Upgrade and absolute calibration of the JET scintillator-based fast-ion loss detector

 View Online  Export Citation RESEARCH ARTICLE | APRIL 23 2021 Upgrade and absolute calibration of the JET scintillatorbased fast-ion loss detector  Special Collection: Proceedings of the 23rd Topical Conference on High-Temperature Plasma Diagnostics J. F. Rivero-Rodriguez  ; C. Perez Von Thun; M. Garcia-Muñoz ; P. Beaumont; V. Kiptily ; J. Garcia-Lopez; V. Goloborodko; M. C. Jimenez-Ramos; M. Rodriguez-Ramos; K. Schoepf; V. Yavorskij; JET Contributors Rev. Sci. Instrum. 92, 043553 (2021) https://doi.org/10.1063/5.0043760 Articles You May Be Interested In A rotary and reciprocating scintillator based fast-ion loss detector for the MAST-U tokamak Rev. Sci. Instrum. (September 2018) Modeling the response of a fast ion loss detector using orbit tracing techniques in a neutral beam promptloss study on the DIII-D tokamak Rev. Sci. Instrum. (October 2010) Self-adaptive diagnostic of radial fast-ion loss measurements on the ASDEX Upgrade tokamak (invited) Rev. Sci. Instrum. (May 2021) 23 January 2025 07:00:30 Review of Scientific Instruments ARTICLE scitation.org/journal/rsi Upgrade and absolute calibration of the JET scintillator-based fast-ion loss detector Cite as: Rev. Sci. Instrum. 92, 043553 (2021); doi: 10.1063/5.0043760 Submitted: 11 January 2021 •Accepted: 1 April 2021 • Published Online: 23 April 2021 J. F. Rivero-Rodriguez,1,2,a) C. Perez Von Thun,3M. Garcia-Muñoz,2,4 P. Beaumont,5V. Kiptily,5 J. Garcia-Lopez,2,4 V. Goloborodko,6M. C. Jimenez-Ramos,2M. Rodriguez-Ramos,7K. Schoepf,6V. Yavorskij,6,b) and JET Contributorsc) AFFILIATIONS 1Departamento de Ingenieria Mecanica y Fabricacion, Universidad de Sevilla, Camino de los Descubrimientos s/n, 41092 Sevilla, Spain 2Centro Nacional de Aceleradores (CNA) (Universidad de Sevilla, CSIC, Junta de Andalucia), Seville, Spain 3Institute of Plasma Physics and Laser Microfusion, Hery 23, 01-497 Warsaw, Poland 4Departamento de Fisica Atomica, Molecular y Nuclear, Universidad de Sevilla, 41012 Sevilla, Spain 5CCFE, Culham Science Centre, Abingdon, Oxon OX14 3DB, United Kingdom 6OEAW, Institute for Theoretical Physics, University of Innsbruck, Innsbruck, Austria 7Laboratory for Ion beam Interactions, Ruder Boskovic Institute, Zagreb, Croatia Note: Paper published as part of the Special Topic on Proceedings of the 23rd Topical Conference on High-Temperature Plasma Diagnostics. a)Author to whom correspondence should be addressed: [email protected] b)Deceased. c)See the author list of Joffrin et al., Nucl. Fusion 59, 112021 (2019). ABSTRACT The JET FILD is a scintillator-based Fast-ion Loss Detector optimized to measure fusion-born alpha-particle losses. This work covers its upgrade and absolute calibration in preparation for the following JET DT experiments. A fast scintillator material (TG-Green) has been installed in the JET FILD. A heater jacket is installed around the fiber bundle, responsible for transmitting the light from the scintillator plate, to anneal the fiber obscuring due to neutron damage. The JET FILD has been upgraded with a 1 Mpx camera and 2MHz photomultiplier data acquisition hardware. Full-orbit simulations give an estimate of the shading effects on the scintillator plate of the first wall structures and provide a synthetic signal of the JET FILD. A detector instrument function enables absolute values of fast-ion losses using calibration factors. The calibration factors are made available in a shot-to-shot basis for the characterized species and energies and with corrections for the diagnostic conditions. The fast acquisition system sets the Nyquist frequency (1MHz) above the typical mode frequencies (≈102kHz), thus making it possible to identify MHD-induced fast-ion losses. Published under license by AIP Publishing. https://doi.org/10.1063/5.0043760 I. INTRODUCTION The Joint European Torus (JET) is currently the only operational tokamak that can perform DT operations, thus being able to explore the reactor-like conditions for future devices where ignition is envisaged, such as ITER or DEMO. One of the main objectives of the JET DT experimental campaign starting in 2021 (DTE2) is to studythefusion-productα-particleconfinementandtransport.1JET is equipped with state-of-the-art fast particle diagnostics that have been successfully exploited in DD operations for decades. However, the expected fusion power of 15MW and the budget of 1.7 ×1021 14MeV neutrons in the DTE2 campaign make it necessary to prepare the JET plasma facing components and diagnostics. In this work, the JET scintillator-based fast-ion loss detector (JET FILD)2has been calibrated and upgraded for the DT experimental campaign, similar to the calibrations and upgrades carried Rev. Sci. Instrum. 