Characteristics of plasma grid bias in large-scaled negative ion sourcea) M. Kisaki, K. Tsumori, K. Ikeda, H. Nakano, M. Osakabe, K. Nagaoka, Y. Takeiri, and O. Kaneko Citation: Review of Scientific Instruments 85, 02B131 (2014); doi: 10.1063/1.4854295 View online: http://dx.doi.org/10.1063/1.4854295 View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/85/2?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Influence of electric field penetration by a three-electrode beam extraction system on hydrogen negative ion source plasmaa) Rev. Sci. Instrum. 85, 02A720 (2014); 10.1063/1.4833918 Spatial distribution of the plasma parameters in a radio-frequency driven negative ion sourcea) Rev. Sci. Instrum. 85, 02B104 (2014); 10.1063/1.4826540 Measurement of Electron Density near Plasma Grid of Largescaled Negative Ion Source by Means of Millimeter Wave Interferometer AIP Conf. Proc. 1390, 374 (2011); 10.1063/1.3637408 Characteristics of the RF Negative Ion Source Using a Mesh Grid Bias Method AIP Conf. Proc. 1097, 109 (2009); 10.1063/1.3112503 Effects of bias potential upon H − density near a plasma grid of a negative ion source Rev. Sci. Instrum. 77, 03A513 (2006); 10.1063/1.2165573
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 132.236.27.111 On: Mon, 15 Dec 2014 11:18:07
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 02B131 (2014)
Characteristics of plasma grid bias in large-scaled negative ion sourcea) M. Kisaki, K. Tsumori, K. Ikeda, H. Nakano, M. Osakabe, K. Nagaoka, Y. Takeiri, and O. Kanekob) National Institute for Fusion Science, Toki 509-5292, Japan
(Presented 10 September 2013; received 22 September 2013; accepted 19 November 2013; published online 14 January 2014) The electron density was measured at various bias voltages to understand how the plasma grid bias affects the electron near the plasma grid in large-scaled negative ion sources. It was found that the response of the electron to the bias voltage changes depending on negative ion production processes. The electron density remarkably decreases with increasing the bias voltage in the purevolume plasma. On the other hand, the electron density depends on the bias voltage weakly in the Cs-seeded plasma. In addition, it was observed that the response of the co-extracted electron current to the bias voltage has similar trend to that of the electron density. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4854295] I. INTRODUCTION
II. EXPERIMENTAL SETUP
Negative-ion-based neutral beam injection (N-NBI) system is a powerful tool for the plasma heating because of the relatively high neutralization coefficient at the high energy region. In negative hydrogen ion sources, the negative hydrogen ion is extracted with the electron, so-called, co-extracted electron. The co-extracted electron causes the extra consumption of electric power and the heat loading on acceleration grids. Two methods, electron suppression magnet and plasma grid bias, are mainly implemented to the negative ion sources in order to suppress these drawbacks. The strong magnetic field is created by the permanent magnets embedded in the extraction grid.1 The co-extracted electron is strongly deflected by the magnetic field and impinges on the extraction grid before full acceleration. In addition, the plasma grid is electrically biased relative to the plasma chamber2 and it has been experimentally observed that the co-extracted electron current gradually decreases with increasing the bias voltage, while the negative ion current remains constant at the low bias voltage. However, it has not been well understood how the electron in the plasma chamber responses to the bias voltage, because the plasma is weakly magnetized and the Langmuir probe has difficulty in obtaining the electron density precisely at such circumstance. Then, it is important to study the electron behavior in the ion source to reduce the co-extracted electron current more effectively. Recently, we utilized a surface wave probe (SWP), which is based on the resonant spectroscopy, and obtained the electron density successfully.3 In this study, the electron density measurement was carried out by the SWP at various bias voltages in the pure-volume and Cs-seeded plasmas. We will discuss effects of the grid bias on electrons in the different negative ion production processes.
