- Email: [email protected]
Neutralizer experiment of KSTAR NBTS system
Vacuum 84 (2010) 568–572
Contents lists available at ScienceDirect
Vacuum journal homepage: www.elsevier.com/locate/vacuum
Neutralizer experiment of KSTAR NBTS system Tae-Seong Kim a, b, *, Sang-Ryul in b, Doo-Hee Chang b, Dae-Sik Chang b, Beon-Yeol Kim b, Seung-Ho Jeong b, Byung-Hoon Oh b, Byung-Joo Yoon b, Jung-Tae Jin b, Kwng-Won Lee b, Chang-Suk Seo b, Woo-Sub Song b, Jinchoon Kim c a
University of Science and Technology, Republic of Korea Korea Atomic Energy Research Institute, Republic of Korea c ProScience, USA b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 October 2008 Received in revised form 9 February 2009 Accepted 6 June 2009
A general neutralizer experiment was carried out to establish the operational coil current of the bending magnet, the neutralization efficiency of an ion beam with various parameter (beam energy, gas flow rate). Suggested operational coil current which was already studied for the bending magnet, was confirmed through a thermocouple installed at the middle of each ion dump and WFC (water flow calorimetry) system. The calculated optimum coil currents agree with the experimental coil current to penetrate each ion dump. Neutralization efficiency was measured by the WFC system. It depends on the gas flow rate and beam energy and a gas injection more than 800 sccm was needed to attain a equilibrium neutralization. Maximum efficiency was measured at more than 60% at 40 keV and less than 30% at the 80 keV of the hydrogen beam. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Neutral beam Positive ion source Neutralization efficiency Bending magnet Hydrogen beam Accelerator KSTAR NBI NBTS
1. Introduction A neutral beam test-stand system (NBTS) is being developed and tested as an auxiliary heating and current drive system for the Korea Superconducting Tokamak Advanced Research (KSTAR). The design requirements of the long pulse ion source for the KSTAR NBI system were a deuterium ion beam of 120 kV/65A, a beam pulse length of 300 s and an initial test beam of 90 kV/45A for 20 s with hydrogen ions. The prototype beamline components have been developed for an NB total power of 8 MW (4 MW from hydrogen beams), which originates from the deuterium ion beam, and is injected to the core plasmas of KSTAR, with three ion sources in a beamline. The key points of this system are to extract an ion beam from the ion source and to transmit it into the tokamak as much as possible at a high potential (120 kV). The efficiency of the transmitted beam power is a very important factor and the extracted beam from the ion source must be a neutral beam to avoid an electromagnetic
* Corresponding author. E-mail address: [email protected] (T.-S. Kim). 0042-207X/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.vacuum.2009.06.052
effect from the tokamak. Thus, as well as increasing the ion beam currents and power, increasing the neutralization efficiency is important to heat the plasma in the tokamak. Actually the neutralization efficiency has been determined by the cross section area of the beam species particles and the total gas line density for the beam energy, up to now, thus it seems to be sufficient only to study it theoretically, thus most studies to develop this system have focused on the development of a large ion beam current and a high energy. In order to inject a neutral beam power satisfactorily, however, one must confirm the neutralization efficiency during the experiments comparing by the calculation data. In this paper, general neutralizer experiment for the KSTAR NBTS including the optimum operation of the bending magnet and neutralization efficiency at each beam energy (40–80 keV) are discussed. 2. Experimental layouts and operation conditions 2.1. Prototype long pulse ion source (LPIS) Performance of the ion source and neutralization efficiency of energetic ions are the key issues. Among these, the prototype LPIS
T.-S. Kim et al. / Vacuum 84 (2010) 568–572
569
Fig. 1. Assembly drawing of the KSTAR NB prototype ion source.
consists of a magnetic cusp bucket plasma generator and a set of tetrode accelerators with 568 circular copper apertures [1] (Fig. 1). Hydrogen gas is injected into the arc plasma generator and ionized, and then the positive ions are accelerated through a potential gradient to form an ion beam. For a beam energy (40–80keV) grid potentials of gradient, suppressor and ground grids are 81% of full energy, 2 kV and 0 kV. A negative potential is applied to the
suppressor grid to keep the electrons, produced outside the accelerator section, from entering it where they would be accelerated into the plasma generator. The arc discharge of the LPIS was operated up to an arc power of 60 kW (about 90 V/650 A) depending on the optimum beam optics. Hydrogen gas was injected at 160–560 sccm into the arc plasma generator through a system of mass-flow controllers (MFCs). All of the components
Fig. 2. Layout of the KSTAR NBTS system.
