Fundamental studies on the Cs dynamics under ion source conditionsa) R. Friedl and U. Fantz Citation: Review of Scientific Instruments 85, 02B109 (2014); doi: 10.1063/1.4830215 View online: http://dx.doi.org/10.1063/1.4830215 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 Cesium dynamics in long pulse operation of negative hydrogen ion sources for fusiona) Rev. Sci. Instrum. 83, 02B110 (2012); 10.1063/1.3670347 Simulation of cesium injection and distribution in rf-driven ion sources for negative hydrogen ion generationa) Rev. Sci. Instrum. 81, 02A706 (2010); 10.1063/1.3258607 Tungsten filament material and cesium dynamic equilibrium effects on a surface converter ion sourcea) Rev. Sci. Instrum. 79, 02A514 (2008); 10.1063/1.2819327 Plasma diagnostic tools for optimizing negative hydrogen ion sources Rev. Sci. Instrum. 77, 03A516 (2006); 10.1063/1.2165769 Analysis of plasma dynamics of a negative ion source based on probe measurements J. Appl. Phys. 96, 4107 (2004); 10.1063/1.1787619
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: 131.94.16.10 On: Sat, 20 Dec 2014 06:22:18
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 02B109 (2014)
Fundamental studies on the Cs dynamics under ion source conditionsa) R. Friedl1,2,b) and U. Fantz1,2 1 AG Experimentelle Plasmaphysik (EPP), Institute of Physics, University of Augsburg, 86135 Augsburg, Germany 2 Max-Planck-Institut für Plasmaphysik, EURATOM Association, Boltzmannstraße 2, 85748 Garching, Germany
(Presented 10 September 2013; received 6 September 2013; accepted 24 September 2013; published online 20 November 2013) The performance of surface conversion based negative hydrogen ion sources is mainly determined by the caesium dynamics. Therefore, fundamental investigations in vacuum and plasma are performed at a flexible laboratory setup with ion source parameters. Studies on the influence of Cs on the plasma parameters of H2 and D2 plasmas showed that ne and Te in the bulk plasma are not affected by relevant amounts of Cs and no isotopic differences could be observed. The coating of the vessel surfaces with Cs, however, leads to a considerable gettering of hydrogen atoms from the plasma volume and to the decrease of ne close to a sample surface due to the formation of negative ions. [http://dx.doi.org/10.1063/1.4830215] I. INTRODUCTION
For the neutral beam injection systems of ITER powerful negative ion sources are required.1 The surface production mechanism for negative hydrogen ions is utilized in order to achieve the ambitious target parameters. Atomic and ionic hydrogen particles from a low-temperature hydrogen plasma are converted into negative ions at a low work function surface, which is, therefore, covered with the alkali metal caesium.2, 3 Caesium is introduced into the ion source by evaporation from a Cs source and via multiple ad- and desorption processes at the inner surfaces of the ion source caesium adsorbs at the first grid of the extraction system, the so-called plasma grid. Due to the high chemical reactivity of Cs the Cs layer upon the plasma grid is very susceptible to impurities from the residual gases. Thus, a steady replenishment of the caesium layer is indispensable, which is accomplished by continuous Cs evaporation. The reliability of a stable Cs covering of the plasma grid resulting in a steady low work function converter surface drastically depends on the Cs dynamics in the hydrogen plasma as well as in the vacuum phases between the plasma pulses.4–6 In order to investigate individual effects related to the Cs dynamics observed at the IPP prototype ion source for negative hydrogen ions,7–11 fundamental studies at a flexible laboratory experiment are performed. The ICP setup (inductively coupled plasma) has comparable vacuum conditions to the ion sources (i.e., a background pressure of about 10−6 mbar) and the plasma parameters of a 10 Pa H2 plasma at 250 W RF power are comparable to those of the plasma in front of the plasma grid (ne close to 1017 m−3 , Te ≈ 1–2 eV, nH ≈ 1019 m−3 ). Furthermore, the typical Cs densities12, 13 are in the ion source relevant range of 1015 –1016 m−3 . The experiment is moreover equipped with a comprehensive set
a) Contributed paper, published as part of the Proceedings of the 15th Interna-
