March 13, 2016 at 17:46Quiet thermonuclear revolution
There probably is not a single field of human activity so full of disappointments and rejected heroes as attempts to create thermonuclear energy. A hundred concepts of reactors, dozens of teams that have consistently become favorites of the public and state budgets, and finally sort of decided on the winner in the form of tokamaks. And here again, the achievements of Novosibirsk scientists revive interest around the world in a concept brutally trampled in the 80s. And now in more detail.
GDL open trap with impressive results
Among the whole variety of proposals, how to extract energy from thermonuclear fusion is most oriented towards stationary confinement of a relatively loose thermonuclear plasma. For example, the ITER project and more broadly - toroidal traps tokamaki and stellarators - from here. They are toroidal because it is the simplest form of a closed vessel of magnetic fields (due to the hedgehog combing theorema spherical vessel cannot be made). However, at the dawn of research in the field of controlled thermonuclear fusion, the favorites were not traps of complex three-dimensional geometry, but attempts to keep the plasma in the so-called open traps. These are usually also cylindrical-shaped magnetic vessels in which the plasma is well retained in the radial direction and flows from both ends. The idea of the inventors here is simple - if the heating of a new plasma by a thermonuclear reaction proceeds faster than the heat flow from the ends, then God bless him, with the openness of our vessel, the energy will be generated, and the leak will still occur in a vacuum vessel and the fuel will walk in the reactor until it burns out.
The idea of an open trap is a magnetic cylinder with corks / mirrors at the ends and expanders behind them.
In addition, in all open traps, one or another method is used to delay the plasma from escaping through the ends — and the simplest here is to sharply increase the magnetic field at the ends (put magnetic “plugs” in Russian terminology or “mirrors” in the west), while incident charged particles will, in fact, spring from the mirror mirrors and only a small fraction of the plasma will pass through them and fall into special expanders.
And a slightly less schematic depiction of the heroine of today - a vacuum chamber is added in which the plasma flies, and all other equipment.
The first experiment with a “mirror” or “open” trap - Q-cucumber was put in 1955 at the American Lawrence Livermore National Laboratory. For many years, this laboratory has become a leader in the development of the concept of TCB based on open traps (OL).
The first experiment in the world - an open trap with magnetic mirrors Q-cucumber
Compared to closed competitors, the advantages of OL can be written much simpler geometry of the reactor and its magnetic system, which means low cost. So, after the fall of the first favorite of the TCB - Z-pinch reactors, open traps receive maximum priority and funding in the early 60s, as promising a quick solution for little money.
Early 60s, Table Top Trap
However, the very Z-pinch did not resign by chance. His funeral was associated with the manifestation of the nature of plasma - instabilities that destroyed plasma formations when trying to compress the plasma with a magnetic field. And this particular feature, poorly studied 50 years ago, immediately began to annoyingly interfere with open trap experimenters. The flute instabilities make the magnetic system more complicated by introducing, in addition to simple round solenoids, “Joffe sticks”, “baseball traps” and “yin-yang coils” and reduce the ratio of the magnetic field pressure to the plasma pressure (parameter β).
“Baseball” superconducting magnet traps Baseball II, mid 70s
In addition, plasma leakage proceeds differently for particles with different energies, which leads to plasma nonequilibrium (i.e., the non-Maxwellian particle velocity spectrum), which causes a number of unpleasant instabilities. These instabilities, in turn, “swaying” the plasma accelerate its escape through the end mirror cells. In the late 60s, simple versions of open traps reached the limit in temperature and density of the plasma being held, and these figures were much orders of magnitude less than those needed for a thermonuclear reaction. The problem was mainly in the rapid longitudinal cooling of electrons, which then lost energy and ions. New ideas were needed.
Successful TMX-U Ambipolar Trap
Physicists are proposing new solutions, primarily related to improving longitudinal plasma confinement: ambipolar confinement, corrugated traps, and gas-dynamic traps.
Interestingly, all of these concepts by which experimental facilities were built required further complication of open trap engineering. First of all, here for the first time in the TCB complex neutron-beam accelerators appear that heat the plasma (in the first installations, heating was achieved by a conventional electric discharge) and modulate its density in the installation. Radio frequency heating, which first appeared at the turn of the 60s / 70s in tokamaks, is added. Large and expensive Gamma-10 units are under construction in Japan, TMX in the USA, AMBAL-M, GOL and GDL in the Novosibirsk INP.