570
T.-S. Kim et al. / Vacuum 84 (2010) 568–572
Fig. 5. The result of the bending magnet scanning through the BTR code.
Fig. 3. Calculated Magnet working current.
including the plasma generator and accelerator were cooled by using a water supply system with a water pressure up to 10 bar. 2.2. Neutral beam test-stand system (NBTS) KSTAR NBTS system is equipped with a 60 m3 vacuum chamber, an ion source (plasma generator and accelerator), and one set of beamline components including a source exit scraper/optical multi-channel analyzer (OMA) chamber, a neutralizer, a bending magnet, ion dumps, a calorimeter, and 4 105 L/s cryopumps. Two cryopumps were used for this experiment (Fig. 2). A multispecies þ þ hydrogen beam (i.e., Hþ 1 , H2 , and H3 ) which was generated from an ion source undergoes atomic collision reactions in the neutralizer and OMA chamber (also neutralizer), and then it is separated electromagnetically into an ion beam and a neutral beam by the bending magnet component. The energetic neutral beam travels further downstream to the calorimeter, while the ion beam is dumped into the ion dumps. If this NBTS is installed in the KSTAR, the neutral beam will be injected into the tokamak without a calorimeter. Additional gases into the middle of the neutralizer cell for an effective neutralization were injected at 100–1200 sccm. The limit of the gas flow rate at the plasma generator and neutralizer cell was
Fig. 4. The result of the bending magnet scanning through the T/C at 60 keV.
determined by a stable beam extraction. To obtain the pressure data during the experiment, a baratron (CDG), ion and cold cathode gauge were installed at the OMA chamber, the neutralizer cell and main vacuum chamber. The beamline components are also actively cooled by using a water cooling system of 2 MW. Typical T-type (beamline) and K-type (accelerator) thermocouples were installed on the outlets of the accelerator and the beamline components for a measurement of the cooling water temperature, and the output signals from the thermocouples were collected to the PXI (PCI eXtensions for Instruments) DAQ (Data Acquisition) system through an isolation amplifier and an optical signal transmission [2]. By using this system, Water-Flow Calorimetry (WFC) calculation was possible. The WFC is typically used for measuring a extracted beam power [2]. By monitoring the temperature rise of the cooling water, one can estimate the total deposited energy as
Q ¼
ZN
mcp DTðtÞdt
0
where m is the flow rate of the water, it was assumed as constant during the experiment, cp is the specific heat of the water (4.2 kJ/ kg C at room temperature), DT is the time-dependent water temperature difference.
Fig. 6. The result of the bending magnet scanning through the WFC system.
T.-S. Kim et al. / Vacuum 84 (2010) 568–572
571
Table 1 Measured neutralization efficiency in the various beam energy. NBTS components
40 keV
50 keV
60 keV
70 keV
80 keV
Calorimeter (%) Full ion dump (%) Half ion dump (%) Third ion dump (%) Bending magnet (%) Neutralizer (%) OMA chamber (%)
46.05 21.78 0.86 0.83 2.45 9.45 7.41
41.69 30.72 0.88 1.49 1.99 8.88 7.57
33.68 36.09 1.34 1.41 1.81 8.80 8.21
27.81 43.95 1.50 1.65 1.75 9.05 8.31
20.43 51.43 1.51 1.66 1.59 8.79 8.26
Total (%)
88.88
93.24
91.37
94.04
93.69
3. Experiment result and analyses 3.1. Optimum operation of bending magnet A fraction of the ion beam from the ion source is changed into a neutral beam by passing it through the neutralizer. There is a maximum attainable neutralizing efficiency depending on the ion species and the beam energy, regardless of how thick the gas target is. [3] The ion beam unconverted in the neutralizer, which can’t penetrate the magnetic barrier of the tokamak, should be removed by the bending magnet (BM) so as not to deposit excessive heat loads to the structures surrounding the beam passage. The ions deflected in the BM are delivered to the ion dump system. The higher the beam energy, the lower the neutral efficiency, and the greater the ion beam power. þ Each ion species (Hþ, Hþ 2 , H3 ) was penetrated into three different ion dumps according to their velocity where the bending degree of the ion beam penetration changes sensitively depending on the operation at coil current of the magnet. Therefore it is necessary to find the optimum coil current which makes all of the ion beam only aside from the neutral beam penetrate to the ion dumps (full energy dump, half energy dump, third energy dump). According to the design of the bending magnet and the ion dump, it is found that the optimum value is to bend of the full energy beam by 60 [4]. Recent research was already studied the optimum operation coil current and calculated it theoretically (Fig. 3). To verify this calculated value, we installed a thermocouple at the middle of each ion dump plate. The experiment was carried out at 60 keV and a 0.5 s short pulse to avoid melting the thermocouple and Fig. 4 shows the result at 360 A of the coil current, we
Fig. 7. Neutralization efficiency as function of the gas line density.
