Sensitivity of an antineutrino monitor for remote nuclear reactor discovery L. KnealeS. T. WilsonyT. Appleyard J. Armitage N. Holland and M. Malek University of Sheeld Sheeld S10 2TN United Kingdom

2025-05-03 0 0 3.52MB 31 页 10玖币

侵权投诉

Sensitivity of an antineutrino monitor for remote nuclear reactor discovery

L. Kneale,∗S. T. Wilson,†T. Appleyard, J. Armitage, N. Holland, and M. Malek

University of Sheﬃeld, Sheﬃeld S10 2TN, United Kingdom

(Dated: February 28, 2023)

Antineutrinos from a nuclear reactor comprise an unshieldable signal which carries information

about the core. A gadolinium-doped, water-based Cherenkov detector could detect reactor antineu-

trinos for mid- to far-ﬁeld remote reactor monitoring for non-proliferation applications.

Two novel and independent reconstruction and analysis pathways have been developed and applied

to a number of representative reactor signals to evaluate the sensitivity of a kiloton-scale, gadolinium-

doped Cherenkov detector as a remote monitor.

The sensitivity of four detector conﬁgurations to nine reactor signal combinations was evaluated

for a detector situated in Boulby Mine, close to the Boulby Underground Laboratory in the UK.

It was found that a 22 m detector with a gadolinium-doped, water-based liquid scintillator ﬁll is

sensitive to a ∼3 GWth reactor at a standoﬀ of ∼150 km within two years in the current reactor

landscape. A larger detector would be required to achieve a more timely detection or to monitor

smaller or more distant reactors.

I. INTRODUCTION

Emerging antineutrino detection technology has opened up the possibility of using a water-based detector as a

remote monitoring tool for nuclear non-proliferation. While near-ﬁeld reactor observation with surface-deployed,

plastic scintillator detectors has been demonstrated [1, 2] and recently investigated with a view to planned reactor

monitoring [3], an underground water-based detector allows the technology to be scaled to larger detector sizes for

mid- to far-ﬁeld monitoring. Aggregate detection of reactor antineutrinos has been achieved for the ﬁrst time in pure

water [4] and applications of antineutrino detection to nuclear non-proliferation have been explored in [5]. A remote

monitor of this type may be used in a collaborative project as part of a remote monitoring toolkit for veriﬁcation of

declared reactor activity. There is interest in the safeguarding and policy communities in a neutrino detector as a

future tool to safeguard advanced reactors and as part of future nuclear deals [6], with a reduction in the perceived

intrusion of on-site inspections at nuclear facilities cited as one of the key beneﬁts. Although the high cost, large size

and underground location of large water-based neutrino detectors limit their practical application for non-cooperative

reactor monitoring, these concerns could be somewhat mitigated by adoption of a neutrino monitor at an early stage

in the construction of an advanced reactor or development of a nuclear deal.

This paper provides a ﬁrst attempt to evaluate the sensitivity of a neutrino detector to real reactor signals, and

analytical and statistical methods that could be employed to draw conclusions about the operation of a reactor in a

realistic nuclear landscape.

The challenge of remote reactor monitoring with a water-based detector is that the low-energy antineutrinos from a

reactor are at the very limit of the energy threshold. Backgrounds (e.g., from other reactors and natural radioactivity

of detector components) further reduce sensitivity to the signal. This issue requires a multi-faceted approach to

improving the sensitivity. This paper presents two new complementary reconstruction-analysis pathways for signal-

to-background discrimination, which have been developed for optimal sensitivity to a remote reactor. The methods

presented in this paper represent a comprehensive treatment of reactor antineutrino detection to evaluate sensitivity

to nine real reactor signals with four diﬀerent detector conﬁgurations close to the Science & Technology Facilities

Council (STFC) Boulby Underground Laboratory in the UK.

The paper is structured as follows. Section II presents the fundamentals of reactor antineutrino emission and

detection in water-based media. The site at Boulby is detailed in Section III, along with discussion of the detector

conﬁgurations and signals evaluated. In Section IV, shared signal and background simulations are detailed and the

reconstruction-analysis pathways are discussed in depth in Section V. Results for the sensitivity of detector-signal

conﬁgurations are presented and discussed in Section VI before concluding in Section VII.

