed D-T fusion. No machines have exceeded breakeven with D-T fusion.
The main problem (as I see it) with fusion has nothing to do with plasma, but has to do with materials.
I disagree with you about the plasmas not being a main problem, however I agree that the materials side also requires a lot of research and development.
Checking a couple of journals reveals that the JT-60 discharge duration can be as long as ~65 seconds while the ELMy H-mode, which ITER will operate in, lasted for about 30 seconds. The article might be referring to this or it may be referring to some of the 30 s discharges that JT-60U has, I'm not sure.
Something else that is interesting is that there are plans to further upgrade JT-60U (U is for upgrade from JT-60) to JT-60SC which will include superconducting magnetic field coils. I haven't been able to find a timeline, but I do know that the design for it is complete.
India has proposed a combined hybrid fission/fusion reactor as have many others. The funding for India's new SST-1 http://www.ipr.res.in/sst1/SST-1.html superconducting tokamak was secured under this pretext.
For the ~ 380 s discharges on Tore Supra the plasma parameters are:
Plasma current = 0.5 - 0.7 MA Toroidal magnetic field = 3.4 T Line averaged electron density = 1.5 10^19 m^-3 Central ion temperature = 1.5 keV Central electron temperature = 4.5 keV
Certainly these parameters are quite good. They aren't what JT-60 or JET can get, but then the machines are designed for very different purposes. The source of Tore Supra's lengthy discharges are it's superconducting magnetic field coils.
Also, Tore Supra has achieved discharge durations in excess of 200 seconds since 2002 and has more recently had shots in excess of 360 seconds (6 minutes). Of course Tore Supra has a significant advantage over most other tokamaks in that it has superconducting toroidal field coils, giving it a steady state toroidal magnetic field. My experience in working with these machines is that on most of them the toroidal magnetic fields seriously handicap their performance due to the massive power requirements of generating a typically > 1 T magnetic field. This brings in the requirement for expensive and bulky power supplies and energy storage devices such as large capacitor banks and flywheel generators. The Joint European Torus has two 3750 MJ flywheel generators that can supply as much as 800 MW (400 MW each) of peak power which are used for the toroidal and poloidal magnetic field coils. The magnetic field coils on JET consume more than half of the power required to run the machine. The remaining power is drawn directly from the grid. Using superconducting magnetic coils greatly reduces the power required to run a tokamak and extends the time over which the discharge can last. Tore Supra has a goal of, I believe 1000 second discharges, which is similar to what ITER will be aiming for.
Well, the reasons given in that article are actually good reasons to go with nuclear.
Reactor safety is basically a non-issue for third generation nuclear reactors (which have passive safety systems). As far as waste management goes, the thing to wait for is the fourth generation nuclear reactors which offer the possibility of burning actinides, which would significantly reduce the amount of high-level waste.
It seems that when groups bring up the fact that mining of uranium ore causes environmental damage they ignore the fact that coal is also mined and that mountain-top removal has a massive environmental impact. The scale of mining required to remove coal is monstrous compared with uranium, considering that uranium has about 3.6 million times the energy density of coal (90 000 000 MJ/kg for uranium compared with 23-29 MJ/kg for coal).
IMO, nuclear represents a step forward, but certainly not a permanent solution. At this point, many "greener" technologies are not suitable for use in as many locations as nuclear.
This is possibly the most confused article I've ever seen here. Somebody doesn't understand relativity, but decided to include implications of it in an article about the cellular effects of radiation.
As I browse the responses it seems obvious that almost nobody actually read the article, nor has understanding of what is even going on. I guess it doesn't help that this is yet another/. summary that is poorly written (intentionally???). That and most/. posters are political experts on topics such as freedom of speech and expression.
$100B might be an overshoot (well, it probably is). The latest estimate I've seen for the cost of building the current machine is about $5 Billion euro, about $6 billion US. I have an estimate from the first scaling down that I know of, which was to 8.14 m major radius (1.5 GW power, 1000s burn time) and it was projected to cost - $10 Billion 2005 dollars to build. An important consideration in the size difference of the machines is that most of the systems scale with volume rather than the major radius. I would guess that such a machine would cost several times more just to construct. I'm not too sure how the operating costs would change though.