Fig. 8. Measured neutralization efficiency in the several beam energy.
could elucidate the maximum the temperature of thermocouple which was set up at the third ion dump. As the bending degree of third energy ion beam is highly changeable, we could make a decision on the optimum value. There is another way to find the optimum coil current. It is by using the Beam Transport with Reionization (BTR) code which was used as the design basis for the beamline components in the NBTS system and the WFC system. Comparing operation coil current in the BTR code with actual operation coil current, the absolute value is different but we can examine the tendency of the heat load in each dump relatively. Automatically bending the degree of the ion beam is calculated by using the BTR code. Therefore if we can find the tendency at which the beam was bended to 60 , we can confirm that the calculated coil current is reasonable. In the BTR code calculation, the coil current which makes the beam bend to 60 is the point that is just before a rapid increase in the heat load of a half energy ion dump due to an incoming full energy ion beam (Fig. 5). This experiment was carried out carefully due to the possibility of a dangerous leak of the beam to the other components, the result is shown in Fig. 6. At 360 A of the magnet coil current, a similar tendency to the result of the BTR code appeared. Although this way provides a rough
Fig. 9. Measured neutralization efficiency in the various beam energy.
572
T.-S. Kim et al. / Vacuum 84 (2010) 568–572
value, it seems to be a simple method to find safety operation coil current. 3.2. Neutralization efficiency The ion beam extracted from the ion source is to be neutral with an efficiency determined by the balance between the neutralization and ionization, which depends on the beam energy, the interaction of cross-sections and the gas line density (average pressure travel length) experienced by the beam particles. This experiment was carried out at various gas flow rates and several beam energy from 40 keV to 80 keV. Gas line density (pressure integral) along the beam trajectory was calculated by the vacuum network [5,6] and by measuring the pressure at the components. The neutralization efficiency was measured by WFC in the NBTS. Heat load in each component is shown in Table 1 measured by WFC with various beam energies and the total heat load of all components was 88– 94%, which seems to be reasonable by considering the heat load of accelerator and the measuring error.
Neutalization Efficiency ¼
4. Conclusion We have presented an operational coil current for a bending magnet, the neutralization efficiency of an ion beam at various parameters (beam energy, gas flow rate) and the relationship between the effective gas flow rate, the neutralization efficiency and the neutral gas line density. Suggested operational coil current of the bending magnet was proved experimentally through a thermocouple installed at the middle of each ion dump and the WFC system. In the experiment using a thermocouple at a short pulse, we could obtain the maximum temperature difference of the thermocouple at the third energy dump and the tendency of a rapid
Heat loadðcalorimeterÞ Heat loadðcalorimeter þ Full; Half; Third energy dumpÞ
Because of a large error of the measurement in the half and third energy dump and missing the beam power in the other components, the neutralization efficiency was mostly measured as follows.
Neutalization Efficiency ¼
the neutralization efficiency of the beam be attained. Neutralization efficiency was measured to be more than 60% at 40 keV and less than 30% at 80 keV. The efficiency has an error of maximum 20% by the calculation of the pressure and the beam species ratio. In order to increase the accuracy of the data it is desirable correct calculation of the pressure and the beam species ratio.