∗Corresponding author (she/her/hers): e.kneale@sheﬃeld.ac.uk

†Corresponding author (he/him/his): stephen.wilson@sheﬃeld.ac.uk

arXiv:2210.11224v2 [physics.ins-det] 27 Feb 2023

II. REACTOR ANTINEUTRINOS

All nuclear power reactors generate antineutrinos, emitting an isotropic ﬂux of O(1020) s−1from a 1 GWth reac-

tor [7]. This is produced by the ﬁssion of 235U, 238U, 239Pu and 241Pu into neutron-rich nuclei, which then undergo a

series of βdecays to stability. This releases, on average, 6 antineutrinos per ﬁssion at energies up to ∼10 MeV. Al-

though the interaction cross section of antineutrinos with matter is very small, the enormous number of antineutrinos

released means that the antineutrino signal from a reactor can be seen in a variety of detectors.

The reactor antineutrino ﬂux is dependent on the reactor thermal power and core composition, on the nuclear

physics of the ﬁssion of the isotopes in the core and on neutrino oscillations, which alter the ﬂux with distance from

a reactor. The precise composition of the core of a reactor and the time evolution of the core (burnup), including the

refueling frequency, depend on the reactor type.

The calculation of the antineutrino spectrum from a reactor is described in [8]. The antineutrino spectrum from a

ﬁssioning isotope in a reactor is related to the power output and composition of a reactor core by

Φ¯νe,i(E¯νe) = Pth

piλi(E¯νe)

,(1)

where Pth is the thermal power of the core, piis the fraction of the thermal power resulting from the ﬁssion of isotope

i,Qiis the average thermal energy emitted per ﬁssion and λi(E¯νe) is the emission energy spectrum in antineutrinos

per ﬁssion for ﬁssioning isotope ias a function of antineutrino energy E¯νegiven by

λi(E¯νe) = exp 6

j=1

ajEj−1

¯νe!,(2)

where the coeﬃcients ajare ﬁt parameters from the Huber-Mueller predictions [9, 10], which are derived from

measurements of the βspectra from nuclear ﬁssion.

Information about the power and core composition of a reactor is carried by the outgoing particles from antineu-

trino interactions in a detector. A change in the number of antineutrinos emitted by a reactor can be caused by a

change in the core composition or reactor thermal power, or indeed by both. The SONGS1 gadolinium-doped liquid-

scintillator antineutrino detector, located 25 m from the 3.56 GWth San Onofre Nuclear Generating Station (SONGS),

demonstrated that the detected antineutrino ﬂux from a reactor reﬂects the operating power and fuel evolution of the

core [11] and as such could be used for discovery, monitoring and veriﬁcation of nuclear reactor operations.

A. Reactor antineutrino detection

Antineutrinos from a reactor can be detected via their inverse βdecay (IBD) interaction with protons in water or

a hydrogenated liquid:

νe+ p −→ e++ n.

Information about the energy of the incoming antineutrino is carried by the outgoing positron, while neutron tagging

can help to reject backgrounds and lower the detectable energy threshold. IBD is the principle interaction by which

antineutrinos can be detected for reactor monitoring in a water-based Cherenkov detector.

The IBD cross section is O(10−44)Eepecm2[12], where Eeand peare the positron energy and momentum. Although

small, it is relatively high compared to the cross sections of other antineutrino interactions in matter and it has been

calculated to within 1% accuracy at low energies. At the time of this study, the most accurate cross section in the

MeV to GeV range was given in [13]. The cross section has since been recalculated with reduced uncertainty in [14].

IBD is the dominant interaction of antineutrinos with energies of less than a few tens of MeV and has a low

threshold energy Ethr which can be expressed in terms of the proton, neutron and positron rest masses mp,mnand

meas approximately

Ethr ≈(mn+me)2−m2

2mp

≈1.8 MeV (3)

in the laboratory frame.