The Gamma-10 magnetic system and plasma heating diagram illustrates well how far they have gone from simple OL solutions to the 80s.
In parallel, in 1975, on the 2X-IIB trap, American researchers were the first in the world to reach a symbolic ion temperature of 10 kV, which is optimal for the thermonuclear burning of deuterium and tritium. It should be noted that in the 60s and 70s they went under the sign of a pursuit of the desired temperature in any way, because the temperature determines whether the reactor will work at all, while the other two parameters - the density and rate of leakage of energy from the plasma (or more often this is called the "retention time") can be compensated by increasing the size of the reactor. However, despite the symbolic achievement, 2X-IIB was very far from what would be called a reactor - the theoretical allocated power would be 0.1% of the plasma spent on holding and heating. A serious problem remained the low electron temperature - of the order of 90 eV against a background of 10 kev ions, associated with
Elements of the now-defunct ambipolar trap AMBAL-M
In the early 80s, there was a peak in the development of this branch of the TCB. The peak of development is the US MFTF project worth $ 372 million (or 820 million at today's prices, which brings the project closer in value to a machine like Wendelstein 7-X or tokamak K-STAR).
The superconducting magnetic modules of the MFTF ...
And the casing of its 400 ton end superconducting magnet
It was an ambipolar trap with superconducting magnets, including masterpiece terminal “yin-yang”, numerous systems and heating plasma diagnostics, a record in all respects. It was planned to achieve Q = 0.5, i.e. the energy output of the thermonuclear reaction is only half the cost of maintaining the operation of the reactor. What results did this program achieve? It was closed by a political decision in a state close to ready for launch.
The terminal "Yin-Yang" MFTF during installation in a 10-meter vacuum chamber of the installation. Its length was to reach 60 meters.
Despite the fact that this decision, shocking from all sides, is very difficult to explain, I will try.
By 1986, when the MFTF was ready to launch in the horizon of TCB concepts, a star of another favorite was lit. A simple and cheap alternative to “bronzed” open traps, which by this time had become too complicated and expensive against the background of the initial concept of the early 60s. All these superconducting magnets of puzzle configurations, fast neutral injectors, powerful radio frequency plasma heating systems, puzzle instability suppression schemes - it seemed that such complex installations will never become the prototype of a thermonuclear power plant.
JET in the initial limited configuration and copper coils.
So tokamaki. In the early 80s, these machines reached plasma parameters sufficient for burning a thermonuclear reaction. In 1984, the European JET tokamak was launched, which should show Q = 1, and it uses simple copper magnets, its cost is only 180 million dollars. In the USSR and France, superconducting tokamaks are being designed, which almost do not spend energy on the operation of the magnetic system. At the same time, physicists working on open traps for years cannot make progress in increasing plasma stability, electron temperature, and promises of MFTF achievements are becoming increasingly vague. The next decades, by the way, will show that the bet on tokamaks turned out to be relatively justified - it was these traps that reached the level of capacities and Q, interesting to power engineers.
The success of open traps and tokamaks by the beginning of the 80s on the map of the “triple parameter”. JET will reach a point slightly above TFTR 1983 in 1997.
The MFTF decision finally undermines the position of this area. Although experiments at the Novosibirsk INP and at the Japanese Gamma-10 facility are ongoing, quite successful programs of the predecessors TMX and 2X-IIB are being closed in the USA.
The end of the story? Not. Literally before our eyes, in 2015, an amazing quiet revolution is taking place. Researchers from the Institute of Nuclear Physics. Budkers in Novosibirsk, who successively improved the GDL trap (by the way, it should be noted that ambipolar rather than gasdynamic traps were superior in the West) suddenly achieve plasma parameters that were predicted to be “impossible” by skeptics in the 80s.
Once again GDL. Green cylinders sticking out in different directions are neutral injectors, which are discussed below.
The three main problems that buried open traps are MHD stability in an axisymmetric configuration (requiring complex magnets), nonequilibrium ion distribution function (micro instability), and low electron temperature. In 2015, the GDL, with a beta value of 0.6, reached an electron temperature of 1 keV. How did this happen?
