Tomographic analysis of the nonthermal x-ray bursts during disruption instability in the T-10 tokamak P. V. Savrukhin, A. I. Ermolaeva, E. A. Shestakov, and A. V. Khramenkov Citation: Review of Scientific Instruments 85, 103508 (2014); doi: 10.1063/1.4898333 View online: http://dx.doi.org/10.1063/1.4898333 View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/85/10?ver=pdfcov Published by the AIP Publishing Articles you may be interested in First results obtained from the soft x-ray pulse height analyzer on experimental advanced superconducting tokamak Rev. Sci. Instrum. 81, 063501 (2010); 10.1063/1.3443572 Measurements of the nonthermal x-ray emission in the T-10 tokamak using CdTe detectors Rev. Sci. Instrum. 73, 4243 (2002); 10.1063/1.1519938 Investigation of coupling of magnetohydrodynamic modes by soft x-ray computer tomography on the WT-3 tokamak Phys. Plasmas 9, 3378 (2002); 10.1063/1.1486449 Tangential x-ray imaging system for analysis of the small-scale modes in the T-10 tokamak Rev. Sci. Instrum. 72, 1668 (2001); 10.1063/1.1344175 X-ray line diagnostics on the Tore Supra tokamak Rev. Sci. Instrum. 70, 308 (1999); 10.1063/1.1149326
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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 103508 (2014)
Tomographic analysis of the nonthermal x-ray bursts during disruption instability in the T-10 tokamak P. V. Savrukhin,1,2,a) A. I. Ermolaeva,1 E. A. Shestakov,1 and A. V. Khramenkov1 1 2
National Research Center “Kurchatov Institute”, 123182 Moscow, Russia ITER RF Domestic Agency, Institution Project Center ITER” 123182 Moscow, Russia
(Received 3 August 2014; accepted 6 October 2014; published online 22 October 2014) Non-thermal x-ray radiation (Eγ up to 150 keV) is measured in the T-10 tokamaks during disruption instability using two sets of CdTe detectors (10 vertical and 7 horizontal view detectors). Special narrow cupper tubes collimators with lead screening and CdTe detectors integrated with amplifiers inside metallic containers provides enhanced spatial resolution of the system (r ∼ 3 cm) and assures protection from the parasitic hard x-ray (Eγ up to 1.5 MeV) and electromagnetic loads during disruption. Spatial localization of the nonthermal x-ray emissivity is reconstructed using tomographic Cormack technique with SVD matrix inversion. Analysis indicated appearance of an intensive nonthermal x-ray bursts during initial stage of the disruptions at high density. The bursts are characterized by repetitive spikes (2–3 kHz) of the x-ray emissivity from the plasma core area. Analysis indicated that the spikes can be connected with acceleration of the non-thermal electrons in enhanced longitudinal electric fields induced during energy quench at the disruption instability. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4898333] I. INTRODUCTION
Acceleration of the non-thermal electron beams is one of the critical problems of disruption instability in tokamaks due to damage of the plasma-facing constructions during the beam-wall interaction.1 The electron beams acceleration is typically connected in the case with enhanced longitudinal electric fields (E ∼ 1–5 V/m) created during disruption due to the plasma cooling and influx of impurities. While general mechanisms of the acceleration are described in the literature,1 details of the initial stage of the process are still under study. This is connected in part with difficulties of analysis of the nonthermal electrons localization within the hot bulk plasma (see, review by Coda2 ). Moreover, experiments3–5 have indicated that the nonthermal electrons during a disruption are often characterized by non-uniform distribution across the plasma column. Such uneven localization of the beams can be connected with enhanced generation of the nonthermal electrons during reconnection of the magnetic field lines in areas of the magnetohydrodynamic (MHD) modes localization.6, 7 Analysis of spatial and temporal evolution of the localized nonthermal electrons (Eγ ∼ 15–150 keV) is one of the primarily task of the present experiment. Spatial evolution of the plasma perturbations during disruption instability in tokamaks is typically identified using tomographic reconstruction of the local x-ray emissivity from the chord-integrated x-ray measurements. The x-ray measurements in T-10 were initially based on arrays of Si detectors placed at multiple locations around the torus (see Ref. 7). However, Si surface-barrier detectors are typically charactera) Author to whom correspondence should be addressed. Electronic mail:
