Open Access
Add to BibSonomy
Abstract
We herein report on using a compact and low cost scintillator-camera based neutron detection system for quantitative time-of-flight imaging applications. While powerful pulsed neutron sources emerge and enable unprecedented scientific achievements, one bottleneck is the availability of suitable detectors that provide high count- and high frame- rate capabilities. For imaging applications the achievable spatial resolution/pixel size is obviously another key characteristic. While major effort was so far directed towards the development of neutron counting type imaging detectors, this work demonstrates that a camera based detector system as commonly employed at steady state sources can also be used if a suitable camera is utilized. This is demonstrated at the ESS test beamline (V20) at Helmholtz-Zentrum Berlin by recording the time-of-flight transmission spectrum of steel samples using a CMOS camera at 1 kHz frame rate, revealing the characteristic Bragg edge pattern. This ‘simple’ setup in the current state presents a useful option of neutron detection and has the potential to overcome many of the existing limitations and could provide a reliable alternative for neutron detector technology in general, given that the camera and scintillator technology keep up the current development speed.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Neutron Imaging is now routinely taking advantage of wavelength selective and wavelength dispersive measurements (i.e. using a single wavelength or a broad wavelength band) [1–3]. While additional instrumentation is required for wavelength dispersive measurements at steady state neutron sources, this is inherent to spallation sources using time-of-flight (ToF) when employing a suitable detector. Especially new imaging installations at powerful neutron sources take significant advantage of the contrast mechanism that became possible due to the wavelength selectivity and promise even more to come [4].
The key requirement to fully exploit the potential at pulsed sources is the availability of suitable detectors with high time resolution (frame rate) and minimal readout times, high count rate capabilities, high detection efficiencies and a high spatial resolution (small pixel size). Moreover, a reasonably large Field of View (FoV) is beneficial for many applications and experience shows that this can be efficiently tuned using a camera based system [5–8]. Presently, a compromise between these capabilities needs to be found as not all requirements can be fulfilled by the same detector type at the same time and hence a selection of detectors is usually offered at user facilities at pulsed sources.
Currently utilized ToF imaging detectors are of counting type and include micro-pattern detectors, such as the μNID (Micro-pixel chamber based Neutron Imaging Detector based on 3He) [9,10] and the nGEM (boroncoated Gas Electron Multiplier) [11], pixelated 6Li-glass scintillator detectors (coupled to a multi-anode photomultiplier) [12], GP2 (a gadolinium converter with a CMOS readout sensor) [13] and those based on borated Micro Channel Plates (MCPs), that can be coupled to a delay line readout [14] or most prominently coupled to a Timepix readout [15]. Neutron-sensitive MCP’s have also been coupled to a fast phosphor screen and CMOS camera [16], but no ToF capability has yet been presented. A sophisticated camera type detector is under development in Japan, where a neutron color image intensifier [17] is coupled to a short afterglow phosphor screen, a high speed gated optical image intensifier unit and a high framerate CMOS camera [18,19].
The standard solution for imaging detectors at steady state sources is rather simple: The incoming neutrons are converted into visible light by a scintillator and the light output is recorded on the other side using a light sensitive camera. Typical exposure times for neutron images are on the order of several seconds [1,8] with the standard solution so far being based on scientific CCD cameras. Such camera detector systems can be gated to acquire images for very short exposure times and hence be used for ToF imaging applications [2,20], however at the cost of very low efficiency since only a fraction of the neutron spectrum can be used within the repetition rate.