This study has been done with the NIFS 1/3-scaled negative ion source for Large Helical Device (LHD) (Fig. 1(a)). The negative ion source consists of two parts: plasma generator and accelerator. The hydrogen plasma is generated by the arc discharge with filaments. The accelerator is composed of the plasma grid (PG), extraction grid (EG), steering grid (SG), and grounded grid (GG). Each grid is divided into two segments and 17 (horizontal) × 14 (vertical) apertures are drilled on a segment. The PG is electrically insulated from the plasma chamber by the bias flange in order to apply the bias voltage between the PG and plasma chamber. Fig. 1(b) shows the schematic illustration of the experimental setup. The SWP is installed on the bias flange and movable along the axis perpendicular to the PG with the drive mechanism. Fig. 1(c) shows a cross-sectional view of the SWP. The SWP consists of an alumina tube with a diameter of 3 mm and a semi-rigid cable whose inner conductor with a length of 5 mm is exposed at the end. Radio frequency power is applied to the semi-rigid cable and the frequency is linearly increased from 300 kHz to 3.0 GHz. The ratio of the reflected power to the applied power is measured by a network analyzer. At certain frequency, an electrostatic wave is excited at the interface between the alumina tube and the plasma at the region where the inner conductor of the co-axial cable is exposed and the incident power is strongly absorbed. The electron density can be estimated from the resonant frequency.
a) Contributed paper, published as part of the Proceedings of the 15th Interna-
tional Conference on Ion Sources, Chiba, Japan, September 2013. b) [email protected]
0034-6748/2014/85(2)/02B131/3/$30.00
III. EXPERIMENTAL RESULTS
The electron densities along the axis perpendicular to the PG were obtained at different bias voltages. Fig. 2(a) shows the electron density as a function of the distance from the PG surface in the pure-volume plasma. The bias voltage, Vbias , varies from 1 V to 9 V, where the PG is positively biased relative to the arc chamber. The arc power, Parc , is 50 kW. In negative ion sources, the electron seems to be diffused from the driver region by the ion-driven ambipolar diffusion because the positive ion can traverse the filter field due to its
85, 02B131-1
© 2014 AIP Publishing LLC
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 132.236.27.111 On: Mon, 15 Dec 2014 11:18:07
02B131-2
Kisaki et al.
Rev. Sci. Instrum. 85, 02B131 (2014)
(a)
(b) Cs tank 14 apertures
Arc chamber Filaments
PG
17 apertures
Bias flange (Insulator)
PG EG SG Drive mechanism
SWP y x z
GG
RF power Network analyzer
PC
5 mm
(c)
Alumina tube
Semi-rigid cable
3 mm
FIG. 1. (a) Cross-sectional view of NIFS 1/3-scaled negative ion source for LHD, (b) schematic drawing of diagnostic system, and (c) cross-sectional view of surface wave probe.
(a) 4
Electron density [10
17
-3
m ]
Vbias = 1 V Vbias = 3 V Vbias = 5 V Vbias = 9 V
3
w/o Cs
2
1
0 0
5
10
15
20
25
30
Distance from PG [mm]
larger Larmor radius and the electron follows it to keep the charge neutrality. Hence, the electron density monotonically decreases toward the PG in Fig. 2(a). Fig. 3 shows the dependence of the plasma potential on the bias voltage. The plasma potential is smaller than the bias voltage and it rises with increasing the bias voltage. This implies that the ion diffusion is reduced by applying the bias voltage and it results in the reduction in the electron diffusion to the extraction region. Then, the electron density in the pure-volume plasma significantly decreases with increasing the bias voltage as in Fig. 2(a). However, the Cs-seeded plasma shows the different properties in Fig. 2(b). The electron density in the extraction region is decreased even at low bias voltage by seeding the Cs, because the Cs seeding enhances the negative ion production from the PG surface4 and the electron diffusion to the extraction region is reduced in order to maintain the charge
(b) 4
3
10
w/ Cs
Plasma potential [V]
Electron density [10
17
-3
m ]
Vbias = 1 V Vbias = 3 V Vbias = 5 V Vbias = 9 V
2
1
8 6 4 2
0 0
5
10
15
20
25
30
Distance from PG [mm]
0 0
2
4
6
8
10
Bias voltage [V] FIG. 2. Electron density distributions at different bias voltages (a) without Cs and (b) with Cs.