tional Conference on Ion Sources, Chiba, Japan, September 2013. b) Electronic mail: [email protected]
0034-6748/2014/85(2)/02B109/3/$30.00
of simultaneously operable diagnostics for Cs and its vacuum and plasma environment. Major issues for negative ion sources involve the commonly observed reduction of the co-extracted electron current due to evaporation of Cs into the ion source7, 9 and the higher co-extracted electron current for operation in deuterium.11 Therefore, the influence of Cs seeding on the plasma parameters of hydrogen and deuterium discharges is investigated in the laboratory experiment. Cs can thereby affect the discharge by a direct influence on the plasma parameters of the bulk plasma, which was mainly studied in Friedl and Fantz,14 or by an indirect influence due to the coating of the vessel surfaces. II. EXPERIMENTAL SETUP
The laboratory experiment ACCesS (Augsburg Comprehensive Cesium Setup) consists of a stainless steel vessel with 15 cm in diameter and 10 cm in height. Plasmas are generated via inductive RF coupling (frequency 27.12 MHz, max. 600 W RF power) using a planar coil located on top of the vessel. The experiment is described in detail in Friedl and Fantz14 and a top view is shown in Figure 1. Cs seeding is performed by evaporation from a Cs dispenser oven15 placed at the bottom plate. The flexible setup allows for the attachment of a variety of peripheral equipments and diagnostics via several ports at the vessel walls and at the exchangeable bottom plate. For the present investigations a Langmuir probe system is used to measure local plasma parameters (here: electron density ne ), optical emission spectroscopy is applied for the determination of global plasma parameters (here: electron temperature Te and atomic hydrogen density nH ) via two perpendicular lines of sight (LOS), white light absorption spectroscopy (WABS) is used to determine the line-of-sight integrated Cs density and via a residual gas analyzer (RGA) the impurity content is monitored. Details on the diagnostic methods are given in Friedl and Fantz.14
85, 02B109-1
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: 131.94.16.10 On: Sat, 20 Dec 2014 06:22:18
02B109-2
R. Friedl and U. Fantz
Rev. Sci. Instrum. 85, 02B109 (2014) 16
6x10
(b)
1.0
11 mm
0.9
11 mm distance to surface 5 mm
0.8 16
10
33 % nH / ne
5 mm distance to surface
Rel. electron density
(vessel walls at 75 mm)
-3
Electron density ne [m ]
(a) without sample surface
0.7
−
15
5x10
14
0 10
15
16
10
10 -3
FIG. 1. Top view of the experimental setup including the arrangement of the diagnostics and the sample surface.
Near the center of the vessel a sample holder is located which is capable of holding samples with 3×3 cm2 size. The sample holder includes a feedthrough for the Langmuir probe, which enables measurements close to the sample surface. Since the position of the Langmuir probe can be varied, the plasma parameters at different distances above the surface can be analyzed. The measurements represent an average over the length of the probe tip of 8 mm, whereas the maximal distance is given in the graphs. As a first step, a stainless steel sample is applied to reduce the amount of different surface materials within the experiment. III. RESULTS A. Investigations within the bulk plasma
H2
2.5
D2
2 1 2.4 2.2 2.0 1.8 1.6 0
(c)
nD
2.0 nH
(b)
1.5
H2 1.0 D2 14
15
0 10 10 -3 Cs density [m ]
16
10
14
15
0 10 10 -3 Cs density [m ]
-3
3.0 (a)
19
10 8 6 4
Atomic hydrogen density [10 m ]
Te [eV]
16
-3
ne [10 m ]
The influence of Cs seeding on the plasma parameters of the bulk plasma (i.e., without the sample holder present) is already presented in Friedl and Fantz14 for H2 operation. Figure 2 recapitulates the results and includes the measurements performed with deuterium. It can be seen that Cs densities up to nearly 1016 m−3 within the discharge have no influence on the electron density or temperature while for both parameters no isotopic difference can be observed: ne ≈ 3.5 × 1016 m−3 , Te ≈ 2.0 eV. The atomic density, however, is higher in deuterium which can be attributed to the higher dissociation cross section.16 Evaporation of Cs leads for both isotopes to a considerable decrease of the atomic hydrogen density accompanied by a strong hysteresis concerning the Cs density in the bulk plasma.
