Figure
1.
Layout of CRAFT NNBI facility (i.e., CRANE site).
| Citation: |
Hanlin WANG, Xiaolan ZOU, Xiaojie WANG, Yves PEYSSON, Miaohui LI, Chenbin WU, Chao ZHANG, Fukun LIU, Joan DECKER. Optimization of X-mode electron cyclotron current drive in high-electron-temperature plasma in the EAST tokamak[J]. Plasma Science and Technology, 2023, 25(2): 025105. DOI: 10.1088/2058-6272/ac910e
|
A discharge with electron temperature up to 14 keV has been achieved in EAST. Analysis of the electron cyclotron current drive (ECCD) efficiency at high electron temperature under EAST parameters is presented using C3PO/LUKE code. Simulation results show that the ECCD efficiency of X-mode increases with central electron temperature up to 10 keV and then starts to decrease above 10 keV, at a specific magnetic field and toroidal angle. The efficiency degradation is due to the presence of the third harmonic extraordinary (X3) downshifted absorption at the low field side (LFS); even the cold resonance of X3 mode is located outside the plasma. As the electron temperature increases from 5 to 20 keV, the X3 absorption increases from 0.9% to 96.4%. The trapping electron effect at the LFS produces a reverse Ohkawa current. The competition between the Fisch–Boozer current drive and the Ohkawa current drive results in a decrease in ECCD efficiency. ECCD efficiency optimization is achieved through two methods. One is to increase the toroidal angle, leading to X2 mode predominating again over X3 mode and the electron resonance domain of X2 mode moving far from the trapped/passing boundary. The second one is to increase the magnetic field to move away the X3 resonance layer from the plasma, hence less EC power absorbed by X3 mode.
Comprehensive Research Facility for Fusion Technology (CRAFT) is one of national large science and technology infrastructures of China, which aims to explore and master the key technologies, standards and methods for the construction and operation of future fusion reactors (especially for the China Fusion Engineering Test Reactor, CFETR). CRAFT program is divided into two research systems (for the superconducting magnet and the divertor, respectively) and includes 20 test facilities or prototypes totally. Neutral beam injection (NBI) is an effective auxiliary heating method for magnetically confined plasma, which can inject energic deuterium (or its isotopes) atomic beam into the fusion device and deliver the kinetic energy and momentum to the plasma. NBI has made great contributions in the milestones of fusion development [1–8] and is still promising in future fusion reactors for not only plasma heating but also current drive, plasma rotation, burning control, and so on [9–14]. Hence, the research on NBI technology is included in the CRAFT program.
As the size of the fusion device enlarges, the required injecting beam energy also increases to attain the core plasma heating and current drive (e.g., 125 keV for JET [15], 500 keV for JT-60SA [16], 1000 keV for ITER [17]). For the beam energy is larger than 150 keV, the negative ion source based NBI (NNBI) system is more practical than the positive ion source based NBI (PNBI) system, because the neutralization efficiency of negative ion beam can stay above 55% but that is below 30% for positive ion beam [18]. However, according to the research and operation experiences of the only two routine NNBI systems for JT-60U tokamak [19] and LHD heliotron [20] respectively, the NNBI system is more complex and challenging than the PNBI system in views of the negative ion production, negative ion extraction and acceleration, multiply higher acceleration voltage, multiply larger beam size, negative ion neutralization, residual ion removal, and so on. Moreover, the demanded operating parameters of future NNBI system are much higher than the achievements worldwide.
CRAFT NNBI facility, also called CRANE (Comprehensive Research Facility for NNBI Experiment), formally started construction in 2019 and planned to be completed in 2025. The aim of the CRAFT NNBI facility is to explore and master the key technologies for future CFETR NBI system [21, 22]. An acceptance mark for the construction completion of CRAFT NNBI facility is to attain a hydrogen neutral beam with the particle energy of 200–400 keV, beam power of > 2 MW, and pulse duration of > 100 s.
The CRAFT NNBI facility mainly contains two testbeds: HONOR (Hefei Open-facility for Negative-ion Source Research) focuses on the negative ion source research for fusion, CANBE (CFETR Advance Neutral Beam Equipment) is the full-function prototype for the CFETR neutral beam injector. There are other supporting systems, for example, the acceleration voltage power supply, cooling plant, cryogenic plant. The HONOR and CANBE started the first operations in April 2023 and December 2023, respectively.
This paper mainly reports the design and developments that have been carried out to realize the CRAFT NNBI facility, and highlights the experimental results made in their first operations.
The layout of the whole CRAFT NNBI facility is shown in figure 1. HONOR and CANBE are designed to be operated individually. So there are independent control systems, high voltage (HV) platform or tank, assembly areas for the two testbeds. The electrical power system, cooling water plant and cryogenic plant can also support the simultaneous operation of the two testbeds. Considered the construction cost, the acceleration voltage power supply is shared by HONOR and CANBE. There are several research and development units for NNBI relevant technologies, for example, caesium (Cs) dynamics, laser neutralizer, HV insulation in vacuum.