FIG. 3. Response of plasma potential to bias voltage.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 132.236.27.111 On: Mon, 15 Dec 2014 11:18:07
Kisaki et al.
02B131-3
Rev. Sci. Instrum. 85, 02B131 (2014)
10
10 w/o Cs Parc = 50 kW Vext = 3.3 kV Vacc = 53 kV
8 6
8 6
2
2
-3
m ]
4
16
4
Electron density [10
Co-extracted electron current [A]
(a)
0 0
2
4
6
8
10
0 12
Bias voltage [V]
10
10 w/ Cs Parc = 50 kW Vext = 5.6 kV Vacc = 64 kV
8 6
8 6
2
2
0 0
2
4
6
8
10
-3
4
15
4
Electron density [10 m ]
Co-extracted electron current [A]
(b)
0 12
Bias voltage [V] FIG. 4. Response of co-extracted electron current and electron density to bias voltage (a) without Cs and (b) with Cs. Filled circle and square show co-extracted electron current and electron density, respectively.
neutrality. The negative ions near the PG surface also mitigate the effect of the grid bias on the electron and the electron density depends on the bias voltage weakly in the surface production process. Figs. 4(a) and 4(b) show the co-extracted electron current and electron density as a function of the bias voltage in the plasmas with and without the Cs, respectively. In Figs. 4(a) and 4(b), the extraction voltages, Vext , are 3.3 kV and 5.6 kV, and the acceleration voltages, Vacc , are 53 kV and 64 kV, respectively. The co-extracted electron current is evaluated from the difference between the currents flowing into the extraction and acceleration power supplies. Here, we assumed that the co-extracted electron is dissipated on the EG by the electron suppression field and cannot pass through the EG.5 In Fig. 4, the co-extracted electron current decreased
only by a factor of 4 by seeding the Cs, though the electron density became smaller by an order of magnitude. This is because the beam extraction was carried out at higher extraction voltage in the Cs-seeded plasma. As shown in Fig. 4, the response of the co-extracted electron current to the bias voltage has similar trend to that of the electron density and the co-extracted electron current becomes lower by seeding the Cs. This suggests that the co-extracted electron current from the pure-volume plasma is significantly suppressed at higher bias voltage as the grid bias decreases the electron diffusion to the extraction region effectively. On the other hand, in the Cs-seeded plasma the electron diffusion is mainly controlled by the negative ion near the PG. Hence, the co-extracted electron current slightly decreases with increasing the bias voltage. IV. CONCLUSION
The electron density measurement in the large-scaled negative ion source was carried out and the effect of the grid bias on the electron was investigated in the pure-volume and Cs-seeded plasmas. It was found that the plasma grid bias decreases the electron diffusion to the extraction region in the pure-volume plasma and this causes the significant reductions of the electron density and the co-extracted electron current. On the other hand, in the Cs-seeded plasma the negative ion controls the electron diffusion to the PG and the grid bias weakly effects the electron. Hence, the co-extracted electron current slightly decreases with increasing the bias voltage in the Cs-seeded plasma. ACKNOWLEDGMENTS
Authors would like to thank technical staff of LHD-NBI group for excellent operations of NBI test stand. This study was supported by JSPS KAKENHI Grant No. 25249134 and NIFS (NIFS13ULRR015). 1 L.
M. Lea, J. T. Holmes, M. F. Thornton, and G. O. R. Naylor, Rev. Sci. Instrum. 61, 409 (1990). 2 T. Inoue, G. D. Ackerman, W. S. Cooper, M. Hanada, J. W. Kwan, Y. Ohara, Y. Okumura, and M. Seki, Rev. Sci. Instrum. 61, 496 (1990). 3 M. Kisaki, K. Tsumori, H. Nakano, K. Ikeda, M. Osakabe, K. Nagaoka, M. Shibuya, M. Sato, H. Sekiguchi, S. Komada, T. Kondo, H. Hayashi, E. Asano, Y. Takeiri, and O. Kaneko, Rev. Sci. Instrum. 83, 02B113 (2012). 4 K. Tsumori, H. Nakano, M. Kisaki, K. Ikeda, K. Nagaoka, M. Osakabe, Y. Takeiri, O. Kaneko, M. Shibuya, E. Asano, T. Kondo, M. Sato, S. Komada, H. Sekiguchi, N. Kameyama, T. Fukuyama, S. Wada, and A. Hatayama, Rev. Sci. Instrum. 83, 02B116 (2012). 5 M. Hanada, Y. Fujiwara, K. Miyamoto, N. Miyamoto, Y. Okumura, and K. Watanabe, Rev. Sci. Instrum. 69, 947 (1998).