Heat load of calorimeterðBM turn onÞ Heat load of calorimeterðBM turn off Þ
The result of measured neutralization efficiency is shown in Fig. 7–9. Neutralization efficiency was measured to be more than 60% at 40 kV and less than 30% at 80 keV. Gas line density needs to be more than 1.0 1016/cm2 to attain a 95% equilibrium neutralization [7] and additional gas has to be injected at more than 800 sccm to satisfy it. Measured efficiency was higher than the calculated data at the low beam energy (40–60 keV). It seems to be caused by different beam species at different beam energies aside from with larger measurement error in the WFC system at a low beam energy. To maintain the optimum optics of the beam (for a stable beam extraction), the arc discharge power was set from 30 kW to 60 kW according to the beam energy. The beam species ratio is varied by the arc discharge power. The lower injected the arc power was, the larger þ generated the molecule beam species (i.e., Hþ 2 , and H3 ) ratio was. Because lower energy particles from molecule beam species have higher cross section area than fast particle from atomic beam species,
increase of the temperature was about to start in the half energy dump. The calculated optimum coil currents agree with experimental coil current to penetrate each ion dump as we expected. Neutralization efficiency was measured by the WFC system. It depends on the gas flow rate and the beam energy, and a gas injection more than 800 sccm was needed to attain an equilibrium neutralization. Maximum efficiency was measured to be more than 60% at 40 keV and less than 30% at 80 keV of the hydrogen beam. Since it closely depends on the beam species ratio and the gas line density, if the species and pressure at the beamline component can be corrected, the accuracy of the experimental data would be increased. References [1] Oh Byung-Hoon, Chang Doo-Hee, Ho Jeong Seung, Lee Kwang-Won, In SangRyul, Yoon Byung-Joo, et al. Rev Sci Instrum 2008;79:02C101. [2] Chang Doo-Hee, Seo Min-Seok, oh Byung-Hoon, Kim Jinchoon, Kim Tae-Seong. J Korean Phys Soc 2006;50(1). [3] In SR, Shim HJ. Neutral beam evolution in the KSTAR NBI test stand. J Korean Vac Sci Technol 2003;7(1):1. [4] IN SR, Yoon BJ, Kim BY, Kim TS. J Korean Phys Soc 2006;49:S320–3. [5] Hellborg Rangar, Kiisk Madis, Person Per, Stenstorom Kristina. Vacuum 2005;78:427. [6] In SR. J Korean Vacuum Soc 1999;8. [7] Kim J, Haselton HH. J Appl Phys June 1979;50(6).
Contents lists available at ScienceDirect
Vacuum journal homepage: www.elsevier.com/locate/vacuum
Neutralizer experiment of KSTAR NBTS system Tae-Seong Kim a, b, *, Sang-Ryul in b, Doo-Hee Chang b, Dae-Sik Chang b, Beon-Yeol Kim b, Seung-Ho Jeong b, Byung-Hoon Oh b, Byung-Joo Yoon b, Jung-Tae Jin b, Kwng-Won Lee b, Chang-Suk Seo b, Woo-Sub Song b, Jinchoon Kim c a
University of Science and Technology, Republic of Korea Korea Atomic Energy Research Institute, Republic of Korea c ProScience, USA b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 October 2008 Received in revised form 9 February 2009 Accepted 6 June 2009
A general neutralizer experiment was carried out to establish the operational coil current of the bending magnet, the neutralization efficiency of an ion beam with various parameter (beam energy, gas flow rate). Suggested operational coil current which was already studied for the bending magnet, was confirmed through a thermocouple installed at the middle of each ion dump and WFC (water flow calorimetry) system. The calculated optimum coil currents agree with the experimental coil current to penetrate each ion dump. Neutralization efficiency was measured by the WFC system. It depends on the gas flow rate and beam energy and a gas injection more than 800 sccm was needed to attain a equilibrium neutralization. Maximum efficiency was measured at more than 60% at 40 keV and less than 30% at the 80 keV of the hydrogen beam. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Neutral beam Positive ion source Neutralization efficiency Bending magnet Hydrogen beam Accelerator KSTAR NBI NBTS
1. Introduction A neutral beam test-stand system (NBTS) is being developed and tested as an auxiliary heating and current drive system for the Korea Superconducting Tokamak Advanced Research (KSTAR). The design requirements of the long pulse ion source for the KSTAR NBI system were a deuterium ion beam of 120 kV/65A, a beam pulse length of 300 s and an initial test beam of 90 kV/45A for 20 s with hydrogen ions. The prototype beamline components have been developed for an NB total power of 8 MW (4 MW from hydrogen beams), which originates from the deuterium ion beam, and is injected to the core plasmas of KSTAR, with three ion sources in a beamline. The key points of this system are to extract an ion beam from the ion source and to transmit it into the tokamak as much as possible at a high potential (120 kV). The efficiency of the transmitted beam power is a very important factor and the extracted beam from the ion source must be a neutral beam to avoid an electromagnetic