92, 043553 (2021); doi: 10.1063/5.0043760 92, 043553-1 Published under license by AIP Publishing 23 January 2025 07:00:30 Review of Scientific Instruments ARTICLE scitation.org/journal/rsi out in TFTR.3,4 It has been optimized to detect and provide absolute measurements of fusion-product 3.5MeV α-particle losses, among other fast-ion species. Its temporal resolution and signal-to-noise ratio have been improved, enabling the detection of the MHD phenomena that cause the losses, in the order of 102kHz. The FILD is a unique diagnostic to detect fast-ion losses and resolve their velocityspace and loss frequency. It has been successfully employed to identify fast-ion losses in most large tokamaks, such as TFTR,5,6 ASDEX Upgrade,7–9 NSTX,10 and DIII-D,11 among others. This paper is structured as follows: The FILD working principle, its original status, and the upgrades are presented in Sec. II. Section III describes the numerical tools developed to construct a synthetic signal, accounting for the different fast-ion populations in the JET plasmas. The absolute calibration of the FILD diagnostic is described in Sec. IV. The data analyses enabled by the diagnostic upgrade and the calibration are given in Sec. V. Section VI summarizes the current status and outlook of the diagnostic. II. DIAGNOSTIC UPGRADE The Fast-Ion Loss Detector (FILD)5works as a magnetic spectrometer, collimating the escaping fast-ions that reach the probe head onto a scintillator plate. Their velocity-space, i.e., gyroradius and pitch angle [arccos(v∥/v)], is inferred by their strike point on the scintillator plate, as illustrated in Fig. 1. The light is recorded by a camera and a set of photomultiplier tubes (PMTs). The JET FILD front face is radially 5mm behind the plasma limiters shown in Fig. 1(a) and 30○below the midplane in standard plasma shapes. This position has been found to be a good compromise between low heat loads from the plasma and high signal intensity. The diagnostic measurement range, mainly determined by the probe head geometry, is 30–130mm gyroradius (15% resolution) and 35○–85○ pitch angle (5% resolution). This makes it possible to measure the FIG. 1. (a) FILD probe head and neighbor elements at the JET tokamak. (b) FILD probe head and the escaping fast-ion orbit reaching the aperture. (c) Scintillator plate where the collimated ions are dispersed using the tokamak magnetic field and characterized by their gyroradius and pitch angle.12 TABLE I. Species, energy, gyroradius (B=3.5T), and fusion reaction of the fusionborn fast-ions in JET operation. Species Energy (Mev) Gyroradius (mm) Reaction α-particles 3.5 77 D +T Protons 3.0 72 D +D Tritium 1.0 72 D +D 3He 0.8 32 D +D main fusion-born fast-ions in JET, as shown in Table I, among other fast-ions produced by auxiliary heating. The main limitation of the JET FILD was the slow decay time (2ms) of its original scintillator material (P56). This impedes the identification of the MHD phenomena that cause fast-ion losses and limit any further enhancement of the data acquisition. For this reason, a fast scintillator material (TG-Green) with a similar scintillator yield and faster decay time (490ns)13 was installed, replacing the old one.TheTG-Greenemitsat540nm(greenlight).Therefore,abandpass filter (510–570nm) was included at the bottom end of the optical system to prevent the detection of fiber optic radioluminescence generated by neutron radiation in DT experiments, whose wavelength is out of this range,14,15 while letting the scintillator emission through. Thenewscintillatormaterialenablestheuseofhighersampling