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 132.236.27.111 On: Mon, 15 Dec 2014 11:18:07
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 132.236.27.111 On: Mon, 15 Dec 2014 11:18:07
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 02B131 (2014)
Characteristics of plasma grid bias in large-scaled negative ion sourcea) M. Kisaki, K. Tsumori, K. Ikeda, H. Nakano, M. Osakabe, K. Nagaoka, Y. Takeiri, and O. Kanekob) National Institute for Fusion Science, Toki 509-5292, Japan
(Presented 10 September 2013; received 22 September 2013; accepted 19 November 2013; published online 14 January 2014) The electron density was measured at various bias voltages to understand how the plasma grid bias affects the electron near the plasma grid in large-scaled negative ion sources. It was found that the response of the electron to the bias voltage changes depending on negative ion production processes. The electron density remarkably decreases with increasing the bias voltage in the purevolume plasma. On the other hand, the electron density depends on the bias voltage weakly in the Cs-seeded plasma. In addition, it was observed that the response of the co-extracted electron current to the bias voltage has similar trend to that of the electron density. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4854295] I. INTRODUCTION
II. EXPERIMENTAL SETUP
Negative-ion-based neutral beam injection (N-NBI) system is a powerful tool for the plasma heating because of the relatively high neutralization coefficient at the high energy region. In negative hydrogen ion sources, the negative hydrogen ion is extracted with the electron, so-called, co-extracted electron. The co-extracted electron causes the extra consumption of electric power and the heat loading on acceleration grids. Two methods, electron suppression magnet and plasma grid bias, are mainly implemented to the negative ion sources in order to suppress these drawbacks. The strong magnetic field is created by the permanent magnets embedded in the extraction grid.1 The co-extracted electron is strongly deflected by the magnetic field and impinges on the extraction grid before full acceleration. In addition, the plasma grid is electrically biased relative to the plasma chamber2 and it has been experimentally observed that the co-extracted electron current gradually decreases with increasing the bias voltage, while the negative ion current remains constant at the low bias voltage. However, it has not been well understood how the electron in the plasma chamber responses to the bias voltage, because the plasma is weakly magnetized and the Langmuir probe has difficulty in obtaining the electron density precisely at such circumstance. Then, it is important to study the electron behavior in the ion source to reduce the co-extracted electron current more effectively. Recently, we utilized a surface wave probe (SWP), which is based on the resonant spectroscopy, and obtained the electron density successfully.3 In this study, the electron density measurement was carried out by the SWP at various bias voltages in the pure-volume and Cs-seeded plasmas. We will discuss effects of the grid bias on electrons in the different negative ion production processes.
This study has been done with the NIFS 1/3-scaled negative ion source for Large Helical Device (LHD) (Fig. 1(a)). The negative ion source consists of two parts: plasma generator and accelerator. The hydrogen plasma is generated by the arc discharge with filaments. The accelerator is composed of the plasma grid (PG), extraction grid (EG), steering grid (SG), and grounded grid (GG). Each grid is divided into two segments and 17 (horizontal) × 14 (vertical) apertures are drilled on a segment. The PG is electrically insulated from the plasma chamber by the bias flange in order to apply the bias voltage between the PG and plasma chamber. Fig. 1(b) shows the schematic illustration of the experimental setup. The SWP is installed on the bias flange and movable along the axis perpendicular to the PG with the drive mechanism. Fig. 1(c) shows a cross-sectional view of the SWP. The SWP consists of an alumina tube with a diameter of 3 mm and a semi-rigid cable whose inner conductor with a length of 5 mm is exposed at the end. Radio frequency power is applied to the semi-rigid cable and the frequency is linearly increased from 300 kHz to 3.0 GHz. The ratio of the reflected power to the applied power is measured by a network analyzer. At certain frequency, an electrostatic wave is excited at the interface between the alumina tube and the plasma at the region where the inner conductor of the co-axial cable is exposed and the incident power is strongly absorbed. The electron density can be estimated from the resonant frequency.