[email protected]
0034-6748/2014/85(10)/103508/4/$30.00
ized by reduced sensitivity to the non-thermal radiation at energies higher than Eγ > 15 keV. In order to increase sensitivity to the non-thermal radiation (with simultaneous retention of the high temporal resolution) the cadmium telluride (CdTe) detectors are used in modern experiments.2, 8–13 The cadmium based detectors are characterized by high photon absorption efficiency due to the high atomic number (Z = 48/52) and high density (6 g/cm3 ) and high resistivity because of the wide band gap (1.4 eV at 300 K).14 The tomographic reconstruction technique based on CdTe detectors was realized previously in two-camera systems built in TORE SUPRA15 and TCV.11, 12 The work performed on TORE SUPRA15 was concerned with analysis of the nonthermal electrons during experiments with application of the Low-Hybrid waves (see also experiments with tangential hard x-ray imaging in PBX-M16 ), while evolution of the fast electrons during a disruption instability was not analyzed in details. Essential new results of the present experiments in T-10 tokamak is evaluation of spatial and temporal evolution of the localized nonthermal electrons (Eγ ∼ 15–150 keV) during initial stage of the disruption instability. Lack of the detailed tomographic studies of the nonthermal x-ray fluxes during disruption instability is generally connected with complications due to high parasitic electromagnetic pick-up on detecting system and strong interference of the hard x-ray and gamma ray fluxes. To minimize such effects x-ray cameras with narrow tube collimators and lead screening and special amplifiers integrated with detectors are used in the T-10 experiments. This paper is organized as follows. Experimental setup and diagnostic technique are described in Sec. II. Section III represents results of analysis of the nonthermal electrons in plasma with high density in the T-10 tokamak (major and minor radii R0 = 1.5 m, a = 0.35 m accordingly).
85, 103508-1
© 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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
103508-2
Savrukhin et al.
Rev. Sci. Instrum. 85, 103508 (2014)
II. DIAGNOSTIC SYSTEM AND TOMOGRAPHIC RECONSTRUCTION TECHNIQUE
Nonthermal x-ray radiation is measured in the T-10 tokamak using two arrays of CdTe detectors placed in the same toroidal location. The CdTe detectors are arranged in two cameras placed in vertical and “horizontal” ports (Fig. 1). The system provides measurements of the x-ray intensity along 10 vertical and 6 “horizontal” chords with spatial resolution δr = 3–5 cm in the plasma area r/a < 1. Spatial resolution can be adjusted by reciprocating of the camera arrays to the plasma vessel on shot to shot basis. Detectors are separated from the vacuum vessel with Al foil (thickness δ = 2 mm) providing low cut-off energy of order of Eγ ∼ 15 keV. Additional Al and Pb filters (δ = 2–6 mm) are used for selection of the nonthermal x-ray radiation energy (Eγ > 15 keV). Selection of the detectors line-of-sight is provided with guide-tubes with rectangular cross-section (10 mm × 5 mm) placed in metallic cameras filled with lead pebbles. Additional shielding from the hard x-ray radiation (Eγ > 0.5–2 MeV) is provided by outer lead blocks. The CdTe detectors are standard semiconductors with sensitive area 50 mm2 (5 × 10 mm) and sensitive region thickness δ ∼ 1 mm. Operation voltage of the detectors 10–40 V, leakage current 10–8 A at room temperature. The detectors are attached to the preamplifiers with conversion rate 107 – 108 V/A and frequency bandwidth up to 100 kHz. Operation of the detectors is regulated with the use of remote control and power systems placed inside the T-10 control room. Relative calibration of the detectors is provided at Laboratory
CdTe detectors
(b)
(d)
Lead shielding
L
r
p
R
R0
(a)
(c)
FIG. 1. (a) and (b) Schematic view of the CdTe arrays in the T-10 tokamak. (c) and (d) Images of the vertical and horizontal CdTe arrays with collimator box and additional lead shielding. Also shown, polar coordinates used for tomographic reconstruction of the x-ray emissivity.