While the use of CMOS (Complementary Metal Oxide Semiconductor) cameras to be used in neutron imaging detectors started to be mentioned only a few years back [6,7], they are now routinely being used for fast/dynamic ‘white beam’ imaging applications [21]. A major advantage of such detector system is its modularity, e.g. it can accommodate scintillators of different types and thickness and the resolution/FoV can be tuned by the use of optical lenses. The two most commonly employed scintillating materials for ‘white beam’ imaging detectors are ‘ZnS:Ag/6LiF’ and ‘Gadox’ (Gd2O2S). While ‘ZnS:Ag/6LiF’ has a relatively low neutron capture from the 6Li, it offers a high light output from the photon emission of ZnS. ‘Gadox’ on the other hand has a much higher neutron capture from Gd, but a much lower photon output. For each scintillator type, the sensitivity (detection efficiency) and the spatial resolution can be adjusted by the thickness. Scintillation based detectors are also commonly utilized at existing time-of-flight scattering instruments, where the two most common scintillating materials are based on ‘GS20:Ce glass’ or ‘ZnS:Ag/6LiF’ [22,23]. Several strategies of collecting the light output of such detectors are being employed and determine the detector performance. For detectors used in scattering experiments, one option is to connect the scintillator screen directly to a photomultiplier tube which ensures a good light collection, but poses limitations on the detector size, geometry and spatial resolution. Fiber optic tapers or Anger camera systems are also commonly utilized and while they offer a higher degree of flexibility in terms of detector size, geometry and position resolution, they drastically reduce the amount of light that can be collected and hence the overall performance. The herein presented experiments demonstrate for the first time that a commonly used scintillator based imaging detector system can be suitable for time-of-flight neutron imaging when a high frame rate CMOS camera is employed.
When considering fast imaging applications (which includes ToF imaging) just as other ToF scattering applications, the effect of scintillator decay time and afterglow need to be considered. Afterglow is known to be a prominent phenomenon for ‘ZnS:Ag/6LiF’ scintillators, while ‘Gadox’ produces much less afterglow but it is known to also be sensitive to gammas. For the herein presented experiments ‘ZnS:Ag/6LiF’ was chosen as a scintillator for the following reasons: (i) It produces a very high light output of ~150,000 photons per neutron (ii) the primary decay time is relatively fast with ≈200 ns (iii) the secondary decay that is recognized as afterglow is long and shows a relatively slow decay that can be corrected for. The high light output was required in the present study for detecting the start of each individual pulse emerging from the source pulse choppers in order to subsequently add several pulses to one (histogram) for increasing the signal to noise ratio. This was necessary mainly because the utilized CMOS camera does not have the option of synchronizing to an external trigger.
CCD cameras have long been the gold standard for scientific imaging but CMOS cameras now also offer high signal to noise ratios and have the advantage of significantly higher frame rate capabilities. Widely available CMOS and sCMOS cameras allow to increase the frame rate through the selection of a ‘region of interest’ sub-window, such that the field of view can be traded off to achieve much higher temporal resolution. These cameras have become very popular to perform e.g. dynamic imaging and ultra-fast tomography measurements using neutrons [21] and x-rays [24], while becoming technically more mature and cheaper at the same time. The developments led to the motivation of building a neutron detector with a CMOS camera and scintillator to exploit the applicability for ToF imaging, which is successfully demonstrated in this manuscript.
2. Detector system
The detector system was custom built, similar to those reported previously [8,25], consisting of a light tight housing, in which the neutrons are converted by a scintillating screen into visible light, which is reflected by a mirror into the optical lens of a camera system (see Fig. 1).
Fig. 1 The camera system is based on the same principle as those currently used as a standard for neutron imaging (adapted from [8]). The main difference is the use of a fast CMOS camera instead of a CCD camera and the subsequent data processing procedure.
The utilized monochrome CMOS camera is a build for deep-sky astrophotography (ZWO ASI 183MM Pro monochrome camera) and available for approximately 1 kEuro. The light-sensitive CMOS has a diameter of 15.9 mm diameter and 5496 x 3672 pixels, with a pixel size of 2.4 µm. It offers fast USB3 connectivity and a memory buffer to minimize amp-glow. According to the specifications, the exposure time can be set between 32 µs to infinity. In the currently demonstrated experiments, an exposure time of 1 ms, corresponding to a continuous imaging frequency of 1 kHz, was utilized.
The utilized lens was a AF-S NIKKOR 35 mm 1:1,4G. The effective pixel size was 18.2 µm with a corresponding Field of View (FoV) of ~100x67 mm2. However, the attainable frame rate for the CMOS camera also scales with the size of the camera chip area that is recorded and an optional binning of multiple pixel into one. A smaller chip area and a higher binning allows for faster measurements. The 400 µm thick ‘ZnS:Ag/6LiF’ scintillator has a light emission at around 450 nm, a fast primary decay time of ≈200 ns and a relatively slow secondary decay, also referred to as “afterglow” that possibly gives rise to what is characterized to as the dark field intensity IDark in neutron imaging. Figure 2 shows the transmission image through a stainless steel tensile sample with a characteristic shape [26,27], including the as recorded images in which the beam profile due to the neutron guide structure is visible. In order to obtain the transmission value T of the sample in every pixel, a measurement with and without the sample in the beam is carried out to obtain:
The corresponding intensity with the sample is ISample and without is IOpenBeam, whereas IDark is the measured intensity without the neutron beam. The latter will be especially important in the present study due to the possibly expected scintillator afterglow. The normalized image is shown in Fig. 2(c). The full FoV was recorded in slow acquisition mode (in this case 100 images à 5 sec were recorded) whereas a smaller region of interest was selected for the high frame rate measurements (1 kHz). This is schematically depicted in Fig. 2(d).Fig. 2 Neutron radiography of a stainless steel sample showing (a) the as recorded image of the sample and (b) the Open Beam image without a sample. (c) The normalized transmission image. (d) The FoV (50mm x 1.1mm; effective pixel size of 36.4 µm) utilized for the high frame rate measurements with images taken at a continuous frame rate of 1 ms.
Besides the trade-off between frame rate and readout area of the camera chip, the utilized camera mainly suffers from not having the option to accept an external trigger signal that can be used to synchronize the image acquisition to the repetition rate of the pulsed neutron source. In the herein presented results we have used a software routine to find the start of each neutron pulse arriving at the detector and thereby aligned the consecutive pulses in order to accumulate the signal over thousands of individual pulses. This is described in the next section.
3. Data
The data for the presented results were obtained at the time-of-flight test beamline of the European Spallation Source located at the Helmholtz Zentrum Berlin [28,29]. While the beamline offers a sophisticated chopper system including Wavelength Frame Multiplication, a simple time-of-flight profile was produced for the herein presented work. The source pulse choppers were set to produce a neutron pulse with a length of ≈2.86 ms with a repetition rate of 14 Hz (which mimics the long pulse time structure of the European Spallation Source that is currently under construction in Lund, Sweden). Since it is well known that ‘ZnS:Ag/6LiF’ scintillator screens possess a slowly decaying afterglow and the fact that the dark signal of the utilized camera is sensitive to the ambient temperature, it was decided to restrict the neutron wavelength band to a certain time interval within the repetition rate of 14Hz (71.4ms). No neutron signal is hence expected between the arrival time of the slowest neutrons (≈42 ms) and the arrival time of the fastest neutrons (≈9 ms). At the detector position, this results in a pulse length of about 33 ms and an intermission of around 38 ms. This was achieved with a double disk chopper at a distance of 10 m from the source pulse chopper acting as a wavelength band chopper. The corresponding chopper settings are schematically represented as a time-of-flight diagram in Fig. 3. The time-of-flight neutron spectrum was recorded using a 3He beam monitor, which was synchronized via a TTL signal to the frequency of the source pulse choppers, and shows that no neutrons arrive for a time interval of ≈38 ms.
Fig. 3 (a) The beamline setup depicted in a time-of-flight diagram. The 2.86 ms long source pulse was created by a double disk chopper and the wavelength band was defined by another double disk chopper at a distance of 10 m downstream from there, while the detector was positioned at a distance of 26 m. (b) The time-of-flight spectrum as recorded using a 3He beam monitor (after 1/v efficiency correction) at the detector position. It can be seen that all neutrons after ≈42 ms are blocked by the wavelength band chopper. This region is used to correct for possible scintillator afterglow and temperature dependent ‘dark’ signal of the camera system.