The negative ion source is the most critical and challenging component in an NNBI system. The technology for the negative ions production, extraction, acceleration and steering is still immature especially in the long-pulse and high-power operation. Intuitively, the required parameters of ITER negative ion source (1000 keV, 40 A and 1 h D− beam) have never been reached simultaneously until now. HONOR devotes to tackle the key physics and engineering issues of the large-scale negative ion source for fusion application.
An overall view of HONOR and the HV platform are shown in figure 2. Due to the potential advantages of low maintenance, simple structure and high stability and repeatability in the long-term operation, radio frequency (RF) driven plasma source is more promising for future NBI system and it has been chosen for the ITER and CFETR NNBI systems [23–26]. Two RF generators on the HV platform are used to produce the plasma with the power of 200 kW and the frequency of 1–1.5 MHz for each [27]. Another large power supply on the HV platform is the extraction voltage power supply of 16 kV/80 A for negative ions extraction, which is based on the pulse step modulation (PSM) technology [28–30]. There are some small power supplies assembled in a unit on the HV platform: two bias power supplies of 100 V/600 A for changing the plasma potential and suppressing the co-extracted electrons [31, 32]; one magnetic filter power supply of 15 V/6000 A to produce the filter field across the source for lowering the electron temperature and reducing the loss of negative ions [33, 34]; one starter filament power supply including two voltage regulators for the filament current and filament bias respectively. The high-temperature water supply is also placed on the HV platform, which can provide the high-pressure hot water up to 200 °C for optimizing the Cs surface condition and enhancing the negative ion production. Note that, HONOR can only operate at the acceleration voltage up to 200 kV, so the insulating supports and the isolation transformer are designed to insulate against the 200 kV voltage.
A view of the HONOR in-vessel components is shown in figure 3. The rectangular beamline vacuum vessel is 5 m in length, 4 m in width and 4 m in height [35]. The top plate of the vacuum vessel can be removed for the installation of inner components. The front port of the vacuum vessel is connected to a gate valve with the opening of 2 m in diameter. The major inner components are the beam dump and the cryopumps for the ion beam operation only. The beam dump adopts an ITER heating neutral beam (HNB) like design that consists of two multi-pipe arrays [36–38]. There are two cryopumps of cryo-absorption type installed on the two lateral walls. The cryopump is of modular structure, and each unit is designed to have a pumping speed of 500000 L/s with the pumping opening of 2.3 m (height) × 1.0 m (width) [39–44]. One cryopump contains 4 units (displayed in figure 3) and another one contains 2 units to make room for beam diagnostics. 4 turbo molecular pumps (with pumping speed of 2200 L/s for each) and 2 units of rotary vane pump and Roots pump are used as the backing pumps.
The most important feature of HONOR is the application of various plasma and beam diagnostic systems. Table 1 lists the diagnostic systems which have been equipped already. By installing a special diagnostic flange in front of the plasma grid (PG) in the negative ion source, the key parameters of the plasma (for example, electron density, electron temperature, Cs density, negative ion density) in the extraction region can be measured and analyzed via different diagnostic techniques. By using the carbon fibre composite (CFC) dump downstream the negative ion source, the multi-beamlet footprints can be measured via the infrared (IR) camera, and then, the multi-beamlet divergences and orbits can be evaluated. The other beam diagnostic techniques (e.g., W-wire dump, visible light camera, secondary electron emission) can give the profile, uniformity, divergence of the whole beam. Hence, the plasma and the beam parameters can be comparably investigated in different designs or operation conditions.
| Plasma diagnostics | Objects | Beam diagnostics | Objects |
| Langmuir probe [45] | Electron density, electron temperature, plasma potential | Carbon dump with infra-red camera [47] |
Beamlet divergence, beamlet alignment |
| Optical emission spectroscopy |
Hα, Hβ, Hγ, Cs+, impurities’ line density | Tungsten wire dump with camera [48] |
Beam uniformity, beam divergence |
| Cavity ring-down spectroscopy [46] |
H− line density | Doppler shift spectroscopy [49] |
Beam energy, beam divergence, stripping loss |
| Laser absorption spectroscopy |
Cs0 line density | Visible light camera | Beam uniformity, beam divergence |
| Microwave interferometer |
Electron line density | Secondary electron emission [50] |
Beam uniformity, beam divergence |
| Water flow calorimetry [51] |
Power load on grid |
DownLoad:
CSV
As the CFETR NBI prototype, CANBE can carry out the whole working process of an NNBI injector including the negative ion beam launch, negative ion beam neutralization, residual ions removal, neutral beam transport to the beam dump, except the beam transport in the drift and injection tubes. Significantly, CANBE adopts an electric deflection technique for removal of the powerful residual ions that is the same with the ITER NBI design. CANBE is the first NBI equipment that can test the electric deflection technique ahead of the ITER NBI system and its test facility [52–54].
