Effects of roughness and temperature on low-energy hydrogen positive and negative ion reflection from silicon and carbon surfacesa) N. Tanaka, S. Kato, T. Miyamoto, M. Nishiura, K. Tsumori, Y. Matsumoto, T. Kenmotsu, A. Okamoto, S. Kitajima , M. Sasao, M. Wada, and H. Yamaoka Citation: Review of Scientific Instruments 85, 02C311 (2014); doi: 10.1063/1.4855455 View online: http://dx.doi.org/10.1063/1.4855455 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 Effects of adsorption and roughness upon the collision processes at the convertor surface of a plasma sputter negative ion sourcea) Rev. Sci. Instrum. 83, 02A722 (2012); 10.1063/1.3673624 3D modeling of the electron energy distribution function in negative hydrogen ion sourcesa) Rev. Sci. Instrum. 81, 02A703 (2010); 10.1063/1.3273075 Energy distribution of negative carbon ion beam extracted from a plasma-sputter-type negative ion source Rev. Sci. Instrum. 71, 1122 (2000); 10.1063/1.1150403 Negative ion enhancement in caesium-seeded hydrogen discharges—a volume or surface effect? Appl. Phys. Lett. 75, 2737 (1999); 10.1063/1.125133 Effects of filament positions on the arc discharge characteristics of a negative hydrogen ion source for neutral beam injector Rev. Sci. Instrum. 70, 2338 (1999); 10.1063/1.1149760
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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 02C311 (2014)
Effects of roughness and temperature on low-energy hydrogen positive and negative ion reflection from silicon and carbon surfacesa) N. Tanaka,1,b) S. Kato,2 T. Miyamoto,2 M. Nishiura,3 K. Tsumori,3 Y. Matsumoto,4 T. Kenmotsu,5 A. Okamoto,6 S. Kitajima,6 M. Sasao,7 M. Wada,2 and H. Yamaoka8 1
Institute of Laser Engineering, Osaka University, Suita, Osaka 565-0871, Japan Graduate School of Science and Engineering, Doshisha University, Kyotanabe, Kyoto 610-0321, Japan 3 National Institute for Fusion Science, Toki, Gifu 509-5292, Japan 4 Tokushima Bunri University, Yamashiro, Tokushima 770-8514, Japan 5 Faculty of Life and Medical Sciences, Doshisha University, Kyotanabe, Kyoto 610-0321, Japan 6 School of Engineering, Tohoku University, Aoba, Sendai, Miyagi 980-8579, Japan 7 Organization for Research Initiatives and Development, Doshisha University, Kyotanabe, Kyoto 610-0321, Japan 8 RIKEN SPring-8 Center, Mikazuki, Sayo, Hyogo 679-5148, Japan 2
(Presented 11 September 2013; received 7 September 2013; accepted 26 November 2013; published online 7 January 2014) Angle-resolved energy distribution functions of positive and negative hydrogen ions produced from a rough-finished Si surface under 1 keV proton irradiation have been measured. The corresponding distribution from a crystalline surface and a carbon surface are also measured for comparison. Intensities of positive and negative ions from the rough-finished Si are substantially smaller than those from crystalline Si. The angular distributions of these species are broader for rough surface than the crystalline surface. No significant temperature dependence for positive and negative ion intensities is observed for all samples in the temperature range from 300 to 400 K. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4855455] I. INTRODUCTION
Low-energy particle interaction with matters has attracted much attention in areas covering from fundamental science to technological applications including surface characterisation, surface atom manipulation, nanotechnology, and film growth.1 Hydrogen ion-solid surface interaction leads to formation of both positive and negative ions at the surface, and is of particular importance in wall conditioning of plasma confinement devices. Despite the large amount of research efforts, the basic process of the beam interaction with matter has not been fully understood yet. We have studied the fundamental processes of beam-surface interactions by measuring the positive and negative ions scattered from the solid surface by injecting low-energy (1–3 keV) light ion beams.2 Both positive and negative ions from molybdenum,3 tungsten,4 vanadium alloy,5 and carbon nano wall6 were observed. The intensities of negative ions formed on these samples were the same order of magnitude. Similar negative ion yields were also observed in the graphite, Mg, Al, Ag, Si, and diamond.7–12 The negative ion yields from such high work function surfaces of graphite, Si, and diamond are too large to be explained by conventional surface charge transfer models.13 We also found that the surface structure affected the reflection property.14 The results suggest that the surface structure is an important factor determining the intensities of scattered positive and negative ions. On the other hand, most experiments of the beam-surface interaction have been performed at room temperature. For the 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)/02C311/4/$30.00