0.5 0
16
10
FIG. 2. (a) Electron density, (b) electron temperature, and (c) atomic hydrogen density for H2 and D2 plasmas at 10 Pa and 250 W with varying Cs density. The arrows indicate the onset of the Cs evaporation.
Cs density [m ]
14
0 10
15
16
10
10 -3
Cs density [m ]
FIG. 3. Electron density depending on the Cs density in a H2 plasma at 10 Pa and 250 W for different distances from a stainless steel surface. (a) Absolute values. (b) Relative values. The arrows indicate the onset of the Cs evaporation.
B. Investigations in proximity to a surface
Figure 3 shows the electron density measured by the Langmuir probe in different distances to the stainless steel sample surface with increasing Cs density in the H2 plasma. At the initial electron densities prior to the evaporation of caesium it can be seen that ne decreases approaching the surface. This is attributable to the additional plasma boundary and the resulting plasma profile towards this boundary. Increasing the Cs density within the discharge leads to decreasing electron densities in the vicinity of the surface and the influence increases with decreasing distance to the surface. Plotting the electron densities relative to the initial values shows that at 11 mm distance the maximal Cs density of 3 × 1015 m−3 yields a decrease in ne of about 10 %. At the same Cs content the electron density at 5 mm distance is already decreased by 20 %. Further increasing the Cs density up to 8 × 1015 m−3 decreases the electron density at 5 mm distance from the surface down to 75 % from the initial value. Decreasing the Cs density reveals again a hysteresis.
C. Discussion
Obviously, caesium densities of several 1015 m−3 have no influence on the electron density or temperature of the bulk plasma. From low-temperature plasma physics it would be expected that the admixture of a heavy atomic particle to the hydrogen discharge results in an increasing electron density and a decreasing electron temperature, which was confirmed in Friedl and Fantz14 by the usage of xenon as a substitute for Cs (comparable masses). However, much higher Cs densities of about 1017 m−3 would be required to reveal these effects. For the plasma in front of the plasma grid in negative ion sources with typical pressures of several 10−1 Pa, this limit is shifted to about 6 × 1015 m−3 . Since these values are rarely achieved in negative ion sources10, 13 a direct influence of Cs on the plasma parameters of the plasma in front of the plasma grid can be excluded. On the other hand, the electron density close to a surface as well as the atomic hydrogen density in the plasma are decreased due to the evaporation of Cs. The latter can be
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: 131.94.16.10 On: Sat, 20 Dec 2014 06:22:18
02B109-3
R. Friedl and U. Fantz
explained by the getter effect of Cs (see also Friedl and Fantz14 ) which arises from the coating of the vessel surfaces with caesium. This coating also leads to the decrease of the surface work functions17 and thus to the enhanced formation of negative hydrogen ions at the surfaces, especially at the dedicated sample surface. Consequently, electrons in front of the surface are repelled to maintain quasineutrality. Thus, it is justified to attribute the observed decrease of ne to the increase of the density of negative hydrogen ions: nH− /ne = 33 % at a distance of 5 mm from the stainless steel sample surface and a Cs density of 8 × 1015 m−3 . Furthermore, the lowering of the work function of the vessel walls and the resulting formation of negative hydrogen ions leads to the observed decrease of the plasma potential in Friedl and Fantz14 due to a varied flux balance between the bulk plasma and the vessel walls. Hence, it can be stated that the reduction of the coextracted electron current due to Cs seeding in negative ion sources does not arise from effects of Cs in the plasma volume but from the repression of electrons from the extraction system due to the formation of negative ions at the low work function surface. Furthermore, no difference between hydrogen and deuterium discharges is observed in the bulk plasma, neither for the influence of Cs seeding. Thus, the present investigations give no indication for an increased co-extracted electron current in deuterium, as long as effects in the plasma volume are considered. However, it cannot be ruled out that isotopic effects with or without caesium occur at the plasma boundary. Thus, investigations on Cs seeding to D2 discharges in proximity of a sample surface will be performed.
IV. CONCLUSIONS
Systematic investigations on the influence of evaporating Cs into hydrogen and deuterium discharges were performed at a dedicated laboratory experiment. The bulk plasma is not
Rev. Sci. Instrum. 85, 02B109 (2014)
affected by relevant densities of Cs within the discharge independently of the isotope. Furthermore, the electron density and temperature of the bulk plasma are virtually equal for both isotopes. Close to a stainless steel surface the electron density is decreased due to the formation of negative hydrogen ions. The results can be transferred to negative ion sources: The reduction of the co-extracted electron current due to Cs seeding is explained by the repression of electrons from the extraction area. The higher co-extracted electron current for D2 discharges cannot be explained by basic isotopic differences within the bulk plasma. Isotope specific effects close to a cesiated surface are to be investigated in a next step. 1 R.
Hemsworth, H. Decamps, J. Graceffa, B. Schunke, M. Tanaka, M. Dremel, A. Tanga, H. P. L. De Esch, F. Geli, J. Milnes, T. Inoue, D. Marcuzzi, P. Sonato, and P. Zaccaria, Nucl. Fusion 49, 045006 (2009). 2 M. Bacal, Nucl. Fusion 46, S250 (2006). 3 Y. I. Belchenko, G. I. Dimov, and V. G. Dudnikov, Nucl. Fusion 14, 113 (1974). 4 R. Gutser, D. Wünderlich, U. Fantz, and the N-NBI Team, Plasma Phys. Controlled Fusion 53, 105014 (2011). 5 U. Fantz, P. Franzen, and D. Wünderlich, Chem. Phys. 398, 7 (2012). 6 W. Kraus, U. Fantz, P. Franzen, M. Fröschle, B. Heinemann, R. Riedl, and D. Wünderlich, Rev. Sci. Instrum. 83, 02B104 (2012). 7 E. Speth, H. D. Falter, P. Franzen, U. Fantz, M. Bandyopadhyay, S. Christ, A. Encheva, M. Fröschle, D. Holtum, B. Heinemann, W. Kraus, A. Lorenz, Ch. Martens, P. McNeely, S. Obermayer, R. Riedl, R. Süss, A. Tanga, R. Wilhelm, and D. Wünderlich, Nucl. Fusion 46, S220 (2006). 8 U. Fantz, P. Franzen, W. Kraus, M. Berger, S. Christ-Koch, M. Fröschle, R. Gutser, B. Heinemann, C. Martens, P. McNeely, R. Riedl, E. Speth, and D. Wünderlich, Plasma Phys. Controlled Fusion 49, B563 (2007). 9 L. Schiesko, P. McNeely, U. Fantz, P. Franzen, and NNBI Team, Plasma Phys. Controlled Fusion 53, 085029 (2011). 10 C. Wimmer, U. Fantz, and NNBI-Team, AIP Conf. Proc. 1515, 246 (2013). 11 U. Fantz, L. Schiesko, D. Wünderlich, and NNBI Team, AIP Conf. Proc. 1515, 187 (2013). 12 U. Fantz, R. Gutser, and C. Wimmer, Rev. Sci. Instrum. 81, 02B102 (2010). 13 U. Fantz and C. Wimmer, J. Phys. D: Appl. Phys. 44, 335202 (2011). 14 R. Friedl and U. Fantz, AIP Conf. Proc. 1515, 255 (2013). 15 U. Fantz, R. Friedl, and M. Fröschle, Rev. Sci. Instrum. 83, 123305 (2012). 16 R. Celiberto, R. K. Janev, A. Laricchiuta, M. Capitelli, J. M. Wadehra, and D. E. Atems, At. Data Nucl. Data Tables 77, 161 (2001). 17 R. Gutser, C. Wimmer, and U. Fantz, Rev. Sci. Instrum. 82, 023506 (2011).