An overall view of CANBE and the HV tank is shown in figure 4. The components and their design parameters inside the CANBE HV tank are similar to that of HONOR HV platform, including RF generator, extraction voltage power supply, magnetic filter power supply, bias power supply, high-temperature water supply. The difference is that the CANBE HV tank is design to insulate against 400 kV voltage.
A view of the CANBE in-vessel components is shown in figure 5. There are two vacuum vessels connecting via the gate valve. One vacuum vessel is for negative ion source (5×5×5 m3) and the other one is for beamline (12×4×4 m3). It means that a vacuum insulated negative ion source can be operated inside the vacuum vessel for better insulation and lower stripping loss, like ITER negative ion source [55]. The acceptance beam energy of CRAFT NNBI facility is 200–400 keV, which is no necessary to use a vacuum insulated negative ion source. However, according to the operation experience of the ITER neutral beam test facility SPIDER [56–58], several problems arose and powerful negative ion beam was hardly produced during the source operation immersed in the vacuum. So the vacuum vessels for negative ion source are prepared on both HONOR and CANBE for future research on the vacuum insulated negative ion source.
A gas neutralizer with the length of 3 m is used for the neutralization of the negative ion beam [59–61]. The neutralizer is divided into two narrow beam channels to reduce the gas inlet for forming optimal gas target. The maximal cross section of each channel is 1.7 m (height) × 0.3 m (width) which can be changed according to different negative ion source.
An electrical residual ion dump (ERID) is applied to CANBE following the ITER NBI design [62]. Although the previous NBI systems all use the magnetic field to deflect the residual ions onto the dumps, a bending magnet for removing the energic and widespread ions in ITER or CFETR NBI (i.e., 1 MeV D+ or D− with a cross section of 0.6 m in width and 1.5–2 m in height) is enormous and uneconomical. Hence, the ERID concept is more realisable. The ERID of CANBE also has two beam channels (0.31 m in width for each) forming by three ion dumps. The middle dump is connected to the positive HV which absorbs the negative ions; the two lateral dumps are grounded which collect the positive ions. Actually, two identical ERIDs have been manufactured, each of which is 1.5 m in length. One can assembly the two modules together to form the 3 m long channels, or just use one module according to the experiments. Thus, ERID can provide with more convenient experimental plan, as the first facilities to test the electric deflection technique.
The beam dump of CANBE is the same with that of HONOR. The cryopumps are also installed on the two lateral walls in the beamline vacuum vessel of CANBE. Each cryopump comprises 8 units with a total length of 8 m. The positions of each component in the beamline vacuum vessel are shown in figure 6.
The science and technology of negative ion source is developed step by step in the CRAFT project. Four negative ion sources are being or will be tested, as shown in figure 7: the number of RF driver (and also source size) is increased from one, two, to four with 2×2 arrangement or 1×4 arrangement; the accelerator is upgraded from single-stage acceleration of 50 kV, single-stage acceleration of 200 kV, to two-stage acceleration of 400 kV working in air or vacuum. Therefore, the major key issues for future CFETR negative ion source about the multi-driver, large-area, high-current, multi-acceleration and vacuum immersed operation are included in the development progress of those four CRAFT negative ion sources.
The single-driver negative ion source has been put into operation since 2022 on a previous testbed HUNTER and successfully achieved the repeatable long-pulse (> 100 s) extractions of negative ions (H−) with an available current density (> 150 A/m2) [63–66]. The single-driver negative ion source is used to understand and optimize the physical issues on RF plasma discharge and power transfer, plasma transport in magnetic field, Cs dynamics and conditioning, plasma characteristics near the PG [67–74]. The dual-driver negative ion source is the first source operating on the CRAFT NNBI facility and its experimental results are discussed in section 3. Both of the quad-driver negative ion sources are being manufactured, as shown in figure 8 [75–77]. The 2×2 driver source will work in air and can take the ITER half-size source ELISE as reference [78–80], thus it deems to be easier to reach. Besides studying the operation in vacuum, the 1×4 driver source with large aspect ratio is more favourable to the beam neutralization, although its non-uniformity of plasma or beam is probably more difficult to deal with.
A whole system commissioning was carried out in the first operation of the CRAFT NNBI facility, to test and debug the problems when all the sub-systems working together. By using the acceleration voltage power supply, both of HONOR HV platform and CANBE HV tank could hold a high voltage of 200 kV for 100 s.