temperature effect on the negative ion yield from surfaces other than alkali metal seeded materials, only some experimental results have been reported.15–17 In this paper we report the surface roughness effect on the reflection properties at Si targets for a 1 keV H+ beam injection. Reflection from Si and carbon surfaces are measured for comparison. Temperature effects of these materials are also studied. II. EXPERIMENT
Figure 1(a) shows a schematic diagram of the experimental setup. A compact multi-cusp magnetic field type ion source is used to produce positive ions. The discharge current is typically about 0.3–0.5 A to obtain a few tens of nA at the target position for a beam energy of 1–2 keV. Deflector and einzel lens systems are installed to improve the beam optics. The beam travels through a water-cooled selector magnet to be separated from ions of different mass. The analyzer chamber is connected to the beam line by a 5 cm long and 5 mm inner diameter tube, which keeps high vacuum in analyzer chamber, where the residual pressure was kept below 2 × 10−5 Pa. The beam collimation system made of a pair of slits in front of the target defines a beam size of 1 (horizontal) mm × 5 (vertical) mm. Incident H+ beam energy was 1 keV throughout this series of the experiments. The analyzer system consists of a target, a magnet momentum analyzer with a MCP (multi-channel plate) detector, and a target heating system. A water-cooled, 90◦ -bending magnet used as a momentum analyzer is installed on a table which can rotate approximately from −60◦ to 60◦ . In many other experiments, the measurements have been performed under the limitation of α + β = const., where α
85, 02C311-1
© 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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
02C311-2
(a)
Tanaka et al.
Rev. Sci. Instrum. 85, 02C311 (2014) Magnetic Momentum Analyzer (MMA)
Ion source Thermocouple Sector magnet
Deflectors Einzel lens Collimator
(b) Incident angle α
Reflection angle Magnetic momentum analyzer
β Target
Sample holder & heater
MCP
Analyzer chamber
MCP
FIG. 1. (a) Schematic diagram of the experimental setup. Extracted ion beam is transported to the analyzer chamber. Both positive and negative ions produced by reflection of incident proton at the surface are measured by the magnetic momentum analyzer installed in vacuum. The target temperature is controlled by a small heater behind the target. (b) Definition of incident angle (α) and reflection angle (β).
and β are incident and reflection angles with respect to the target plane as shown in Fig. 1(b). Contrast to this restriction, our system has an advantage that the angles of the target and the analyzer are controlled independently. Furthermore, we can measure both positive and negative ions produced from the target by changing the polarity of the analyzer magnet current during a single scan. The analyzer energy resolution is theoretically estimated to be about E/δE = 45 in a few keV energy range, where E and δE are the energy and energy spread, respectively. However, the experimentally measured energy resolution E/δ is lower than the theoretical estimation. The possible explanations for having a lower resolution are energy and spatial distributions of the incident ion beam. We prepared crystalline Si surface, rough-finished Si crystal, and carbon surface. The surface of single crystal Si was roughened by a sand paper to prepare a rough-finished Si target. Figure 2 shows field-emission type secondary electron microscope (FE-SEM) images of rough-finished Si and carbon plate. Both material surfaces have micrometer-order structures. The rough-finished Si surface shows hill-like structure. Carbon plate has a fine structure in the form of small laminae and consists of smaller grains compared to roughfinished Si. Laser-scanning microscopy shows a few to 10 μm surface roughness for the rough-finished Si.
Rough-finished Si (x 8000)
Carbon plate (x 8000)
FIG. 2. FE-SEM images of carbon plate and rough-finished Si targets. The rough-finished Si surface consists of hill-like structure. Carbon sample shows a fine structure in the form of small laminae and consists of smaller grains compared to rough-finished Si.
FIG. 3. Intensities of reflected H+ and H− ions as functions of the reflection angle β at the incident angles from α = 10◦ to 50◦ for (a) the crystalline Si surface and (b) rough-finished Si. They are plotted from α = 10◦ to 30◦ for the carbon plate in (c). The intensity of rough-finished Si and carbon plate is reduced and approximately one fourth of that of crystalline Si surface.