The camera detector system was operated as standalone without any reference signal from the beamline. Figure 4(a) shows individual neutron pulses for a stripe section of the detector as indicated in Fig. 2(d), with dimensions of 50 mm x 1.1 mm. The 14 Hz repetition rate is clearly visible and the neutron imaging signal is significantly above the noise and dark field signal. The raw signal was smoothed (Fig. 4(b)) in order to improve the automatic detection of the start of the individual neutron spectra for subsequent alignment and adding up into one spectrum (i.e. histogram), which is presented in Fig. 4(c).
Fig. 4 (a) The raw signal without a sample in the beam, recorded with 1 kHz frame rate for a region of ≈50 mm x 1.1 mm (horizontal stripe as depicted in Fig. 2(b)), reveals individual pulses of the 14 Hz pulsed neutron beam. (b) The smoothed signal allows detecting individual pulse positions with respect to time for successive alignment and integration of subsequent raw data pulses; needed since the utilized camera does not provide an external trigger input. (c) Integrating (raw) ‘Open Beam’ data for 25 minutes, corresponding to 21000 individual pulses, yields data with high statistics that allow for spatially resolved time-of-flight imaging. The secondary y-axis shows the neutrons counts after dark-field correction, correctly resembling the spectrum recorded with the 3He monitor (see Fig. 3 (b)). It can be noted that the utilized CMOS camera possesses a fairly high constant offset pedestal in addition to temperature dependent readout noise that was much smaller than the measured signal. (d) The attenuated neutron spectrum transmitted through a BCC iron plate for which the normalized transmission spectrum is presented in Fig. 5. The alignment of the individual pulses was also performed for a region without the sample covering the detector.
As can be seen from Figs. 4(a)–4(d), a nearly constant offset value is present in the detection system even during the time when no neutrons arrive at the detector (compare Fig. 3(b)). For the presented settings, the average offset value is ≈660. It is worth noting that this offset value is fairly constant within the time where neutrons arrive, for both the Open Beam and Sample measurements (i.e. even at different neutron flux). However, this offset value showed slight dependency from the camera chip temperature rather than the primary signal from the pulse amplitude, which allows the conclusion that the afterglow is negligible. The following procedure was applied to obtain the ToF transmission spectra in every pixel:
- - Utilizing an area (Region of Interest, ROI) of the detector where no sample is located (the same ROI for the Open Beam and the Sample measurement) and smoothing the raw time signal recorded in this area as shown in Fig. 4(b). Based on this signal, a common time for each subsequent neutron pulse is determined via a threshold.
- - Aligning the raw spectra within every pixel by the previously determined common time and summing them into a resulting spectrum for each pixel and measurement (in this case of the ‘Sample’ and the ‘Open Beam’). The result is one image stack of 71 images per measurement spanning over repetition rate of 71.4 ms.
- - Determining the dark field intensity IDark in every pixel by calculating the average gray value in the pixel for the time window when no neutrons arrive at the detector within the repetition rate (namely 41-71 ms).
- - Assigning a ‘dark intensity corrected’ value to each pixel of the image stacks (i.e. ‘Sample’ and ‘Open Beam’ stack).
- - Calculating the transmission value for each pixel value of the image stack (see Eq. (1)).