bench using led-diode with amplitude modulation and standard x-ray tube (Eγ < 45 keV) and in situ in the T-10 tokamak by changing of the detectors positions in reference T-10 plasma discharges. The system allows measurements of the x-ray fluxes 105 –109 cm−2 s−1 in energy range 2.5—150 keV. Data storage is provided by two independent acquisition systems based on CAMAC ADC modules (8 μs, 8 bit) and ADC card (3 μs, 12 bit) realized in PCI frame. Data analysis incorporates set of programs for calibration, visualization, and modeling of the signals in the MATLAB computing environment. Reconstruction of the local x-ray emissivity g(r,θ ) from measured chord-integrated x-ray intensity I(p,ψ) is based on analytic method17 of inversion of the integral equation: I(p,ψ) = L(p,ψ ) G(r,θ ) dL. The reconstruction is considered in polar coordinates: r,θ (see Fig. 1(c)). Each detector is characterized by impact parameter (p) and pitch angle (ψ) between the perpendicular line to the detector line of sight L and equatorial plane θ = 0◦ . The reconstruction is based on splitting of the x-ray intensities I(p,ψ) and g(r,θ ) in Fourier rows along the poloidal coordinates and calculation of the Chebishev polynomial expansions using SVD inversion technique (see Ref. 7). III. EXPERIMENTAL RESULTS
The non-thermal x-ray intensity perturbations are analyzed using CdTe diagnostic system in the T–10 tokamak in ohmically heated plasma with relatively high density (central line-averaged electron density, ne up to 3.5 × 1019 m–3 ). Typical evolution of the plasma parameters observed in the experiments is shown in Fig. 2. Additional gas puffing at the quasi-stationary stage of the discharge leads to increase of the electron density (see, ne at t > 330–370 ms in Fig. 2(a)) and is accompanied by abrupt growth of the m = 2, n = 1 magnetic perturbations at t = 350 ms (m, n are poloidal and toroidal mode numbers). Subsequent slowing down of the mode rotation (mode “locking”) and growth of the magnetic perturbations lead to the discharge termination at t ∼ 754–770 ms. The density limit disruption is initiated with an energy quench accompanied by bursts of the non-thermal x-ray radiation (see Fig. 2(b)). The non-thermal x-ray bursts are observed with CdTe detectors as well as by the hard x-ray NaI monitor. Appearance of the non-thermal x-ray burst at the energy quench is a typical feature of all disruptions under study while intensity of the burst is different in various plasma conditions. It is observed that amplitude of the non-thermal x-ray bursts is reduced up to 5–15 times in plasma with increased rate of additional gas puffing. Contour plot of the non-thermal x-ray intensity measured using vertical array of the CdTe detectors is shown in Fig. 2(c). The non-thermal x-ray bursts are observed most clearly at the central CdTe detectors placed in vertical array characterized by direct view of the rail limiter. This indicates indirectly interaction of the non-thermal beams with the inner surface of the limiter at the initial stage of the disruption. Typical feature of the T-10 experiments is modulation of the nonthermal x-ray bursts with typical repetition rate 2.5–3 kHz.
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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
Savrukhin et al.
Rev. Sci. Instrum. 85, 103508 (2014)
Density limit
3 1
Thermal quench
Gas detector (2-10keV)
NaI (0.5-3.0 MeV) Non-thermal x-ray bursts
0
Ixray (a.u.)
100
0 0
Non-thermal x-ray bursts 1/ ~ 2.5-3 kHz
450
CdTe09 (on-axis)
452
1.7
454
456
458
Time , (ms)
Nonthermal x-ray contour plot
0 1
0 20 0 -20 1
“Quiescent” m=2 mode Voltage spikes CdTe05 (on-axis)
0 820
840
(b) + 0.3
CdTe16 (off-axis) 464
(c)
Hard x-ray bursts
NaI (0.5-3.0 MeV)
t1
Non-thermal x-ray bursts
Time, (ms)
880
(b)
Ixray
Height, (m)
0 1
0 1
Bp2 (a.u.)
Locked mode
0 1
(a)
Current collapse
IHXR (a.u.)
0 4
T10-64771 0.2
Uloop (V)
I p, Bp2, , 19 3 (a.u.) (10 m- ) (MA)
IHXR, ISXR, (a.u.) (a.u.)
Current collapse
350 Time, (ms) 400
Major radii, (m)
(a)
Ip (MA)
T10-64040 0.2
Ixray (a.u.)
103508-3
- 0.3 1.2
1.5
1.3 454
456
458
Time, (ms)
464
FIG. 2. (a) and (b) Time evolution of the plasma parameters in the T-10 tokamak in discharge with density limit disruption. Here Ip is plasma current, ne is electron density, δBp2 is poloidal magnetic field perturbations, and IHXR , Isxr, and Ixray are intensities of the hard (0.5–3.0 MeV), soft (2–10 keV), and nonthermal (15–150 keV) x-ray intensities measured using NaI monitor, multi-wire gas detector, and vertical CdTe detectors. (c) Contour plot of the nonthermal x-ray intensity measured using vertical array of the CdTe detectors.