* Corresponding author. E-mail address: [email protected] (T.-S. Kim). 0042-207X/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.vacuum.2009.06.052
effect from the tokamak. Thus, as well as increasing the ion beam currents and power, increasing the neutralization efficiency is important to heat the plasma in the tokamak. Actually the neutralization efficiency has been determined by the cross section area of the beam species particles and the total gas line density for the beam energy, up to now, thus it seems to be sufficient only to study it theoretically, thus most studies to develop this system have focused on the development of a large ion beam current and a high energy. In order to inject a neutral beam power satisfactorily, however, one must confirm the neutralization efficiency during the experiments comparing by the calculation data. In this paper, general neutralizer experiment for the KSTAR NBTS including the optimum operation of the bending magnet and neutralization efficiency at each beam energy (40–80 keV) are discussed. 2. Experimental layouts and operation conditions 2.1. Prototype long pulse ion source (LPIS) Performance of the ion source and neutralization efficiency of energetic ions are the key issues. Among these, the prototype LPIS
T.-S. Kim et al. / Vacuum 84 (2010) 568–572
569
Fig. 1. Assembly drawing of the KSTAR NB prototype ion source.
consists of a magnetic cusp bucket plasma generator and a set of tetrode accelerators with 568 circular copper apertures [1] (Fig. 1). Hydrogen gas is injected into the arc plasma generator and ionized, and then the positive ions are accelerated through a potential gradient to form an ion beam. For a beam energy (40–80keV) grid potentials of gradient, suppressor and ground grids are 81% of full energy, 2 kV and 0 kV. A negative potential is applied to the
suppressor grid to keep the electrons, produced outside the accelerator section, from entering it where they would be accelerated into the plasma generator. The arc discharge of the LPIS was operated up to an arc power of 60 kW (about 90 V/650 A) depending on the optimum beam optics. Hydrogen gas was injected at 160–560 sccm into the arc plasma generator through a system of mass-flow controllers (MFCs). All of the components
Fig. 2. Layout of the KSTAR NBTS system.
570
T.-S. Kim et al. / Vacuum 84 (2010) 568–572
Fig. 5. The result of the bending magnet scanning through the BTR code.
Fig. 3. Calculated Magnet working current.
including the plasma generator and accelerator were cooled by using a water supply system with a water pressure up to 10 bar. 2.2. Neutral beam test-stand system (NBTS) KSTAR NBTS system is equipped with a 60 m3 vacuum chamber, an ion source (plasma generator and accelerator), and one set of beamline components including a source exit scraper/optical multi-channel analyzer (OMA) chamber, a neutralizer, a bending magnet, ion dumps, a calorimeter, and 4 105 L/s cryopumps. Two cryopumps were used for this experiment (Fig. 2). A multispecies þ þ hydrogen beam (i.e., Hþ 1 , H2 , and H3 ) which was generated from an ion source undergoes atomic collision reactions in the neutralizer and OMA chamber (also neutralizer), and then it is separated electromagnetically into an ion beam and a neutral beam by the bending magnet component. The energetic neutral beam travels further downstream to the calorimeter, while the ion beam is dumped into the ion dumps. If this NBTS is installed in the KSTAR, the neutral beam will be injected into the tokamak without a calorimeter. Additional gases into the middle of the neutralizer cell for an effective neutralization were injected at 100–1200 sccm. The limit of the gas flow rate at the plasma generator and neutralizer cell was
Fig. 4. The result of the bending magnet scanning through the T/C at 60 keV.
determined by a stable beam extraction. To obtain the pressure data during the experiment, a baratron (CDG), ion and cold cathode gauge were installed at the OMA chamber, the neutralizer cell and main vacuum chamber. The beamline components are also actively cooled by using a water cooling system of 2 MW. Typical T-type (beamline) and K-type (accelerator) thermocouples were installed on the outlets of the accelerator and the beamline components for a measurement of the cooling water temperature, and the output signals from the thermocouples were collected to the PXI (PCI eXtensions for Instruments) DAQ (Data Acquisition) system through an isolation amplifier and an optical signal transmission [2]. By using this system, Water-Flow Calorimetry (WFC) calculation was possible. The WFC is typically used for measuring a extracted beam power [2]. By monitoring the temperature rise of the cooling water, one can estimate the total deposited energy as
Q ¼
ZN
mcp DTðtÞdt
0
where m is the flow rate of the water, it was assumed as constant during the experiment, cp is the specific heat of the water (4.2 kJ/ kg C at room temperature), DT is the time-dependent water temperature difference.
