The article discusses the reasons why controlled thermonuclear fusion has not found industrial application so far.
When in the fifties of the last century the Earth was shaken by powerful explosions thermonuclear bombsit seemed that before peaceful use fusion energies very little is left: one or two decades. There were also grounds for such optimism: only 10 years had elapsed from the moment the atomic bomb was used to the creation of a reactor that produced electricity.
But the task of curbing thermonuclear fusion turned out to be extraordinarily difficult. Decades passed one after another, and access to unlimited supplies of energy was never obtained. During this time, humanity, burning fossil resources, polluted the atmosphere with emissions and overheated it with greenhouse gases. The disasters at Chernobyl and Fukushima-1 have discredited nuclear power.
What prevented the mastery of such a promising and safe process of thermonuclear fusion, which could forever remove the problem of providing mankind with energy?
Initially, it was clear that for the reaction to proceed, it is necessary to bring the hydrogen nuclei closer together so that the nuclear forces could form the nucleus of a new element - helium with the release of a significant amount of energy. But hydrogen nuclei are repelled from each other by electrical forces. An assessment of the temperatures and pressures at which a controlled thermonuclear reaction begins has shown that no material can withstand such temperatures.
For the same reasons, pure deuterium, an isotope of hydrogen, was rejected. After spending billions of dollars and decades of time, scientists were finally able to ignite a thermonuclear flame for a very short time. It remains to learn how to hold the fusion plasma for a long time. It was necessary to move from computer modeling to the construction of a real reactor.
At this stage, it became clear that the efforts and funds of a separate state would not be enough for the construction and operation of experimental and experimental industrial installations. Within the framework of international cooperation, it was decided to implement a project of an experimental thermonuclear reactor worth more than $ 14 billion.
But in 1996, the United States ceased its participation and, accordingly, funding for the project. For some time, the implementation went on at the expense of Canada, Japan and Europe, but the matter did not come to the construction of the reactor.
The second project, also international, is being implemented in France. Long-term retention of plasma occurs due to a special form of the magnetic field - in the form of a bottle. The basis of this method was laid by Soviet physicists. The first installation of the "Tokamak" type should give at the output more energy than is spent on igniting and holding the plasma.
By 2012, the installation of the reactor should have been completed, but there is still no information on successful operation. Perhaps the economic upheavals of recent years have made adjustments to the plans of scientists.
Difficulties in achieving controlled thermonuclear fusion gave rise to a lot of speculation and false reports about the so-called "Cold" thermonuclear nuclear fusion reaction. Despite the fact that no physical capabilities or laws have yet been found, many researchers argue about its existence. After all, the stakes are too high: from Nobel Prizes for scientists to the geopolitical domination of a state that has mastered this technology and gained access to energy abundance.
But every such message turns out to be exaggerated or downright false. Serious scientists are skeptical about the existence of such a reaction.
The real possibilities of mastering fusion and starting commercial operation of thermonuclear reactors are postponed to the middle of the 21st century. By this time, it will be possible to select the necessary materials and work out its safe operation. Since such reactors will operate with very low density plasma, safety of fusion power plants will be much higher than nuclear power plants.
Any violation in the reaction zone will immediately "extinguish" the thermonuclear flame. But you should not neglect safety measures: the unit power of the reactors will be so great that an accident even in the heat extraction circuits can lead to casualties and environmental pollution. The only thing left to do is to wait 30-40 years and see the era of energy abundance. If we live, of course.
Innovative projects using modern superconductors in the near future will make it possible to carry out controlled thermonuclear fusion, some optimists say. Experts, however, predict that practical implementation will take several decades.
Why is it so difficult?
Fusion energy is considered a potential source. It is the pure energy of the atom. But what is it and why is it so difficult to achieve? First, you need to understand the difference between classical and thermonuclear fusion.
Atomic fission means that radioactive isotopes - uranium or plutonium - are fissioned and converted into other highly radioactive isotopes, which must then be buried or reprocessed.
Synthesis is that two isotopes of hydrogen - deuterium and tritium - merge into a single whole, forming non-toxic helium and a single neutron, without producing radioactive waste.
Control problem
The reactions that take place on the sun or in a hydrogen bomb are thermonuclear fusion, and engineers are faced with a daunting task - how to control this process at a power plant?
This is what scientists have been working on since the 1960s. Another experimental thermonuclear fusion reactor called Wendelstein 7-X began work in the northern German city of Greifswald. It is not yet designed to create a reaction - it is just a special design that is being tested (stellarator instead of tokamak).
High energy plasma
All thermonuclear installations have a common feature - an annular shape. It is based on the idea of \u200b\u200busing powerful electromagnets to create a strong electromagnetic field in the shape of a torus - an inflated bicycle tube.
This electromagnetic field must be so dense that when it is heated in a microwave oven to one million degrees Celsius, a plasma should appear in the very center of the ring. It is then ignited so that fusion can begin.
Demonstration of possibilities
Two similar experiments are currently underway in Europe. One of them is Wendelstein 7-X, which recently generated its first helium plasma. The other is ITER, a huge experimental fusion plant in the south of France that is still under construction and will be ready to launch in 2023.
It is assumed that real nuclear reactions will take place on ITER, however, only for a short period of time and certainly no longer than 60 minutes. This reactor is just one of many steps towards putting nuclear fusion into practice.
Fusion reactor: smaller and more powerful
Several designers recently announced a new design for the reactor. According to a group of MIT students and representatives of the arms manufacturer Lockheed Martin, thermonuclear fusion can be carried out in installations that are much more powerful and smaller than ITER, and they are ready to do it within ten years.