Net energy output is the primary goal, amongst many other things. The goal of an ignited plasma with infinite power gain is what was scaled back when the machine size was scaled down from a 10 m major radius to the 6.2 m radius that the design is for now due to the ridiculous cost that such a machine would have (> 100 billion most likely). It isn't precluded from achieving ignition, it is just that a more modest goal of Q = 10 as a proof of concept. If you can achieve Q = 10 then it becomes more realistic to attempt to get Q = 1000 or infinity rather than going right after it once you achieve Q = 1.
It's more sustainable, but with the amount of deuterium in the oceans it doesn't make much difference. At current power consumption it's estimated that there is enough deuterium for about millions of years. Fusion could be used to make a large powerplant (>1 GW) in a very small area.
This doesn't explain why other nations are even more involved in tokamak research than the US (China, India, EU).
Working in the field, I get all kinds of people coming up to me with "Did you hear about the device that (insert something about outperforming a tokamak)." At current, the only other devices achieving plasma performance like tokamaks are other large toroidal devices with helical fields like stellarators. A lot of people claim to be outperforming them in their backyard, though. I read Physics of Plasmas, Plasma Physics and Controlled Fusion, Nuclear Fusion and Physical Review Letters on a daily basis and I've seen no mention of them. I do find lots of mention of it on random websites.
To quote F.F. Chen: "A plasma is a quasineutral gas o charged and neutral particles which exhibits collective behaviour".
A burning plasma is a nearly fully-ionized gas in which the fusion power captured by the plasma keeps the plasma hot. It can also be called a self-heating plasma.
Break even (or the equivalent of break even) was achieved on the JT-60U tokamak in Japan in the late 90's (1998 I think). No fusion occured because tritium wasn't used in the reactor since I don't believe that JT-60U is equipped to handle tritium (for reasons of radiation). The performance of the plasma, being the energy confinement time, fuel density and ion temperature, was such that the equivalent energy gain had tritium been present would have been 1.25.
Some of the problems with the current generation of machines are the use of copper coils rather than superconducting niobium-tin coils as copper coils require a tremendous amount of power to generate the magnetic fields necessary to confine the plasma (typically 3-5 Tesla at the machine major radius). The coils on the small tokamak I work on are copper and require a few tens of kilowatts of power to generate a 0.7 Tesla magnetic field we use and we have the benefit of having very small coils. The largest machines, where the copper coils are much larger (about 3m diameter, roughly compared to.4m on our machine) require hundreds of megawatts to generate their magnetic fields. Superconducting coils present the ability to greatly reduced the power required to operate the machine.
The problems with plasma performance are generally centered around energy loss from the plasma, through particles, heat or EM radiation. Radiation isn't a big problem, but particle and heat loss are. The plasma is very turbulent and this turbulence leads to what is referred to as anomalous losses, anomalous in the sense that they are not well explained by theory and are orders of magnitude larger than what is predicted by theory. These losses can be reduced by elaborate modes of operation, generally referred to as H-modes (H meaning high confinement). There are some other drawbacks to these modes, but without getting into much detail, the scaling of confinement with various parameters of the machines shows that a machine the size of ITER (http://www.iter.org/ should have a plasma performance that is good enough to achieve a fusion power gain of 10, that is 50 MW of heating to the plasma and 500 MW of fusion power output. The ideal would be to be able to turn off the plasma heating, but if ITER works as predicted it will be very good.
There is also some concern over what will happen to the alpha-particles produced after they give up there energy to other species in the plasma. They have to be removed as they will degrade the plasma performance. I believe that there is an idea of a way to remove them, but this is outside of my area of research.
There are other problems with an actual power producing machine. Most of these are engineering problems and have to do with such things as building some sort of lithium blanket that can withstand being bombarded with 14 MeV neutrons, breeding and extracting the tritium fuel and handling the tritium fuel.
Actually, what it means is that the ratio of input power to output power is infinite. This is because an ideal tokamak would have a burning plasma in it that was heating itself with the alpha-particles produced by fusion. It's written to be misleading.
The whole thing is unclassified. Many countries have tokamaks. I don't think that the article was trying to imply that China was the first or anything like that. It was trying to say that China was going to be first to have a working device.