Fig. 6. The result of the bending magnet scanning through the WFC system.
T.-S. Kim et al. / Vacuum 84 (2010) 568–572
571
Table 1 Measured neutralization efficiency in the various beam energy. NBTS components
40 keV
50 keV
60 keV
70 keV
80 keV
Calorimeter (%) Full ion dump (%) Half ion dump (%) Third ion dump (%) Bending magnet (%) Neutralizer (%) OMA chamber (%)
46.05 21.78 0.86 0.83 2.45 9.45 7.41
41.69 30.72 0.88 1.49 1.99 8.88 7.57
33.68 36.09 1.34 1.41 1.81 8.80 8.21
27.81 43.95 1.50 1.65 1.75 9.05 8.31
20.43 51.43 1.51 1.66 1.59 8.79 8.26
Total (%)
88.88
93.24
91.37
94.04
93.69
3. Experiment result and analyses 3.1. Optimum operation of bending magnet A fraction of the ion beam from the ion source is changed into a neutral beam by passing it through the neutralizer. There is a maximum attainable neutralizing efficiency depending on the ion species and the beam energy, regardless of how thick the gas target is. [3] The ion beam unconverted in the neutralizer, which can’t penetrate the magnetic barrier of the tokamak, should be removed by the bending magnet (BM) so as not to deposit excessive heat loads to the structures surrounding the beam passage. The ions deflected in the BM are delivered to the ion dump system. The higher the beam energy, the lower the neutral efficiency, and the greater the ion beam power. þ Each ion species (Hþ, Hþ 2 , H3 ) was penetrated into three different ion dumps according to their velocity where the bending degree of the ion beam penetration changes sensitively depending on the operation at coil current of the magnet. Therefore it is necessary to find the optimum coil current which makes all of the ion beam only aside from the neutral beam penetrate to the ion dumps (full energy dump, half energy dump, third energy dump). According to the design of the bending magnet and the ion dump, it is found that the optimum value is to bend of the full energy beam by 60 [4]. Recent research was already studied the optimum operation coil current and calculated it theoretically (Fig. 3). To verify this calculated value, we installed a thermocouple at the middle of each ion dump plate. The experiment was carried out at 60 keV and a 0.5 s short pulse to avoid melting the thermocouple and Fig. 4 shows the result at 360 A of the coil current, we
Fig. 7. Neutralization efficiency as function of the gas line density.
Fig. 8. Measured neutralization efficiency in the several beam energy.
could elucidate the maximum the temperature of thermocouple which was set up at the third ion dump. As the bending degree of third energy ion beam is highly changeable, we could make a decision on the optimum value. There is another way to find the optimum coil current. It is by using the Beam Transport with Reionization (BTR) code which was used as the design basis for the beamline components in the NBTS system and the WFC system. Comparing operation coil current in the BTR code with actual operation coil current, the absolute value is different but we can examine the tendency of the heat load in each dump relatively. Automatically bending the degree of the ion beam is calculated by using the BTR code. Therefore if we can find the tendency at which the beam was bended to 60 , we can confirm that the calculated coil current is reasonable. In the BTR code calculation, the coil current which makes the beam bend to 60 is the point that is just before a rapid increase in the heat load of a half energy ion dump due to an incoming full energy ion beam (Fig. 5). This experiment was carried out carefully due to the possibility of a dangerous leak of the beam to the other components, the result is shown in Fig. 6. At 360 A of the magnet coil current, a similar tendency to the result of the BTR code appeared. Although this way provides a rough
Fig. 9. Measured neutralization efficiency in the various beam energy.