The positron carries most of the kinetic energy from the antineutrino and, with good energy resolution, the incident

antineutrino energy can be determined. The energy of the incoming antineutrino E¯νeis related approximately to that

of the positron by

E¯νe≈Ee++Ethr −me.(4)

While the positron emission is almost isotropic, with a slight bias in the backwards direction, the neutron takes on

most of the antineutrino’s momentum and its initial direction is largely parallel to that of the incoming antineutrino.

From the point of emission, the neutron then takes a random walk and thermalizes in the detector medium through

successive scatterings, which knock the neutron oﬀ its original path. Once thermalized, the neutron is captured on a

hydrogen nucleus or on another nucleus, such as gadolinium, added speciﬁcally for its neutron-capture capabilities.

A second signal arising from the de-excitation of the capture nucleus can be detected. This occurs within a short

distance and time of the positron signal and results in a signal of coincident interactions which can be beneﬁcial for

background rejection. The time and distance between the positron and neutron events are dependent on the medium

in which the interaction takes place.

B. Reactor antineutrino detection media

Water Cherenkov and scintillator detectors are the two principal types of antineutrino detection technology. Scin-

tillator detectors are a proven technology for reactor antineutrino detection but are not readily scalable for mid- to

far-ﬁeld monitoring. A nascent water Cherenkov technology - gadolinium doping - presents the possibility of a scal-

able reactor antineutrino detector. The combination of the two technologies into a water-based scintillator technology

promises to exploit the best features of each method.

The principle of using gadolinium (Gd) to delve into lower-energy neutrino detection was ﬁrst introduced by [15]

and developed by [16]. Gadolinium has a very high thermal neutron capture cross section (48,800 barns (b) for

natural Gd compared to ∼0.3 b for hydrogen) and a relatively high-energy subsequent gamma cascade of ∼8 MeV

(mean total energy) compared to a single 2.2 MeV gamma from the capture on hydrogen. This gives a more easily

detectable correlated signal from the inverse βdecay reaction [16]. With a concentration of 0.1% Gd ions, ∼90% of

the neutrons capture on Gd [16]. Most of the remaining neutrons capture onto the hydrogen in the water.

In 0.1% gadolinium-doped ultra-pure water (Gd-H2O), the neutron thermalizes and captures after a mean time of

∼30 µs and mean distance of ∼6 cm. The delayed neutron-capture emission is seen in a Gd-H2O Cherenkov detector

with a peak in visible light at ∼4.5 MeV. The peak positron energy from IBD interactions of reactor antineutrinos

is ∼2.5 MeV. In a gadolinium-doped medium, the positron-detection eﬃciency for reactor antineutrinos can be

increased by looking for a positron-like signal (the prompt event) in coincidence with the generally higher-energy (and

thus easier to observe) neutron-capture signal (the delayed event). Positron events at the lower end of the energy

range would otherwise be lost among the background. In this way, Gd can lower the energy threshold of a water

Cherenkov detector to increase sensitivity to the low-energy positrons for reactor antineutrino interactions via IBD.

The emerging Gd-H2O technology has been demonstrated in EGADS (Evaluating Gadolinium’s Actions on Detector

Systems) [17] and has now been deployed in SK-Gd (Super-Kamiokande with Gadolinium) [18] and the ANNIE

(Accelerator Neutrino-Neutron Interaction Experiment) detector [19].

Reactor antineutrinos have been detected with liquid and plastic scintillator detectors. A liquid scintillator is

composed of an organic solvent containing a scintillating chemical in solution of the type used in the Kamioka Liquid

Scintillator Antineutrino Detector (KamLAND) [20]. In a scintillating medium, the scintillator interacts with incoming

particles, which impart energy to the scintillator. Excited scintillator particles then release this additional energy as

light.

Scintillation detectors bring a high light yield, low-energy sensitivity, and good energy and position resolution.

However, they do not preserve directional information and are limited in size due to light absorption and the cost and

availability of the medium. Water Cherenkov detectors bring directional information and are scalable to very large

detectors. However, they have a low light yield and no sensitivity below the Cherenkov threshold. Combining the

two media into a water-based liquid scintillator (WbLS) [21] provides a solution which can be scaled to large sizes

and results in a higher light yield, sensitivity down to lower energies, improved energy and position resolution, and

directional information from the Cherenkov light, which has beneﬁts for reactor antineutrino detection [22].