$37 million is for an upgrade to an already existing machine. Building a brand new machine takes years, not months and the superconducting coils cost more than $37 million. HT-7 is a medium sized machine, so I find it hard to believe that even with an upgrade to have fully superconducting magnetic coils it will be able to generate anything more than an insignificant amount of fusion. For instance, HT-7 is 1.22m in radius with a plasma current of 90 kA. JET is 2.96m with a plasma current of 4.8 MA and it has not achieved breakeven. The article is very misleading; I'm sure that they mean that the intent of the tokamak device in general is to generate power instead of saying that this specific machine will.
Slashdot Mirror
In early tokamaks, the confinement times of energy and particles were seen to be much lower than what was anticipated from theory and also much lower than what is needed for fusion. The reason for this is attributed to anomalous transport of particles and energy across the magnetic field lines (which are intended to prevent this) due to various instability modes in the plasma.
Also, early tokamaks could not be heated to anywhere near the temperatures required for fusion since they used exclusively inductive current drive to heat the plasma, which is ineffective beyond ~1 keV. Auxiliary heating methods such as wave heating with electron cyclotron, ion cyclotron or lower hybrid waves and neutral beam injection were then adopted in order to further heat the plasma beyond the ohmic limit. Initially these were observed to further reduce the confinement time from the ohmic limit. Later the ASDEX tokamak observed an increase in the confinment time back to ohmic levels with auxiliary neutral beam heating. This regime of operation is referred to as H-mode. It required the auxiliary heating power to exceed a certain minimum threshold. http://prola.aps.org/abstract/PRL/v49/i19/p1408_1
Additionally, it has been roughly observed that the confinement times in tokamaks scale with the plasma parameters such as plasma current, toroidal magnetic field, plasma volume, etc. ITER was designed using a database of other tokamaks, their parameters and their performance, in order to scale the machine to the desired performance http://www.iop.org/EJ/abstract/0029-5515/32/2/I11. Superconducting magnetic systems will allow ITER to have long pulse durations like Tore Supra (> 6 minutes) with better plasma performance than any of the big tokamaks now (JT-60, JET, DIII-D, T-15).
Iran has several tokamaks already.
There is certainly the risk of proliferation from fusion technology. IMO proliferation will just become a bigger and bigger concern even without fusion since as technology progresses it will become easier and easier and accessible to more countries.
Are you working at Culham? We recently got a post doc from there. I'm working on a very small tokamak in Canada.
This isn't one of ITER's goals. There are other projects that are designed to address these issues. IFMIF is designed to address the environmental, safety and economic concerns of fusion power http://www.frascati.enea.it/ifmif/. Sometime after ITER and IFMIF there would be DEMO which would first replicate ITER's performance and the preliminary track would then be to produce 1 GW of electric power with DEMO. From there it would be PROTO, a prototype reactor. Concurrent to ITER there are several projects planned such as IGNITOR http://www.frascati.enea.it/ignitor/ and FIRE http://fire.pppl.gov/
Fusion plasmas today already put out more energy then we put in, but we can't turn that energy into electricity yet.
Do you have a source for this? I know of only two tokamaks that have performed D-T fusion, JET (16.1 MW, Q = 0.6) and TFTR (10.7 MW, Q = 0.27). JT-60 has achieved plasma performance which corresponds with Q = 1.25, however JT-60 is not designed to handle tritium fuel and hence has never performed D-T fusion. No machines have exceeded breakeven with D-T fusion.
The main problem (as I see it) with fusion has nothing to do with plasma, but has to do with materials.
I disagree with you about the plasmas not being a main problem, however I agree that the materials side also requires a lot of research and development.
What is this record for? Discharge duration only? The peak Q obtained in JT-60U to date is 1.25.
A UHV chamber of that size would be something to behold. I predict that it would be impossible given current technology.
Checking a couple of journals reveals that the JT-60 discharge duration can be as long as ~65 seconds while the ELMy H-mode, which ITER will operate in, lasted for about 30 seconds. The article might be referring to this or it may be referring to some of the 30 s discharges that JT-60U has, I'm not sure. Something else that is interesting is that there are plans to further upgrade JT-60U (U is for upgrade from JT-60) to JT-60SC which will include superconducting magnetic field coils. I haven't been able to find a timeline, but I do know that the design for it is complete.
