|
内容記述 |
Beams of hydrogen and deuterium negative ions are used for plasma heating by a neutral beam injection for thermo-nuclear fusion experiments. Volume-type negative ion sources are utilized in experiment of large fusion devices in Japan such as the large helical device (LHD) and JT-60. For development of high efficient H<SUP>-</SUP> ion sources, H<SUP>-</SUP> production and transport mechanism in extraction resion of ion sources should be investigated.<br /> For H<SUP>-</SUP> density and velocity measurement in an ion source plasma, the laser- photodetachment method with Langmuir probe (PD-LP) has been widely used. Since the distributions of H<SUP>-</SUP> density and H<SUP>-</SUP> velocity inside ion sources can be obtained from this method, these parameters are useful to study the negative ion production mechanism and the extraction modeling. However it is difficult to construct an exact model for beam extraction based on the photodetachment method with a PD-LP. Therefore H<SUP>-</SUP> transport mechanism in ion sources has been studied with numerical analysis mainly, because there is no experimental method to measure the relation of H<SUP>-</SUP> ions inside and extracted from ion sources. Another experimental method suitable to study the negative ion transport in an ion source is desirable.<br /><br /> In this thesis, a new diagnostic method, laser aided H<SUP>-</SUP> beam current measurement (PD-FC： photodetachment method with Faraday cup), is developed and utilized to study the H<SUP>-</SUP> transport inside an H<SUP>-</SUP> ion source. This method consists of the followings, (1) A pulsed laser beam is irradiated into the plasma, the photodetachment reaction (H<SUP>-</SUP>＋hv → H ＋e<SUP>-</SUP>) occurs along the laser beam. (2) Steady state transport of H<SUP>-</SUP> is hindered by localized depletion of H<SUP>-</SUP>, and the change in the extracted H<SUP>-</SUP> current is detected by a Faraday cup. The detected change in the H<SUP>-</SUP> current should include information on the H<SUP>-</SUP> density, the velocity, and the transport. It can be also applicable to a strongly magnetized plasma as it directly detects negative ions. In this thesis, the H<SUP>-</SUP> transport near the plasma electrode (PH) inside the plasma is studied using both the conventional method with PD-LP and the new method.<br /><br /> This thesis consists of 7 chapters. The background of this study, principle of measurement, and the extperimental setup for this study are described in chapter 1, 2, and 3, respectively. Chapter 4, 5, and 6 are for the experimental results and discussion. The new diagnostic method is applied to a compact negative ion source, and three experimental studies have been carried out depending on the distance between the PE and the laser beam axis, defined as D, and on whether Cs vapor is added or not.<br /><br /> In chapter 4, the time evolution of H<SUP>-</SUP> beam current was analyzed in the case of D＝0, where the laser beam with semicircular cross section passes the region right front of the beam extraction aperture contacting with PE. The decrease of H<SUP>-</SUP> beam current after photodetachment, △I<SUB>H</SUB>-, becomes larger as the laser power density is increased. It is confirmed that the laser power is large enough to destruct all the H<SUP>-</SUP> in the beam path.<br /> Next, to study H<SUP>-</SUP> transport near the extraction hole, the time evolution of H<SUP>-</SUP> beam current after photodetachment is analyzed on the basis of the ballistic theory. From the recoverv time of H<SUP>-</SUP> beam current, H<SUP>-</SUP> velocity is derived, and then H<SUP>-</SUP> density is estimated from the obtained H<SUP>-</SUP> velocity. The measured H<SUP>-</SUP> velocities obtained with both methods agree well, while H<SUP>-</SUP> density measured with the Faraday cup is larger than that by the Langmuir probe in high gas pressure condition. The reasons for this disagreement are considered to be due to the influence of collision with H<SUB>2</SUB>. molecules in plasma, because ballistic theory does not hold in high gas pressure condition.<br /><br /> In chapter 5, the characteristics of H<SUP>-</SUP> beam current after photodetachment with D>0 are discussed. Here we cannot adopt the ballistic theory used in chapter 4, because in the experimental conditions the shortest mean-free-path and the Larmor radius are both comparable to D. Therefore we must treat collisions of H<SUP>-</SUP> ions with other ions, electrons, atoms, and molecules, and the gyro motions during the travel of H<SUP>-</SUP> ions from the plasma to the extraction hole of PE. This situation is simulated by the Monte-Carlo method. The calculated results are compared with the experimental results, and then we can make sure whether the transport model is valid or not.<br /> The maximum change of H<SUP>-</SUP> current after photodetachmemt (△I<SUB>H</SUB>-<SUB>max</SUB>) decreases gradually as D increases. At D>15mm the signal of △I<SUB>H</SUB>-<SUB>max</SUB> was too small to distinguish it from the plasma noise. In the plasma used here, we found that the H<SUP>-</SUP> ions in the region within 15 mm from the PE contributed for H<SUP>-</SUP> beam. This tendency is in good agreement with the simulation results, and therefore it is concluded that the presernt H<SUP>-</SUP> transport model is reasonable in a compact negative ion source of this size.<br /> It is found that the sheath potential at PE affects the amount and time behavior of PD-FC signal. Because the kinetic energy is smaller than the typically formed negative sheath potential on the wall, H<SUP>-</SUP> is reflected on the surface of PE. Then some part of extracted beam consists of these reflected ions, which is observed as a tail in PD-FC signal wave form. When the sheath potential is reversed by adjusting the PE bias voltage (V<SUB>b</SUB>) against plasma potential (V<SUB>p</SUB>), H<SUP>-</SUP> becomes hitting PE and does not survive. As a result of experiment, the tail was vanished in the PD-FC signal with V<SUB>p</SUB>＜V<SUB>b</SUB> condition. This change of wave form is also demonstrated by Monte-Carlo calculation which takes the destruction of H<SUP>-</SUP> ions on the PE surface into consideration. For high extraction efficiency of H<SUP>-</SUP> ions from ion sources, a PE should be biased lower than plasma potential to prevent H<SUP>-</SUP> ions from being destructed by collision with a PE. <br /><br /> It is widely known that addition of Cs into volume type negative ion sources make increase of extracted H<SUP>-</SUP> beam current. In chapter 6, to clear the cause of this Cs effect, the same experimental method as in chapter 5 is applied for cesium added hydrogen plasma. By introducing Cs vapor, the extracted H<SUP>-</SUP> current increases by factor of 2.5. At D～2 mm, △I<SUB>H</SUB>-<SUB>max</SUB> is also enhanced by 2.5 times. The dependence of △I<SUB>H</SUB>-<SUB>max</SUB>. upon D shows that the increase of H<SUP>-</SUP> near the PE contributes to the enhancement of extracted beam in the case of Cs added plasma, which suggests the surface production of H<SUP>-</SUP> on the grid. <br /><br /> In chapter 7, the conclusions are described throughout the studies on the H<SUP>-</SUP> transport inside the ion source. |
要旨・審査要旨 / Abstract, Screening Result (444.8 kB)