After the last step the transmission spectrum can be plotted for every pixel of the image. The time interval between consecutive images is also known, in this case 1 ms. As in any time-of-flight experiment, the source pulse contains neutrons of all energies which will start to propagate with varying speeds, determined by their energies, i.e. wavelengths λ. The neutron’s time-of-flight tTOF is defined defined by:
where LDet is the detector distance from the source pulse choppers, m is the mass of the neutron and h is Planck’s constant. The wavelength resolution of the beamline is determined aswhere τ(λ) is the burst time (length of the source pulse) and tTOF(λ) is the time-of-flight for neutrons of a certain wavelengths. At 4 Å, we hence obtain a resolution of approximately 10% for the utilized settings and flight path, which is worse than the possible precision of the CMOS camera given the frame rate of 1 kHz (1 ms corresponding to a ≈0.15 Å between two consecutive images) resulting in a theoretical resolution of ≈3.75% at 4 Å.In order to demonstrate the capability of the detector for practical applications of scientific and societal merit, we demonstrate that a Bragg edge spectrum can be measured successfully and used for quantitative analysis. For low resolution wavelength settings (such as the presented case using the long source pulse and a relatively short flight path), the achievable results are of almost comparable quality to established time-of-flight detectors. Bragg edges are visible in the transmission spectrum of polycrystalline materials because neutrons are being scattered away from the direct beam due to Bragg scattering at atomic sets of lattice planes within a crystal structure. The method has become a standard tool at wavelength resolved neutron imaging beamlines [3]. The transmission through a 10 mm thick ferritic steel plate (having a body centred cubic, BCC, crystal structure) was measured using the CMOS camera system and under similar experimental parameters also with the ‘GP2’ detector developed in the UK. ‘GP2’ is a 100k pixel time-of-flight (ToF) neutron camera, which combines a gadolinium converter film and a CMOS readout sensor; details can be found in [13,30] and it’s use at V20 in [31]. The corresponding transmission spectra are depicted in Fig. 5, showing a good agreement between the two measurements. As can be seen, the utilized frame rate of the ‘GP2’ detector offers a timing resolution that allows excessive oversampling of the spectrum, while the 1 kHz frame rate of the CMOS camera provides only a few data points on the Bragg edges, resulting in less sharper features (compare e.g. the strong Bragg edge visible at ≈4.05 Å). Nonetheless, a framerate of 1 kHz provides enough data points to clearly define features in the spectrum and can be deemed sufficient to enable many fruitful studies.
Fig. 5 Transmission measurement of a ferritic (BCC) steel sample recorded by the scintillator coupled CMOS camera system described in this work. A measurement of the same sample is included for comparison that was obtained using the same chopper setup and flight path with the newly developed GP2 ToF imaging detector [13,30]. The 1 kHz frame rate of the CMOS camera allowed a sampling rate of 1 ms (≈0.15Å), resulting in a theoretical resolution of ≈3.75% at 4 Å.
This is furthermore underlined by investigating small regions of interest (ROI) in a tensile sample made from metastable stainless steel that has locally undergone a phase transformation from Austenite to Martensite. The investigated sample is well known as it has been studied several times and at several beamlines [26,27]. The Bragg-cut-off’s (when no more diffraction occurs) for the two phases, namely the austenitic Bragg edge corresponding to fcc(111) and the martensitic Bragg edge corresponding to bcc(110), have a slightly different position. This can also be observed even with the relaxed resolution in the presented measurement, as depicted in Fig. 6. The presented data was acquired within 60 minutes for the sample and 20 minutes for the Open Beam measurement and clearly demonstrates the suitability for relevant applications.
Fig. 6 Demonstration that the herein utilized ToF camera system is suitable to distinguish crystalline phases by Bragg edge analysis.
4. Discussion and conclusion
The work presented herein is the first demonstration that a ‘standard’ neutron imaging detector, a scintillator coupled to a CMOS camera, can be used efficiently for quantitative ToF imaging. The setup is very cost effective and versatile when compared to other ToF detector solutions. It has the added advantage of being also usable as a general white beam detector with a large FoV and due to the modularity even for very high spatial resolution. While the utilized detector is only a prototype, it already presents an attractive solution for many applications that take advantage of wavelength dependent image contrast. While applications that require high wavelength resolution are currently not possible given the limited frame rate, applications that require low and moderate resolution can readily be undertaken. Moreover, while the wavelength dependent transmission that many materials exhibit often remains unutilized in ‘white beam’ neutron imaging (which still presents the majority of use cases at operating beamlines), the presented solution can efficiently take advantage of this additional information, by e.g. dual wavelength contrast or high duty cycle measurements [32].