a) Contributed paper, published as part of the Proceedings of the 15th Interna-
tional Conference on Ion Sources, Chiba, Japan, September 2013. b) [email protected]
0034-6748/2014/85(2)/02B131/3/$30.00
III. EXPERIMENTAL RESULTS
The electron densities along the axis perpendicular to the PG were obtained at different bias voltages. Fig. 2(a) shows the electron density as a function of the distance from the PG surface in the pure-volume plasma. The bias voltage, Vbias , varies from 1 V to 9 V, where the PG is positively biased relative to the arc chamber. The arc power, Parc , is 50 kW. In negative ion sources, the electron seems to be diffused from the driver region by the ion-driven ambipolar diffusion because the positive ion can traverse the filter field due to its
85, 02B131-1
© 2014 AIP Publishing LLC
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 132.236.27.111 On: Mon, 15 Dec 2014 11:18:07
02B131-2
Kisaki et al.
Rev. Sci. Instrum. 85, 02B131 (2014)
(a)
(b) Cs tank 14 apertures
Arc chamber Filaments
PG
17 apertures
Bias flange (Insulator)
PG EG SG Drive mechanism
SWP y x z
GG
RF power Network analyzer
PC
5 mm
(c)
Alumina tube
Semi-rigid cable
3 mm
FIG. 1. (a) Cross-sectional view of NIFS 1/3-scaled negative ion source for LHD, (b) schematic drawing of diagnostic system, and (c) cross-sectional view of surface wave probe.
(a) 4
Electron density [10
17
-3
m ]
Vbias = 1 V Vbias = 3 V Vbias = 5 V Vbias = 9 V
3
w/o Cs
2
1
0 0
5
10
15
20
25
30
Distance from PG [mm]
larger Larmor radius and the electron follows it to keep the charge neutrality. Hence, the electron density monotonically decreases toward the PG in Fig. 2(a). Fig. 3 shows the dependence of the plasma potential on the bias voltage. The plasma potential is smaller than the bias voltage and it rises with increasing the bias voltage. This implies that the ion diffusion is reduced by applying the bias voltage and it results in the reduction in the electron diffusion to the extraction region. Then, the electron density in the pure-volume plasma significantly decreases with increasing the bias voltage as in Fig. 2(a). However, the Cs-seeded plasma shows the different properties in Fig. 2(b). The electron density in the extraction region is decreased even at low bias voltage by seeding the Cs, because the Cs seeding enhances the negative ion production from the PG surface4 and the electron diffusion to the extraction region is reduced in order to maintain the charge
(b) 4
3
10
w/ Cs
Plasma potential [V]
Electron density [10
17
-3
m ]
Vbias = 1 V Vbias = 3 V Vbias = 5 V Vbias = 9 V
2
1
8 6 4 2
0 0
5
10
15
20
25
30
Distance from PG [mm]
0 0
2
4
6
8
10
Bias voltage [V] FIG. 2. Electron density distributions at different bias voltages (a) without Cs and (b) with Cs.
FIG. 3. Response of plasma potential to bias voltage.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 132.236.27.111 On: Mon, 15 Dec 2014 11:18:07
Kisaki et al.