Different type of the plasma disruption is represented in Fig. 3. In the case, large-scale m = 2, n = 1 magnetic perturbations are induced in plasma with relatively low density at the beginning of the discharge. In contrast with the high density limit disruption magnetic perturbations are characterized by almost unchanged rotation frequency and amplitude of the m = 2, n = 1 mode through the whole duration of the plasma discharge. Hard x-ray radiation intensity is modulated in the case with the same frequency as one of the m = 2 “quiescent” mode. Results of the tomographic reconstruction of the nonthermal x-ray intensity perturbations during the disruption are shown in Fig. 3(b). The x-ray emissivity is increased within the plasma core. The x-ray radiation appears in the case mainly as a result of the forward bremsstrahlung from the residual plasma. During the disruption the x-ray radiation spikes appear also at the plasma periphery which is associated with interaction of the nonthermal electrons with the limiters. It should be pointed out that Cormack tomographic reconstruction is based on splitting of the x-ray intensities in Fourier rows along the poloidal coordinates. The harmonics
Major radii, (m) 1.8
FIG. 3. (a) Time evolution of the plasma parameters in the T-10 tokamak in discharge with density limit disruption. Here Ip is plasma current, δBp2 is poloidal magnetic field perturbation, Uloop is loop voltage, and IHXR, Ixray are intensities of the hard (0.5–3.0 MeV) and nonthermal (>100 keV) xray intensities measured using NaI monitor and vertical CdTe detector. (b) Tomographic images of the nonthermal x-ray emissivity reconstructed during disruption (t1 = 857.17 ms).
m = 0, cosθ , sinθ , cos2θ are considered in present experiments with two arrays of the detectors. Such harmonics are typically used in the tomographic analysis of x-ray intensity in tokamaks due to low-m helical symmetry of the plasma perturbations observed in the experiments. Plasma in T-10 has circular cross-section and such symmetry may be applicable to the non-thermal x-ray radiation; however, more sophisticated tomographic reconstruction without additional assumptions (such as constant emission on flux surfaces) is required for detailed analysis of the localized non-thermal electron beams (see Ref. 12). Space resolution is also restricted in present experiments by limited number of detectors in “horizontal” array. Nevertheless, the local emissivity images shown in Fig. 3(b) can qualitatively indicate position of the nonthermal electrons beams during disruption in T-10. In conclusion, installation of the 2D CdTe detector arrays in the T-10 tokamak provides tomographic reconstruction of the non-thermal x-ray emissivity during disruption instability. Analysis has indicated appearance of the nonthermal x-ray bursts during initial stage of the disruptions at high density. The bursts are characterized by repetitive spikes (2–3 kHz) of the x-ray emissivity from the plasma core area. Analysis indicated that the spikes can be connected with acceleration of the non-thermal electrons in enhanced longitudinal electric fields induced during energy quench at the disruption instability.
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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
103508-4
Savrukhin et al.
ACKNOWLEDGMENTS
The author is thankful for V. M. Trukhin and A. V. Sushkov for technical support of the gas detector and x-ray PHA system during the experiments. The work is supported in part by ROSATOM No. N.4h.44.90.13.1101. 1 ITER
Expert Group on Disruption, Plasma Control, and MHD and ITER Physics Basis Editors, Nucl. Fusion 39, 2251 (1999). 2 S. Coda, Rev. Sci. Instrum. 79, 10F501 (2008). 3 I. Entrop, R. Jaspers, N. J. Lopez-Cardozo et al., Plasma Phys. Controlled Fusion 41, 377 (1999). 4 R. D. Gill, B. Alper, A. W. Edwards et al., Nucl. Fusion 40, 163 (2000). 5 I. Klimanov, A. Fasoli, T. P. Goodman, and the TCV team, Plasma Phys. Controlled Fusion 49, L1 (2007). 6 E. Priest and T. Forbes, Magnetic Reconnection, (Cambridge University Press, 2000).
Rev. Sci. Instrum. 85, 103508 (2014) 7 P.
V. Savrukhin, Phys. Rev. Lett. 86, 3036 (2001). Imbeaux and Y. Peysson, Phys. Rev. Lett. 84, 2873 (2000). 9 Y. Peysson, S. Coda, and F. Imbeaux, Nucl. Instrum. Methods Phys. Res., Sect. A 458, 269 (2001). 10 R. O’Connell, D. J. Den Hartog, C. B. Forest, and R. W. Harvey, Rev. Sci. Instrum. 74, 2001 (2003). 11 J. Kamleitner, S. Coda, J. Decker, J. P. Graves, and S. Gnesin, “Study of suprathermal electron dynamics by energy-resolved tomography of hard Xray emission on the TCV tokamak,” in Proceedings of the 40th EPS Conference on Plasma Physics, Espoo, Finland, 2013, Europhysics Conference Abstracts, vol. 37D, p. 5.127. 12 S. Gnesin, S. Coda, J. Decker, and Y. Peysson, Rev. Sci. Instrum. 79, 10F504 (2008). 13 P. V. Savrukhin, Rev. Sci. Instrum. 73, 4243 (2002). 14 T. Takahashi and S. Watanabe, IEEE Trans. Nucl. Sci. 48, 950 (2001). 15 Y. Peysson and F. Imbeaux, Rev. Sci. Instrum. 70, 3987 (1999). 16 S. Von Goeler, S. Jones, R. Kaita et al., Rev. Sci. Instrum. 66, 585 (1995). 17 A. M. Cormack, J. Appl. Phys. 34, 2722 (1963). 8 F.