India has proposed a combined hybrid fission/fusion reactor as have many others. The funding for India's new SST-1 http://www.ipr.res.in/sst1/SST-1.html superconducting tokamak was secured under this pretext.
For the ~ 380 s discharges on Tore Supra the plasma parameters are:
Plasma current = 0.5 - 0.7 MA
Toroidal magnetic field = 3.4 T
Line averaged electron density = 1.5 10^19 m^-3
Central ion temperature = 1.5 keV
Central electron temperature = 4.5 keV
Certainly these parameters are quite good. They aren't what JT-60 or JET can get, but then the machines are designed for very different purposes. The source of Tore Supra's lengthy discharges are it's superconducting magnetic field coils.
Also, Tore Supra has achieved discharge durations in excess of 200 seconds since 2002 and has more recently had shots in excess of 360 seconds (6 minutes). Of course Tore Supra has a significant advantage over most other tokamaks in that it has superconducting toroidal field coils, giving it a steady state toroidal magnetic field. My experience in working with these machines is that on most of them the toroidal magnetic fields seriously handicap their performance due to the massive power requirements of generating a typically > 1 T magnetic field. This brings in the requirement for expensive and bulky power supplies and energy storage devices such as large capacitor banks and flywheel generators. The Joint European Torus has two 3750 MJ flywheel generators that can supply as much as 800 MW (400 MW each) of peak power which are used for the toroidal and poloidal magnetic field coils. The magnetic field coils on JET consume more than half of the power required to run the machine. The remaining power is drawn directly from the grid. Using superconducting magnetic coils greatly reduces the power required to run a tokamak and extends the time over which the discharge can last. Tore Supra has a goal of, I believe 1000 second discharges, which is similar to what ITER will be aiming for.
Well, the reasons given in that article are actually good reasons to go with nuclear. Reactor safety is basically a non-issue for third generation nuclear reactors (which have passive safety systems). As far as waste management goes, the thing to wait for is the fourth generation nuclear reactors which offer the possibility of burning actinides, which would significantly reduce the amount of high-level waste. It seems that when groups bring up the fact that mining of uranium ore causes environmental damage they ignore the fact that coal is also mined and that mountain-top removal has a massive environmental impact. The scale of mining required to remove coal is monstrous compared with uranium, considering that uranium has about 3.6 million times the energy density of coal (90 000 000 MJ/kg for uranium compared with 23-29 MJ/kg for coal). IMO, nuclear represents a step forward, but certainly not a permanent solution. At this point, many "greener" technologies are not suitable for use in as many locations as nuclear.
This is possibly the most confused article I've ever seen here. Somebody doesn't understand relativity, but decided to include implications of it in an article about the cellular effects of radiation.
As I browse the responses it seems obvious that almost nobody actually read the article, nor has understanding of what is even going on. I guess it doesn't help that this is yet another /. summary that is poorly written (intentionally???). That and most /. posters are political experts on topics such as freedom of speech and expression.
$100B might be an overshoot (well, it probably is). The latest estimate I've seen for the cost of building the current machine is about $5 Billion euro, about $6 billion US. I have an estimate from the first scaling down that I know of, which was to 8.14 m major radius (1.5 GW power, 1000s burn time) and it was projected to cost - $10 Billion 2005 dollars to build. An important consideration in the size difference of the machines is that most of the systems scale with volume rather than the major radius. I would guess that such a machine would cost several times more just to construct. I'm not too sure how the operating costs would change though.
Net energy output is the primary goal, amongst many other things. The goal of an ignited plasma with infinite power gain is what was scaled back when the machine size was scaled down from a 10 m major radius to the 6.2 m radius that the design is for now due to the ridiculous cost that such a machine would have (> 100 billion most likely). It isn't precluded from achieving ignition, it is just that a more modest goal of Q = 10 as a proof of concept. If you can achieve Q = 10 then it becomes more realistic to attempt to get Q = 1000 or infinity rather than going right after it once you achieve Q = 1.
It's more sustainable, but with the amount of deuterium in the oceans it doesn't make much difference. At current power consumption it's estimated that there is enough deuterium for about millions of years. Fusion could be used to make a large powerplant (>1 GW) in a very small area.