The idea of \u200b\u200bthe new design is to use modern high-temperature superconductors in electromagnets, which show their properties when cooled with liquid nitrogen, rather than conventional ones, for which a new, more flexible technology will allow a complete redesign of the reactor.
Klaus Hesch, in charge of technology at the Karlsruhe Institute of Technology in southwestern Germany, is skeptical. It supports the use of new high-temperature superconductors for new reactor designs. But, according to him, it is not enough to develop something on a computer, taking into account the laws of physics. It is necessary to take into account the challenges that arise when translating an idea into practice.
Science fiction
According to Hesh, the MIT student model shows only the possibility of a project. But it's actually a lot of science fiction. The project assumes that serious technical problems of thermonuclear fusion have been solved. But modern science has no idea how to solve them.
One such problem is the idea of \u200b\u200bcollapsible coils. In the MIT design model, the electromagnets can be disassembled to get inside the plasma-containing ring.
This would be very useful, because one could access and replace objects in the internal system. But in reality, superconductors are made of ceramic material. Hundreds of them must be intertwined in a sophisticated way to form the correct magnetic field. And this is where more fundamental difficulties arise: the connections between them are not as simple as those of copper cables. Nobody has even thought about concepts that would help solve such problems.
Too hot
High temperatures are also a problem. In the core of the fusion plasma, the temperature will reach about 150 million degrees Celsius. This extreme heat stays in place - right in the center of the ionized gas. But even around it, it is still very hot - from 500 to 700 degrees in the reactor zone, which is the inner layer of a metal tube, in which tritium will be "reproduced", which is necessary for nuclear fusion to occur.
Has an even bigger problem - the so-called power release. This is the part of the system that receives spent fuel from the fusion process, mainly helium. The first metal components that get hot gas are called "divertors". It can heat up to over 2000 ° C.
Divertor problem
In order for the installation to withstand such temperatures, engineers are trying to use the metallic tungsten used in old-fashioned incandescent bulbs. The melting point of tungsten is about 3000 degrees. But there are other limitations as well.
In ITER, this can be done, because heating does not occur constantly in it. The reactor is expected to operate only 1-3% of the time. But this is not an option for a power plant that needs to operate 24/7. And, if someone claims to be able to build a smaller reactor with the same power as ITER, it is safe to say that they have no solution to the divertor problem.
Power plant in a few decades
Nevertheless, scientists are optimistic about the development of thermonuclear reactors, although it will not be as fast as some enthusiasts predict.
ITER should show that controlled thermonuclear fusion can actually produce more energy than would be expended to heat the plasma. The next step will be the construction of an entirely new hybrid demonstration power plant that actually generates electricity.
Engineers are already working on its design. They will have to learn from ITER, which is slated to launch in 2023. Given the time required for design, planning and construction, it seems unlikely that the first fusion power plant will be launched much earlier than the mid-21st century.
Rossi's cold fusion
In 2014, an independent test of the E-Cat reactor concluded that the device produced an average of 2800 watts of output power over a period of 32 days while consuming 900 watts. This is more than any chemical reaction can produce. The result speaks either of a breakthrough in thermonuclear fusion, or of outright fraud. The report has disappointed skeptics who question whether the test was truly independent and speculate that the test results could be falsified. Others set out to figure out the "secret ingredients" that allow Rossi's fusion to replicate the technology.
Is Rossi a crook?
Andrea is imposing. He publishes proclamations to the world in unique English in the comments section of his website, the pretentiously titled Journal of Nuclear Physics. But his previous unsuccessful attempts included an Italian project for converting garbage into fuel and a thermoelectric generator. Petroldragon, a waste-to-energy project, has failed in part because the illegal disposal of waste is controlled by Italian organized crime, which has filed criminal charges against him for violating waste regulations. He also created a thermoelectric device for the US Army Corps of Engineers, but during testing, the gadget produced only a fraction of the declared power.
Many do not trust Russia, and the editor-in-chief of the New Energy Times called him a felon with a series of unsuccessful energy projects behind him.
Independent verification
Rossi signed a contract with the American company Industrial Heat to conduct a year-long secret testing of a 1-MW cold fusion plant. The device was a shipping container packed with dozens of E-Cats. The experiment had to be monitored by a third party who could confirm that there was indeed heat generation. Rossi claims to have spent most of the last year practically living in a container and overseeing operations for over 16 hours a day to prove the commercial viability of the E-Cat.
The test ended in March. Rossi's supporters anxiously awaited the report of the observers, hoping to justify their hero. But in the end they got a lawsuit.
Trial
In a statement to a Florida court, Rossi claims the test was successful and an independent arbiter confirmed that the E-Cat reactor produces six times more energy than it consumes. He also claimed that Industrial Heat had agreed to pay him $ 100 million - $ 11.5 million upfront after a 24-hour trial (ostensibly for licensing rights so the company could sell the technology in the US) and another $ 89 million after successfully completing an extended trial. within 350 days. Rossi accused IH of carrying out a "fraudulent scheme" aimed at stealing his intellectual property. He also accused the company of misappropriating E-Cat reactors, illegally copying innovative technologies and products, functionality and designs, and improperly seeking a patent for its intellectual property.
Goldmine
Elsewhere, Rossi claims that, amid one of his demonstrations, IH received $ 50-60 million from investors and another $ 200 million from China after a replay involving top Chinese officials. If this is true, then a lot more than a hundred million dollars are at stake. Industrial Heat dismissed these claims as unfounded and will actively defend itself. More importantly, she claims that "for over three years, she has been working to validate the results that Rossi allegedly achieved with his E-Cat technology, and all to no avail."