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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 103508 (2014)
Tomographic analysis of the nonthermal x-ray bursts during disruption instability in the T-10 tokamak P. V. Savrukhin,1,2,a) A. I. Ermolaeva,1 E. A. Shestakov,1 and A. V. Khramenkov1 1 2
National Research Center “Kurchatov Institute”, 123182 Moscow, Russia ITER RF Domestic Agency, Institution Project Center ITER” 123182 Moscow, Russia
(Received 3 August 2014; accepted 6 October 2014; published online 22 October 2014) Non-thermal x-ray radiation (Eγ up to 150 keV) is measured in the T-10 tokamaks during disruption instability using two sets of CdTe detectors (10 vertical and 7 horizontal view detectors). Special narrow cupper tubes collimators with lead screening and CdTe detectors integrated with amplifiers inside metallic containers provides enhanced spatial resolution of the system (r ∼ 3 cm) and assures protection from the parasitic hard x-ray (Eγ up to 1.5 MeV) and electromagnetic loads during disruption. Spatial localization of the nonthermal x-ray emissivity is reconstructed using tomographic Cormack technique with SVD matrix inversion. Analysis indicated appearance of an intensive nonthermal x-ray bursts during initial stage of the disruptions at high density. The bursts are characterized by repetitive spikes (2–3 kHz) of the x-ray emissivity from the plasma core area. Analysis indicated that the spikes can be connected with acceleration of the non-thermal electrons in enhanced longitudinal electric fields induced during energy quench at the disruption instability. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4898333] I. INTRODUCTION
Acceleration of the non-thermal electron beams is one of the critical problems of disruption instability in tokamaks due to damage of the plasma-facing constructions during the beam-wall interaction.1 The electron beams acceleration is typically connected in the case with enhanced longitudinal electric fields (E ∼ 1–5 V/m) created during disruption due to the plasma cooling and influx of impurities. While general mechanisms of the acceleration are described in the literature,1 details of the initial stage of the process are still under study. This is connected in part with difficulties of analysis of the nonthermal electrons localization within the hot bulk plasma (see, review by Coda2 ). Moreover, experiments3–5 have indicated that the nonthermal electrons during a disruption are often characterized by non-uniform distribution across the plasma column. Such uneven localization of the beams can be connected with enhanced generation of the nonthermal electrons during reconnection of the magnetic field lines in areas of the magnetohydrodynamic (MHD) modes localization.6, 7 Analysis of spatial and temporal evolution of the localized nonthermal electrons (Eγ ∼ 15–150 keV) is one of the primarily task of the present experiment. Spatial evolution of the plasma perturbations during disruption instability in tokamaks is typically identified using tomographic reconstruction of the local x-ray emissivity from the chord-integrated x-ray measurements. The x-ray measurements in T-10 were initially based on arrays of Si detectors placed at multiple locations around the torus (see Ref. 7). However, Si surface-barrier detectors are typically charactera) Author to whom correspondence should be addressed. Electronic mail:
[email protected]
0034-6748/2014/85(10)/103508/4/$30.00
ized by reduced sensitivity to the non-thermal radiation at energies higher than Eγ > 15 keV. In order to increase sensitivity to the non-thermal radiation (with simultaneous retention of the high temporal resolution) the cadmium telluride (CdTe) detectors are used in modern experiments.2, 8–13 The cadmium based detectors are characterized by high photon absorption efficiency due to the high atomic number (Z = 48/52) and high density (6 g/cm3 ) and high resistivity because of the wide band gap (1.4 eV at 300 K).14 The tomographic reconstruction technique based on CdTe detectors was realized previously in two-camera systems built in TORE SUPRA15 and TCV.11, 12 The work performed on TORE SUPRA15 was concerned with analysis of the nonthermal electrons during experiments with application of the Low-Hybrid waves (see also experiments with tangential hard x-ray imaging in PBX-M16 ), while evolution of the fast electrons during a disruption instability was not analyzed in details. Essential new results of the present experiments in T-10 tokamak is evaluation of spatial and temporal evolution of the localized nonthermal electrons (Eγ ∼ 15–150 keV) during initial stage of the disruption instability. Lack of the detailed tomographic studies of the nonthermal x-ray fluxes during disruption instability is generally connected with complications due to high parasitic electromagnetic pick-up on detecting system and strong interference of the hard x-ray and gamma ray fluxes. To minimize such effects x-ray cameras with narrow tube collimators and lead screening and special amplifiers integrated with detectors are used in the T-10 experiments. This paper is organized as follows. Experimental setup and diagnostic technique are described in Sec. II. Section III represents results of analysis of the nonthermal electrons in plasma with high density in the T-10 tokamak (major and minor radii R0 = 1.5 m, a = 0.35 m accordingly).