III. RESULTS AND DISCUSSION
Figures 3(a)–3(c) show the intensities of the reflected H+ and H− ions as functions of β at the incident angles from α = 10◦ to 50◦ for the crystalline Si surface and roughfinished Si, and from α = 10◦ to 30◦ for the carbon surface. The intensity was derived from each spectrum by integrating the Gaussian fit. The negative ion intensity is always smaller than the positive ion intensity. The intensities decrease and the width of the angle distribution increases for increasing α. It indicates multiple scattering of the hydrogen particles inside the surface, because the penetration depth increases at higher incident angle and number of scattering also increases, resulting in the reduction of reflected intensities at each reflection angle β and wider angular distribution. Figure 3 clearly shows the effect of the surface roughness on the angle distribution and intensity of reflected ions. The intensity for the rough-finished Si target is strongly reduced and approximately one-fourth of that for single crystalline Si surface. The angular distribution is wider for both ion species for all incident angles. The intensity in the carbon surface is also much smaller and is the same order of magnitude as the rough-finished Si target. The width of angular distribution is also slightly larger than that for crystalline Si surface. It can be due to the effects of the surface orientation of the
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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
02C311-3
Tanaka et al.
Rev. Sci. Instrum. 85, 02C311 (2014)
FIG. 4. (a) and (b) Peak energies in the energy distribution functions of H+ and H− ions as a function of the reflection angle β at the incident angles from α = 10◦ to 50◦ for (a) the crystalline Si surface and (b) the rough-finished Si.
rough-finished Si and carbon surface. They have many microscopic angled surfaces as seen in the FE-SEM image in Fig. 2. They can cause the deviation in the incident angle with respect to the beam direction, which is much small for the single crystalline Si, and may result in wider angle distribution of the reflected particles. We observe large yield of negative ions, which may have the same order as the positive ion yield, although we do not measure the absolute fraction of the yield. Such large negative ion yield has been observed for other materials, but it cannot be understood by a conventional surface charge transfer model because of large work functions as described in the Introduction.7–9, 11, 12, 18 Recent quantum mechanical study suggested the interference effect on charge exchange of hydrogen on surface to explain such large fraction of negative hydrogen ion scattering.19, 20 When hydrogen atom is close to the surface, through the ion-surface interaction, the affinity level can shift closer to the Fermi level of the solid target. Thus, the quantum mechanical interference could lead to a large negative ion fraction. Figures 4(a) and 4(b) show the energy of the reflected H+ and H− ions as a function of β for the crystalline Si and roughfinished Si. Both ion species had shown a similar trend, and clear difference between two samples is not observed. The reflected energy decreases monotonically with β. This monotonic decrease of the energy (E) with β is described by the formula based on the classical binary collision, i.e., ⎡ 1/2 ⎤2 2 M12 M 2 ⎣cos θ ± ⎦ , − sin2 θ E/E0 = (M1 + M2 )2 M1 where M1 is the mass of the projectile, M2 the mass of the target, θ the laboratory scattering angle defined as the angle with respect to the incident beam trajectory, and E0 is the primary beam energy.1 However, we note that the change in the
FIG. 5. Temperature dependence of the reflected ion intensities for (a) the crystalline Si surface, (b) that for the rough-finished Si, and (c) that for the carbon surface. Incident angle, reflection angle, and H+ beam energy are α = 10◦ , β = 10◦ , and 1 keV, respectively. The crystalline Si and roughfinished Si crystals did not have clear temperature dependence, while the carbon surface showed decrease of H+ and increase of H− intensities with increasing temperature. Note that ordinate scales from 1 for crystalline Si, and 0.1 in rough-finished Si and carbon surfaces.
energy at α = 10◦ for the crystalline Si shows a broad peak, not monotonic decrease, at β = 10◦ as shown in the reflection angle dependence of the H− energy in Fig. 4(a). This suggests that above simple formula is valid for the neutral particle scattering, but does not include the charge transfer process. Thus, it cannot describe the behaviour of beam energy shown in Fig. 4. In particular at grazing incident angles we may have to take into account the quantum mechanical process. Temperature dependence of the reflected ion intensities for the three samples is shown in Fig. 5. These three samples do not show significant temperature dependence for both the reflected H+ and H− ion intensities. This indicates that the sample surfaces are kept clean without contamination such as water.21 On the other hand, for the carbon surface it seems that the H+ (H− ) intensity slightly increases (decreases) with increasing temperature. We have confirmed reproducibility of this trend. Note that the incident beam intensity is kept constant during the measurement, the reflected beam intensity is normalized by the target current, and the reflected ion intensities of these three samples do not show time-dependence. The
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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
02C311-4
Tanaka et al.
temperature-induced change in the reflected ion intensity may be due to the hydrogen retention. A small quantity of the injected hydrogen atoms is absorbed possibly changing the densities of states in the vicinity of the surface layer. However, at present the origin of this temperature-dependent phenomenon only observed for the carbon sample remains unclear and requires further study.