Until now, both of 2-unit and 4-unit cryopumps have been installed on the HONOR as design; a full-size cryopump with 8 units has been installed on CANBE, but another one just has 2 units due to the deferred delivery. The cryopump with only 2 units in either HONOR or CANBE vacuum vessel was tested to have a pumping speed of > 100000 L/s, as shown in figure 9. The gas was puffed into from the negative ion source. The pumping speed reduced a bit with the increasing gas inflow.
During the commissioning, a high-power and long-pulse RF plasma discharge (140 kW for 110 s) was obtained via the dual-driver negative ion source; a H− beam current of ~ 1 A was successfully extracted for 10 s in the volume mode (i.e., without Cs seeding). The design parameters of the dual-driver negative ion source are listed in table 2.
| Items | Design values |
| RF power | 140 kW |
| RF frequency | 1 MHz (±5% adjustable) |
| RF driver | Φ 24 cm, 16 cm (height) |
| Expansion chamber | 90 (length) × 50 (width) × 20 (depth) cm3 |
| PG aperture | Φ 14 mm, 6×16×4 |
| Extraction voltage | −8 kV |
| Acceleration voltage | 50/200 kV |
| Ion current | 15 A (~ 255 A/m2) |
| Pulse duration | > 100 s |
DownLoad:
CSV
The dual-driver negative ion source of CRAFT is based on the design and experimental results of the single-driver negative ion source. In order to demonstrate the physical designs and operation methods for a RF driven negative ion source, the dual-driver negative ion source is required to replicate the results of single-driver source that is long-pulse negative ions extraction. Hence, the dual-driver negative ion source was equipped with a 50 keV accelerator at first [81]. The experiments were carried out with the 2-unit cryopump only in the HONOR.
A long-pulse extraction of H− has been achieved repeatably and the typical waveforms are shown in figure 10. The RF coils were connected in series between two drivers and the RF power was 50 kW in this shot. The filling pressure was 0.4 Pa, the bias voltage was 14 V, and the PG current for magnetic filter was 1000 A. The extraction voltage was 5.4 kV, the acceleration voltage was 50 kV. The bias voltage was generally applied 3 s leading the beam extraction pulse. Because the bias voltage was already lower than the floating potential, the bias current should be negative, but the unipolar power supply could not display the negative current. Note that, the floating potential was not measured simultaneously during the beam extraction because it requires the high-voltage isolation. But it was measured during the shots of the plasma only.
Here, the current load on the extraction grid (EG), IEG, was regarded as the co-extracted electron current, because almost all the electrons could be deflected onto the EG [82]; the current output of the acceleration voltage power supply, IACC, was regarded as the extracted H− current, although it included part of stripping electrons (assumed to be < 10% of the acceleration current). There were 20 times of breakdown (i.e., the burrs of the acceleration voltage signal, VACC) during the pulse duration of 105 s and the pause before the restart of beam extraction was 200 ms. The breakdowns were all due to the over-current of the high voltage power supply, which were the sparks between two grid electrodes essentially. But the over-current signals were not displayed in the waveforms because the high-speed data acquisition system was not ready [83]. Among them, 17 times were related to the extraction voltage and 3 times were due to the acceleration voltage. Furthermore, the breakdowns were more frequent after 45 s and the reasons were not determined (maybe outgas from the grids by particle bombardment or Cs pollution on the grids). The IEG was gradually increased from 0.8 A to 1.1 A and IACC was increased from 7.3 A (~ 123 A/m2) to 8 A (~ 135 A/m2), which induced an electron-to-ion ratio of < 0.15.
By increasing the RF power and extraction voltage, a high-current extraction of negative ions was also tested. The typical waveforms are shown in figure 11. The RF power was 97 kW, the filling pressure was 0.4 Pa, the bias voltage was 15 V, and the PG current for magnetic filter was 1100 A. The extraction voltage was 8.0 kV, the acceleration voltage was 47 kV. The IACC was almost constant at 14.7 A (~ 248 A/m2) and IEG was below 6.0 A, which induced an electron-to-ion ratio of < 0.41. There were 9 times of breakdowns during the pulse duration of 30 s and all of them were related to the extraction voltage. Note that, only the room-temperature water was used for the PG in these experiments. For Cs seeding, the temperature of the Cs delivery pipe was constant at a value between 220 °C and 260 °C in general, the temperature of the Cs oven was increased in the step of 10 °C when the electron-to-ion ratio was almost unchanged [84].
Since the extraction of negative ions has been verified, the dual-driver negative ion source was upgraded with 200 keV accelerator. The major differences were that the acceleration gap was enlarged from 20 mm to 90 mm and the height of main insulator was increased from 100 mm to 400 mm. The target was to achieve a high-energy and high-power negative ion beam. The waveforms of the highest beam power in the first operation of HONOR are shown in figure 12. The RF power was 69 kW, the filling pressure was 0.4 Pa, the bias voltage was 16 V, and the PG current for magnetic filter was 1100 A. The extraction voltage was 5.9 kV, the acceleration voltage was 130 kV. The IACC was constant at 8.9 A and IEG was 5.0 A. There were 2 times of breakdowns during the pulse duration of 10 s, and the pause before restart of beam extraction was changed to 1 s during the high energy operation. Because the breakdown rate was hardly below 0.2 count/s (breakdown rate = number of breakdown / pulse duration), the long-pulse and high-energy acceleration of negative ions was not attempted.
Until the end of the first operation on HONOR (from July 2023 to November 2023), the acceleration voltage of the dual-driver negative ion source was hardly higher than 140 kV even decreasing extracted ion current. Several creepage paths were found on the outside surface of the insulator between PG and EG. But the reason of the limited acceleration voltage was not sure. After upgrading to 200 keV accelerator, the electron-to-ion ratio of the dual-driver negative ion source showed a general increase in different levels of extracted ion current, even with a higher heating temperature for the Cs injection system. The reasons were not determined. But one guess was that the back-streaming ions with higher energy changed the Cs re-distribution or impurities released from the back plate of the source.
Considered the performance of the dual-driver negative ion source hitting a bottleneck, the first operation of HONOR was closed and the source was installed on CANBE for its first operation. The target was to demonstrate a whole operation process of negative ion based neutral beam injector. The operation value of acceleration voltage was limited to 120 kV to avoid too much breakdowns. The cross section of either beam channel of the neutralizer was changed to 0.6 m × 0.18 m to match the beam profile from the dual-driver negative ion source. However, the structure of ERID was hard to be changed, which caused the width of beam channels of ERID (0.31 m) being larger than that of neutralizer. To ensure the removal of the residual ions, the ERID with 3 m in length was used. As mentioned above, the experiments could be carried out only with a 2-unit cryopump on CANBE. It resulted in the accelerated beam current being hardly higher than ~ 6 A when the ERID was on work, because the HV breakdown on the ERID became frequent.
The typical waveforms of the neutral beam purification with the ERID are shown in figure 13 (displaying three shots: Nos. 2496, 2498 and 2499). The operation parameters of negative ion source were the same in these three shots. The RF power was 42 kW, the filling pressure was 0.4 Pa, the bias voltage was 25 V, and the PG current for magnetic filter was 2000 A. The extraction voltage was 4.5 kV, the acceleration voltage was 105 kV. The IACC was 3.4 A and IEG was 1.5 A. For No. 2496 and No. 2498, the ERID was off work and on work, respectively. For No. 2498 and No. 2499, the ERID was both on work, but the gas input from the neutralizer was 0 and 100 mL/min, respectively.
The applied voltage to ERID VERID was gradually increased to 4.6 kV to ensure that the residual ions were fully deflected onto the ERID. The judgement was that the power load on the beam dump had little increase when raising the VERID. Generally, the ratio of the power loads on the beam dump (i.e., calorimeter) with and without the removal of residual ions (i.e., with and without the VERID here) was used to evaluate the neutralization efficiency [85–88], and then, the average neutralization efficiency was 35.7% without the makeup gas and 43.1% with gas input of 100 mL/min from neutralizer. Note that, the gas input to the negative ion source was constant at 1750 mL/min.
The ERID current IERID (i.e., the output current of the ERID power supply) was already higher than the IACC, which was the solid evidence of the production of secondary plasma inside the ERID. Because the maximal residual ion beam current could not be higher than the initial ion beam current even if there was no neutralization. Moreover, when adding the gas from the neutralizer (shot No. 2499), the IERID was increased but the neutralization efficiency was higher (i.e., less residual ions) comparing to the shot without gas from neutralizer (shot No. 2498). Therefore, the increased IERID was due to the secondary plasma. Just because of the increased IERID, the breakdowns became frequent when trying to raise the gas inlet rate from neutralizer.
Although no higher beam energy or beam power and no maximal neutralization efficiency was achieved, the first operation of CANBE (from November 2023 to January 2024) was closed to modify and assemble the large cryopumps.
This paper gives an overview of the CRAFT NNBI facility, called CRANE, and its two testbeds: HONOR for the negative ion source research and CANBE as the CFETR NBI injector prototype. CRANE also hosts some necessary supporting systems for the testbeds, such as high voltage power supply, cryogenic plant, cooling plant. The most outstanding features of HONOR are the flexible assembly and the various diagnostics for source plasma and beam. The CANBE is designed to have a comparable size and full working process to the CFETR NBI injector, especially the unconfirmed technique of the electrical removal of powerful residual ions.
The CRANE building construction was started in 2019 and completed in 2021. The component integration and commissioning of HONOR and CANBE were finished in 2022 and 2023, respectively. The power supply, vacuum pumping, diagnostics and control have been implemented and tested. Both of two testbeds have possessed the experimental ability of negative ion beam acceleration.