85, 103508-1
© 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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
103508-2
Savrukhin et al.
Rev. Sci. Instrum. 85, 103508 (2014)
II. DIAGNOSTIC SYSTEM AND TOMOGRAPHIC RECONSTRUCTION TECHNIQUE
Nonthermal x-ray radiation is measured in the T-10 tokamak using two arrays of CdTe detectors placed in the same toroidal location. The CdTe detectors are arranged in two cameras placed in vertical and “horizontal” ports (Fig. 1). The system provides measurements of the x-ray intensity along 10 vertical and 6 “horizontal” chords with spatial resolution δr = 3–5 cm in the plasma area r/a < 1. Spatial resolution can be adjusted by reciprocating of the camera arrays to the plasma vessel on shot to shot basis. Detectors are separated from the vacuum vessel with Al foil (thickness δ = 2 mm) providing low cut-off energy of order of Eγ ∼ 15 keV. Additional Al and Pb filters (δ = 2–6 mm) are used for selection of the nonthermal x-ray radiation energy (Eγ > 15 keV). Selection of the detectors line-of-sight is provided with guide-tubes with rectangular cross-section (10 mm × 5 mm) placed in metallic cameras filled with lead pebbles. Additional shielding from the hard x-ray radiation (Eγ > 0.5–2 MeV) is provided by outer lead blocks. The CdTe detectors are standard semiconductors with sensitive area 50 mm2 (5 × 10 mm) and sensitive region thickness δ ∼ 1 mm. Operation voltage of the detectors 10–40 V, leakage current 10–8 A at room temperature. The detectors are attached to the preamplifiers with conversion rate 107 – 108 V/A and frequency bandwidth up to 100 kHz. Operation of the detectors is regulated with the use of remote control and power systems placed inside the T-10 control room. Relative calibration of the detectors is provided at Laboratory
CdTe detectors
(b)
(d)
Lead shielding
L
r
p
R
R0
(a)
(c)
FIG. 1. (a) and (b) Schematic view of the CdTe arrays in the T-10 tokamak. (c) and (d) Images of the vertical and horizontal CdTe arrays with collimator box and additional lead shielding. Also shown, polar coordinates used for tomographic reconstruction of the x-ray emissivity.
bench using led-diode with amplitude modulation and standard x-ray tube (Eγ < 45 keV) and in situ in the T-10 tokamak by changing of the detectors positions in reference T-10 plasma discharges. The system allows measurements of the x-ray fluxes 105 –109 cm−2 s−1 in energy range 2.5—150 keV. Data storage is provided by two independent acquisition systems based on CAMAC ADC modules (8 μs, 8 bit) and ADC card (3 μs, 12 bit) realized in PCI frame. Data analysis incorporates set of programs for calibration, visualization, and modeling of the signals in the MATLAB computing environment. Reconstruction of the local x-ray emissivity g(r,θ ) from measured chord-integrated x-ray intensity I(p,ψ) is based on analytic method17 of inversion of the integral equation: I(p,ψ) = L(p,ψ ) G(r,θ ) dL. The reconstruction is considered in polar coordinates: r,θ (see Fig. 1(c)). Each detector is characterized by impact parameter (p) and pitch angle (ψ) between the perpendicular line to the detector line of sight L and equatorial plane θ = 0◦ . The reconstruction is based on splitting of the x-ray intensities I(p,ψ) and g(r,θ ) in Fourier rows along the poloidal coordinates and calculation of the Chebishev polynomial expansions using SVD inversion technique (see Ref. 7). III. EXPERIMENTAL RESULTS
The non-thermal x-ray intensity perturbations are analyzed using CdTe diagnostic system in the T–10 tokamak in ohmically heated plasma with relatively high density (central line-averaged electron density, ne up to 3.5 × 1019 m–3 ). Typical evolution of the plasma parameters observed in the experiments is shown in Fig. 2. Additional gas puffing at the quasi-stationary stage of the discharge leads to increase of the electron density (see, ne at t > 330–370 ms in Fig. 2(a)) and is accompanied by abrupt growth of the m = 2, n = 1 magnetic perturbations at t = 350 ms (m, n are poloidal and toroidal mode numbers). Subsequent slowing down of the mode rotation (mode “locking”) and growth of the magnetic perturbations lead to the discharge termination at t ∼ 754–770 ms. The density limit disruption is initiated