frequencies in the data acquisition system. Hence, the Photometrics Cascade 512B camera (256 ×36 px2, 20Hz, CCD) was replaced by a Photron Ultima APX-i2 (256 ×128 px2, 30 kHz, CMOS). The sampling rate of the new CMOS camera has not been fully exploited yet. It is currently being operated at 60Hz and maximum exposure time due to the relatively low intensity of the signal in DD operation. However, it is estimated that the signal intensity in DT operation will allow sampling frequencies up to 1 kHz. The PMT’s original data acquisition system (5 kHz sampling rate) was replaced by low noiseTrans-ImpedanceAmplifiers(TIAs)and14-bitBAD2Analog- to-Digital Converters (ADCs). The TIA +BAD2 ADC electronic achieves a very low noise signal (rms =3 nA) with 4 selectable gain levels (125, 250, 500, and 1000 kΩ) and 1MHz bandwidth. The gain level has been set to 1000 kΩin DD experiments and will be revised to the signal intensity in the DT campaign. The light is transmitted to the optical box outside the torus hall through the original JET FILD 95 ×36 coherent fiber bundle, made of the pure fused silica core and fluorine doped silica cladding by Fiberguide Industries. The ≈3×1016 n/s flux that could be achieved in DTE2 will cause damage to the fiber, reducing its transmission in a rate down to −13 dB/km 1010n/cm2.16–18 The obscuration can be reverted by heating up the fiber bundle to 200○C, annealing the radiation damage.19 For this reason, the fiber bundle has been put into a customized heater jacket that will heat up the bundle to maintain its transmissivity. A blind fiber installed inside the heater jacket is used to monitor the changes in light transmissivity due to neutron damage and after the annealing. This is known as the Fiber-optic Attenuation Monitor (FAM). The heater jacket installation and first fiber annealing were successful, but it did not reveal any enhancement of the fiber transmission. This suggests that the fiber bundle was already at optimal conditions. Rev. Sci. Instrum. 92, 043553 (2021); doi: 10.1063/5.0043760 92, 043553-2 Published under license by AIP Publishing 23 January 2025 07:00:30 Review of Scientific Instruments ARTICLE scitation.org/journal/rsi FIG. 2. (a) Synthetic signal of DD fusion-product protons at shot No. 92054 (B= 2.6T at the plate). (b) Shading effect caused by the JET 3D wall (Bt=2.6T and Ip=2MA). III. NUMERICAL TOOLS The JET FILD synthetic signal is constructed with a set of codes, some developed in the scope of this work (SNBI, FIDIT, SEDORP, and FOLD). EfipDesign20 calculates the strike map onto the scintillator plate and the expected resolution and collimating factor (fcol =Nions scintillator/Nions pinhole)21 of the system. SNBI calculates the birthdistributionoftheNBIaccountingfordifferentinjectedspecies (p, T, D). Then, FIDIT solves the Fokker–Planck equation based on the six-time step implicit integrator and finite difference scheme to obtain the distribution function of fast-ions in the plasma. FIDIT accounts for orbital effects, Coulomb collisions, ripple-induced diffusion, and ICRF heating of NBI ions on the second harmonic. The code SEDORP calculates the distribution function of charged fusion products, given the fast-ion distribution from FIDIT or other codes such as TRANSP.22 Finally, FOLD carries out full-orbit backward tracing of ion trajectories from the FILD entrance hole into the plasma and forward tracing from the entrance hole to the scintillator plate to obtain the projection of the fast-ion distribution on the scintillator plate. As a result, the FILD synthetic signal can be constructed, as it is shown in Fig. 2(a), where the signal of discharge No. 92054 produced by DD fusion-product protons is simulated using SNBI +FIDIT +SEDORP +FOLD. Moreover, FOLD uses the backward tracing to resolve the scintillator plate domain shadowed by first wall elements, marked in black in Fig. 2(b). The results show that the shading effect is very sensitive to the shape of the flux surfaces at the plasma boundary. Simulations with similar plasma shapes showed that the shadowing effect is negligible for high magneticfieldsandhighplasmacurrents andusuallybecomes important at Bt<3T and