IV. CONCLUSION
We have studied the effects of surface roughness on the angle distributions and intensities of positive and negative ions formed by proton beam injected onto crystalline Si, rough-finished Si, and carbon surfaces. The intensity of reflected negative ions is always smaller than that of reflected positive ions. The magnitude of negative ion current is of the same order of positive ion current. Such large negative ion yield for the samples with large work functions cannot be explained by a conventional surface charge transfer model. The reflection intensities in the rough-finished Si and carbon are much smaller compared to that in the Si crystal. Angular distributions for the rough-finished Si are wider for both reflected ion species than those for Si crystal. Clear temperature dependence of the reflected ion intensities is not observed for all samples in the measured temperature range from 300 to 440 K. Only the carbon sample seems to show a very weak dependence on the reflected ions.
ACKNOWLEDGMENTS
This work was supported by National Institute for Fusion Science general collaboration Project Nos. NIFS11KBAR001 and NIFS12KBAR004, and partially supported by Japan Society for the Promotion of Science (JSPS) KAKENHI Grant-
Rev. Sci. Instrum. 85, 02C311 (2014)
in-Aid for Scientific Research (A) Grant No. 24246152. We thank M. Kisaki for his kind support. 1 H.
Niehus, W. Heiland, and E. Taglauer, Surf. Sci. Rep. 17, 213 (1993).
2 M. Wada, T. Kenmotsu, Y. Matsumoto, M. Nishiura, M. Sasao, K. Tsumori,
and H. Yamaoka, Plasma Dev. Operat. 17, 132 (2009). Yamaoka, Y. Matsumoto, M. Nishiura, M. Sasao, and M. Wada, J. Nucl. Mater. 337–339, 942 (2005). 4 H. Yamaoka, Y. Matsumoto, M. Nishiura, K. Tsumori, H. Sugawara, S. Takeuchi, K. Shinto, M. Sasao, and M. Wada, Rev. Sci. Instrum. 77, 03C301 (2006). 5 H. Yamaoka, N. Tanaka, Y. Matsumoto, M. Nishiura, K. Tsumori, H. Sugawara, S. Takeuchi, S. Shinto, A. Okamoto, M. Sasao, and M. Wada, J. Nucl. Mater. 363–365, 1304 (2007). 6 N. Tanaka, H. Yamaoka, M. Nishiura, K. Tsumori, T. Nagamura, M. Sasao, T. Kenmotsu, Y. Matsumoto, and M. Wada, J. Nucl. Mater. 390–391, 1035 (2009). 7 H. Verbeek, W. Eckstein, and R. S. Bhattacharya, Surf. Sci. 95, 380 (1980). 8 R. S. Bhattacharya, W. Eckstein, and H. Verbeek, Surf. Sci. 93, 563 (1980). 9 K. Tsumori, W. R. Koppers, R. M. Heeren, M. F. Kadodwafa, J. H. M. Beijersbergen, and A. W. Kleyn, J. Appl. Phys. 81, 6390 (1997). 10 P. Wurz, R. Schletti, and M. R. Aellig, Surf. Sci. 373, 56 (1997). 11 M. Maazouz, A. G. Borisov, V. A. Esaulov, J. P. Gauyacq, L. Guillemot, S. Lacombe, and D. Teillet-Billy, Phys. Rev. B 55, 13869 (1997). 12 M. Maazouz, L. Guillemot, V. A. Esaulov, and D. J. O’Connor, Surf. Sci. 398, 49 (1998). 13 J. Los and J. J. C. Geerlings, Phys. Rep. 190, 133 (1990). 14 T. Kenmotsu, M. Wada, H. Yamaoka, M. Nishiura, and K. Tsumori, J. Plasma Fusion Res. Ser. 8, 442 (2009), available online at http://www.jspf.or.jp/JPFRS/PDF/Vol8/jpfrs2009_08-0442.pdf. 15 P. Kumar, A. Ahmad, C. Pardanaud, M. Carrère, J. M. Layet, G. Cartry, F. Silva, A. Gicquel, and R. Engeln, J. Phys. D: Appl. Phys. 44, 372002 (2011). 16 G. Cartry, L. Schiesko, C. Hopf, A. Ahmad, M. Carrère, J. M. Layet, P. Kumar, and R. Engeln, Phys. Plasmas 19, 063503 (2012). 17 L. Schiesko, M. Carrère, J. M. Layet, and G. Cartry, Plasma Sources Sci. Technol. 19, 045016 (2010). 18 L. Guillemot, S. Lacombe, and V. A. Esaulov, Nucl. Instrum. Methods Phys. Res. B 164–165, 601 (2000). 19 E. Sanchez, L. Guillemot, and V. A. Esaulov, Phys. Rev. Lett. 83, 428 (1999). 20 M. Romero, A. Iglesias-García, and E. C. Goldberg, J. Phys.: Condens. Matter 24, 045004 (2012). 21 J. N. DeFazio and B. L. Peko, J. Vac. Sci. Technol. A 22, 2251 (2004). 3 H.