Avoiding axial (cylindrical) symmetry in the 60s in attempts to defeat flute and other MHD instabilities of the plasma, in addition to complicating magnetic systems, also led to an increase in radial heat loss from the plasma. A group of scientists working with GDL used the idea of the 80s to apply a radial electric field, creating a swirling plasma. This approach led to a brilliant victory - at beta 0.6 (I recall that this ratio of plasma pressure to magnetic field pressure is a very important parameter in the design of any thermonuclear reactor - because the speed and density of energy release are determined by the plasma pressure, and the cost of the reactor is determined power of its magnets), compared with the tokamak 0.05-0.1 plasma is stable.
New measuring instruments - “diagnostics”, allow better understanding of plasma physics in GDL
The second problem with micro instabilities, caused by the lack of ions with low temperatures (which are pulled from the ends of the trap by the ambipolar potential), was solved by tilting the injectors of neutral rays at an angle. This arrangement creates ion density peaks along the plasma trap, which delay the “warm” ions from leaving. A relatively simple solution leads to the complete suppression of micro-instabilities and to a significant improvement in plasma confinement parameters.
Neutron flux from thermonuclear combustion of deuterium in the GDL trap. Black dots - measurements, lines - different calculated values for different levels of micro instabilities. Red line - micro instability suppressed.
Finally, the main “grave digger” is the low temperature of electrons. Although thermonuclear parameters have been achieved for ions in traps, a high electron temperature is the key to keeping hot ions from cooling, and therefore a high Q value. The reason for the low temperature is the high thermal conductivity “along” and the ambipolar potential, which sucks “cold” electrons from expanders at the ends traps inside the magnetic system. Until 2014, the electron temperature in open traps did not exceed 300 eV, and the psychologically important value of 1 kV was obtained in the GDL. It was obtained due to the fine work with the physics of the interaction of electrons in terminal expanders with a neutral gas and plasma absorbers.
This turns the situation upside down. Now, simple traps again threaten the primacy of tokamaks, which have reached monstrous sizes and complexity ( several examples of the complexity of ITER systems ). Moreover, this opinion is not only of scientists from INP, but also of serious American scientists , published in authoritative journals.
More GDL close. Thanks for the photos dedmaxopka
So far, however, the success of the GDL has led to new proposals for installations only at the INP itself. Having won a grant of the Ministry of Education and Science at 650 million rubles, the institute will build several engineering stands in the framework of the promising rector " GDML-U", объединяющего идеи и достижения ГДЛ и способ улучшения продольного удержания ГОЛ. Хотя под влиянием новых результатов образ ГДМЛ меняется, но она остается магистральной идеей в области открытых ловушек.
Где находятся текущие и будущие разработки по сравнению с конкурентами? Токамаки, как известно, достигли значения Q=1, решили множество инженерных проблем, перешлю к строительству ядерных, а не электрических установок и уверено движутся к уже скорее прообразу энергетического реактора с Q=10 и термоядерной мощностью до 700 МВт (ИТЭР). Стеллараторы, отстающие на пару шагов переходят от изучения принципиальной физики и решению инженерных проблем при Q=0.1, но пока не рискуют заходить на поле истинно ядерных установок с термоядерным горением трития. ГДМЛ-U могла бы быть похожа на стелларатор W-7Xby plasma parameters (being, however, a pulsed installation with a discharge duration of several seconds versus the W-7X half-hour operation in the future), however, due to the simple geometry, its cost can be several times lower than the German stellarator.
INP assessment.
There are options for using GDMF as a facility for studying the interaction of plasma and materials (there are, however, quite a lot of such facilities in the world) and as a thermonuclear source of neutrons for various purposes.
Extrapolation of GDMF sizes depending on the desired Q and possible applications.
If tomorrow open traps again become favorites in the race to the TCB, one could expect that due to lower capital investments in each stage, by 2050 they will catch up and overtake tokamaks, becoming the heart of the first thermonuclear power plants. Unless the plasma presents new unpleasant surprises ...