The CRAFT dual-driver negative ion source was applied to the first operations of HONOR and CANBE. By using a 50 keV accelerator, the long-pulse and high-current extractions of negative hydrogen ions have been achieved on HONOR, and the typical values were 55.4 keV, 7.3 A (~ 123 A/m2), 105 s and 55.0 keV, 14.7 A (~ 248 A/m2), 30 s, respectively. The results demonstrated the physical design of the dual-driver negative ion source which was transplanted from the CRAFT single-driver negative ion source. By replacing with a 200 keV accelerator, the MW-class negative ion acceleration has been achieved on HONOR, and the typical values were 135.9 keV, 8.9 A (~ 150 A/m2), 8 s. However, the acceleration voltage could not raise above 140 kV.
By installing dual-driver negative ion source on CANBE, a whole operation process of negative ion production, neutralization and purification has been demonstrated, and the typical values were 109.5 keV, 3.4 A, 8 s with a neutralization efficiency of 43.1%. The electrical removal of powerful residual ions was applied to an NNBI injector for the first time. However, a problem of frequent HV breakdowns on the ERID was exposed when the large cryopump was not fully on work.
According to the results of the first operation of CRANE and the acceptance target of the project (200–400 keV, 2 MW, 100 s, H0), several tasks will be carried out in the next step. The dual-driver negative ion source needs to resolve the HV breakdown problem on the accelerator and has to attain a long-pulse and MW-class negative ion beam. The Cs pollution, beam optics, stray charged particles could be the causes of the HV breakdown or spark down. In future, we will detect the reason and further improve the beam parameters. Two different quad-driver negative ion sources will be assembled and tested. For HONOR, several diagnostic systems need to support the research on the plasma and beam uniformity in the next campaign. For CANBE, the large cryopumps should be fully on work at first.
One of the authors (Hanlin Wang) would like to acknowledge the helpful suggestions by Dr G. Giruzzi. This work is supported by the National Key R & D Program of China (Nos. 2017YFE0300500 and 2017YFE0300503) and the Comprehensive Research Facility for Fusion Technology Program of China (No. 2018-000052-73-01-001228).
| [1] |
Bornatici M et al 1983 Nucl. Fusion 23 1153 doi: 10.1088/0029-5515/23/9/005
|
| [2] |
Fidone I et al 1986 Phys. Fluids 29 803 doi: 10.1063/1.865937
|
| [3] |
Erckmann V and Gasparino U 1994 Plasma Phys. Control. Fusion 36 1869 doi: 10.1088/0741-3335/36/12/001
|
| [4] |
Prater R 2004 Phys. Plasmas 11 2349 doi: 10.1063/1.1690762
|
| [5] |
Fisch N J 1978 Phys. Rev. Lett. 41 873 doi: 10.1103/PhysRevLett.41.873
|
| [6] |
Fisch N J and Boozer A H 1980 Phys. Rev. Lett. 45 720 doi: 10.1103/PhysRevLett.45.720
|
| [7] |
Fisch N J 1981 Phys. Rev. A 24 3245 doi: 10.1103/PhysRevA.24.3245
|
| [8] |
Fisch N J 1987 Rev. Mod. Phys. 59 175 doi: 10.1103/RevModPhys.59.175
|
| [9] |
Riviere A C 1986 Plasma Phys. Control. Fusion 28 1263 doi: 10.1088/0741-3335/28/9A/006
|
| [10] |
Prater R 1990 J. Fusion Energy 9 19 doi: 10.1007/BF01057320
|
| [11] |
Lloyd B 1998 Plasma Phys. Control. Fusion 40 A119 doi: 10.1088/0741-3335/40/8A/010
|
| [12] |
Kirneva N A 2001 Plasma Phys. Control. Fusion 43 A195 doi: 10.1088/0741-3335/43/12A/314
|
| [13] |
Zou X L et al 2003 Nucl. Fusion 43 1411 doi: 10.1088/0029-5515/43/11/013
|
| [14] |
Mantica P et al 2000 Phys. Rev. Lett. 85 4534 doi: 10.1103/PhysRevLett.85.4534
|
| [15] |