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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 02C311 (2014)
Effects of roughness and temperature on low-energy hydrogen positive and negative ion reflection from silicon and carbon surfacesa) N. Tanaka,1,b) S. Kato,2 T. Miyamoto,2 M. Nishiura,3 K. Tsumori,3 Y. Matsumoto,4 T. Kenmotsu,5 A. Okamoto,6 S. Kitajima,6 M. Sasao,7 M. Wada,2 and H. Yamaoka8 1
Institute of Laser Engineering, Osaka University, Suita, Osaka 565-0871, Japan Graduate School of Science and Engineering, Doshisha University, Kyotanabe, Kyoto 610-0321, Japan 3 National Institute for Fusion Science, Toki, Gifu 509-5292, Japan 4 Tokushima Bunri University, Yamashiro, Tokushima 770-8514, Japan 5 Faculty of Life and Medical Sciences, Doshisha University, Kyotanabe, Kyoto 610-0321, Japan 6 School of Engineering, Tohoku University, Aoba, Sendai, Miyagi 980-8579, Japan 7 Organization for Research Initiatives and Development, Doshisha University, Kyotanabe, Kyoto 610-0321, Japan 8 RIKEN SPring-8 Center, Mikazuki, Sayo, Hyogo 679-5148, Japan 2
(Presented 11 September 2013; received 7 September 2013; accepted 26 November 2013; published online 7 January 2014) Angle-resolved energy distribution functions of positive and negative hydrogen ions produced from a rough-finished Si surface under 1 keV proton irradiation have been measured. The corresponding distribution from a crystalline surface and a carbon surface are also measured for comparison. Intensities of positive and negative ions from the rough-finished Si are substantially smaller than those from crystalline Si. The angular distributions of these species are broader for rough surface than the crystalline surface. No significant temperature dependence for positive and negative ion intensities is observed for all samples in the temperature range from 300 to 400 K. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4855455] I. INTRODUCTION
Low-energy particle interaction with matters has attracted much attention in areas covering from fundamental science to technological applications including surface characterisation, surface atom manipulation, nanotechnology, and film growth.1 Hydrogen ion-solid surface interaction leads to formation of both positive and negative ions at the surface, and is of particular importance in wall conditioning of plasma confinement devices. Despite the large amount of research efforts, the basic process of the beam interaction with matter has not been fully understood yet. We have studied the fundamental processes of beam-surface interactions by measuring the positive and negative ions scattered from the solid surface by injecting low-energy (1–3 keV) light ion beams.2 Both positive and negative ions from molybdenum,3 tungsten,4 vanadium alloy,5 and carbon nano wall6 were observed. The intensities of negative ions formed on these samples were the same order of magnitude. Similar negative ion yields were also observed in the graphite, Mg, Al, Ag, Si, and diamond.7–12 The negative ion yields from such high work function surfaces of graphite, Si, and diamond are too large to be explained by conventional surface charge transfer models.13 We also found that the surface structure affected the reflection property.14 The results suggest that the surface structure is an important factor determining the intensities of scattered positive and negative ions. On the other hand, most experiments of the beam-surface interaction have been performed at room temperature. For the 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)/02C311/4/$30.00
temperature effect on the negative ion yield from surfaces other than alkali metal seeded materials, only some experimental results have been reported.15–17 In this paper we report the surface roughness effect on the reflection properties at Si targets for a 1 keV H+ beam injection. Reflection from Si and carbon surfaces are measured for comparison. Temperature effects of these materials are also studied. II. EXPERIMENT
Figure 1(a) shows a schematic diagram of the experimental setup. A compact multi-cusp magnetic field type ion source is used to produce positive ions. The discharge current is typically about 0.3–0.5 A to obtain a few tens of nA at the target position for a beam energy of 1–2 keV. Deflector and einzel lens systems are installed to improve the beam optics. The beam travels through a water-cooled selector magnet to be separated from ions of different mass. The analyzer chamber is connected to the beam line by a 5 cm long and 5 mm inner diameter tube, which keeps high vacuum in analyzer chamber, where the residual pressure was kept below 2 × 10−5 Pa. The beam collimation system made of a pair of slits in front of the target defines a beam size of 1 (horizontal) mm × 5 (vertical) mm. Incident H+ beam energy was 1 keV throughout this series of the experiments. The analyzer system consists of a target, a magnet momentum analyzer with a MCP (multi-channel plate) detector, and a target heating system. A water-cooled, 90◦ -bending magnet used as a momentum analyzer is installed on a table which can rotate approximately from −60◦ to 60◦ . In many other experiments, the measurements have been performed under the limitation of α + β = const., where α
85, 02C311-1
© 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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
02C311-2
(a)
Tanaka et al.