IH does not believe the E-Cat will work, and the New Energy Times sees no reason to doubt it. In June 2011, a representative of the publication visited Italy, interviewed Rossi and filmed a demonstration of his E-Cat. A day later, he announced his serious concerns about the method of measuring heat output. After 6 days, the journalist posted his video on YouTube. Experts from all over the world sent him analyzes, which were published in July. It became clear that this was a hoax.
Experimental confirmation
Nevertheless, a number of researchers - Alexander Parkhomov from the Peoples' Friendship University of Russia and the Martin Fleischman Memory Project (MFPM) - managed to reproduce Rossi's cold thermonuclear fusion. The MFPM report was titled "The End of the Carbon Era is Near." The reason for this admiration was the discovery that cannot be explained otherwise than by a thermonuclear reaction. According to the researchers, Rossi has exactly what he is talking about.
A viable open recipe for cold fusion has the potential to trigger an energetic gold rush. Alternative methods could be found to bypass Rossi's patents and leave him out of the multibillion-dollar energy business.
So perhaps Rossi would have preferred to avoid this confirmation.
Yu.N. Dnestrovsky - Ph.D. Sci., Professor, Institute of Nuclear Fusion,
RRC "Kurchatov Institute", Moscow, Russia
Materials of the International Conference
"THE WAY TO THE FUTURE - SCIENCE, GLOBAL PROBLEMS, DREAMS AND HOPES"
November 26-28, 2007 Institute of Applied Mathematics. M.V. Keldysh RAS, Moscow
Can controlled thermonuclear fusion (CTF) solve the energy problem in the long term? What part of the path for mastering TCB has already been passed and how much is still left to go? What are the challenges ahead? These problems are discussed in this work.
1. Physical prerequisites for TCB
For energy production, it is proposed to use nuclear fusion reactions of light nuclei. Among the many reactions of this type, the most easily feasible reaction is the fusion of deuterium and tritium nuclei.
Here, through denotes a stable helium nucleus (alpha particle), through N - a neutron, in brackets denotes the energy of particles after the reaction,. In this reaction, the energy released per particle with the mass of a neutron is approximately 3.5 MeV. This is about 3-4 times the energy per particle released by the fission of uranium.
What problems arise when trying to implement reaction (1) to obtain energy?
The main problem is that tritium does not exist in nature. It is radioactive, its half-life is approximately 12 years, therefore, if it was once in large quantities on Earth, then nothing remained of it for a long time. The amount of tritium received on the Earth due to natural radioactivity or from cosmic radiation is negligible. A small amount of tritium is produced in the reactions taking place inside an atomic uranium reactor. A collection of such tritium is organized at one of the reactors in Canada, but its production in the reactors is very slow and the production is too expensive.
Thus, the production of energy in a thermonuclear reactor based on reaction (1) must be accompanied by the simultaneous production of tritium in the same reactor. We will discuss below how this can be done.
Both particles, the nuclei of deuterium and tritium, participating in reaction (1), have a positive charge and therefore are repelled from each other by the Coulomb force. To overcome this force, the particles must have high energy. The dependence of the reaction rate (1),, on the temperature of the tritium-deuterium mixture is shown in Fig. 1 on a double logarithmic scale.
It is seen that the probability of reaction (1) rapidly increases with increasing temperature. The reaction rate acceptable for the reactor is achieved at a temperature T\u003e 10 keV. Considering that degrees, then the temperature in the reactor should exceed 100 million degrees. All atoms of a substance at such a temperature must be ionized, and the substance itself in this state is usually called plasma. Recall that according to modern estimates, the temperature in the center of the Sun reaches "only" 20 million degrees.
There are other fusion reactions that are suitable, in principle, for the production of thermonuclear energy. Here we note only two reactions widely discussed in the literature:
Here is an isotope of the helium nucleus with a mass equal to 3, p is a proton (hydrogen nucleus). Reaction (2) is good in that there is as much fuel (deuterium) for it on Earth. The technology for separating deuterium from seawater has been developed and is relatively inexpensive. Unfortunately, the rate of this reaction is noticeably lower than the rate of reaction (1) (see Fig. 1), therefore, to implement reaction (2), a temperature of about 500 million degrees is required.
Reaction (3) is currently causing great excitement among people who are engaged in space flights. It is known that there is a lot of the isotope on the Moon, so the possibility of transporting it to Earth is discussed as one of the priority tasks of cosmonautics. Unfortunately, the rate of this reaction (Fig. 1) is also noticeably lower, the rate of reaction (1) and the required temperatures for this reaction are also at the level of 500 million degrees.
To confine plasma with a temperature of about 100 - 500 million degrees, it was proposed to use a magnetic field (I.E. Tamm, A.D. Sakharov). The most promising now seem to be installations in which the plasma has the form of a torus (donut). We denote the large radius of this torus by R, and small through a... To suppress unstable plasma motions, in addition to the toroidal (longitudinal) magnetic field B 0, a transverse (poloidal) field is also required. There are two types of installations in which this magnetic configuration is realized. In devices of the tokamak type, the poloidal field is created by the longitudinal current I flowing in the plasma in the direction of the field. In stellarator-type installations, the poloidal field is created by external helical windings with current. Each of these installations has its own advantages and disadvantages. In a tokamak, the current I must be matched to the field. The stellarator is technically more complex. Today, tokamak installations are more advanced. Although there are also large, successful stellarators available.