Ip<2MA. IV. ABSOLUTE CALIBRATION A. Characterization of the scintillator material Thecharacterizationofthenewscintillatormaterialwascarried out using Ion Beam Analysis (IBA) techniques in a 3 MV tandem accelerator at the Spanish National Center of Accelerators (CNA).23 TG-Green sample plates from the same batch as the JET plate (9 μm, Sarnoff Co.) were employed for the characterization. The sampleswereplacedinascatteringvacuumchamberinthemainline of the accelerator. A fiber optic connected to a spectrometer (previously calibrated), with no intermediate filters, recorded the light emitted by the sample in a solid angle of 2.2 ⋅10−5sr. The incident beam current was monitored in real time to account for the current FIG. 3. Scintillator yield of the TG-Green for α-particles (a), deuterium (b), and protons (c) resolved experimentally and calculated with Birk’s model. fluctuations (3.9 ⋅1010–2.0 ⋅1011ions/cm2s−1). With this, the scintillatoryieldiscalculatedasemitted photonsperincidentionsonthe scintillator material.13 The scintillator yield was resolved for the α-particles, deuterium, and protons in the energy range of 0.5–3MeV, as it is shown in Fig. 3. The scintillator yield was also calculated applying Birk’s model24 to SRIM results25 for comparison. In the case of 3.5MeV αparticles, an experimental yield was not determined. However, due to the fair agreement between the experimental results and Birk’s model, the latter is used to estimate the scintillator yield for 3.5MeV α-particles. In order to measure the scintillator properties at different temperatures, the scattering vacuum chamber was equipped with a temperature-controlled sample holder.26 It was observed that the yield was approximately the same from room temperature (RT) up to 100○C, after which the scintillator yield drastically decreases, as observedinTableII.Thisstudyalsorevealsthationfluencedegrades the scintillator yield while high temperatures anneal the damage. TABLE II. Scintillator yield ratio (YT/YRT) as a function of temperature.26 Temperature (○C) RT 100 200 300 400 500 α-particles 1 1 0.30 0.10 0.017 0.002 Deuterium 1 1 0.41 0.10 0.018 0.002 Protons 1 1 0.47 0.10 0.013 0.001 Rev. Sci. Instrum. 92, 043553 (2021); doi: 10.1063/5.0043760 92, 043553-3 Published under license by AIP Publishing 23 January 2025 07:00:30 Review of Scientific Instruments ARTICLE scitation.org/journal/rsi B. Calibration of the optical system The optical transmissivity and sensor efficiency were characterized in situ. The diagnostic (limiter guide tube +probe head) was extracted from the vacuum vessel and the probe head was removed, leaving direct sight to the optical system. As the probe head could be potentially beryllium contaminated, for health and safety reasons, the limiter guide tube was covered by a vinyl sheet. The calibration of the optical system was carried out with an Ulbricht sphere placed at the front end of the covered limiter guide tube, whose light was transmitted along the optical system and measured by the Photron camera and the PMTs simultaneously. Neutral Density (ND) filters were added to the optical enclosure to modify the light intensity. With this, an instrument function that translates sensor counts into incoming photons from the scintillator plate is produced. The sensor’s instrument function was evaluated for a reference case with a calibration lamp of 5W and a ND filter OD 3.0. The Photron camera image intensifier was set to 640V and 60Hz and the PMT supply voltage to 967.5V. A set of ND filters with different ODs was employed to verify the signal linearity to different light intensities. This verification enables the use of a linear instrument function. The signal linearity is also verified for exposure times of the Photron camera ranging from 60 to 250Hz. The PMT sampling frequency does not have an impact on the signal. A scan in all the detector parameters, i.e., the image intensifier voltage and the PMT supply voltage, provides correction factors to the reference instrument function, in preparation for DT experiments, where a different configuration could be used. A light shield was used to close the light entrance and has measurements of the