Rev. Sci. Instrum. 85, 02C311 (2014) Magnetic Momentum Analyzer (MMA)
Ion source Thermocouple Sector magnet
Deflectors Einzel lens Collimator
(b) Incident angle α
Reflection angle Magnetic momentum analyzer
β Target
Sample holder & heater
MCP
Analyzer chamber
MCP
FIG. 1. (a) Schematic diagram of the experimental setup. Extracted ion beam is transported to the analyzer chamber. Both positive and negative ions produced by reflection of incident proton at the surface are measured by the magnetic momentum analyzer installed in vacuum. The target temperature is controlled by a small heater behind the target. (b) Definition of incident angle (α) and reflection angle (β).
and β are incident and reflection angles with respect to the target plane as shown in Fig. 1(b). Contrast to this restriction, our system has an advantage that the angles of the target and the analyzer are controlled independently. Furthermore, we can measure both positive and negative ions produced from the target by changing the polarity of the analyzer magnet current during a single scan. The analyzer energy resolution is theoretically estimated to be about E/δE = 45 in a few keV energy range, where E and δE are the energy and energy spread, respectively. However, the experimentally measured energy resolution E/δ is lower than the theoretical estimation. The possible explanations for having a lower resolution are energy and spatial distributions of the incident ion beam. We prepared crystalline Si surface, rough-finished Si crystal, and carbon surface. The surface of single crystal Si was roughened by a sand paper to prepare a rough-finished Si target. Figure 2 shows field-emission type secondary electron microscope (FE-SEM) images of rough-finished Si and carbon plate. Both material surfaces have micrometer-order structures. The rough-finished Si surface shows hill-like structure. Carbon plate has a fine structure in the form of small laminae and consists of smaller grains compared to roughfinished Si. Laser-scanning microscopy shows a few to 10 μm surface roughness for the rough-finished Si.
Rough-finished Si (x 8000)
Carbon plate (x 8000)
FIG. 2. FE-SEM images of carbon plate and rough-finished Si targets. The rough-finished Si surface consists of hill-like structure. Carbon sample shows a fine structure in the form of small laminae and consists of smaller grains compared to rough-finished Si.
FIG. 3. Intensities of reflected H+ and H− ions as functions of the reflection angle β at the incident angles from α = 10◦ to 50◦ for (a) the crystalline Si surface and (b) rough-finished Si. They are plotted from α = 10◦ to 30◦ for the carbon plate in (c). The intensity of rough-finished Si and carbon plate is reduced and approximately one fourth of that of crystalline Si surface.
III. RESULTS AND DISCUSSION
Figures 3(a)–3(c) show the intensities of the reflected H+ and H− ions as functions of β at the incident angles from α = 10◦ to 50◦ for the crystalline Si surface and roughfinished Si, and from α = 10◦ to 30◦ for the carbon surface. The intensity was derived from each spectrum by integrating the Gaussian fit. The negative ion intensity is always smaller than the positive ion intensity. The intensities decrease and the width of the angle distribution increases for increasing α. It indicates multiple scattering of the hydrogen particles inside the surface, because the penetration depth increases at higher incident angle and number of scattering also increases, resulting in the reduction of reflected intensities at each reflection angle β and wider angular distribution. Figure 3 clearly shows the effect of the surface roughness on the angle distribution and intensity of reflected ions. The intensity for the rough-finished Si target is strongly reduced and approximately one-fourth of that for single crystalline Si surface. The angular distribution is wider for both ion species for all incident angles. The intensity in the carbon surface is also much smaller and is the same order of magnitude as the rough-finished Si target. The width of angular distribution is also slightly larger than that for crystalline Si surface. It can be due to the effects of the surface orientation of the
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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
02C311-3
Tanaka et al.