2. Conditions for the tokamak reactor
We will indicate here only two necessary conditions that determine the "window" in the space of parameters of the reactor tokamak plasma. There are, of course, many other conditions that reduce this "window", but they are still not so essential.
1). For a reactor to be commercially viable (not too large), the power density P of the released energy must be large enough
Here n 1 and n 2 are the densities of deuterium and tritium - the energy released in one act of reaction (1). Condition (4) limits the densities n 1 and n 2 from below.
2). For a plasma to be stable, the plasma pressure must be noticeably less than the pressure of the longitudinal magnetic field For a plasma with a reasonable geometry, this condition has the form
For a given magnetic field, this condition limits the plasma density and temperature from above. If the reaction requires an increase in temperature (for example, from reaction (1) to go to reactions (2) or (3)), then in order to fulfill condition (5), the magnetic field must be increased.
What magnetic field is needed to implement TCF? Let us first consider a reaction of type (1). Let us assume for simplicity that n 1 \u003d n 2 \u003d n / 2, where n is the plasma density. Then at temperature condition (1) gives
Using condition (5), we find the lower bound for the magnetic field
In the toroidal geometry, the longitudinal magnetic field decreases as 1 / r as the distance from the main axis of the torus increases. The field is the field at the center of the plasma meridional section. The field will be larger on the inner circumference of the torus. With an aspect ratio
R/ a~ 3 the magnetic field inside the coils of the toroidal field turns out to be 2 times greater. Thus, to fulfill the conditions (4-5), the longitudinal field coils must be made of a material capable of operating at a magnetic field of the order of 13-14 Tesla.
For stationary operation of a tokamak reactor, the conductors in the coils must be made of superconducting material. Some properties of modern superconductors are shown in Fig. 2.
Currently, several tokamaks with superconducting windings have been built in the world. The very first tokamak of this type (tokamak T-7), built in the USSR in the seventies, used niobium-titanium (NbTi) as a superconductor. The same material was used in the large French tokamak Tore Supra (mid-1980s). Figure 2 shows that at the temperature of liquid helium, the magnetic field in a tokamak with such a superconductor can reach 4 Tesla. For the international tokamak reactor ITER, it was decided to use a niobium-tin superconductor with great capabilities, but also with a more complex technology. This superconductor is used in the Russian T-15 facility, launched in 1989. It can be seen from Fig. 2 that in ITER at a helium temperature of the order of the magnetic field in plasma with a large margin can reach the required field values \u200b\u200bof 6 Tesla.
For reactions (2) and (3), conditions (4) - (5) turn out to be much more stringent. To satisfy condition (4), the plasma temperature in the reactor T must be 4 times higher and the plasma density n 2 times higher than in the reactor based on reaction (1). As a result, the plasma pressure increases 8 times, and the required magnetic field increases 2.8 times. This means that the magnetic field on a superconductor should reach values \u200b\u200bof 30 Tesla. So far, no one has worked with such fields in a large volume in a stationary mode. Figure 2 shows that there is hope to create in the future a superconductor for such a field. However, at present, conditions (4) - (5) for reactions of type (2) - (3) in a tokamak installation cannot be realized.
3. Tritium production
In a tokamak reactor, the plasma chamber must be surrounded by a thick layer of materials that protect the toroidal field windings from the destruction of superconductivity by neutrons. This layer, about a meter thick, is called blanket. Here, in the blanket, the heat released by neutrons during deceleration should be removed. In this case, some of the neutrons can be used to produce tritium inside the blanket. The most suitable nuclear reaction for such a process is the following reaction, which goes with the release of energy
Here is the isotope of lithium with mass 6. Since the neutron is a neutral particle, there is no Coulomb barrier and reaction (8) can proceed at a neutron energy that is noticeably less than 1 MeV. For efficient production of tritium, the number of reactions of type (8) must be sufficiently large, and for this there must be a large number of reacting neutrons. To increase the number of neutrons, materials in which the neutron multiplication reactions take place must be located here in the blanket. Since the energy of the primary neutrons obtained in reaction (1) is high (14 MeV), and reaction (8) requires low-energy neutrons, then, in principle, the number of neutrons in the blanket can be increased by 10-15 times and, thereby , close the balance for tritium: for each act of reaction (1) receive one or more acts of reaction (8). Can this balance be realized in practice? The answer to this question requires detailed experiments and calculations. The ITER reactor is not required to provide itself with fuel, but experiments will be carried out on it to clarify the problem of tritium balance.
How much tritium is required to operate the reactor? Simple estimates show that a reactor with a thermal capacity of 3 GW (electrical capacity of the order of 1 GW) will require 150 kg of tritium per year. This is approximately one times less than the weight of fuel oil required for the annual operation of a thermal power plant of the same capacity.
By virtue of (8), the primary "fuel" for the reactor is the lithium isotope. Is there a lot of it in nature? Natural lithium contains two isotopes
It is seen that the content of the isotope in natural lithium is quite high. Lithium reserves in the Earth at the current level of energy consumption will last for several thousand years, and in the ocean for tens of millions of years. Estimates based on formulas (8) - (9) show that natural lithium needs to be mined 50-100 times more than tritium is required. Thus, one reactor with the discussed capacity will require 15 tons of natural lithium per year. This is 10 5 times less than the required fuel oil for a thermal power plant. Although significant energy is required for isotope separation in natural lithium, the additional energy released in reaction (8) can compensate for these costs.