Song S D et al 2012 Nucl. Fusion 52 033006 doi: 10.1088/0029-5515/52/3/033006
|
| [16] |
Ryter F et al 2001 Phys. Rev. Lett. 86 5498 doi: 10.1103/PhysRevLett.86.5498
|
| [17] |
Cui Z Y et al 2013 Nucl. Fusion 53 093001 doi: 10.1088/0029-5515/53/9/093001
|
| [18] |
Seol J et al 2010 Nucl. Fusion 50 105008 doi: 10.1088/0029-5515/50/10/105008
|
| [19] |
Yoshida M et al 2009 Phys. Rev. Lett. 103 065003 doi: 10.1103/PhysRevLett.103.065003
|
| [20] |
Giruzzi G et al 2004 Phys. Rev. Lett. 93 255002 doi: 10.1103/PhysRevLett.93.255002
|
| [21] |
Harvey R W et al 1997 Nucl. Fusion 37 69 doi: 10.1088/0029-5515/37/1/I06
|
| [22] |
Zohm H et al 2013 Nucl. Fusion 53 073019 doi: 10.1088/0029-5515/53/7/073019
|
| [23] |
Wenninger R et al 2015 Nucl. Fusion 55 063003 doi: 10.1088/0029-5515/55/6/063003
|
| [24] |
Angioni C et al 2003 Nucl. Fusion 43 455 doi: 10.1088/0029-5515/43/6/308
|
| [25] |
Nowak S et al 2014 Nucl. Fusion 54 033003 doi: 10.1088/0029-5515/54/3/033003
|
| [26] |
Yuan Y et al 2016 Phys. Plasmas 23 062503 doi: 10.1063/1.4953605
|
| [27] |
Humphreys D A et al 2006 Phys. Plasmas 13 056113 doi: 10.1063/1.2173606
|
| [28] |
Lazaros A, Maraschek M and Zohm H 2007 Phys. Plasmas 14 042505 doi: 10.1063/1.2720372
|
| [29] |
Wang X G et al 2015 Phys. Plasmas 22 022512 doi: 10.1063/1.4913352
|
| [30] |
ITER Physics Basis Editors et al 1999 Nucl. Fusion 39 2137 doi: 10.1088/0029-5515/39/12/301
|
| [31] |
Leuterer F et al 2001 Fusion Eng. Des. 53 485 doi: 10.1016/S0920-3796(00)00524-X
|
| [32] |
Ikeda Y et al 2001 Fusion Eng. Des. 53 351 doi: 10.1016/S0920-3796(00)00511-1
|
| [33] |
Wang X J et al 2015 Fusion Eng. Des. 96–97 181 doi: 10.1016/j.fusengdes.2015.03.042
|
| [34] |
Wan Y X et al 2017 Nucl. Fusion 57 102009 doi: 10.1088/1741-4326/aa686a
|
| [35] |
Xu L Q et al 2020 Nucl. Fusion 60 106027 doi: 10.1088/1741-4326/abadab
|
| [36] |
Smith G R, Cohen R H and Mau T K 1987 Phys. Fluids 30 3633 doi: 10.1063/1.866448
|
| [37] |
Farina D and Figini L 2010 Nucl. Fusion 50 095007 doi: 10.1088/0029-5515/50/9/095007
|
| [38] |
Poli E et al 2013 Nucl. Fusion 53 013011 doi: 10.1088/0029-5515/53/1/013011
|
| [39] |
Gong X Z et al 2017 Plasma Sci. Technol. 19 032001 doi: 10.1088/2058-6272/19/3/032001
|
| [40] |
Wei W et al 2014 Chin. Phys. B 23 055201 doi: 10.1088/1674-1056/23/5/055201
|
| [41] |
Li J C et al 2016 Chin. Phys. B 25 045201 doi: 10.1088/1674-1056/25/4/045201
|
| [42] |
Peysson Y and Decker J 2008 AIP Conf. Proc. 1069 176 doi: 10.1063/1.3033701
|
| [43] |
Peysson Y et al 2011 Plasma Phys. Control. Fusion 53 124028 doi: 10.1088/0741-3335/53/12/124028
|
| [44] |
Peysson Y, Decker J and Morini L 2012 Plasma Phys. Control. Fusion 54 045003 doi: 10.1088/0741-3335/54/4/045003
|
| [45] |
Peysson Y and Decker J 2014 Fusion Sci. Technol. 65 22 doi: 10.13182/FST13-643
|
| [46] |
Brambilla M 1998 Kinetic Theory of Plasma Waves (Oxford: Clarendon Press)
|
| [47] |
Ohkawa T 1976 Steady-state operation of tokamaks by r-f heating GA-A13847 San Diego: General Atomics (https://library.psfc.mit.edu/catalog/online_pubs/tech_reports/GA-A13847.pdf)
|
| [48] |
Decker J 2003 AIP Conf. Proc. 694 447 doi: 10.1063/1.1638075
|
| [49] |
Giruzzi G 1987 Nucl. Fusion 27 1934 doi: 10.1088/0029-5515/27/11/019
|
| [50] |
Li J C et al 2015 Phys. Plasmas 22 062512 doi: 10.1063/1.4923303
|
| [1] | Imran Ali KHAN, G MURTAZA. Effect of kappa distribution on the damping rate of the obliquely propagating magnetosonic mode[J]. Plasma Science and Technology, 2018, 20(3): 35302-035302. DOI: 10.1088/2058-6272/aaa457 |
