High spatial resolution fiber based miniature detector for neutron flux mapping in a subcritical assembly and neutron generator

1. Introduction

Accelerator driven subcritical (ADS) systems are attracting worldwide attention due to their role in the incineration of transuranic elements from spent nuclear fuel, energy production and utilization of thorium, besides superior safety characteristics. Research on ADS is being pursued in several countries and involves mainly feasibility study with the use of experimental test facilities [1–3]. Neutron flux profile is one of the critical parameters in ADS as it provides system characteristics. Similarly neutron flux measurements play an important role for research reactors and neutron generators. Neutron energies ranging from thermal (0.025 eV) to fast (typically 14 MeV) are the characteristic of these types of systems. Such systems have typical problems of high neutron flux which tends to saturate conventional detector easily and the presence of neutron-gamma mixed field. In addition the compact designs of these systems such as reactor core require the flux measurement in very narrow spaces with high spatial resolution ( ∼ mm ). Conventionally, various detectors are used for measuring real time neutron flux such as ³He or BF₃ proportional counters and fission chambers. However, in some cases, the dimensional constraint such as the narrow gap between fuel assemblies restricts the use of these detectors to measure the real time neutron flux profile with high spatial resolution. In recent years, scintillation based detectors coupled to an optical fiber have also been used for neutron flux measurement [4–6]. They have found applications in several fields such as flux mapping in reactors, tritium production rate measurement [7] in D-T neutron field and medical application of boron neutron capture therapy (BNCT). These detectors utilize neutron convertor-scintillator coupled to the tip of a plastic or quartz optical fiber at one end with the other end coupled to a photomultiplier tube (PMT). The neutron interaction with neutron convertor nuclei produces charged particle which interacts with scintillator (such as – ZnS(Ag)), resulting in visible light. This light is guided to PMT through the long optical fiber and corresponding electronic signals generated are processed with associated electronics such as amplifier, discriminator and counter. Typically neutron converter materials such as ¹⁰B, ⁶Li (for thermal neutrons) or ²³⁸U, ²³²Th (for fast neutrons) in combination with ZnS(Ag) or glass scintillator are used for these detectors. The possibility to fabricate these detectors in smaller size with very small ( ∼ mm ) active area gives unique advantages of high spatial resolution for neutron flux measurement and low sensitivity towards gamma over conventional gas based detectors.

In this paper, we present the design and development of optical fiber based high spatial resolution neutron (thermal/fast) detector. The thermal neutron detector (TND) is developed using 6 LiF + ZnS ( Ag ) scintillator whereas fast neutron detector (FND) is developed using ThO 2 + ZnS ( Ag ) . The thermal neutron flux profile with TND is measured in the narrow experimental channels of subcritical (BRAHMMA) facility [8,9] driven by D-D/D-T neutrons [10]. BRAHMMA is an experimental thermal subcritical assembly for studying the physics of accelerator driven systems on low power scales. The assembly consists of natural uranium as fuel arranged in a square lattice, high density polyethylene (HDPE) as moderator and beryllium oxide (BeO) as reflector. Periodic variation in the flux profile is observed at the fuel lattice and moderator positions of the subcritical assembly, which was not observed with a miniature ³He detector developed in the past [11]. This is due to the high spatial resolution ( ∼ 1 mm ) of the developed detector. The neutron flux distribution measurement of DC accelerator based D-T neutron generator using FND is also presented.

2. Detector design

An optical fiber based miniature detector was designed using suitable neutron converter materials (thermal-⁶LiF or fast-ThO₂) along with scintillator (ZnS(Ag)), plastic optical fiber (1 mm core diameter) and a photomultiplier tube (PMT). The cross-sectional dimension of the detector was decided to keep in mind the physical space constraint of the subcritical assembly and the available neutron flux. The requirement was such that the detector should be able to provide sufficient number of counts and at the same time should not get saturated at the available flux range ( 10 5 – 10 7 n cm − 2 sec − 1 ). The thermal (or fast) detector was designed by depositing a uniform layer of 6 LiF + ZnS ( Ag ) (or ThO 2 + ZnS ( Ag ) ) material (thickness ∼ 300 micro meters) on an aluminum disk of 3 mm diameter and thickness of 1 mm. The neutron converter and scintillator material were mixed in equal proportion. This disk was carefully glued on the tip of fiber at one end using optical glue. The other end of the fiber (20 meter long fiber) was coupled to a small size PMT. A thin aluminum cap covered the scintillator for light shielding and reflection. Figure 1(a) and Fig. 1(b) shows the schematic and photograph of the detector respectively.

OPEN IN VIEWER

Fig. 1.

(a) Schematic of fiber based detector and allied electronics (b) Photograph of the detector assembly.