4. A Brief History of TCF Research
Historically, the secret Report of I.E. Tamm and A.D. Sakharov, released in March-April 1950, is considered the first study on TCF in our country. It was published later in 1958. The report contained an overview of the main ideas on the confinement of hot plasma by a magnetic field in a toroidal installation and an estimate of the size of a fusion reactor. Surprisingly, the ITER tokamak currently under construction is close in its parameters to the predictions of the historical Report.
Hot plasma experiments began in the USSR in the early fifties. At first, these were small installations of various types, straight and toroidal, but already in the middle of the decade, the joint work of experimenters and theoreticians led to installations that received the name "tokamak". From year to year, the size and complexity of the installations increased, and in 1962 the T-3 installation was launched with dimensions R \u003d 100 cm, a \u003d 20 cm and a magnetic field of up to four Tesla. The experience accumulated over a decade and a half has shown that in an installation with a metal chamber, well-cleaned walls and a high vacuum (up to mm Hg), it is possible to obtain a clean, stable plasma with a high electron temperature. L.A. Artsimovich reported on these results at the International Conference on Plasma Physics and Controlled Fusion in 1968 in Novosibirsk. After that, the direction of tokamaks was recognized by the world scientific community and installations of this type began to be built in many countries.
Tokamaks of the next, second, generation (T-10 in the USSR and PLT in the USA) began working with plasma in 1975. They showed that the hopes generated by the first generation tokamaks are being confirmed. And in tokamaks with large dimensions, it is possible to work with stable and hot plasma. However, even then it became clear that it was impossible to create a small-sized reactor and that it was necessary to increase the size of the plasma.
The design of the third generation tokamaks took about five years and construction began in the late seventies. In the next decade, they were consistently put into operation and by 1989, 7 large tokamaks were operating: TFTR and DIII - D in the USA, JET (the largest) in united Europe, ASDEX - U in Germany, TORE - SUPRA in France, JT 60 - U in Japan and T-15 in the USSR. These installations were used to obtain the temperature and plasma density required for the reactor. Of course, so far they were obtained separately, separately for temperature and separately for density. The TFTR and JET installations allowed the possibility of working with tritium, and for the first time they obtained a noticeable thermonuclear power P DT (in accordance with reaction (1)), comparable to the external power introduced into the plasma P aux. The maximum power P DT at the JET facility in the 1997 experiments reached 16 MW at a power P aux of about 25 MW. A sectional view of the JET unit and an internal view of the chamber are shown in Fig. 3 a, b. Here, for comparison, the size of a person is shown.
At the very beginning of the 1980s, a joint work of an international group of scientists (Russia, USA, Europe, Japan) began to design the next (fourth) generation tokamak - the INTOR reactor. At this stage, the task was to look at the bottlenecks of the future installation without creating a complete project. However, by the mid-80s it became clear that a more complete task should be set, including the creation of a project. At the suggestion of E.P. Velikhov, after lengthy negotiations at the level of the leaders of states (M.S. Gorbachev and R. Reagan), an Agreement was signed in 1988 and work began on the project of the ITER tokamak reactor. The work was carried out in three stages with interruptions and took a total of 13 years. The diplomatic history of the ITER project itself is dramatic, has led to dead ends more than once and deserves a separate description (see, for example, a book). Formally, the project was completed in July 2000, but it was still necessary to select a site for construction and develop the Construction Agreement and the ITER Charter. All together it took almost 6 years, and finally, in November 2006, the Agreement on the construction of ITER in Southern France was signed. The construction itself is expected to take about 10 years. Thus, it will take about 30 years from the start of negotiations to the receipt of the first plasma in the ITER fusion reactor. This is already comparable to the time of a person's active life. These are the realities of progress.
In terms of its linear dimensions, the ITER is approximately twice the size of the JET facility. According to the project, the magnetic field in it \u003d 5.8 Tesla, and the current I \u003d 12-14 MA. It is assumed that the thermonuclear power will reach the value introduced into the plasma for heating, will be about 10.
5. Development of plasma heating means.
In parallel with the growth in the size of the tokamak, the technology of plasma heating devices was developed. Three different heating methods are currently used:
- Ohmic heating of the plasma by the current flowing through it.
- Heating by beams of hot neutral particles of deuterium or tritium.
- Heating by electromagnetic waves in different frequency ranges.
Ohmic heating of plasma in a tokamak is always present, but it is insufficient for heating to thermonuclear temperatures of the order of 10-15 keV (100-150 million degrees). The fact is that as the electrons are heated, the plasma resistance rapidly decreases (inversely proportional), therefore, at a fixed current, the applied power also decreases. As an example, let us point out that in the JET setup with a current of 3-4 MA it is possible to heat the plasma only up to ~ 2 - 3 keV. In this case, the plasma resistance is so small that a current of several million amperes (MA) is maintained by a voltage of 0.1 - 0.2 V.
Hot neutral beam injectors appeared for the first time on the American PLT facility in 1976-77, and since then they have passed a long technological development path. A typical injector now has a particle beam with an energy of 80 - 150 keV and a power of up to 3 - 5 MW. In a large installation, usually up to 10 - 15 injectors of different capacities are installed. The total power of the beams captured by the plasma reaches 25 - 30 MW. This is comparable to the capacity of a small thermal power plant. At ITER, it is planned to install injectors with a particle energy of up to 1 MeV and a total power of up to 50 MW. There are no such beams yet, but intensive development is underway. In the ITER Agreement, Japan assumed responsibility for these developments.
It is now believed that plasma heating by electromagnetic waves is effective in three frequency ranges:
- heating of electrons at their cyclotron frequency f ~ 170 GHz;
- heating of ions and electrons at the ion cyclotron frequency f ~ 100 MHz;
- heating at an intermediate (lower-hybrid) frequency f ~ 5 GHz.