with an energy quench accompanied by bursts of the non-thermal x-ray radiation (see Fig. 2(b)). The non-thermal x-ray bursts are observed with CdTe detectors as well as by the hard x-ray NaI monitor. Appearance of the non-thermal x-ray burst at the energy quench is a typical feature of all disruptions under study while intensity of the burst is different in various plasma conditions. It is observed that amplitude of the non-thermal x-ray bursts is reduced up to 5–15 times in plasma with increased rate of additional gas puffing. Contour plot of the non-thermal x-ray intensity measured using vertical array of the CdTe detectors is shown in Fig. 2(c). The non-thermal x-ray bursts are observed most clearly at the central CdTe detectors placed in vertical array characterized by direct view of the rail limiter. This indicates indirectly interaction of the non-thermal beams with the inner surface of the limiter at the initial stage of the disruption. Typical feature of the T-10 experiments is modulation of the nonthermal x-ray bursts with typical repetition rate 2.5–3 kHz.
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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
Savrukhin et al.
Rev. Sci. Instrum. 85, 103508 (2014)
Density limit
3 1
Thermal quench
Gas detector (2-10keV)
NaI (0.5-3.0 MeV) Non-thermal x-ray bursts
0
Ixray (a.u.)
100
0 0
Non-thermal x-ray bursts 1/ ~ 2.5-3 kHz
450
CdTe09 (on-axis)
452
1.7
454
456
458
Time , (ms)
Nonthermal x-ray contour plot
0 1
0 20 0 -20 1
“Quiescent” m=2 mode Voltage spikes CdTe05 (on-axis)
0 820
840
(b) + 0.3
CdTe16 (off-axis) 464
(c)
Hard x-ray bursts
NaI (0.5-3.0 MeV)
t1
Non-thermal x-ray bursts
Time, (ms)
880
(b)
Ixray
Height, (m)
0 1
0 1
Bp2 (a.u.)
Locked mode
0 1
(a)
Current collapse
IHXR (a.u.)
0 4
T10-64771 0.2
Uloop (V)
I p, Bp2, , 19 3 (a.u.) (10 m- ) (MA)
IHXR, ISXR, (a.u.) (a.u.)
Current collapse
350 Time, (ms) 400
Major radii, (m)
(a)
Ip (MA)
T10-64040 0.2
Ixray (a.u.)
103508-3
- 0.3 1.2
1.5
1.3 454
456
458
Time, (ms)
464
FIG. 2. (a) and (b) Time evolution of the plasma parameters in the T-10 tokamak in discharge with density limit disruption. Here Ip is plasma current, ne is electron density, δBp2 is poloidal magnetic field perturbations, and IHXR , Isxr, and Ixray are intensities of the hard (0.5–3.0 MeV), soft (2–10 keV), and nonthermal (15–150 keV) x-ray intensities measured using NaI monitor, multi-wire gas detector, and vertical CdTe detectors. (c) Contour plot of the nonthermal x-ray intensity measured using vertical array of the CdTe detectors.
Different type of the plasma disruption is represented in Fig. 3. In the case, large-scale m = 2, n = 1 magnetic perturbations are induced in plasma with relatively low density at the beginning of the discharge. In contrast with the high density limit disruption magnetic perturbations are characterized by almost unchanged rotation frequency and amplitude of the m = 2, n = 1 mode through the whole duration of the plasma discharge. Hard x-ray radiation intensity is modulated in the case with the same frequency as one of the m = 2 “quiescent” mode. Results of the tomographic reconstruction of the nonthermal x-ray intensity perturbations during the disruption are shown in Fig. 3(b). The x-ray emissivity is increased within the plasma core. The x-ray radiation appears in the case mainly as a result of the forward bremsstrahlung from the residual plasma. During the disruption the x-ray radiation spikes appear also at the plasma periphery which is associated with interaction of the nonthermal electrons with the limiters. It should be pointed out that Cormack tomographic reconstruction is based on splitting of the x-ray intensities in Fourier rows along the poloidal coordinates. The harmonics
Major radii, (m) 1.8
FIG. 3. (a) Time evolution of the plasma parameters in the T-10 tokamak in discharge with density limit disruption. Here Ip is plasma current, δBp2 is poloidal magnetic field perturbation, Uloop is loop voltage, and IHXR, Ixray are intensities of the hard (0.5–3.0 MeV) and nonthermal (>100 keV) xray intensities measured using NaI monitor and vertical CdTe detector. (b) Tomographic images of the nonthermal x-ray emissivity reconstructed during disruption (t1 = 857.17 ms).