572
T.-S. Kim et al. / Vacuum 84 (2010) 568–572
value, it seems to be a simple method to find safety operation coil current. 3.2. Neutralization efficiency The ion beam extracted from the ion source is to be neutral with an efficiency determined by the balance between the neutralization and ionization, which depends on the beam energy, the interaction of cross-sections and the gas line density (average pressure travel length) experienced by the beam particles. This experiment was carried out at various gas flow rates and several beam energy from 40 keV to 80 keV. Gas line density (pressure integral) along the beam trajectory was calculated by the vacuum network [5,6] and by measuring the pressure at the components. The neutralization efficiency was measured by WFC in the NBTS. Heat load in each component is shown in Table 1 measured by WFC with various beam energies and the total heat load of all components was 88– 94%, which seems to be reasonable by considering the heat load of accelerator and the measuring error.
Neutalization Efficiency ¼
4. Conclusion We have presented an operational coil current for a bending magnet, the neutralization efficiency of an ion beam at various parameters (beam energy, gas flow rate) and the relationship between the effective gas flow rate, the neutralization efficiency and the neutral gas line density. Suggested operational coil current of the bending magnet was proved experimentally through a thermocouple installed at the middle of each ion dump and the WFC system. In the experiment using a thermocouple at a short pulse, we could obtain the maximum temperature difference of the thermocouple at the third energy dump and the tendency of a rapid
Heat loadðcalorimeterÞ Heat loadðcalorimeter þ Full; Half; Third energy dumpÞ
Because of a large error of the measurement in the half and third energy dump and missing the beam power in the other components, the neutralization efficiency was mostly measured as follows.
Neutalization Efficiency ¼
the neutralization efficiency of the beam be attained. Neutralization efficiency was measured to be more than 60% at 40 keV and less than 30% at 80 keV. The efficiency has an error of maximum 20% by the calculation of the pressure and the beam species ratio. In order to increase the accuracy of the data it is desirable correct calculation of the pressure and the beam species ratio.
Heat load of calorimeterðBM turn onÞ Heat load of calorimeterðBM turn off Þ
The result of measured neutralization efficiency is shown in Fig. 7–9. Neutralization efficiency was measured to be more than 60% at 40 kV and less than 30% at 80 keV. Gas line density needs to be more than 1.0 1016/cm2 to attain a 95% equilibrium neutralization [7] and additional gas has to be injected at more than 800 sccm to satisfy it. Measured efficiency was higher than the calculated data at the low beam energy (40–60 keV). It seems to be caused by different beam species at different beam energies aside from with larger measurement error in the WFC system at a low beam energy. To maintain the optimum optics of the beam (for a stable beam extraction), the arc discharge power was set from 30 kW to 60 kW according to the beam energy. The beam species ratio is varied by the arc discharge power. The lower injected the arc power was, the larger þ generated the molecule beam species (i.e., Hþ 2 , and H3 ) ratio was. Because lower energy particles from molecule beam species have higher cross section area than fast particle from atomic beam species,
increase of the temperature was about to start in the half energy dump. The calculated optimum coil currents agree with experimental coil current to penetrate each ion dump as we expected. Neutralization efficiency was measured by the WFC system. It depends on the gas flow rate and the beam energy, and a gas injection more than 800 sccm was needed to attain an equilibrium neutralization. Maximum efficiency was measured to be more than 60% at 40 keV and less than 30% at 80 keV of the hydrogen beam. Since it closely depends on the beam species ratio and the gas line density, if the species and pressure at the beamline component can be corrected, the accuracy of the experimental data would be increased. References [1] Oh Byung-Hoon, Chang Doo-Hee, Ho Jeong Seung, Lee Kwang-Won, In SangRyul, Yoon Byung-Joo, et al. Rev Sci Instrum 2008;79:02C101. [2] Chang Doo-Hee, Seo Min-Seok, oh Byung-Hoon, Kim Jinchoon, Kim Tae-Seong. J Korean Phys Soc 2006;50(1). [3] In SR, Shim HJ. Neutral beam evolution in the KSTAR NBI test stand. J Korean Vac Sci Technol 2003;7(1):1. [4] IN SR, Yoon BJ, Kim BY, Kim TS. J Korean Phys Soc 2006;49:S320–3. [5] Hellborg Rangar, Kiisk Madis, Person Per, Stenstorom Kristina. Vacuum 2005;78:427. [6] In SR. J Korean Vacuum Soc 1999;8. [7] Kim J, Haselton HH. J Appl Phys June 1979;50(6).