GDL open trap with impressive results
Among the whole variety of proposals, how to extract energy from thermonuclear fusion is most oriented towards stationary confinement of a relatively loose thermonuclear plasma. For example, the ITER project and more broadly - toroidal traps tokamaki and stellarators - from here. They are toroidal because it is the simplest form of a closed vessel of magnetic fields (due to the hedgehog combing theorema spherical vessel cannot be made). However, at the dawn of research in the field of controlled thermonuclear fusion, the favorites were not traps of complex three-dimensional geometry, but attempts to keep the plasma in the so-called open traps. These are usually also cylindrical-shaped magnetic vessels in which the plasma is well retained in the radial direction and flows from both ends. The idea of the inventors here is simple - if the heating of a new plasma by a thermonuclear reaction proceeds faster than the heat flow from the ends, then God bless him, with the openness of our vessel, the energy will be generated, and the leak will still occur in a vacuum vessel and the fuel will walk in the reactor until it burns out.
The idea of an open trap is a magnetic cylinder with corks / mirrors at the ends and expanders behind them.
In addition, in all open traps, one or another method is used to delay the plasma from escaping through the ends — and the simplest here is to sharply increase the magnetic field at the ends (put magnetic “plugs” in Russian terminology or “mirrors” in the west), while incident charged particles will, in fact, spring from the mirror mirrors and only a small fraction of the plasma will pass through them and fall into special expanders.
And a slightly less schematic depiction of the heroine of today - a vacuum chamber is added in which the plasma flies, and all other equipment.
The first experiment with a “mirror” or “open” trap - Q-cucumber was put in 1955 at the American Lawrence Livermore National Laboratory. For many years, this laboratory has become a leader in the development of the concept of TCB based on open traps (OL).
The first experiment in the world - an open trap with magnetic mirrors Q-cucumber
Compared to closed competitors, the advantages of OL can be written much simpler geometry of the reactor and its magnetic system, which means low cost. So, after the fall of the first favorite of the TCB - Z-pinch reactors, open traps receive maximum priority and funding in the early 60s, as promising a quick solution for little money.
Early 60s, Table Top Trap
However, the very Z-pinch did not resign by chance. His funeral was associated with the manifestation of the nature of plasma - instabilities that destroyed plasma formations when trying to compress the plasma with a magnetic field. And this particular feature, poorly studied 50 years ago, immediately began to annoyingly interfere with open trap experimenters. The flute instabilities make the magnetic system more complicated by introducing, in addition to simple round solenoids, “Joffe sticks”, “baseball traps” and “yin-yang coils” and reduce the ratio of the magnetic field pressure to the plasma pressure (parameter β).
“Baseball” superconducting magnet traps Baseball II, mid 70s
In addition, plasma leakage proceeds differently for particles with different energies, which leads to plasma nonequilibrium (i.e., the non-Maxwellian particle velocity spectrum), which causes a number of unpleasant instabilities. These instabilities, in turn, “swaying” the plasma accelerate its escape through the end mirror cells. In the late 60s, simple versions of open traps reached the limit in temperature and density of the plasma being held, and these figures were much orders of magnitude less than those needed for a thermonuclear reaction. The problem was mainly in the rapid longitudinal cooling of electrons, which then lost energy and ions. New ideas were needed.
Successful TMX-U Ambipolar Trap
Physicists are proposing new solutions, primarily related to improving longitudinal plasma confinement: ambipolar confinement, corrugated traps, and gas-dynamic traps.
- Ambipolar confinement is based on the fact that electrons “leak” out of an open trap 28 times faster than deuterium and tritium ions, and a potential difference arises at the ends of the trap — positive from ions inside and negative outside. If amplification of a field with a dense plasma is made at the ends of the setup, then the ambipolar potential in a dense plasma will keep the internal less dense contents from scattering.
- Corrugated traps create a “ribbed” magnetic field at the end, where heavy ions fly apart due to “friction” against trap fields locked in “troughs”.
- Finally, gas-dynamic traps create a magnetic field analogue of a vessel with a small hole, from which the plasma flows at a lower speed than in the case of “mirror-plugs”.
Interestingly, all of these concepts by which experimental facilities were built required further complication of open trap engineering. First of all, here for the first time in the TCB complex neutron-beam accelerators appear that heat the plasma (in the first installations, heating was achieved by a conventional electric discharge) and modulate its density in the installation. Radio frequency heating, which first appeared at the turn of the 60s / 70s in tokamaks, is added. Large and expensive Gamma-10 units are under construction in Japan, TMX in the USA, AMBAL-M, GOL and GDL in the Novosibirsk INP.
The Gamma-10 magnetic system and plasma heating diagram illustrates well how far they have gone from simple OL solutions to the 80s.
In parallel, in 1975, on the 2X-IIB trap, American researchers were the first in the world to reach a symbolic ion temperature of 10 kV, which is optimal for the thermonuclear burning of deuterium and tritium. It should be noted that in the 60s and 70s they went under the sign of a pursuit of the desired temperature in any way, because the temperature determines whether the reactor will work at all, while the other two parameters - the density and rate of leakage of energy from the plasma (or more often this is called the "retention time") can be compensated by increasing the size of the reactor. However, despite the symbolic achievement, 2X-IIB was very far from what would be called a reactor - the theoretical allocated power would be 0.1% of the plasma spent on holding and heating. A serious problem remained the low electron temperature - of the order of 90 eV against a background of 10 kev ions, associated with
Elements of the now-defunct ambipolar trap AMBAL-M
In the early 80s, there was a peak in the development of this branch of the TCB. The peak of development is the US MFTF project worth $ 372 million (or 820 million at today's prices, which brings the project closer in value to a machine like Wendelstein 7-X or tokamak K-STAR).
The superconducting magnetic modules of the MFTF ...
And the casing of its 400 ton end superconducting magnet
It was an ambipolar trap with superconducting magnets, including masterpiece terminal “yin-yang”, numerous systems and heating plasma diagnostics, a record in all respects. It was planned to achieve Q = 0.5, i.e. the energy output of the thermonuclear reaction is only half the cost of maintaining the operation of the reactor. What results did this program achieve? It was closed by a political decision in a state close to ready for launch.
The terminal "Yin-Yang" MFTF during installation in a 10-meter vacuum chamber of the installation. Its length was to reach 60 meters.
Despite the fact that this decision, shocking from all sides, is very difficult to explain, I will try.
By 1986, when the MFTF was ready to launch in the horizon of TCB concepts, a star of another favorite was lit. A simple and cheap alternative to “bronzed” open traps, which by this time had become too complicated and expensive against the background of the initial concept of the early 60s. All these superconducting magnets of puzzle configurations, fast neutral injectors, powerful radio frequency plasma heating systems, puzzle instability suppression schemes - it seemed that such complex installations will never become the prototype of a thermonuclear power plant.
JET in the initial limited configuration and copper coils.
So tokamaki. In the early 80s, these machines reached plasma parameters sufficient for burning a thermonuclear reaction. In 1984, the European JET tokamak was launched, which should show Q = 1, and it uses simple copper magnets, its cost is only 180 million dollars. In the USSR and France, superconducting tokamaks are being designed, which almost do not spend energy on the operation of the magnetic system. At the same time, physicists working on open traps for years cannot make progress in increasing plasma stability, electron temperature, and promises of MFTF achievements are becoming increasingly vague. The next decades, by the way, will show that the bet on tokamaks turned out to be relatively justified - it was these traps that reached the level of capacities and Q, interesting to power engineers.
The success of open traps and tokamaks by the beginning of the 80s on the map of the “triple parameter”. JET will reach a point slightly above TFTR 1983 in 1997.
The MFTF decision finally undermines the position of this area. Although experiments at the Novosibirsk INP and at the Japanese Gamma-10 facility are ongoing, quite successful programs of the predecessors TMX and 2X-IIB are being closed in the USA.
The end of the story? Not. Literally before our eyes, in 2015, an amazing quiet revolution is taking place. Researchers from the Institute of Nuclear Physics. Budkers in Novosibirsk, who successively improved the GDL trap (by the way, it should be noted that ambipolar rather than gasdynamic traps were superior in the West) suddenly achieve plasma parameters that were predicted to be “impossible” by skeptics in the 80s.
Once again GDL. Green cylinders sticking out in different directions are neutral injectors, which are discussed below.