Many of the present technical limitations might be overcome within the next few years as even more powerful cameras become available. Even though the frame rates of available cameras are high, they would need to be even higher, one or two orders of magnitude, when aiming for high wavelength resolution measurements. The fact that the readout speed is inversely proportional to the readout area presents a significant drawback of currently available cameras, however this may also significantly improve in the near future. The use of a frame-transfer CMOS sensors [24] could already now present a viable and improved option.
A major advantage lies in the fact that very high neutron count rate capabilities become possible with such a camera based detector. Another main challenge lies in the scintillator technology (concerning primary decay time and afterglow), but once again the modularity of the detector will prove to be a key benefit. The synchronization challenge of the presented implementation is specific to the utilized camera and will be easy to overcome when employing a camera that accepts an external trigger signal.
Last but not least, the fact that the detector can be realized very cost effective (especially when compared to commonly utilized ToF detectors) one could also cover a large area, e.g. aiming to detect scattered neutrons. Instruments with immediate benefit from this could be for example ToF Laue diffraction. Also for the imaging case this could be useful, e.g. by enabling scattering corrections that are required for an improved determination of attenuation coefficients. Employing several of such detectors simultaneously could represent a viable alternative to other scattering correction approaches [33,34] as it would in contrast also allow for in situ studies. In summary, we can conclude that the demonstrated solution represents a viable detector for low to medium wavelength resolution ToF neutron imaging, and might even be a candidate to introduce a paradigm shift for ToF neutron detection in general.
Acknowledgment
The authors acknowledge the presentation by Burkhard Schillinger (TUM/FRM2) during the Neuwave9 workshop, where the herein utilized CMOS camera was presented.
References
1. N. Kardjilov, I. Manke, R. Woracek, A. Hilger, and J. Banhart, “Advances in neutron imaging,” Mater. Today 21(6), 652–672 (2018). [CrossRef]
2. W. Kockelmann, G. Frei, E. H. Lehmann, P. Vontobel, and J. R. Santisteban, “Energy-selective neutron transmission imaging at a pulsed source,” Nucl. Instrum. Methods Phys. Res. A 578(2), 421–434 (2007). [CrossRef]
3. R. Woracek, J. Santisteban, A. Fedrigo, and M. Strobl, “Diffraction in neutron imaging—A review,” Nucl. Instrum. Methods Phys. Res. A 878, 141–158 (2018). [CrossRef]
4. E. H. Lehmann and B. Schillinger, “How the NEUWAVE workshop series has pushed neutron imaging developments,” Neutron News 29(1), 25–31 (2018). [CrossRef]
5. M. Morgano, E. Lehmann, and M. Strobl, “Detectors requirements for the ODIN beamline at ESS,” Phys. Procedia 69, 152–160 (2015). [CrossRef]
6. B. Schillinger and E. Lehmann, “Scientific Reviews: Detectors for Neutron Imaging,” Neutron News 17(1), 19–21 (2006). [CrossRef]
7. E. Lehmann, A. Tremsin, C. Grünzweig, I. Johnson, P. Boillat, and L. Josic, “Neutron imaging—Detector options in progress,” J. Instrum. 6(01), C01050 (2011). [CrossRef]
8. N. Kardjilov, M. Dawson, A. Hilger, I. Manke, M. Strobl, D. Penumadu, F. Kim, F. Garcia-Moreno, and J. Banhart, “A highly adaptive detector system for high resolution neutron imaging,” Nucl. Instrum. Methods Phys. Res. A 651(1), 95–99 (2011). [CrossRef]
9. J. D. Parker, M. Harada, H. Hayashida, K. Hiroi, T. Kai, Y. Matsumoto, T. Nakatani, K. Oikawa, M. Segawa, and T. Shinohara, “Development of energy-resolved neutron imaging detectors at RADEN,” arXiv preprint arXiv:1806.09752 (2018).