For the last two frequency ranges, there have long been powerful sources of radiation, and the main problem here is to properly match the sources (antennas) with the plasma to reduce the effects of wave reflection. At a number of large installations, due to the high skill of the experimenters, it was possible to introduce up to 10 MW of power into the plasma in this way.
For the first, most high-frequency range, the problem initially consisted in the development of powerful radiation sources with a wavelength of l ~ 2 mm. The pioneer here was the Institute of Applied Physics in Nizhny Novgorod. For half a century of purposeful work, it was possible to create radiation sources (gyrotrons) with a power of up to 1 MW in a stationary mode. These are the devices that will be installed at the ITER. In gyrotrons, technology is brought to the level of art. The resonator, in which the waves are excited by an electron beam, has dimensions of the order of 20 cm, and the required wavelength is 10 times less. Therefore, it is required to resonantly invest up to 95% of the power into one and very high spatial harmonic, and no more than 5% into all the others. In one of the ITER gyrotrons, a harmonic with numbers (number of nodes) along the radius \u003d 25 and along the angle \u003d 10 is used as such a selected harmonic. To output radiation from the gyrotron, a polycrystalline diamond disk 1.85 mm thick and 106 mm in diameter is used as a window. Thus, to solve the problem of plasma heating, it was necessary to develop the production of giant artificial diamonds.
6. Diagnostics
At a plasma temperature of 100 million degrees, no measuring device can be inserted into the plasma. It will evaporate without having time to transmit reasonable information. Therefore, all measurements are indirect. Currents, fields and particles outside the plasma are measured, and then, using mathematical models, the recorded signals are interpreted.
What is actually being measured?
First of all, these are currents and voltages in the circuits surrounding the plasma. Local probes measure electric and magnetic fields outside the plasma. The number of such probes can be up to several hundred. From these measurements, solving inverse problems, it is possible to restore the shape of the plasma, its position in the chamber, and the magnitude of the current.
Both active and passive methods are used to measure plasma temperature and density. An active method is understood as a method when any radiation (for example, a laser beam or a beam of neutral particles) is injected into the plasma, and the scattered radiation is measured, which carries information about the plasma parameters. One of the difficulties of the problem is that, as a rule, only a small fraction of the injected radiation is scattered. So when using a laser to measure the temperature and density of electrons, only 10 -10 of the energy of the laser pulse is scattered. When a neutral beam is used to measure the ion temperature, the intensity, shape, and position of optical lines appearing when plasma ions are recharged at the neutrals of the beam are measured. The intensity of these lines is very low and high-sensitivity spectrometers are required to analyze their shape.
Passive methods are defined as methods that measure radiation constantly emanating from a plasma. In this case, electromagnetic radiation is measured in different frequency ranges or fluxes and spectra of outgoing neutral particles. This includes hard and soft X-ray, ultraviolet, optical, infrared and radio measurements. Measurements of spectra as well as positions and shapes of individual lines are interesting. The number of spatial channels in some diagnostics reaches several hundred. The signal registration frequency reaches several MHz. Each self-respecting installation has a set of 25-30 diagnostics. At the ITER tokamak reactor, only at the initial stage, it is supposed to have several dozen passive and active diagnostics.
7. Plasma mathematical models
The problems of mathematical modeling of plasma can be roughly divided into two groups. The first group includes the problems of interpretation of the experiment. They are, as a rule, incorrect and require the development of regularization methods. Here are some examples of tasks for this group.
- Reconstruction of the plasma boundary from magnetic (probe) measurements of fields outside the plasma. This problem leads to Fredholm integral equations of the first kind or to strongly degenerate linear algebraic systems.
- Processing chord measurements. Here we come to integral equations of the first kind of mixed Volterra-Fredholm type.
- Processing of measurements of spectral lines. Here it is required to take into account the instrumental functions, and we again come to the Fredholm integral equations of the first kind.
- Processing of noisy time signals. It uses various spectral expansions (Fourier, wave-years), calculations of correlations of various orders.
- Analysis of particle spectra. Here we are dealing with nonlinear integral equations of the first kind.
The following figures illustrate some of the above examples. Figure 4 shows the temporal behavior of soft X-ray signals at the MAST facility (England), measured along chords with collimated detectors.
The installed diagnostics register over 100 such signals. Sharp peaks on the curves correspond to rapid internal movements (“breakdowns”) of the plasma. The two-dimensional structure of such movements can be found using tomographic processing of a large number of signals.
Figure 5 shows the spatial distribution of the electron pressure for two pulses from the same MAST setup.
The spectra of the scattered radiation of the laser beam are measured at 300 points along the radius. Each point in Fig. 5 is the result of a complex processing of the energy spectrum of the photons recorded by the detectors. Since only a small part of the laser beam energy is scattered, the number of photons in the spectrum is small and the restoration of the temperature from the spectrum width turns out to be an incorrect problem.
The second group includes the actual problems of modeling the processes occurring in plasma. Hot plasma in a tokamak has a large number of characteristic times, the extreme of which differ by 12 orders of magnitude. Therefore, the expectation that models can be created containing "all" processes in the plasma are vain. We have to use models that are valid only in a fairly narrow band of characteristic times.
