By being confined in a doughnut-type magnetic field container, high-temperature plasma that aims for nuclear fusion power generation does not touch the vacuum vessel. Nevertheless, it is not the case that a high-temperature plasma never touches the vacuum vessel. If the vacuum vessel is made of stainless steel, during plasma experiments irons, which are the constituent element of the stainless steel, will flow in the plasma. Because irons absorb energy from the plasma and that energy is emitted in the form of light, plasma performance will greatly degrade. Thus, many nuclear fusion devices around the world cover the entire surface of the stainless steel vacuum vessel with carbon tiles. Of course, the carbons released from the carbon tiles mix inside the plasma, but because carbon, unlike iron, is not a heavy element we can reduce its influence upon the plasma performance. However, in the Large Helical Device (LHD), despite that the surface of the vacuum vessel chamber is composed almost entirely of stainless steel, deterioration of plasma performance due to iron ions that has been observed in other experimental devices has not been observed at all. Why is that? Solving this mystery is the topic of this Research Update.
The magnetic field lines that confine the doughnut-type plasma are broadly divided into two types. One type is magnetic field lines in which most of the plasma exists and which circulate infinitely around the doughnut. The other type is magnetic field lines that exist in the plasma’s edge area (on the side near the vacuum vessel), and which eventually reach the heat absorption board called the divertor plate that is placed in one area of the vacuum vessel. An important characteristic of the LHD is that the magnetic field lines in a peripheral area twist in a complicated manner, and that the distance that the magnetic field lines traverse to reach the divertor plate is long. The longest magnetic field line is more than one kilometer, and this length is more than ten times the length in other experimental devices. Typically, electrons and ions move along the magnetic field lines in the plasma’s edge region and eventually reach the divertor plate. Reflecting that movement, as a result, from plasma to divertor plate, the density gradient and the temperature gradient of electrons and ions occur along the magnetic field lines. Here, “gradient” expresses how different are the density and the temperature over distance. In general, in the plasma’s edge region, when the density gradient increases, impurities such as iron ions move from the plasma toward the divertor plate. Conversely, when the temperature gradient increases, impurities proceed from the divertor plate toward the plasma. In the LHD, because the length of the twisting magnetic field lines in the edge region is long, the temperature gradient between the plasma and the divertor plate becomes smaller naturally. On the other hand, when we inject hydrogen gas, which is the fuel of the plasma, because the length for gas to ionize is short, the hydrogen gas becomes hydrogen ions while moving through the edge region, and density increases there. Thus, the density gradient increases along the field lines, and impurities that had been emitted from the vacuum vessel chamber’s surface and the divertor plate and then moved to the plasma side change direction, and are pushed back to the divertor plate. Moreover, when the density is raised further impurities increasingly tend to move toward the divertor plate, and finally impurities stop invading the plasma. This phenomenon is called “impurity shielding.” In the LHD, the impurity shielding effect functions extremely well, and this is because the amount of impurities in the plasma decreases.
Recently in the LHD, we have become able to precisely measure the two-dimensional structure of light strength in the extreme ultraviolet region (the wavelength area between ultraviolet rays and X-rays) emitted by ion impurities. And we have become able to research in even greater detail the movements of impurity ions that exist non-uniformly in the plasma edge. Thus, by comparing those observation results with numerical simulations, it is possible to directly compare and investigate the movement of impurities in the plasma edge with the structure of the twisted magnetic field lines. Further, by slightly changing the structure of the twisting magnetic field lines of the peripheral region by the magnetic field made by the coils placed in the upper part and in the lower part of the LHD, it is becoming increasingly clear that we can augment the impurity shielding more actively. These achievements will become important progress toward a more stable and steady maintenance of LHD plasma. We anticipate still further developments in research.