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: 131.94.16.10 On: Sat, 20 Dec 2014 06:22:18
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: 131.94.16.10 On: Sat, 20 Dec 2014 06:22:18
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 02B109 (2014)
Fundamental studies on the Cs dynamics under ion source conditionsa) R. Friedl1,2,b) and U. Fantz1,2 1 AG Experimentelle Plasmaphysik (EPP), Institute of Physics, University of Augsburg, 86135 Augsburg, Germany 2 Max-Planck-Institut für Plasmaphysik, EURATOM Association, Boltzmannstraße 2, 85748 Garching, Germany
(Presented 10 September 2013; received 6 September 2013; accepted 24 September 2013; published online 20 November 2013) The performance of surface conversion based negative hydrogen ion sources is mainly determined by the caesium dynamics. Therefore, fundamental investigations in vacuum and plasma are performed at a flexible laboratory setup with ion source parameters. Studies on the influence of Cs on the plasma parameters of H2 and D2 plasmas showed that ne and Te in the bulk plasma are not affected by relevant amounts of Cs and no isotopic differences could be observed. The coating of the vessel surfaces with Cs, however, leads to a considerable gettering of hydrogen atoms from the plasma volume and to the decrease of ne close to a sample surface due to the formation of negative ions. [http://dx.doi.org/10.1063/1.4830215] I. INTRODUCTION
For the neutral beam injection systems of ITER powerful negative ion sources are required.1 The surface production mechanism for negative hydrogen ions is utilized in order to achieve the ambitious target parameters. Atomic and ionic hydrogen particles from a low-temperature hydrogen plasma are converted into negative ions at a low work function surface, which is, therefore, covered with the alkali metal caesium.2, 3 Caesium is introduced into the ion source by evaporation from a Cs source and via multiple ad- and desorption processes at the inner surfaces of the ion source caesium adsorbs at the first grid of the extraction system, the so-called plasma grid. Due to the high chemical reactivity of Cs the Cs layer upon the plasma grid is very susceptible to impurities from the residual gases. Thus, a steady replenishment of the caesium layer is indispensable, which is accomplished by continuous Cs evaporation. The reliability of a stable Cs covering of the plasma grid resulting in a steady low work function converter surface drastically depends on the Cs dynamics in the hydrogen plasma as well as in the vacuum phases between the plasma pulses.4–6 In order to investigate individual effects related to the Cs dynamics observed at the IPP prototype ion source for negative hydrogen ions,7–11 fundamental studies at a flexible laboratory experiment are performed. The ICP setup (inductively coupled plasma) has comparable vacuum conditions to the ion sources (i.e., a background pressure of about 10−6 mbar) and the plasma parameters of a 10 Pa H2 plasma at 250 W RF power are comparable to those of the plasma in front of the plasma grid (ne close to 1017 m−3 , Te ≈ 1–2 eV, nH ≈ 1019 m−3 ). Furthermore, the typical Cs densities12, 13 are in the ion source relevant range of 1015 –1016 m−3 . The experiment is moreover equipped with a comprehensive set
a) Contributed paper, published as part of the Proceedings of the 15th Interna-
tional Conference on Ion Sources, Chiba, Japan, September 2013. b) Electronic mail: [email protected]
0034-6748/2014/85(2)/02B109/3/$30.00