02B131-3
Rev. Sci. Instrum. 85, 02B131 (2014)
10
10 w/o Cs Parc = 50 kW Vext = 3.3 kV Vacc = 53 kV
8 6
8 6
2
2
-3
m ]
4
16
4
Electron density [10
Co-extracted electron current [A]
(a)
0 0
2
4
6
8
10
0 12
Bias voltage [V]
10
10 w/ Cs Parc = 50 kW Vext = 5.6 kV Vacc = 64 kV
8 6
8 6
2
2
0 0
2
4
6
8
10
-3
4
15
4
Electron density [10 m ]
Co-extracted electron current [A]
(b)
0 12
Bias voltage [V] FIG. 4. Response of co-extracted electron current and electron density to bias voltage (a) without Cs and (b) with Cs. Filled circle and square show co-extracted electron current and electron density, respectively.
neutrality. The negative ions near the PG surface also mitigate the effect of the grid bias on the electron and the electron density depends on the bias voltage weakly in the surface production process. Figs. 4(a) and 4(b) show the co-extracted electron current and electron density as a function of the bias voltage in the plasmas with and without the Cs, respectively. In Figs. 4(a) and 4(b), the extraction voltages, Vext , are 3.3 kV and 5.6 kV, and the acceleration voltages, Vacc , are 53 kV and 64 kV, respectively. The co-extracted electron current is evaluated from the difference between the currents flowing into the extraction and acceleration power supplies. Here, we assumed that the co-extracted electron is dissipated on the EG by the electron suppression field and cannot pass through the EG.5 In Fig. 4, the co-extracted electron current decreased
only by a factor of 4 by seeding the Cs, though the electron density became smaller by an order of magnitude. This is because the beam extraction was carried out at higher extraction voltage in the Cs-seeded plasma. As shown in Fig. 4, the response of the co-extracted electron current to the bias voltage has similar trend to that of the electron density and the co-extracted electron current becomes lower by seeding the Cs. This suggests that the co-extracted electron current from the pure-volume plasma is significantly suppressed at higher bias voltage as the grid bias decreases the electron diffusion to the extraction region effectively. On the other hand, in the Cs-seeded plasma the electron diffusion is mainly controlled by the negative ion near the PG. Hence, the co-extracted electron current slightly decreases with increasing the bias voltage. IV. CONCLUSION
The electron density measurement in the large-scaled negative ion source was carried out and the effect of the grid bias on the electron was investigated in the pure-volume and Cs-seeded plasmas. It was found that the plasma grid bias decreases the electron diffusion to the extraction region in the pure-volume plasma and this causes the significant reductions of the electron density and the co-extracted electron current. On the other hand, in the Cs-seeded plasma the negative ion controls the electron diffusion to the PG and the grid bias weakly effects the electron. Hence, the co-extracted electron current slightly decreases with increasing the bias voltage in the Cs-seeded plasma. ACKNOWLEDGMENTS
Authors would like to thank technical staff of LHD-NBI group for excellent operations of NBI test stand. This study was supported by JSPS KAKENHI Grant No. 25249134 and NIFS (NIFS13ULRR015). 1 L.
M. Lea, J. T. Holmes, M. F. Thornton, and G. O. R. Naylor, Rev. Sci. Instrum. 61, 409 (1990). 2 T. Inoue, G. D. Ackerman, W. S. Cooper, M. Hanada, J. W. Kwan, Y. Ohara, Y. Okumura, and M. Seki, Rev. Sci. Instrum. 61, 496 (1990). 3 M. Kisaki, K. Tsumori, H. Nakano, K. Ikeda, M. Osakabe, K. Nagaoka, M. Shibuya, M. Sato, H. Sekiguchi, S. Komada, T. Kondo, H. Hayashi, E. Asano, Y. Takeiri, and O. Kaneko, Rev. Sci. Instrum. 83, 02B113 (2012). 4 K. Tsumori, H. Nakano, M. Kisaki, K. Ikeda, K. Nagaoka, M. Osakabe, Y. Takeiri, O. Kaneko, M. Shibuya, E. Asano, T. Kondo, M. Sato, S. Komada, H. Sekiguchi, N. Kameyama, T. Fukuyama, S. Wada, and A. Hatayama, Rev. Sci. Instrum. 83, 02B116 (2012). 5 M. Hanada, Y. Fujiwara, K. Miyamoto, N. Miyamoto, Y. Okumura, and K. Watanabe, Rev. Sci. Instrum. 69, 947 (1998).
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 132.236.27.111 On: Mon, 15 Dec 2014 11:18:07