The three main problems that buried open traps are MHD stability in an axisymmetric configuration (requiring complex magnets), nonequilibrium ion distribution function (micro instability), and low electron temperature. In 2015, the GDL, with a beta value of 0.6, reached an electron temperature of 1 keV. How did this happen?
Avoiding axial (cylindrical) symmetry in the 60s in attempts to defeat flute and other MHD instabilities of the plasma, in addition to complicating magnetic systems, also led to an increase in radial heat loss from the plasma. A group of scientists working with GDL used the idea of the 80s to apply a radial electric field, creating a swirling plasma. This approach led to a brilliant victory - at beta 0.6 (I recall that this ratio of plasma pressure to magnetic field pressure is a very important parameter in the design of any thermonuclear reactor - because the speed and density of energy release are determined by the plasma pressure, and the cost of the reactor is determined power of its magnets), compared with the tokamak 0.05-0.1 plasma is stable.
New measuring instruments - “diagnostics”, allow better understanding of plasma physics in GDL
The second problem with micro instabilities, caused by the lack of ions with low temperatures (which are pulled from the ends of the trap by the ambipolar potential), was solved by tilting the injectors of neutral rays at an angle. This arrangement creates ion density peaks along the plasma trap, which delay the “warm” ions from leaving. A relatively simple solution leads to the complete suppression of micro-instabilities and to a significant improvement in plasma confinement parameters.
Neutron flux from thermonuclear combustion of deuterium in the GDL trap. Black dots - measurements, lines - different calculated values for different levels of micro instabilities. Red line - micro instability suppressed.
Finally, the main “grave digger” is the low temperature of electrons. Although thermonuclear parameters have been achieved for ions in traps, a high electron temperature is the key to keeping hot ions from cooling, and therefore a high Q value. The reason for the low temperature is the high thermal conductivity “along” and the ambipolar potential, which sucks “cold” electrons from expanders at the ends traps inside the magnetic system. Until 2014, the electron temperature in open traps did not exceed 300 eV, and the psychologically important value of 1 kV was obtained in the GDL. It was obtained due to the fine work with the physics of the interaction of electrons in terminal expanders with a neutral gas and plasma absorbers.
This turns the situation upside down. Now, simple traps again threaten the primacy of tokamaks, which have reached monstrous sizes and complexity ( several examples of the complexity of ITER systems ). Moreover, this opinion is not only of scientists from INP, but also of serious American scientists , published in authoritative journals.
More GDL close. Thanks for the photos dedmaxopka
So far, however, the success of the GDL has led to new proposals for installations only at the INP itself. Having won a grant of the Ministry of Education and Science at 650 million rubles, the institute will build several engineering stands in the framework of the promising rector " GDML-U", объединяющего идеи и достижения ГДЛ и способ улучшения продольного удержания ГОЛ. Хотя под влиянием новых результатов образ ГДМЛ меняется, но она остается магистральной идеей в области открытых ловушек.
Где находятся текущие и будущие разработки по сравнению с конкурентами? Токамаки, как известно, достигли значения Q=1, решили множество инженерных проблем, перешлю к строительству ядерных, а не электрических установок и уверено движутся к уже скорее прообразу энергетического реактора с Q=10 и термоядерной мощностью до 700 МВт (ИТЭР). Стеллараторы, отстающие на пару шагов переходят от изучения принципиальной физики и решению инженерных проблем при Q=0.1, но пока не рискуют заходить на поле истинно ядерных установок с термоядерным горением трития. ГДМЛ-U могла бы быть похожа на стелларатор W-7Xby plasma parameters (being, however, a pulsed installation with a discharge duration of several seconds versus the W-7X half-hour operation in the future), however, due to the simple geometry, its cost can be several times lower than the German stellarator.
INP assessment.
There are options for using GDMF as a facility for studying the interaction of plasma and materials (there are, however, quite a lot of such facilities in the world) and as a thermonuclear source of neutrons for various purposes.
Extrapolation of GDMF sizes depending on the desired Q and possible applications.
If tomorrow open traps again become favorites in the race to the TCB, one could expect that due to lower capital investments in each stage, by 2050 they will catch up and overtake tokamaks, becoming the heart of the first thermonuclear power plants. Unless the plasma presents new unpleasant surprises ...