10. J. Parker, K. Hattori, H. Fujioka, M. Harada, S. Iwaki, S. Kabuki, Y. Kishimoto, H. Kubo, S. Kurosawa, K. Miuchi, T. Nagae, H. Nishimura, T. Oku, T. Sawano, T. Shinohara, J. Suzuki, A. Takada, T. Tanimori, and K. Ueno, “Neutron imaging detector based on the μPIC micro-pixel chamber,” Nucl. Instrum. Methods Phys. Res. A 697, 23–31 (2013). [CrossRef]
11. S. Uno, T. Uchida, M. Sekimoto, T. Murakami, K. Miyama, M. Shoji, E. Nakano, and T. Koike, “Development of a two-dimensional gaseous detector for energy-selective neutron radiography,” Phys. Procedia 37, 600–605 (2012). [CrossRef]
12. H. Sato, O. Takada, S. Satoh, T. Kamiyama, and Y. Kiyanagi, “Development of material evaluation method by using a pulsed neutron transmission with pixel type detectors,” Nucl. Instrum. Methods Phys. Res. A 623(1), 597–599 (2010). [CrossRef]
13. D. Pooley, J. Lee, M. Brouard, J. John, W. Kockelmann, N. Rhodes, E. Schooneveld, I. Sedgwick, R. Turchetta, and C. Vallance, “Development of the “GP2” detector: modification of the PImMS CMOS sensor for energy-resolved neutron radiography,” IEEE Trans. Nucl. Sci. 64(12), 2970–2981 (2017). [CrossRef]
14. ProxiVision, “http://www.proxivision.de/products/neutron-detector.html”, retrieved http://www.proxivision.de/products/neutron-detector.html.
15. A. Tremsin, J. Vallerga, J. McPhate, and O. Siegmund, “Optimization of high count rate event counting detector with Microchannel Plates and quad Timepix readout,” Nucl. Instrum. Methods Phys. Res. A 787, 20–25 (2015). [CrossRef]
16. S. D. Pinto, R. Ortega, S. Ritzau, D. Pasquale, B. Laprade, S. Mrotek, S. Gardell, Z. Zhou, J. Plomp, L. Eijck, H. Bilheux, and I. Dhiman, “Neutron imaging and tomography with MCPs,” JINST 12(12), C12006 (2017). [CrossRef]
17. K. Nittoh, C. Konagai, T. Noji, and K. Miyabe, “New feature of the neutron color image intensifier,” Nucl. Instrum. Methods Phys. Res. A 605(1-2), 107–110 (2009). [CrossRef]
18. K. Mochiki, K. Ishizuka, K. Morikawa, T. Kamiyama, and Y. Kiyanagi, “Development of a New High-Frame-Rate Camera for Pulsed Neutron Transmission Spectroscopic Radiography,” Phys. Procedia 69, 143–151 (2015). [CrossRef]
19. T. Uragaki, J. Koide, J. Kawarabayashi, K.-i. Mochiki, Y. Matsumoto, Y. Su, K. Hiroi, T. Shinohara, and T. Kai, “Evaluation of High-frame-rate Camera with Digital Accumulation System Combined with Neutron Color Image Intensifier for Energy Resolved Neutron Imaging,” in Proceedings of the International Conference on Neutron Optics (NOP2017) (2018), 011027. [CrossRef]
20. M. Strobl, A. Hilger, M. Boin, N. Kardjilov, R. Wimpory, D. Clemens, M. Mühlbauer, B. Schillinger, T. Wilpert, C. Schulz, K. Rolfs, C. M. Davies, N. O’Dowd, P. Tiernan, and I. Manke, “Time-of-flight neutron imaging at a continuous source: Proof of principle using a scintillator CCD imaging detector,” Nucl. Instrum. Methods Phys. Res. A 651(1), 149–155 (2011). [CrossRef]
21. C. Tötzke, N. Kardjilov, I. Manke, and S. E. Oswald, “Capturing 3D Water Flow in Rooted Soil by Ultra-fast Neutron Tomography,” Sci. Rep. 7(1), 6192 (2017). [CrossRef] [PubMed]
22. N. Rhodes, “Scientific reviews: status and future development of neutron scintillation detectors,” Neutron News 17(1), 16–18 (2006). [CrossRef]