| [2] | Wei ZHONG (钟伟), AoXU (徐翱), Yunlong LIU (刘云龙), Lei CHEN (陈磊). Visualization of particulates distribution from electrode erosion[J]. Plasma Science and Technology, 2018, 20(2): 25502-025502. DOI: 10.1088/2058-6272/aa9327 |
| [3] | Xifeng CAO (曹希峰), Guanrong HANG (杭观荣), Hui LIU (刘辉), Yingchao MENG (孟颖超), Xiaoming LUO (罗晓明), Daren YU (于达仁). Hybrid–PIC simulation of sputtering product distribution in a Hall thruster[J]. Plasma Science and Technology, 2017, 19(10): 105501. DOI: 10.1088/2058-6272/aa7940 |
| [4] | Wenjun YANG (杨文军), Guoqiang LI (李国强), Yanqiang HU (胡砚强), Songlin LIU (刘松林), Hang LI (李航), Xiang GAO (高翔). Effects of kinetic profiles on neutron wall loading distribution in CFETR[J]. Plasma Science and Technology, 2017, 19(8): 85102-085102. DOI: 10.1088/2058-6272/aa6795 |
| [5] | Y WANG (王宇), G ZHAO (赵高), C NIU (牛晨), Z W LIU (刘忠伟), J T OUYANG (欧阳吉庭), Q CHEN (陈强). Reversal of radial glow distribution in helicon plasma induced by reversed magnetic field[J]. Plasma Science and Technology, 2017, 19(2): 24003-024003. DOI: 10.1088/2058-6272/19/2/024003 |
| [6] | CHENG Anyi (成安义), ZHANG Qiyong (张启勇), FU Bao (付豹), LU Xiaofei (陆小飞). Process Design of Cryogenic Distribution System for CFETR CS Model Coil[J]. Plasma Science and Technology, 2016, 18(2): 202-205. DOI: 10.1088/1009-0630/18/2/18 |
| [7] | XU Xiufeng (徐修峰), LI Shiping (李世平), CAO Hongrui (曹宏睿), XIAO Rui (肖锐), et al.. A Simulation Study on the Flash X-Ray Spectra Spatial Distribution[J]. Plasma Science and Technology, 2013, 15(11): 1165-1168. DOI: 10.1088/1009-0630/15/11/16 |
| [8] | CHENG Xian (程显), DUAN Xiongying (段雄英), LIAO Minfu (廖敏夫), et al.. The Voltage Distribution Characteristics of a Hybrid Circuit Breaker During High Current Interruption[J]. Plasma Science and Technology, 2013, 15(8): 800-806. DOI: 10.1088/1009-0630/15/8/16 |
| [9] | WU Xingquan (伍兴权), LUO Jiarong (罗家融), WU Bin (吴斌), WANG Jinfang (王进芳), HU Chundong (胡纯栋). Calculation of Shinethrough and its Power Density Distribution for EAST NBI[J]. Plasma Science and Technology, 2013, 15(5): 480-484. DOI: 10.1088/1009-0630/15/5/17 |
| [10] | SONG Yushou(宋玉收), YAN Qiang(颜强), JING Tian(井田), XI Yinyin(席印印), LIU Huilan(刘辉兰). The Distortion of Energy Deposit Distribution of 12C Ions in Water[J]. Plasma Science and Technology, 2012, 14(7): 665-669. DOI: 10.1088/1009-0630/14/7/22 |
Figures(20) / Tables(1)
Supported by: Beijing Renhe Information Technology Co., Ltd. 
| Plasma diagnostics | Objects | Beam diagnostics | Objects |
| Langmuir probe [45] | Electron density, electron temperature, plasma potential | Carbon dump with infra-red camera [47] |
Beamlet divergence, beamlet alignment |
| Optical emission spectroscopy |
Hα, Hβ, Hγ, Cs+, impurities’ line density | Tungsten wire dump with camera [48] |
Beam uniformity, beam divergence |
| Cavity ring-down spectroscopy [46] |
H− line density | Doppler shift spectroscopy [49] |
Beam energy, beam divergence, stripping loss |
| Laser absorption spectroscopy |
Cs0 line density | Visible light camera | Beam uniformity, beam divergence |
| Microwave interferometer |
Electron line density | Secondary electron emission [50] |
Beam uniformity, beam divergence |
| Water flow calorimetry [51] |
Power load on grid |
DownLoad:
CSV
| Items | Design values |
| RF power | 140 kW |
| RF frequency | 1 MHz (±5% adjustable) |
| RF driver | Φ 24 cm, 16 cm (height) |
| Expansion chamber | 90 (length) × 50 (width) × 20 (depth) cm3 |
| PG aperture | Φ 14 mm, 6×16×4 |
| Extraction voltage | −8 kV |
| Acceleration voltage | 50/200 kV |
| Ion current | 15 A (~ 255 A/m2) |
| Pulse duration | > 100 s |
DownLoad:
CSV