background noise. Correction factors accounting for the relative sensitivity of each PMT and additional vignetting effect of the optical system are applied to the reference instrument function. The effect of the heater jacket on the fiber transmissivity was measured after the first annealing. Since no enhancement of the transmissivity was observed, the reference instrument function has not been modified. Spectrometric characterization of the calibration lamp and the bandpass and ND filters in the wavelength range of the bandpass filter was carried out to compute correction factors for the instrument function. The vinyl sheet that covered the limiter guide tube was alsocharacterized,givingan additionalattenuationfactorduring the calibration pulses that needs to be compensated in the instrument function. C. Calculation of FILD instrument functions The collimating factor calculated in Sec. III, the scintillator yield from Sec. IV A, and the optical transmissivity and sensor efficiency from Sec. IVBare combined in an instrument function that translates the sensor counts into the ion flux incident on the FILD, following the description in Ref. 27. The instrument function is computed for the scintillator yield at room temperature and the reference optical calibration. It is then corrected for the specific conditions of each pulse (scintillator temperature, sensor parameters, and fiber transmissivity). It provides calibration factors (CFs) for each sensor and each calibrated species andenergyinashot-to-shotbasis.CFsmultiplythesensors’rawdata to translate them into the ion flux, Γtransmitted ions =CF ⋅Γmeasured signal . (1) To obtain absolute measurements of the losses, the CF of the specific sensor and the measured species and energy should be selected. As will be observed in Sec. V, careful analysis of the JET FILD signal is required to distinguish the measured species and energy since only the gyroradius and pitch angle of the losses could be inferred from the JET FILD data. The measured species will depend on the experiment, the heating scheme, and the existing species in the plasma. A correction factor of the fiber transmissivity is used based on theFAMdata.Ithasbeenobservedthatherethetemperaturecorrection does not play an important role because the scintillator plate is keptbelow70○C,thanks totheJETFILDactivecooling.Corrections due to ion fluence and temperature annealing on the scintillator yield are currently not considered in the instrument function but willbeincluded iftheJET FILDsignal is severelyaffected during DT experiments. V. DATA ANALYSIS The diagnostic instrument function can be employed in any discharge since 2017, when the upgraded JET FILD became operative. As an example, the CFs of discharge No. 92394 are resolved. This shot consisted of an Ip=2.2MA, Bt=2.8T, deuterium plasma with a low concentration of hydrogen (1%–3%), 26MW of NBI, and 5MW of ICRH heating. There, fast-ion losses, correlated with n=1 fishbones, were identified.28–30 Figure 4 shows the CMOS camera frame and the PMT14 time trace during the fishboneactivity. Theanalysis of theCMOS camerareflects that thespot at 11.5cm gyroradius is fusion-product 3MeV protons (B=2.18T at the plate). Therefore, the CFs of protons at 3MeV were chosen to translate the sensor counts into the ion flux. The absolute measurements are normalized to the area on the scintillator plate subtendedby theCMOS camera andPMT14. Itcan be observedthat bothsensorsestimatebroadlyconsistentfluxesofthefusion-product losses. Fourier analyses of the data up to 1MHz are provided with the upgraded PMT data acquisition system. This enables the identification of the MHD phenomena that are driving the losses by comparingthe JET FILD signal with the magnetic pickupcoil signal, as shown in Fig. 5, where the JET FILD PMT10 signal can be correlated with n=1,2,3 NTMs on the pickup coils in discharge No. 92054. FIG. 4. (a) JET FILD CMOS camera frame showing the fishbone-driven fusionproduct losses and the field of view of PMT14 in green. (b) Time trace of JET FILD PMT14 showing the peaks correlated with fishbones. Rev. Sci. Instrum. 