m = 0, cosθ , sinθ , cos2θ are considered in present experiments with two arrays of the detectors. Such harmonics are typically used in the tomographic analysis of x-ray intensity in tokamaks due to low-m helical symmetry of the plasma perturbations observed in the experiments. Plasma in T-10 has circular cross-section and such symmetry may be applicable to the non-thermal x-ray radiation; however, more sophisticated tomographic reconstruction without additional assumptions (such as constant emission on flux surfaces) is required for detailed analysis of the localized non-thermal electron beams (see Ref. 12). Space resolution is also restricted in present experiments by limited number of detectors in “horizontal” array. Nevertheless, the local emissivity images shown in Fig. 3(b) can qualitatively indicate position of the nonthermal electrons beams during disruption in T-10. In conclusion, installation of the 2D CdTe detector arrays in the T-10 tokamak provides tomographic reconstruction of the non-thermal x-ray emissivity during disruption instability. Analysis has indicated appearance of the nonthermal x-ray bursts during initial stage of the disruptions at high density. The bursts are characterized by repetitive spikes (2–3 kHz) of the x-ray emissivity from the plasma core area. Analysis indicated that the spikes can be connected with acceleration of the non-thermal electrons in enhanced longitudinal electric fields induced during energy quench at the disruption instability.
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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
103508-4
Savrukhin et al.
ACKNOWLEDGMENTS
The author is thankful for V. M. Trukhin and A. V. Sushkov for technical support of the gas detector and x-ray PHA system during the experiments. The work is supported in part by ROSATOM No. N.4h.44.90.13.1101. 1 ITER
Expert Group on Disruption, Plasma Control, and MHD and ITER Physics Basis Editors, Nucl. Fusion 39, 2251 (1999). 2 S. Coda, Rev. Sci. Instrum. 79, 10F501 (2008). 3 I. Entrop, R. Jaspers, N. J. Lopez-Cardozo et al., Plasma Phys. Controlled Fusion 41, 377 (1999). 4 R. D. Gill, B. Alper, A. W. Edwards et al., Nucl. Fusion 40, 163 (2000). 5 I. Klimanov, A. Fasoli, T. P. Goodman, and the TCV team, Plasma Phys. Controlled Fusion 49, L1 (2007). 6 E. Priest and T. Forbes, Magnetic Reconnection, (Cambridge University Press, 2000).
Rev. Sci. Instrum. 85, 103508 (2014) 7 P.
V. Savrukhin, Phys. Rev. Lett. 86, 3036 (2001). Imbeaux and Y. Peysson, Phys. Rev. Lett. 84, 2873 (2000). 9 Y. Peysson, S. Coda, and F. Imbeaux, Nucl. Instrum. Methods Phys. Res., Sect. A 458, 269 (2001). 10 R. O’Connell, D. J. Den Hartog, C. B. Forest, and R. W. Harvey, Rev. Sci. Instrum. 74, 2001 (2003). 11 J. Kamleitner, S. Coda, J. Decker, J. P. Graves, and S. Gnesin, “Study of suprathermal electron dynamics by energy-resolved tomography of hard Xray emission on the TCV tokamak,” in Proceedings of the 40th EPS Conference on Plasma Physics, Espoo, Finland, 2013, Europhysics Conference Abstracts, vol. 37D, p. 5.127. 12 S. Gnesin, S. Coda, J. Decker, and Y. Peysson, Rev. Sci. Instrum. 79, 10F504 (2008). 13 P. V. Savrukhin, Rev. Sci. Instrum. 73, 4243 (2002). 14 T. Takahashi and S. Watanabe, IEEE Trans. Nucl. Sci. 48, 950 (2001). 15 Y. Peysson and F. Imbeaux, Rev. Sci. Instrum. 70, 3987 (1999). 16 S. Von Goeler, S. Jones, R. Kaita et al., Rev. Sci. Instrum. 66, 585 (1995). 17 A. M. Cormack, J. Appl. Phys. 34, 2722 (1963). 8 F.
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: 131.156.157.31 On: Wed, 26 Nov 2014 14:36:59