The developed neutron detectors were calibrated for neutrons and also tested for gamma sensitivity. For thermal neutron sensitivity measurement of TND, a thermal neutron assembly was set up using HDPE surrounded by graphite on all sides. A neutron source of Am-Be (500 mCi) was placed at the centre of the assembly. The radiation shielding of assembly was provided using borated polyethylene and cadmium. Measurement of thermal neutron flux at the center of the assembly was carried out using the gold (Au) foil activation method [12]. The signal counts were also recorded by placing the tip of the TND at the same location as the foil. A typical neutron signal observed on an oscilloscope is shown in Fig. 2. The sensitivity was calculated taking the ratio of detector count rate (cps) to the measured flux ( n cm − 2 sec − 1 ) via activation method. It was found to be 3 × 10 − 4 cps per unit neutron flux. Unit neutron flux corresponds to neutron field of one neutron per unit area per second. The gamma sensitivity of TND was tested using a 33 mCi gamma source of ¹³⁷Cs kept at ∼ 1 cm from the tip. The count rate observed was nearly background (3–5 cps) level indicating very low gamma sensitivity which is due to the small thickness of the detector.

OPEN IN VIEWER

Fig. 2.

Neutron signal from TND observed on an oscilloscope.

TND was used for thermal neutron distribution profile (axial and radial) measurements in a zero power ADS system – BRAHMMA developed at BARC, India [8]. The subcritical core consists of natural uranium as fuel with HDPE as moderator and BeO (Beryllium oxide) as reflector. The fuel rods are arranged in a 13 × 13 square HDPE lattice with a 48 mm pitch (centre-to-centre distance between two fuel rods) as depicted in Fig. 3. The fuel rods are of 1000 mm active length and 35.5 mm diameter. The system is covered with cadmium (1.5 mm) to isolate the system from scattered neutrons from the surroundings. The central 3 × 3 part of the lattice ( 144 mm × 144 mm ) of the subcritical assembly serves as the cavity for inserting the neutron source. The neutron producing target is located at the centre of the core. Seven experimental channels (EC) are located at different axial (along the fuel rod length) and radial (perpendicular to the fuel rod) positions as shown in Fig. 3. The relative positions of the experimental channels are such that their influence on each other is minimized. Three axial experimental channels (EC1, EC2 and EC3) of diameter 10 mm are provided at radial distances of 122 mm, 238 mm and 265 mm respectively for measurement. EC1 is close to the source, whereas EC3 is near the reflector. Four experimental channels (EC4, EC5, EC6, and EC7) of diameter 7.5 mm are provided in the moderator/reflector region along the radial direction. Channels EC5, EC6, and EC7 are located in the mid-elevation plane and run up to the cavity only. Channel EC4 is located above the mid-elevation plane and covers the full length of the moderator. A thermal neutron flux profile measurement was carried out both in axial and in radial directions using TND.

OPEN IN VIEWER

Fig. 3.

(a) Front view (b) side view of BRAHMMA. The positions of the axial and radial experimental channels are also shown.

4.2. Axial profile measurement

In a similar fashion, a thermal neutron flux distribution profiling across the axial direction was measured in the EC1 channel in the experimental ADS assembly. The whole length along the axial direction was scanned every 5 cm. The experimental data were plotted against scanned positions as shown in Fig. 5. The results were compared with results of previously measured data using a ³He detector [11]. It was observed that axial profiles from both the detectors were of similar nature and the peak position of flux was at the center of the assembly. Peak flux at the center was expected due to the fact that source neutrons (D-T/D-D) were emitted in the middle of the cavity which was along the central axis of the assembly. The flux range across the axial/radial position measured was ( 0.1 – 2 ) × 10 6 n cm − 2 sec − 1 using TND.

In the above flux profile measurement the observation for each data position was carried out ten times and averaged over it. The estimated error bars in the data are around ± 3.5 % .

OPEN IN VIEWER

Fig. 4.

Radial thermal neutron flux distribution profile in subcritical assembly measured with (a) TND (b) ³He detector (c) simulated.

OPEN IN VIEWER

Fig. 5.

Axial thermal neutron flux distribution profile in subcritical assembly measured with (a) TND.

Neutron (thermal/fast) detectors of high spatial resolution have been developed for experiments at BRAHMMA. The sensitivity of the thermal and fast (14MeV) neutron detectors was 3 × 10 − 4 cps and 3.6 × 10 − 6 cps per unit neutron flux respectively. Using these detectors, measurements of thermal neutron flux distribution profiles (axial and radial) in channels of a subcritical assembly and fast neutron distributions across a D-T neutron source were carried out successfully. The result of axial flux distribution was comparable with the result of ³He detector. In addition, flux variations in the radial direction were observed at the fuel lattice position, which was not observed with miniature ³He detector. Future work in this direction will be carried out in assembling of a large number of such detectors and using them together in providing three dimensional information of neutron (thermal/fast) field in assemblies of interest.