The main models include:
- Plasma gyrokinetic description.Here, the unknown is the ion distribution function, which depends on six variables: three spatial coordinates in toroidal geometry, longitudinal and transverse velocities, and time. Averaging methods are used to describe electrons in such models. To solve this problem, giant codes have been developed in a number of foreign centers. Calculation on them takes a lot of time on supercomputers. There are no such codes in Russia now, in the rest of the world there are about a dozen of them. Currently, gyrokinetic codes describe plasma processes in the time range of 10 -5 -10 -2 sec. This includes the development of instabilities and the behavior of plasma turbulence. Unfortunately, these codes do not yet provide a reasonable picture of plasma transport. Comparison of the results of calculations with experiment is still at the initial stage.
- Magnetohydrodynamic (MHD) description of plasma.In this area, a number of centers have created codes for linearized 3D models. They are used to study plasma stability. As a rule, the boundaries of instabilities in the parameter space and increment values \u200b\u200bare sought. Nonlinear codes are developing in parallel.
Note that over the past 2 decades, the attitude of physicists to plasma instabilities has changed markedly. In the 1950s and 1960s, plasma instabilities were discovered "almost every day." But over time, it became clear that only some of them lead to partial or complete destruction of the plasma, and the rest only increase (or do not increase) the transfer of energy and particles. The most dangerous instability leading to complete destruction of the plasma is called "stall instability" or simply "stall". It is nonlinear and develops when more elementary linear MHD modes associated with individual resonant surfaces intersect in space and, thereby, destroy magnetic surfaces. Attempts to describe the stripping process have led to the creation of nonlinear codes. Unfortunately, so far none of them is able to describe the picture of plasma destruction.
In today's plasma experiments, in addition to the quench instability, a small number of instabilities are considered dangerous. Here we will name only two of them. This is the so-called RWM mode, associated with the finite conductivity of the chamber walls and the attenuation of the plasma stabilizing currents in it, and the NTM mode associated with the formation of magnetic islands on resonant magnetic surfaces. To date, several three-dimensional MHD codes in toroidal geometry have been created to study these types of perturbations. An active search is under way for methods to suppress these instabilities, both at an early stage and at the stage of developed turbulence.
- Description of plasma transport, thermal conductivity and diffusion.The classical (based on paired particle collisions) theory of transport in a toroidal plasma was created about forty years ago. This theory has been called "neoclassical". However, already at the end of the 60s, experiments showed that the transfer of energy and particles in plasma is much larger than neoclassical (by 1 - 2 orders of magnitude). For this reason, the usual transport in experimental plasma is called “anomalous”.
Many attempts have been made to describe anomalous transport through the development of turbulent cells in plasma. The usual path taken in the last decade in many laboratories around the world is as follows. It is assumed that the primary cause of the anomalous transport is drift-type instabilities associated with the temperature gradients of ions and electrons or with the presence of trapped particles in the toroidal geometry of the plasma. The calculation results for such codes lead to the following picture. If the temperature gradients exceed a certain critical value, then the developing instability leads to plasma turbulence and a sharp increase in energy fluxes. It is assumed that these fluxes grow in proportion to the distance (in some metric) between the experimental and critical gradients. In the last decade, several transport models have been built along this path to describe energy transfer in tokamak plasma. However, attempts to compare calculations based on these models with experiment do not always lead to success. To describe the experiments, it is necessary to assume that different instabilities play the main role in the transfer in different discharge regimes and at different spatial points of the plasma cross section. As a result, the prediction is not always reliable.
The matter is further complicated by the fact that over the past quarter century many signs of plasma "self-organization" have been discovered. An example of such an effect is shown in Fig. 6 a, b.
Figure 6a shows the plasma density profiles n (r) for two MAST discharges with the same currents and magnetic fields, but with different deuterium gas supply rates to maintain the density. Here r is the distance to the central axis of the torus. It is seen that the density profiles are very different in shape. Figure 6b shows the electron pressure profiles for the same pulses, normalized at the point - the electron temperature profile. It is seen that the “wings” of the pressure profiles are in good agreement. It follows from this that the profiles of the electron temperature are, as it were, "adjusted" to make the pressure profiles the same. But this means that the transfer coefficients are "adjusted", that is, they are not functions of the local plasma parameters. This picture as a whole is called self-organization. The discrepancy between the pressure profiles in the central part is explained by the presence of periodic MHD oscillations in the central zone of the discharge with a higher density. The pressure profiles on the wings are the same despite this nonstationarity.
In our works, it is assumed that the effect of self-organization is determined by the simultaneous action of many instabilities. It is impossible to single out the main instability among them; therefore, the description of the transfer should be associated with some variational principles that are realized in plasma due to dissipative processes. As such a principle, it is proposed to use the principle of minimum magnetic energy proposed by Kadomtsev. This principle makes it possible to single out some special profiles of current and pressure, which are usually called canonical. In transport models, they play the same role as critical gradients. The models built along this path make it possible to reasonably describe the experimental profiles of temperature and plasma density in different operating modes of the tokamak.
8. Path to the future. Hopes and dreams.
For more than half a century of research on hot plasma, a noticeable fraction of the path to a fusion reactor has been covered. At present, the most promising seems to be the use of tokamak installations for this purpose. In parallel, although with a delay of 10-15 years, the direction of stellarators is developing. Which of these installations will ultimately prove to be more suitable for a commercial reactor cannot be said now. This can only be resolved in the future.
Progress in research on CTF since the 1960s is shown in Fig. 7 on a double logarithmic scale.
The field of plasma physics has blossomed out of the desire to bottle up a star. Over the past several decades, this field has grown in countless directions, from astrophysics to space weather and nanotechnology.
As our general understanding of plasma grew, so did our ability to maintain fusion conditions for more than a second. Earlier this year, a new superconducting fusion reactor in China was able to hold a plasma at a temperature of 50 million degrees Celsius for a record 102 seconds. The Wendelstein X-7 Stellarator, which went live in Germany for the first time last fall, is expected to break that record and hold plasma for up to 30 minutes at a time.
The recent NSTX-U update looks modest in comparison to these monsters: the experiment can now hold plasma for five seconds instead of one. But this is also an important milestone.
“Creating a fusion plasma that lasts only five seconds may not seem like a very long process, but in plasma physics, five seconds can be compared to its physics in a stable state,” says Myers, referring to the conditions under which a plasma is stable. The ultimate goal is to achieve a stable state of "burning plasma" that can carry out fusion by itself with a little input of energy from the outside. No experiment has yet achieved this.
The NSTX-U will allow Princeton researchers to bridge some of the gaps between what is now known from plasma physics and what will be needed to create a pilot plant capable of achieving steady state combustion and generating clean electricity.
On the one hand, in order to find the best containment materials, we need to better understand what happens between the fusion plasma and the reactor walls. Princeton is exploring the possibility of replacing its reactor walls (carbon graphite) with a liquid lithium "wall" to reduce long-term corrosion.
On top of that, scientists believe that if fusion will help fight global warming, they need to hurry. The NSTX-U will help physicists decide whether to continue developing the spherical tokamak design. Most tokamak reactors are less like an apple in shape and more like a donut, bagel, torus. The unusual shape of the spherical torus allows for more efficient use of the magnetic field of its coils.
“In the long run, we would like to figure out how to optimize the configuration of one of these machines,” says Martin Greenwald, deputy director of the Center for Plasma Science and Fusion at. "To do this, you need to know how the performance of the machine depends on something that you control, like form."
Myers loathes judging how far we are from commercially feasible fusion energy, and is understandable. After all, decades of unrelenting optimism have seriously damaged the field's reputation and reinforced the notion that synthesis is a pipe dream. With all the funding implications.
It was a major blow to the MIT synthesis program that the feds provided support for the Alcator C-Mid tokamak, which produces one of the most powerful magnetic fields and demonstrates synthesized plasma at the highest pressure. Most of the pending research on NSTX-U will depend on continued federal support, which Myers says comes "in a year."
Everyone has to be careful about spending their research dollars, and some synthesis programs have already gobbled up incredible sums. Take, for example, ITER, a huge superconducting fusion reactor currently under construction in France. When international cooperation began in 2005, it was announced as a $ 5 billion 10 year project. After years of failure, the price tag rose to $ 40 billion. According to the most optimistic estimates, the facility will be completed by 2030.
And where the ITER seems to swell like a tumor until it runs out of resources and kills the host, MIT's stripped-down fusion program shows how you can do it all on a much smaller budget. Last summer, a team of MIT graduate students unveiled plans for ARC, a low-cost fusion reactor that will use new high-temperature superconducting materials to generate the same amount of power as ITER, only with a much smaller device. 
“The challenge with fusion is finding a technical route that makes it economically attractive - and that's what we plan to do soon,” says Greenwald, noting that the ARC concept is currently being pursued as part of the Energy Initiative at MIT. "We believe that if fusion is to play a role in global warming, we need to move faster."
“Fusion promises to be the main source of energy - this is, in fact, our ultimate goal,” says Robert Rosner, a plasmophysicist at the University of Chicago and co-founder of the Energy Policy Institute under him. “At the same time, there is an important question: how much are we willing to spend right now. If we cut funding to the point where the next generation of smart kids won't want to do it at all, we can get out of it altogether. ”
A new technique has been developed to effectively slow down the runaway electrons by introducing "heavy" ions, such as neon or argon, into the reactor.
A functional fusion reactor is still a dream come true, but it could eventually come true thanks to a lot of research and experimentation to unlock an unlimited supply of clean energy. The problems that scientists face in obtaining nuclear fusion are undoubtedly serious and really complex, but everything is surmountable. And it seems that one of the main problems has been solved.
Nuclear fusion is not a process invented by mankind, but a process existing in nature initially, a process that feeds our Sun. Deep inside our home star, hydrogen atoms sit together to form helium, which is the propulsion for the process. Thermonuclear fusion releases a tremendous amount of energy, but requires enormous costs to create extremely high pressures and temperatures that are difficult to reproduce in a controlled manner on Earth.
Last year, researchers at the Massachusetts Institute of Technology brought us closer to fusion by placing plasma under conditions with the right pressure, now, two researchers at Chalmers University have opened another piece of the puzzle.
One of the problems that engineers have faced is runaway electrons. These extremely high-energy electrons can suddenly and unexpectedly accelerate to very high speeds, which can destroy the reactor wall without warning.
Doctoral students Linnaeus Heshlov and Ole Emberose have developed a new technique to effectively slow down these runaway electrons by injecting "heavy" ions such as neon or argon into a reactor. As a result, electrons, colliding with a high charge into the nuclei of these ions, slow down and become much more controllable.
“When we can effectively slow down the runaway electrons, we’ll get one step closer to a functional fusion reactor,” says Linnea Heshlov.
The researchers have created a model that can effectively predict electron energy and behavior. Using Mathematical Plasma Simulation, physicists can now effectively control the electron runaway rate without interrupting the fusion process.
“Many people think it will work, but it’s easier to go to Mars than to achieve a fusion,” says Linnea Heshlov. “You could say that we are trying to collect stars here on earth, and it may take a while. It takes incredibly high temperatures, hotter than the center of the sun, for us to successfully merge here on earth. Therefore, I hope that all this is a matter of time. "
based on materials from newatlas.com, translation