92, 043553 (2021); doi: 10.1063/5.0043760 92, 043553-4 Published under license by AIP Publishing 23 January 2025 07:00:30 Review of Scientific Instruments ARTICLE scitation.org/journal/rsi FIG. 5. Spectrogram of the JET magnetic pickup coils (a) and JET FILD PMT10 (b)indischargeNo. 92054 showing that thefrequencyofthefast-ion losses match n=1,2,3 NTMs. VI. CONCLUSION The JET FILD is now conditioned for the next DT experimental campaign at JET. The signal provides measurements of fast-ion losses in terms of ions per second, thanks to the absolute calibration of the scintillator material and the optical system. The upgrade of the data acquisition system and the development of numerical toolsmakeit possibleto identifythe MHDphenomena thatare driving the fast-ion losses. The upgraded diagnostic is operating in DD experiments since 2017, benefitting from the drastic improvement of the detector. JET is equipped with two fast-ion loss detectors conditioned for DT experiments: a scintillator-based FILD, which is presented here, and an array of Faraday cups.31 Both signals may be combined to distinguish some properties of the fast-ion losses, such as the species or the energy, when a single diagnostic is not sufficient. ACKNOWLEDGMENTS This work was carried out within the framework of the EUROfusion Consortium and received funding from the Euratom Research and Training Programme 2014–2018 and 2019–2020 under Grant Agreement No. 633053. The views and opinions expressed herein do not necessarily reflect those of the European Commission. This research received funding from the V Plan Propio de Investigaciónde la Universidad deSevilla (Grant No. PP2016-7145). This scientific work was partly supported by the Polish Ministry of Science and Higher Education within the framework of the scientific financial resources in the year 2020 allocated for the realization of the international co-financed project (No. 5118/H2020/EURATOM/2020/2). DATA AVAILABILITY The data that support the findings of this study are available from the corresponding author upon reasonable request. REFERENCES 1E. Joffrin et al.,Nucl. Fusion 59, 112021 (2019). 2S. Baeumel et al.,Rev. Sci. Instrum. 75, 3563 (2004). 3R. L. Boivin et al.,Rev. Sci. Instrum. 63, 4418 (1992). 4D. S. Darrow et al.,Rev. Sci. Instrum. 66, 476 (1995). 5S. J. Zweben, Nucl. Fusion 29, 825 (1989). 6S. J. Zweben et al.,Nucl. Fusion 30, 1551 (1990). 7M. García-Muñoz et al.,Rev. Sci. Instrum. 80, 053503 (2009). 8J. Gonzalez-Martin et al.,Rev. Sci. Instrum. 89, 10I106 (2018). 9J. Gonzalez-Martin et al.,J. Instrum. 14, C11005 (2019). 10D. S. Darrow, Rev. Sci. Instrum. 79, 023502 (2008). 11R. K. Fisher et al.,Rev. Sci. Instrum. 81, 10D307 (2010). 12J. F. Rivero-Rodriguez et al.,Rev. Sci. Instrum. 89, 10I112 (2018). 13M.C. Jimenez-Ramoset al.,Nucl.Instrum. Methods Phys.Res., Sect. B332,216 (2014). 14A. Moroño et al.,J. Nucl. Mater. 258-263, 1889 (1998). 15F. Sato et al.,Fusion Eng. Des. 41, 207 (1998). 16A. T. Ramsey et al.,Rev. Sci. Instrum. 63, 4735 (1992). 17P. D. Morgan, in Proceedings of the 17th Symposium On Fusion Technology (North-Holland, Amsterdam, 1992). 18H. G. Adler et al.,Rev. Sci. Instrum. 66, 904 (1995). 19W. Tighe et al.,Rev. Sci. Instrum. 66, 907 (1995). 20A. Werner et al.,Rev. Sci. Instrum. 72, 780 (2001). 21J. Galdon-Quiroga et al.,Nucl. Fusion 58, 036005 (2018). 22R. V. Budny et al., Nucl. Fusion 32, 429 (1992); available at https:// iopscience.iop.org/article/10.1088/0029-5515/32/3/I07. 23J. Garcia Lopez et al.,Nucl. Instrum. Methods Phys. Res., Sect. B 161-163, 1137 (2000). 24J. B. Birks, Proc. Phys. Soc. London, Sect. A 64, 874 (1951). 25J. F. Ziegler et al.,Nucl. Instrum. Methods Phys. Res., Sect. B 268, 1818 (2010). 26M. Rodriguez-Ramos et al.,Nucl. Instrum. Methods Phys. Res., Sect. B 403, 7 (2017). 27M. Rodriguez-Ramos et al.,Plasma Phys. Controlled Fusion 59, 105009 (2017). 28V. G. Kiptily et al.,Nucl. Fusion 58, 014003 (2017). 29C. P. von Thun et al.,Nucl. Fusion 50, 084009 (2010). 30C. P. von Thun et al.,Nucl. Fusion 51, 053003 (2011). 31P. J. Bonofiglo et al.,Rev. Sci. Instrum. 91, 093502 (2020). Rev. Sci. Instrum. 92, 043553 (2021); doi: 10.1063/5.0043760 92, 043553-5 Published under license by AIP Publishing 23 January 2025 07:00:30