23. A. Spowart, “Measurement of the absolute scintillation efficiency of granular and glass neutron scintillators,” Nucl. Instrum. Methods 75(1), 35–42 (1969). [CrossRef]
24. M. P. Olbinado, J. Grenzer, P. Pradel, T. D. Resseguier, P. Vagovic, M.-C. Zdora, V. A. Guzenko, C. David, and A. Rack, “Advances in indirect detector systems for ultra high-speed hard X-ray imaging with synchrotron light,” J. Instrum. 13(04), C04004 (2018). [CrossRef]
25. N. Kardjilov, A. Hilger, I. Manke, R. Woracek, and J. Banhart, “CONRAD-2: the new neutron imaging instrument at the Helmholtz-Zentrum Berlin,” J. Appl. Cryst. 49(1), 195–202 (2016). [CrossRef]
26. R. Woracek, D. Penumadu, N. Kardjilov, A. Hilger, M. Boin, J. Banhart, and I. Manke, “3D Mapping of Crystallographic Phase Distribution using Energy-Selective Neutron Tomography,” Adv. Mater. 26(24), 4069–4073 (2014). [CrossRef] [PubMed]
27. R. Woracek, D. Penumadu, N. Kardjilov, A. Hilger, M. Boin, J. Banhart, and I. Manke, “Neutron Bragg Edge Tomography for Phase Mapping,” Phys. Procedia 69, 227–236 (2015). [CrossRef]
28. R. Woracek, T. Hofmann, M. Bulat, M. Sales, K. Habicht, K. Andersen, and M. Strobl, “The test beamline of the European Spallation Source – Instrumentation development and wavelength frame multiplication,” Nucl. Instrum. Methods Phys. Res. A 839, 102–116 (2016). [CrossRef]
29. M. Strobl, M. Bulat, and K. Habicht, “The wavelength frame multiplication chopper system for the ESS test beamline at the BER II reactor—A concept study of a fundamental ESS instrument principle,” Nucl. Instrum. Methods Phys. Res. A 705, 74–84 (2013). [CrossRef]
30. D. Pooley, J. Lee, M. Brouard, R. Farrow, J. John, W. Kockelmann, R. Nickerson, N. Rhodes, E. Schooneveld, and I. Sedgwick, “‘GP2′—An energy resolved neutron imaging detector using a Gd coated CMOS sensor,” in Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC),2015IEEE, (IEEE, 2015), 1–3. [CrossRef]
31. D. E. Pooley, J. W. L. Lee, F. A. Akeroyd, M. Hart, J. J. John, P. M. Kadletz, W. Kockelmann, T. Minniti, C. Moreton-Smith, M. Morgano, N. J. Rhodes, E. M. Schooneveld, I. Sedgwick, C. Vallance, and R. Woracek, “Energy resolved imaging using the GP2 detector: progress in instrumentation, methods and data analysis,” WCNR Proceedings (2019).
32. M. Siegwart, R. Woracek, J. I. M. Damián, A. S. Tremsin, V. Manzi-Orezzoli, M. Strobl, T. J. Schmidt, and P. Boillat, “Distinction between Super-Cooled Water and Ice with High Duty Cycle Time-of-Flight Neutron Imaging,” Rev. Sci. Instrum.submitted.
33. C. Carminati, P. Boillat, F. Schmid, P. Vontobel, J. Hovind, M. Morgano, M. Raventos, M. Siegwart, D. Mannes, C. Gruenzweig, P. Trtik, E. Lehmann, M. Strobl, and A. Kaestner, “Implementation and assessment of the black body bias correction in quantitative neutron imaging,” PLoS One 14(1), e0210300 (2019). [CrossRef] [PubMed]
34. P. Boillat, C. Carminati, F. Schmid, C. Grünzweig, J. Hovind, A. Kaestner, D. Mannes, M. Morgano, M. Siegwart, P. Trtik, P. Vontobel, and E. H. Lehmann, “Chasing quantitative biases in neutron imaging with scintillator-camera detectors: a practical method with black body grids,” Opt. Express 26(12), 15769–15784 (2018). [CrossRef] [PubMed]