Which tiny reactors in Europe? Europe has the largest tokamak on the planet.
This doesn't explain why other nations are even more involved in tokamak research than the US (China, India, EU). Working in the field, I get all kinds of people coming up to me with "Did you hear about the device that (insert something about outperforming a tokamak)." At current, the only other devices achieving plasma performance like tokamaks are other large toroidal devices with helical fields like stellarators. A lot of people claim to be outperforming them in their backyard, though. I read Physics of Plasmas, Plasma Physics and Controlled Fusion, Nuclear Fusion and Physical Review Letters on a daily basis and I've seen no mention of them. I do find lots of mention of it on random websites.
To quote F.F. Chen: "A plasma is a quasineutral gas o charged and neutral particles which exhibits collective behaviour".
A burning plasma is a nearly fully-ionized gas in which the fusion power captured by the plasma keeps the plasma hot. It can also be called a self-heating plasma.
Break even (or the equivalent of break even) was achieved on the JT-60U tokamak in Japan in the late 90's (1998 I think). No fusion occured because tritium wasn't used in the reactor since I don't believe that JT-60U is equipped to handle tritium (for reasons of radiation). The performance of the plasma, being the energy confinement time, fuel density and ion temperature, was such that the equivalent energy gain had tritium been present would have been 1.25. Some of the problems with the current generation of machines are the use of copper coils rather than superconducting niobium-tin coils as copper coils require a tremendous amount of power to generate the magnetic fields necessary to confine the plasma (typically 3-5 Tesla at the machine major radius). The coils on the small tokamak I work on are copper and require a few tens of kilowatts of power to generate a 0.7 Tesla magnetic field we use and we have the benefit of having very small coils. The largest machines, where the copper coils are much larger (about 3m diameter, roughly compared to .4m on our machine) require hundreds of megawatts to generate their magnetic fields. Superconducting coils present the ability to greatly reduced the power required to operate the machine.
The problems with plasma performance are generally centered around energy loss from the plasma, through particles, heat or EM radiation. Radiation isn't a big problem, but particle and heat loss are. The plasma is very turbulent and this turbulence leads to what is referred to as anomalous losses, anomalous in the sense that they are not well explained by theory and are orders of magnitude larger than what is predicted by theory. These losses can be reduced by elaborate modes of operation, generally referred to as H-modes (H meaning high confinement). There are some other drawbacks to these modes, but without getting into much detail, the scaling of confinement with various parameters of the machines shows that a machine the size of ITER (http://www.iter.org/ should have a plasma performance that is good enough to achieve a fusion power gain of 10, that is 50 MW of heating to the plasma and 500 MW of fusion power output. The ideal would be to be able to turn off the plasma heating, but if ITER works as predicted it will be very good.
There is also some concern over what will happen to the alpha-particles produced after they give up there energy to other species in the plasma. They have to be removed as they will degrade the plasma performance. I believe that there is an idea of a way to remove them, but this is outside of my area of research.
There are other problems with an actual power producing machine. Most of these are engineering problems and have to do with such things as building some sort of lithium blanket that can withstand being bombarded with 14 MeV neutrons, breeding and extracting the tritium fuel and handling the tritium fuel.
Actually, what it means is that the ratio of input power to output power is infinite. This is because an ideal tokamak would have a burning plasma in it that was heating itself with the alpha-particles produced by fusion. It's written to be misleading.
The whole thing is unclassified. Many countries have tokamaks. I don't think that the article was trying to imply that China was the first or anything like that. It was trying to say that China was going to be first to have a working device.
$37 million is for an upgrade to an already existing machine. Building a brand new machine takes years, not months and the superconducting coils cost more than $37 million. HT-7 is a medium sized machine, so I find it hard to believe that even with an upgrade to have fully superconducting magnetic coils it will be able to generate anything more than an insignificant amount of fusion. For instance, HT-7 is 1.22m in radius with a plasma current of 90 kA. JET is 2.96m with a plasma current of 4.8 MA and it has not achieved breakeven. The article is very misleading; I'm sure that they mean that the intent of the tokamak device in general is to generate power instead of saying that this specific machine will.
Agreed, JET and TFTR, which both performed deuterium-tritium fusion, had ion temperatures of around 40 keV, which is around 400 million degrees K.