of simultaneously operable diagnostics for Cs and its vacuum and plasma environment. Major issues for negative ion sources involve the commonly observed reduction of the co-extracted electron current due to evaporation of Cs into the ion source7, 9 and the higher co-extracted electron current for operation in deuterium.11 Therefore, the influence of Cs seeding on the plasma parameters of hydrogen and deuterium discharges is investigated in the laboratory experiment. Cs can thereby affect the discharge by a direct influence on the plasma parameters of the bulk plasma, which was mainly studied in Friedl and Fantz,14 or by an indirect influence due to the coating of the vessel surfaces. II. EXPERIMENTAL SETUP
The laboratory experiment ACCesS (Augsburg Comprehensive Cesium Setup) consists of a stainless steel vessel with 15 cm in diameter and 10 cm in height. Plasmas are generated via inductive RF coupling (frequency 27.12 MHz, max. 600 W RF power) using a planar coil located on top of the vessel. The experiment is described in detail in Friedl and Fantz14 and a top view is shown in Figure 1. Cs seeding is performed by evaporation from a Cs dispenser oven15 placed at the bottom plate. The flexible setup allows for the attachment of a variety of peripheral equipments and diagnostics via several ports at the vessel walls and at the exchangeable bottom plate. For the present investigations a Langmuir probe system is used to measure local plasma parameters (here: electron density ne ), optical emission spectroscopy is applied for the determination of global plasma parameters (here: electron temperature Te and atomic hydrogen density nH ) via two perpendicular lines of sight (LOS), white light absorption spectroscopy (WABS) is used to determine the line-of-sight integrated Cs density and via a residual gas analyzer (RGA) the impurity content is monitored. Details on the diagnostic methods are given in Friedl and Fantz.14
85, 02B109-1
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: 131.94.16.10 On: Sat, 20 Dec 2014 06:22:18
02B109-2
R. Friedl and U. Fantz
Rev. Sci. Instrum. 85, 02B109 (2014) 16
6x10
(b)
1.0
11 mm
0.9
11 mm distance to surface 5 mm
0.8 16
10
33 % nH / ne
5 mm distance to surface
Rel. electron density
(vessel walls at 75 mm)
-3
Electron density ne [m ]
(a) without sample surface
0.7
−
15
5x10
14
0 10
15
16
10
10 -3
FIG. 1. Top view of the experimental setup including the arrangement of the diagnostics and the sample surface.
Near the center of the vessel a sample holder is located which is capable of holding samples with 3×3 cm2 size. The sample holder includes a feedthrough for the Langmuir probe, which enables measurements close to the sample surface. Since the position of the Langmuir probe can be varied, the plasma parameters at different distances above the surface can be analyzed. The measurements represent an average over the length of the probe tip of 8 mm, whereas the maximal distance is given in the graphs. As a first step, a stainless steel sample is applied to reduce the amount of different surface materials within the experiment. III. RESULTS A. Investigations within the bulk plasma
H2
2.5
D2
2 1 2.4 2.2 2.0 1.8 1.6 0
(c)
nD
2.0 nH
(b)
1.5
H2 1.0 D2 14
15
0 10 10 -3 Cs density [m ]
16
10
14
15
0 10 10 -3 Cs density [m ]
-3
3.0 (a)
19
10 8 6 4
Atomic hydrogen density [10 m ]
Te [eV]
16
-3
ne [10 m ]
The influence of Cs seeding on the plasma parameters of the bulk plasma (i.e., without the sample holder present) is already presented in Friedl and Fantz14 for H2 operation. Figure 2 recapitulates the results and includes the measurements performed with deuterium. It can be seen that Cs densities up to nearly 1016 m−3 within the discharge have no influence on the electron density or temperature while for both parameters no isotopic difference can be observed: ne ≈ 3.5 × 1016 m−3 , Te ≈ 2.0 eV. The atomic density, however, is higher in deuterium which can be attributed to the higher dissociation cross section.16 Evaporation of Cs leads for both isotopes to a considerable decrease of the atomic hydrogen density accompanied by a strong hysteresis concerning the Cs density in the bulk plasma.
0.5 0
16
10
FIG. 2. (a) Electron density, (b) electron temperature, and (c) atomic hydrogen density for H2 and D2 plasmas at 10 Pa and 250 W with varying Cs density. The arrows indicate the onset of the Cs evaporation.
Cs density [m ]
14
0 10
15
16
10
10 -3
Cs density [m ]
FIG. 3. Electron density depending on the Cs density in a H2 plasma at 10 Pa and 250 W for different distances from a stainless steel surface. (a) Absolute values. (b) Relative values. The arrows indicate the onset of the Cs evaporation.
B. Investigations in proximity to a surface
Figure 3 shows the electron density measured by the Langmuir probe in different distances to the stainless steel sample surface with increasing Cs density in the H2 plasma. At the initial electron densities prior to the evaporation of caesium it can be seen that ne decreases approaching the surface. This is attributable to the additional plasma boundary and the resulting plasma profile towards this boundary. Increasing the Cs density within the discharge leads to decreasing electron densities in the vicinity of the surface and the influence increases with decreasing distance to the surface. Plotting the electron densities relative to the initial values shows that at 11 mm distance the maximal Cs density of 3 × 1015 m−3 yields a decrease in ne of about 10 %. At the same Cs content the electron density at 5 mm distance is already decreased by 20 %. Further increasing the Cs density up to 8 × 1015 m−3 decreases the electron density at 5 mm distance from the surface down to 75 % from the initial value. Decreasing the Cs density reveals again a hysteresis.
C. Discussion
Obviously, caesium densities of several 1015 m−3 have no influence on the electron density or temperature of the bulk plasma. From low-temperature plasma physics it would be expected that the admixture of a heavy atomic particle to the hydrogen discharge results in an increasing electron density and a decreasing electron temperature, which was confirmed in Friedl and Fantz14 by the usage of xenon as a substitute for Cs (comparable masses). However, much higher Cs densities of about 1017 m−3 would be required to reveal these effects. For the plasma in front of the plasma grid in negative ion sources with typical pressures of several 10−1 Pa, this limit is shifted to about 6 × 1015 m−3 . Since these values are rarely achieved in negative ion sources10, 13 a direct influence of Cs on the plasma parameters of the plasma in front of the plasma grid can be excluded. On the other hand, the electron density close to a surface as well as the atomic hydrogen density in the plasma are decreased due to the evaporation of Cs. The latter can be
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: 131.94.16.10 On: Sat, 20 Dec 2014 06:22:18
02B109-3
R. Friedl and U. Fantz
explained by the getter effect of Cs (see also Friedl and Fantz14 ) which arises from the coating of the vessel surfaces with caesium. This coating also leads to the decrease of the surface work functions17 and thus to the enhanced formation of negative hydrogen ions at the surfaces, especially at the dedicated sample surface. Consequently, electrons in front of the surface are repelled to maintain quasineutrality. Thus, it is justified to attribute the observed decrease of ne to the increase of the density of negative hydrogen ions: nH− /ne = 33 % at a distance of 5 mm from the stainless steel sample surface and a Cs density of 8 × 1015 m−3 . Furthermore, the lowering of the work function of the vessel walls and the resulting formation of negative hydrogen ions leads to the observed decrease of the plasma potential in Friedl and Fantz14 due to a varied flux balance between the bulk plasma and the vessel walls. Hence, it can be stated that the reduction of the coextracted electron current due to Cs seeding in negative ion sources does not arise from effects of Cs in the plasma volume but from the repression of electrons from the extraction system due to the formation of negative ions at the low work function surface. Furthermore, no difference between hydrogen and deuterium discharges is observed in the bulk plasma, neither for the influence of Cs seeding. Thus, the present investigations give no indication for an increased co-extracted electron current in deuterium, as long as effects in the plasma volume are considered. However, it cannot be ruled out that isotopic effects with or without caesium occur at the plasma boundary. Thus, investigations on Cs seeding to D2 discharges in proximity of a sample surface will be performed.