Rev. Sci. Instrum. 85, 02C311 (2014)
FIG. 4. (a) and (b) Peak energies in the energy distribution functions of H+ and H− ions as a function of the reflection angle β at the incident angles from α = 10◦ to 50◦ for (a) the crystalline Si surface and (b) the rough-finished Si.
rough-finished Si and carbon surface. They have many microscopic angled surfaces as seen in the FE-SEM image in Fig. 2. They can cause the deviation in the incident angle with respect to the beam direction, which is much small for the single crystalline Si, and may result in wider angle distribution of the reflected particles. We observe large yield of negative ions, which may have the same order as the positive ion yield, although we do not measure the absolute fraction of the yield. Such large negative ion yield has been observed for other materials, but it cannot be understood by a conventional surface charge transfer model because of large work functions as described in the Introduction.7–9, 11, 12, 18 Recent quantum mechanical study suggested the interference effect on charge exchange of hydrogen on surface to explain such large fraction of negative hydrogen ion scattering.19, 20 When hydrogen atom is close to the surface, through the ion-surface interaction, the affinity level can shift closer to the Fermi level of the solid target. Thus, the quantum mechanical interference could lead to a large negative ion fraction. Figures 4(a) and 4(b) show the energy of the reflected H+ and H− ions as a function of β for the crystalline Si and roughfinished Si. Both ion species had shown a similar trend, and clear difference between two samples is not observed. The reflected energy decreases monotonically with β. This monotonic decrease of the energy (E) with β is described by the formula based on the classical binary collision, i.e., ⎡ 1/2 ⎤2 2 M12 M 2 ⎣cos θ ± ⎦ , − sin2 θ E/E0 = (M1 + M2 )2 M1 where M1 is the mass of the projectile, M2 the mass of the target, θ the laboratory scattering angle defined as the angle with respect to the incident beam trajectory, and E0 is the primary beam energy.1 However, we note that the change in the
FIG. 5. Temperature dependence of the reflected ion intensities for (a) the crystalline Si surface, (b) that for the rough-finished Si, and (c) that for the carbon surface. Incident angle, reflection angle, and H+ beam energy are α = 10◦ , β = 10◦ , and 1 keV, respectively. The crystalline Si and roughfinished Si crystals did not have clear temperature dependence, while the carbon surface showed decrease of H+ and increase of H− intensities with increasing temperature. Note that ordinate scales from 1 for crystalline Si, and 0.1 in rough-finished Si and carbon surfaces.
energy at α = 10◦ for the crystalline Si shows a broad peak, not monotonic decrease, at β = 10◦ as shown in the reflection angle dependence of the H− energy in Fig. 4(a). This suggests that above simple formula is valid for the neutral particle scattering, but does not include the charge transfer process. Thus, it cannot describe the behaviour of beam energy shown in Fig. 4. In particular at grazing incident angles we may have to take into account the quantum mechanical process. Temperature dependence of the reflected ion intensities for the three samples is shown in Fig. 5. These three samples do not show significant temperature dependence for both the reflected H+ and H− ion intensities. This indicates that the sample surfaces are kept clean without contamination such as water.21 On the other hand, for the carbon surface it seems that the H+ (H− ) intensity slightly increases (decreases) with increasing temperature. We have confirmed reproducibility of this trend. Note that the incident beam intensity is kept constant during the measurement, the reflected beam intensity is normalized by the target current, and the reflected ion intensities of these three samples do not show time-dependence. The
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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
02C311-4
Tanaka et al.
temperature-induced change in the reflected ion intensity may be due to the hydrogen retention. A small quantity of the injected hydrogen atoms is absorbed possibly changing the densities of states in the vicinity of the surface layer. However, at present the origin of this temperature-dependent phenomenon only observed for the carbon sample remains unclear and requires further study.
IV. CONCLUSION
We have studied the effects of surface roughness on the angle distributions and intensities of positive and negative ions formed by proton beam injected onto crystalline Si, rough-finished Si, and carbon surfaces. The intensity of reflected negative ions is always smaller than that of reflected positive ions. The magnitude of negative ion current is of the same order of positive ion current. Such large negative ion yield for the samples with large work functions cannot be explained by a conventional surface charge transfer model. The reflection intensities in the rough-finished Si and carbon are much smaller compared to that in the Si crystal. Angular distributions for the rough-finished Si are wider for both reflected ion species than those for Si crystal. Clear temperature dependence of the reflected ion intensities is not observed for all samples in the measured temperature range from 300 to 440 K. Only the carbon sample seems to show a very weak dependence on the reflected ions.