WbLS is an emerging detector medium, which is still undergoing optimization and improvement. WbLS cocktails

using the PPO (2,5-diphenyl-oxazole) wavelength-shifting scintillator in a linear alkylbenzene (LAB) solvent have

been produced [21] and gadolinium doping is in development. Pure liquid scintillator is a scintillating material in

solution in an oily organic solvent. In WbLS, the scintillator is dissolved in an oily solvent in the same way. This

solution is then further combined with pure water. The mixing between the oil and water in WbLS is achieved by

the addition of a surfactant which creates micelles with both hydrophilic and hydrophobic surfaces.

The addition of gadolinium brings a further increase in light yield due to the neutron capture on gadolinium and

enhanced background rejection due to the coincident signal pair which occur closer in space and time with gadolinium

doping. Gadolinium-doped water-based liquid scintillator (Gd-WbLS) has the potential to combine the beneﬁts of

liquid scintillator and gadolinium-doped water Cherenkov detectors. The combination of Cherenkov and scintillation

light ultimately brings increased IBD detection, with the added beneﬁt of improved detection quality, particularly

where it is possible to separate the Cherenkov and scintillation components.

Although WbLS is an emerging technology - particularly with the addition of Gd - detailed characterization of

diﬀerent WbLS cocktails has been performed [23–26] and the ANNIE [19] collaboration is currently working towards

a WbLS ﬁll.

III. GD-DOPED CHERENKOV DETECTOR AT BOULBY

The location used for this study is the ICL (Israel Chemicals Ltd) Boulby Mine on the coast near Whitby in North

Yorkshire, UK. Boulby Mine is an ultra-low background environment and hosts the Boulby Underground Laboratory,

which has been home to deep underground physics experiments since the 1990s and is now operated by the UK’s

STFC.

The low-background environment in Boulby Mine can be attributed to low radioactivity rates in Boulby Mine’s

rock salt layers, a low cosmic muon rate and the world’s lowest ambient air radon concentration in an underground

lab (∼3 Bq m−3). Boulby Underground Laboratory is at a depth of 1.1 km underground and an eﬀective depth of

2.8 km water equivalent (km.w.e.) with a ﬂat overburden. This results in a signiﬁcant O(106) reduction in cosmogenic

muons compared to the surface [27].

The detector conﬁgurations used for this study are based on the AIT-NEO (Advanced Instrumentation Testbed -

Neutrino Experiment One) detector, which was investigated for the AIT site at Boulby Mine. In total four conﬁgu-

rations - two detector geometries with each of two ﬁll media - and nine reactor signals at Boulby were considered.

The two detector geometries are upright cylinders, with parameters summarized in Table I. A schematic of the

detector design in Fig. 1 shows the inner PMT support structure which creates an instrumented inner detector

volume within the tank and an uninstrumented buﬀer volume between the inner volume and the tank walls for the

reduction of backgrounds from the tank and surrounding rock.

TABLE I: Summary of detector geometries used in this study.

Tank diameter PMT support Buﬀer Inner PMT Number

and height [m] structure radius [m] width [m] coverage [%] of PMTs

16 5.7 2.3 15 2500

22 9.0 2.0 15 4600

Each of the two detector sizes was evaluated with each of two ﬁll media:

•Gd-H2O with 0.2% Gd2(SO4)3doping (for 0.1% Gd concentration) and

•Gd-WbLS with 0.2% Gd2(SO4)3doping and ∼100 photons per MeV WbLS. The WbLS component gives

approximately 1% of the light yield of pure LAB-based scintillator with 2g/L PPO typically used in large

neutrino experiments such as Daya Bay and SNO+ [28, 29].