IV. CONCLUSIONS
Systematic investigations on the influence of evaporating Cs into hydrogen and deuterium discharges were performed at a dedicated laboratory experiment. The bulk plasma is not
Rev. Sci. Instrum. 85, 02B109 (2014)
affected by relevant densities of Cs within the discharge independently of the isotope. Furthermore, the electron density and temperature of the bulk plasma are virtually equal for both isotopes. Close to a stainless steel surface the electron density is decreased due to the formation of negative hydrogen ions. The results can be transferred to negative ion sources: The reduction of the co-extracted electron current due to Cs seeding is explained by the repression of electrons from the extraction area. The higher co-extracted electron current for D2 discharges cannot be explained by basic isotopic differences within the bulk plasma. Isotope specific effects close to a cesiated surface are to be investigated in a next step. 1 R.
Hemsworth, H. Decamps, J. Graceffa, B. Schunke, M. Tanaka, M. Dremel, A. Tanga, H. P. L. De Esch, F. Geli, J. Milnes, T. Inoue, D. Marcuzzi, P. Sonato, and P. Zaccaria, Nucl. Fusion 49, 045006 (2009). 2 M. Bacal, Nucl. Fusion 46, S250 (2006). 3 Y. I. Belchenko, G. I. Dimov, and V. G. Dudnikov, Nucl. Fusion 14, 113 (1974). 4 R. Gutser, D. Wünderlich, U. Fantz, and the N-NBI Team, Plasma Phys. Controlled Fusion 53, 105014 (2011). 5 U. Fantz, P. Franzen, and D. Wünderlich, Chem. Phys. 398, 7 (2012). 6 W. Kraus, U. Fantz, P. Franzen, M. Fröschle, B. Heinemann, R. Riedl, and D. Wünderlich, Rev. Sci. Instrum. 83, 02B104 (2012). 7 E. Speth, H. D. Falter, P. Franzen, U. Fantz, M. Bandyopadhyay, S. Christ, A. Encheva, M. Fröschle, D. Holtum, B. Heinemann, W. Kraus, A. Lorenz, Ch. Martens, P. McNeely, S. Obermayer, R. Riedl, R. Süss, A. Tanga, R. Wilhelm, and D. Wünderlich, Nucl. Fusion 46, S220 (2006). 8 U. Fantz, P. Franzen, W. Kraus, M. Berger, S. Christ-Koch, M. Fröschle, R. Gutser, B. Heinemann, C. Martens, P. McNeely, R. Riedl, E. Speth, and D. Wünderlich, Plasma Phys. Controlled Fusion 49, B563 (2007). 9 L. Schiesko, P. McNeely, U. Fantz, P. Franzen, and NNBI Team, Plasma Phys. Controlled Fusion 53, 085029 (2011). 10 C. Wimmer, U. Fantz, and NNBI-Team, AIP Conf. Proc. 1515, 246 (2013). 11 U. Fantz, L. Schiesko, D. Wünderlich, and NNBI Team, AIP Conf. Proc. 1515, 187 (2013). 12 U. Fantz, R. Gutser, and C. Wimmer, Rev. Sci. Instrum. 81, 02B102 (2010). 13 U. Fantz and C. Wimmer, J. Phys. D: Appl. Phys. 44, 335202 (2011). 14 R. Friedl and U. Fantz, AIP Conf. Proc. 1515, 255 (2013). 15 U. Fantz, R. Friedl, and M. Fröschle, Rev. Sci. Instrum. 83, 123305 (2012). 16 R. Celiberto, R. K. Janev, A. Laricchiuta, M. Capitelli, J. M. Wadehra, and D. E. Atems, At. Data Nucl. Data Tables 77, 161 (2001). 17 R. Gutser, C. Wimmer, and U. Fantz, Rev. Sci. Instrum. 82, 023506 (2011).
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: 131.94.16.10 On: Sat, 20 Dec 2014 06:22:18