ACKNOWLEDGMENTS
This work was supported by National Institute for Fusion Science general collaboration Project Nos. NIFS11KBAR001 and NIFS12KBAR004, and partially supported by Japan Society for the Promotion of Science (JSPS) KAKENHI Grant-
Rev. Sci. Instrum. 85, 02C311 (2014)
in-Aid for Scientific Research (A) Grant No. 24246152. We thank M. Kisaki for his kind support. 1 H.
Niehus, W. Heiland, and E. Taglauer, Surf. Sci. Rep. 17, 213 (1993).
2 M. Wada, T. Kenmotsu, Y. Matsumoto, M. Nishiura, M. Sasao, K. Tsumori,
and H. Yamaoka, Plasma Dev. Operat. 17, 132 (2009). Yamaoka, Y. Matsumoto, M. Nishiura, M. Sasao, and M. Wada, J. Nucl. Mater. 337–339, 942 (2005). 4 H. Yamaoka, Y. Matsumoto, M. Nishiura, K. Tsumori, H. Sugawara, S. Takeuchi, K. Shinto, M. Sasao, and M. Wada, Rev. Sci. Instrum. 77, 03C301 (2006). 5 H. Yamaoka, N. Tanaka, Y. Matsumoto, M. Nishiura, K. Tsumori, H. Sugawara, S. Takeuchi, S. Shinto, A. Okamoto, M. Sasao, and M. Wada, J. Nucl. Mater. 363–365, 1304 (2007). 6 N. Tanaka, H. Yamaoka, M. Nishiura, K. Tsumori, T. Nagamura, M. Sasao, T. Kenmotsu, Y. Matsumoto, and M. Wada, J. Nucl. Mater. 390–391, 1035 (2009). 7 H. Verbeek, W. Eckstein, and R. S. Bhattacharya, Surf. Sci. 95, 380 (1980). 8 R. S. Bhattacharya, W. Eckstein, and H. Verbeek, Surf. Sci. 93, 563 (1980). 9 K. Tsumori, W. R. Koppers, R. M. Heeren, M. F. Kadodwafa, J. H. M. Beijersbergen, and A. W. Kleyn, J. Appl. Phys. 81, 6390 (1997). 10 P. Wurz, R. Schletti, and M. R. Aellig, Surf. Sci. 373, 56 (1997). 11 M. Maazouz, A. G. Borisov, V. A. Esaulov, J. P. Gauyacq, L. Guillemot, S. Lacombe, and D. Teillet-Billy, Phys. Rev. B 55, 13869 (1997). 12 M. Maazouz, L. Guillemot, V. A. Esaulov, and D. J. O’Connor, Surf. Sci. 398, 49 (1998). 13 J. Los and J. J. C. Geerlings, Phys. Rep. 190, 133 (1990). 14 T. Kenmotsu, M. Wada, H. Yamaoka, M. Nishiura, and K. Tsumori, J. Plasma Fusion Res. Ser. 8, 442 (2009), available online at http://www.jspf.or.jp/JPFRS/PDF/Vol8/jpfrs2009_08-0442.pdf. 15 P. Kumar, A. Ahmad, C. Pardanaud, M. Carrère, J. M. Layet, G. Cartry, F. Silva, A. Gicquel, and R. Engeln, J. Phys. D: Appl. Phys. 44, 372002 (2011). 16 G. Cartry, L. Schiesko, C. Hopf, A. Ahmad, M. Carrère, J. M. Layet, P. Kumar, and R. Engeln, Phys. Plasmas 19, 063503 (2012). 17 L. Schiesko, M. Carrère, J. M. Layet, and G. Cartry, Plasma Sources Sci. Technol. 19, 045016 (2010). 18 L. Guillemot, S. Lacombe, and V. A. Esaulov, Nucl. Instrum. Methods Phys. Res. B 164–165, 601 (2000). 19 E. Sanchez, L. Guillemot, and V. A. Esaulov, Phys. Rev. Lett. 83, 428 (1999). 20 M. Romero, A. Iglesias-García, and E. C. Goldberg, J. Phys.: Condens. Matter 24, 045004 (2012). 21 J. N. DeFazio and B. L. Peko, J. Vac. Sci. Technol. A 22, 2251 (2004). 3 H.
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: 128.138.73.68 On: Mon, 22 Dec 2014 23:50:17