For this study, the reactor landscape around Boulby Mine is used. Several reactors are planned to decommission or

come on line in the coming years and this is reﬂected in the study. Currently, the principal reactor signals at Boulby

are the Advanced Gas-Cooled Reactor (AGR) nuclear power stations at Hartlepool, Heysham and Torness. Other

important sources of reactor antineutrinos at Boulby are the Pressurized Water Reactors (PWRs) at Sizewell B and

at Gravelines in France. New PWR cores at Hinkley Point C are expected to come on line in the coming years. The

detector and reactor locations are shown in Fig. 2.

The six principal reactor signals studied are the AGR signals listed in Table II in order of decreasing signal at

Boulby, along with their published dates for decommissioning at the time of this study.

Table III shows the reactors included in the signal and background for each of the reactor signals listed in Table II,

with the reactor signal and background rates in terms of IBD interactions in a Boulby detector from [8]. ‘World’

reactor backgrounds include all reactors more distant than Torness. World reactor backgrounds are projected to 2026,

with Hinkley Point B oﬀ and Hinkley Point C on, according to published schedules at the time of the study. These

scenarios take into account the expected shutdown of the Hartlepool cores and Heysham 1 in 2024.

FIG. 1: Schematic of the detector design by Jan Boissevain (University of Pennsylvania), showing the tank

supported on a steel truss structure and inner PMT support structure.

TABLE II: The six AGR reactor signals evaluated in this study, including their power in terms of total thermal

capacity, standoﬀ distance from a detector at Boulby and currently planned decommissioning dates [30]. Real

reactor signals and schedules were used to represent authentic cases of remote monitoring.

Signal Number Reactor Standoﬀ Decommissioning

of cores power [GWth] distance [km] date

Hartlepool 1 & 2 2 3.0 26 2024

Hartlepool 1 1 1.5 26 2024

Heysham 1 & 2 4 3.0 149 2024 (1), 2028 (2)

Heysham 2 & Torness 4 1.5, 3.2 149, 187 2028

Heysham 2 2 1.5 149 2028

Torness 2 3.2 187 2028

An extension of the study to the PWR reactor signals at Sizewell B, Hinkley Point C and Gravelines after 2028 is

presented in Section VI B.

IV. SIGNAL AND BACKGROUND SIMULATIONS

Full Monte Carlo detector simulations were carried out with RAT-PAC (Reactor Analysis Tool - Plus Addi-

tional Codes) [32], which has been adapted for AIT-NEO and which is based on the physics simulation framework

GEANT4 [33, 34], the CLHEP physics library [35], the GLG4sim (Generic Liquid-scintillator Anti-Neutrino Detector

or GenericLAND) GEANT4 simulation for neutrino physics [36] and the data analysis framework ROOT [37].

RAT-PAC models the event-by-event detector response to the signal and background. Events are produced for this

study with custom GLG4sim Monte Carlo event generators and particles are propagated in the detector medium with

GEANT4. Light emission and PMT response is managed by GLG4sim. RAT-PAC also handles the triggering and

data acquisition (DAQ) before the data are output in ROOT format.

The MC model for WbLS is described in more detail in [38]. The time proﬁle of the scintillation light was based

on measurements of WbLS [39, 40], and measurements of Gd-WbLS [41] were used for the light yield and scattering

in the model.

For each of the detector conﬁgurations simulated, the large-scale, complex structures (e.g., I-beams, trusses, PMT

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

SensitivityofanantineutrinomonitorforremotenuclearreactordiscoveryL.Kneale,S.T.Wilson,yT.Appleyard,J.Armitage,N.Holland,andM.MalekUniversityofSheeld,SheeldS102TN,UnitedKingdom(Dated:February28,2023)Antineutrinosfromanuclearreactorcompriseanunshieldablesignalwhichcarriesinformationaboutthecore.Aga...

展开>> 收起<<

Sensitivity of an antineutrino monitor for remote nuclear reactor discovery L. KnealeS. T. WilsonyT. Appleyard J. Armitage N. Holland and M. Malek University of Sheeld Sheeld S10 2TN United Kingdom.pdf

共31页,预览5页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Sensitivity of an antineutrino monitor for remote nuclear reactor discovery L. KnealeS. T. WilsonyT. Appleyard J. Armitage N. Holland and M. Malek University of Sheeld Sheeld S10 2TN United Kingdom

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: