Exclusion and Verication of Remote Nuclear Reactors with a 1-Kiloton Gd-Doped Water Detector O. A. Akindele1A. Bernstein1M. Bergevin1S. A. Dazeley1F. Sutanto1A. Mullen1and J. Hecla1

2025-05-06 0 0 1.16MB 15 页 10玖币

侵权投诉

Exclusion and Veriﬁcation of Remote Nuclear Reactors

with a 1-Kiloton Gd-Doped Water Detector

O. A. Akindele,1, ∗A. Bernstein,1M. Bergevin,1S. A. Dazeley,1F. Sutanto,1A. Mullen,1and J. Hecla1

1Lawrence Livermore National Laboratory, Livermore, California 94550, USA

(Dated: October 19, 2022)

To date, antineutrino experiments built for the purpose of demonstrating a nonproliferation ca-

pability have typically employed organic scintillators, were situated as close to the core as possible

- typically a few meters to tens of meters distant and have not exceeded a few tons in size. One

problem with this approach is that proximity to the reactor core require accommodation by the

host facility. Water Cherenkov detectors located oﬀsite, at distances of a few kilometers or greater,

may facilitate non-intrusive monitoring and veriﬁcation of reactor activities over a large area. As

the standoﬀ distance increases, the detector target mass must scale accordingly. This article quan-

tiﬁes the degree to which a kiloton-scale gadolinium-doped water-Cherenkov detector can exclude

the existence of undeclared reactors within a speciﬁed distance, and remotely detect the presence

of a hidden reactor in the presence of declared reactors, by verifying the operational power and

standoﬀ distance using a Feldman-Cousins based likelihood analysis. A 1-kton scale (ﬁducial) water

Cherenkov detector can exclude gigawatt-scale nuclear reactors up to tens of kilometers within a

year. When attempting to identify the speciﬁc range and power of a reactor, the detector energy

resolution was not suﬃcient to delineate between the two.

I. INTRODUCTION

Antineutrino monitoring has been proposed for vari-

ous nonproliferation applications and reactor fuel com-

positions [1–6]. All of these prior studies used ton-scale

scintillator detectors in close proximity (tens of meters or

less) to the reactor core. While short-distance monitor-

ing with scintillator detectors is less intrusive to reactor

operations compared to other veriﬁcation methods, these

detectors still require on-site accommodation. For exam-

ple, space and power must be provided for the equip-

ment, the materials used must be compliant with facility

regulations, and any maintenance on the detector will

require access by a veriﬁcation body. Greater standoﬀ

distances could further reduce the intrusiveness of the

method by permitting deployment outside the reactor

operator’s facility grounds. However, to eﬀectively probe

a larger exclusion area requires a larger target volume.

Ton-scale near-ﬁeld detectors still require on-site compli-

ance. Space and power must be provided for the equip-

ment, the materials used must be shown not to aﬀect

facility operations, and any maintenance on the detector

will require access by the veriﬁcation body.

Antineutrinos are weakly interacting and can be de-

tected at long distances from their source of origin. This

raises the question: can antineutrino detection be used in

the mid-ﬁeld, which we deﬁne here to be approximately

10 to 100 km, to monitor the operation or presence of

nuclear reactors? In this article, we evaluate the ability

to detect antineutrinos at a hypothetical far-ﬁeld deploy-

ment in the presence of other reactors producing a high

antineutrino background.

∗Corresponding author; Email: akindele1@llnl.gov

II. THE BASELINE DETECTOR DESIGN

In this study the hypothetical deployment is located in

the Boulby Underground Laboratory on the eastern side

of the United Kingdom, an operating potash/polyhalite

mine [7]. The mine rock is low in uranium, thorium, and

radon compared to many other underground facilities.

The detector is modeled to be 1.1 km (2.86 km.w.e.) un-

derground, and approximately 26 km from the Hartlepool

Reactor Complex. The Complex houses two 1.5 GWth

Advanced Gas Reactors (AGRs) yielding 3 GWth capac-

ity. The reactor antineutrino background at the Boulby

mine is relatively high due to the presence of a large

number of other operating reactors in the UK and West-

ern Europe. For speciﬁc use cases, design variations may

be employed to maximize the sensitivity of the detector,

such as increased photo-coverage, large target volumes,

or multiple detectors.

This article presents the sensitivity of a gadolinium-

doped water (Gd-H2O) detector, using the WATCHMAN

collaboration’s 2019 baseline Gd-H2O detector design [8].

We refer to this design as ‘the Gd-H2O baseline design’ or

the ‘Gd-H2O detector’ throughout this article. Sensitiv-

ity estimates are provided for exclusion of the existence of

undeclared reactors over a speciﬁed radial distance, and

for determining the presence of a hidden reactor near a

declared reactor facility. The Boulby Underground Lab-

oratory site is used to provide a concrete example of sen-

sitivity in a well-studied background environment [7].

The detection medium is contained in a cylindrical

stainless-steel tank with a 20-meter height and diame-

ter. The tank is ﬁlled with approximately 6 kilotons of

ultra-pure water mixed with gadolinium sulfate, for a to-

tal loading of 0.1% gadolinium by weight. The detector

has two optically separated regions, the muon veto region

and the inner detector. Events occurring in the 3.3 m

thick outer veto region are read out by 226 PMTs, while

arXiv:2210.09391v1 [physics.ins-det] 17 Oct 2022

FIG. 1. An illustration of the Gd-H2O baseline detector design, with a vertical cutaway through the center. The width

and height of the cylindrical tank is 20-meters. In total, the detector will house 3,554 Photo-multiplier Tubes (PMTs), 3,328

as part of the inner detector (20 % photocoverage) facing the ﬁducial volume, and 226 veto PMTs facing the tank (2 %

photocoverage)to reject cosmogenic events from within the veto volume. Other components of the tank include: a top and

bottom hatch, calibration ports, a top deck for data acquisition (DAQ) electronics, and external access locations.

events in the inner detector are read out by 3,328 PMTs.

This equates to a 20% photo-coverage for the inner de-

tector and a 2% photo-coverage for the veto detector. To

reduce backgrounds caused by radiation from the PMTs,

only events which reconstruct within the central ∼1 kton

ﬁducial volume will be considered as candidate antineu-

trinos. An illustration of the detector is shown in Figure

Gadolinium-doped water is an ideal medium for far-

ﬁeld antineutrino monitoring due to its chemical simplic-

ity, low cost, and good optical transparency[9, 10]. The

long attenuation length of the Gd-doped water allows

the medium to be scaled to larger volumes if required,

based on the particular needs of real world applications

[11, 12]. Water is a low toxicity medium that has been

deployed in many other neutrino detectors such as Su-

per Kamiokande[13], and SNO[14]. The addition of the

capture agent gadolinium allows for an enhanced neu-

tron capture signal, favorable for experiments focused

primarily on detecting reactor antineutrinos through in-

verse beta decay (IBD).

In the IBD process, antineutrinos interact with quasi-

free protons in the water, producing a positron-neutron

pair in the ﬁnal state:

νe+p−→ e++n. (1)

The positron is detected as a prompt signal through the

Cherenkov light emitted when its velocity exceeds the

speed of light in the water. This is a multiple threshold

reaction in which the νeenergy must exceed 1.8 MeV

to generate an IBD reaction, and the resulting positron

kinetic energy must exceed ∼253 keV in water to gen-

erate Cherenkov light. The neutron produced through

IBD will elastically scatter oﬀ hydrogen in the detector

until thermalization, after which it can capture on either

a gadolinium or hydrogen nucleus. For gadolinium load-

ing at 0.1%, the average capture time is ∼30 µs after

the prompt positron event [9]. Neutron capture on hy-

drogen account for 9% of captures and will result in a

single 2.2-MeV gamma ray, while captures on isotopes of

gadolinium accounts for 91% of captures and will result

in a cascade of gamma rays summing to ∼8 MeV. The

gamma rays from neutron capture will Compton scatter,

and the Compton electrons may emit Cherenkov light.

Due to the threshold required for Cherenkov emission,

not all of the scattered electrons from gamma rays re-

leased from neutron capture on Gd will contribute to the

signal, but the neutron capture on Gd still produces a

distinctly bright signature.

III. SIMULATION AND RECONSTRUCTION

FRAMEWORK

Various tools are used in the simulation, reconstruc-

tion, and event categorization to predict detector re-

sponse. The Reactor Analysis Tool-Plus Additional

Codes (RAT-PAC)[15], employs Geant4 version 10.4 [16],

10−0 10 20 30 40 50 60 70 80 90 100

Photomultiplier Tube Hit Time [ns]

Number of Events

1 2 3 4 5 6 7 8 9 10

Positron Energy [MeV]

Number of PMTs in 9 ns [n9]

0 1 2 3 4 5 6 7 8

Positron Energy [MeV]

R [meters]∆

FIG. 2. The PMT hit timing distribution from a set of

107events [A], the prompt PMT response to positrons as a

function of energy [B], and the distance between true and

reconstructed vertex as a function of energy [C].

ROOT [17], and C++ to perform high ﬁdelity Monte-

Carlo simulations of the events expected from radiolog-

ical processes, cosmogenically produced particles, and

IBD. Following event production in the detector, all the

subsequent processes are modeled, from optical photon

production and transport through the water, to photon

collection in PMTs.

The accuracy of the complex decay scheme resulting

from neutron capture on Gd, as modeled in Geant4, has

been an area of concern in recent years [18, 19]. Neutron

capture on 157Gd generates the most prominent gamma

ray cascade due to the larger cross-section relative to

the other isotopes. To remedy the modeling concerns of

Geant4, the DANCE [20](Detector for Advanced Neu-

tron Capture Experiments)Collaboration’s gamma ray

production results were incorporated into RAT-PAC us-

ing DICEBOX [21]. The resulting simulation was val-

idated against the WATCHBOY detector’s neutron re-

sponse and demonstrated better agreement with experi-

mental data than the previous 157Gd gamma ray cascade

model [22].

Following the simulation, the position and energy of

each event are reconstructed using Branching Optimiza-

tion Navigating Successive Annealing Iterations (BON-

SAI) [23]. BONSAI uses the timing and the positions

of the PMT hits to reconstruct the vertex position of

each event based on the detected Cherenkov light and

has been used by the Super Kamiokande collaboration

for low energy event reconstruction [24].

Antineutrino interactions are characterized by simu-

lating the response to positron and neutron pairs. A ﬂat

positron energy spectrum was simulated to understand

the energy-dependent detector response. The emitted

Cherenkov light from positron interactions results in ap-

proximately 9 detected photo-electrons (PE) per MeV.

To inprove the energy resolution without increasing con-

tributions from backgrounds, the energy proxy is taken

as the number of PMTs that trigger from the prompt

Cherenkov light in the ﬁrst nine nanoseconds.

The energy-dependent positron response is seen in Fig-

ure 2. Figure 2.A shows that the PMT hit timing dis-

tribution from positrons primarily occurs within a 9 ns

window following a start time deﬁned by the event ver-

tex. The negative times observed and the full width of

the peak are due to the 2 ns PMT timing jitter, while

the late light is caused by scattering in the detector. The

features around 40-70 ns are caused by after-pulsing in

the PMTs. Figure 2.B shows that positrons below 1

MeV do not produce events above the rate of events seen

from the dark rate in the PMTs. Lastly, once the prompt

PMT hits are registered, BONSAI is used to reconstruct

the events. Below 2 MeV, the the true vertex is recon-

structed to 1 meter; above 3 MeV the vertex reconstruc-

tion improves and rapidly converges at 50 cm, as shown

in Figure 2.C.

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：

ExclusionandVericationofRemoteNuclearReactorswitha1-KilotonGd-DopedWaterDetectorO.A.Akindele,1,A.Bernstein,1M.Bergevin,1S.A.Dazeley,1F.Sutanto,1A.Mullen,1andJ.Hecla11LawrenceLivermoreNationalLaboratory,Livermore,California94550,USA(Dated:October19,2022)Todate,antineutrinoexperimentsbuiltforthepurp...

展开>> 收起<<

Exclusion and Verication of Remote Nuclear Reactors with a 1-Kiloton Gd-Doped Water Detector O. A. Akindele1A. Bernstein1M. Bergevin1S. A. Dazeley1F. Sutanto1A. Mullen1and J. Hecla1.pdf

共15页,预览3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！

Exclusion and Verication of Remote Nuclear Reactors with a 1-Kiloton Gd-Doped Water Detector O. A. Akindele1A. Bernstein1M. Bergevin1S. A. Dazeley1F. Sutanto1A. Mullen1and J. Hecla1

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: