Lab ManualFast Neutron Laboratory

February 13, 2017

Contents

1 Background 31.1 The Neutron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.2 Discovery of the neutron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.3 Sources of Free Neutrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.4 Radioactive Neutron Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.4.1 Actinides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.4.2 Actinide/Beryllium Sources . . . . . . . . . . . . . . . . . . . . . . . . . 61.4.3 Am/Be source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.5 Scintillation Light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.6 Scintillator Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1.6.1 Organic Scintillators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101.6.2 Inorganic Scintillators . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

1.7 Scintillator Detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.8 PMT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121.9 Fast-Neutron Detection with Scintillator Detectors . . . . . . . . . . . . . . . . 12

2 Setup and Experimental Techniques 142.1 Neutron Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.2 The Source Testing Facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.3 Source Tank and Shielding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.4 Detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.4.1 The YAP Scintillator Detector . . . . . . . . . . . . . . . . . . . . . . . . 172.4.2 The Liquid Scintillator Detector . . . . . . . . . . . . . . . . . . . . . . . 17

2.5 Data Acquisition and Electronics . . . . . . . . . . . . . . . . . . . . . . . . . . 192.6 Experimental Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.7 Pulse Shape Discrimination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202.8 Energy-Tagged Fast-Neutrons and Time-of-Flight . . . . . . . . . . . . . . . . 22

2

1 Background

In this chapter,the neutron as a fundamental particle is introduced and the theory behinddetecting neutrons from radioactive sources with scintillator detectors is given. The mixed-field neutron/gamma-ray source used in this work is introduced. The intrinsic propertiesof scintillator materials and the different scintillation processes in organic scintillators andinorganic scintillators are presented. The general detector concept of scintillator detectors ispresented. Last, the detection of fast neutrons with scintillator detectors is discussed.

1.1 The Neutron

A neutron is a subatomic particle with no electric charge and a mass of 1.009 u (atomic massunits1), slightly bigger than a proton. Together, these particles form the nucleus and arebound by the strong force that attracts all nucleons separated by about 1 fm. A neutronis formed from quarks, also held together by the strong force. While bond neutrons arestable, free neutrons are unstable, with a half life of about 15 minutes. The strong force isapproximately 137 times stronger than the Coulomb force and much stronger than any otherfundamental force. Most of the mass of a neutron or a proton comes from the strong forcefield energy, the mass of the quarks themselves ads up to only 1% of the total mass.

Figure 1: Quark structure of a neutron [1]

1.2 Discovery of the neutron

Historically, a chain of crucial revelations led to the discovery of the neutron. In the begin-ning of the 20th century the focus of physics turned to the atom. The 1911 Rutherford modeldescribed the atom as having a heavy but small positively charged nucleus surrounded by amuch bigger negatively charged electron cloud. In 1920, Rutherford introduced a new modelto account for mass and charge differences. The nucleus was now seen as consisting of pro-tons and “neutral" particles thought to be made of a proton and an electron bound together.For 10 years, this was the prevailing model of the nucleus. Nuclear electrons were justified

1The unified atomic mass unit (u) is the standard unit used to indicate the mass on the atomic scale.1 u=1.66053904× 10−27 kg

3

because it was known that beta radiation is emitted from the nucleus. However, Heisen-berg’s Uncertainty Principle and inconsistencies regarding particle spins still presented aproblem. In 1931, a deeply penetrating radiation was observed when beryllium, boron, orlithium was bombarded with α-particles. This was initially thought to be gamma-rays, butleft some scientists doubtful. In 1932, James Chadwick performed a series of experiments todetermine the nature of this radiation and concluded that it must consist of uncharged parti-cles having masses essentially equal to that of the proton. These particles were neutrons. In1935, Chadwick won the Nobel Prize for his work. The discovery of neutrons was a big stepforwards for nuclear physics. It opened doors for new experiments and never-before-seenradioactive elements were found. Nuclear fission was discovered consequently leading tothe development of the atom bomb.

1.3 Sources of Free Neutrons

Free neutrons can be obtained in a multitude of ways. They can be found as natural radiationfor example as secondary products from cosmic rays and from fission of natural occurringuranium. These natural neutron sources are often a background source in modern exper-iments, but are unsuited for controlled detector characterization. For controlled neutronirradiations there are essentially three sources: accelerators, nuclear reactors or radioactiveneutron sources.

Figure 2: The Large Hadron Collider[2]

Figure 3: The Watts Bar Nuclear Plant[3]

High flux accelerator based neutron sources are based on a process called spallation (likethe ESS in Lund) using high energy protons (1 GeV) on a heavy target (tangsten, mercury,

4

lead) [4, 5]. Lower flux sources are using a wide range of nuclear reactions between light pro-jectiles (hydrogen, deuterium or tritium) light target isotopes (deuterium, tritium, lithium,beryllium). These accelerators operate with energies of 1 MeV or higher[6].

Nuclear reactors have been used for neutron science from the 1950s. Neutrons are pro-duced in a controlled chain-reaction by fission of nuclear fuel.

1.4 Radioactive Neutron Sources

In radioactive sources, neutrons are produced by nuclear reactions following the radioac-tive decay of an unstable isotope. They are used widely in such diverse fields as land-mine detection [7], education [8], detector characterizations [9] and oil-well logging [10].Two types of radioactive neutron sources (actinide/9Be sources and the spontaneous fissionsource 252Cf) are regular employed to characterize fast-neutron detectors with source-basedneutron-irradiation techniques. In this laboratory actinide/9Be sources will be used.

Neutrons produced by actinide/Be sources have been essential to the field of neutronphysics since the discovery of the neutron by Chadwick [11] in 1932. Chadwick used a polo-nium alpha-particle source and a 9Be target to produce neutrons in his famous experiment.Today, the most commonly employed neutron sources of the actinide/Be type are eithermixed-isotope Pu [12, 13, 14] or 241Am [15, 16, 7]. The Pu-based sources are often legacysources from the 1970s and earlier, while Am/Be sources continue to be supplied commer-cially.

In actinide/Be sources most of the fast neutrons (E> 0.5 MeV) are produced when analpha-particle interacts with 9Be to produce 12C and a fast neutron according to:

9Be (α, n)12 C(∗). (1)

1.4.1 Actinides

Figure 4: Actinides[17]

All but one of the actinides are so called “f-block" elements, corresponding to the filling ofthe 5f electron shell. They have very large atomic and ionic radii and exhibit a wide range ofphysical properties. All actinides are radioactive and release energy upon decay. Many aresynthetic elements [18]. Nuclear weapons tests have released at least six actinides heavierthan plutonium into the environment. As early as 1952, analysis of debris from hydrogen-bomb detonations showed the presence of americium, curium, berkelium, californium, ein-steinium and fermium[19].

5

1.4.2 Actinide/Beryllium Sources

Figure 5: Relative probabilities for the state of the recoiling 12C in an α/Be reaction[20].

Beryllium-based neutron sources exploit a particular nuclear reaction involving an α-particle. The α-particle enters a natural beryllium nucleus and is captured, resulting in a 12Cnucleus forms and a free neutron:

α +9 Be→12 C + n + Q, (2)

Q represents the energy difference of parent and daughter nuclei.The recoiling carbon nucleus can be produced in the ground state, the first, or the second

excited state. The calculations of Vijaya and Kumar[20] suggest that the relative popula-tions of the ground/first/second excited states for the recoiling 12C nucleus are ∼ 35%/ ∼55%/ ∼ 15% (see figure 5).

If the recoiling 12C nucleus is left in its first-excited state, it will promptly decay to theground state via the isotropic emission of a 4.44 MeV gamma-ray. Mowlavi and Koohi-Fayegh[21] as well as Liu et al.[22] have measured R - the 4.44 MeV gamma-ray to neutronratio - for an actinide/Be source to be approximately 0.58.

The α-particle can come from any number of radioactive α-emitters, or from a particleaccelerator that accelerates helium ions. Typical radioactive emitters historically used inneutron sources are 222Rn, 226Ra, 210Po, 239Pu and 241Am, among others. Properties suchas yield, lifetime, spectrum, and also price, depend on the α-particle source. For example,Po/Be α-emitters have a high yield, but due to the short half-life of 210Po - only 138.2 days, it

6

is necessary to make time-dependent corrections for the yield, and the source rather quicklybecomes too weak to use. Chemically, these sources are mixtures, therefore the spectrumvaries with grain size of the powders which are combined. Even after the source is fabricated,the extent of mixing can change for example due to heating. Pu/Be, on the other hand, is anintermetallic compound, and the spectrum consequently does not depend upon grain size.In addition, its relatively long half-life (2.2× 104 years), assures constancy within a life time.

1.4.3 Am/Be source

Figure 6: X.3 capsule

Am/Be radioactive source is a mixture of americium oxide and beryllium. Radioactive241Am is a long-lived alpha emitter that has a half-life of 432.2 years2. Many of Am/Be prop-erties make it well suitable for experimental work. This laboratory uses a (nominal) 18.5 GBqAm/Be source contained in a so-called “X.3" capsule, see figure 6. The manufacturer guar-antees the source for 15 years. After that, the source should be retired. The 241Am alphadecay is:

241Am→ 237Np + α + Qα (3)

The Q-value (in MeV) of this process[23] is:

2The ground state of the daughter nucleus, 237Np, is also unstable, but has a very long half-life of over 2million years.

7

Qα = (M241Am − (M237Np + Mα))(931.494MeV/u)

= (241.056822u− (237.048166u + 4.00260325u))(931.494MeV/u)= 5.638MeV,

(4)

M is the mass.

Fast neutrons are produced when the decay α-particles subsequently interact with 9Be:

α +9 Be→ n +12 C + Q. (5)

Depending on the interaction and its kinematics, 12C and a free neutron may be produced.The resulting free-neutron distribution has a maximum value of about 11 MeV and a sub-structure of peaks, whose energies and relative intensities vary depending upon the proper-ties of the Am/Be source containment capsule and the size of the 241AmO2 and Be particlesin the powders employed. In general, approximately 25% of the neutrons emitted have anenergy of less than ∼ 1 MeV with a mean energy of ∼ 400keV[20]. The average fast neutronenergy is ∼ 4.5 MeV[20]. The unshielded source used in this lab has been independentlydetermined to emit (1.106± 0.015)106 neutrons per second nearly isotropically. The kine-matics and the reaction cross section for the 9Be(α, n) interaction determine the state of therecoiling 12C nucleus produced in the reaction. As already mentioned, almost 60% of theneutrons emitted by an Am/Be source are accompanied by a prompt, time-correlated 4.44MeV gamma-ray.

1.5 Scintillation Light

Scintillation light is a form of luminescence that results from the de-excitation of electronstates which were excited by the interaction with ionizing radiation. Depending on thetime scale of the de-excitation, it is common to divide the emitted scintillation light intotwo groups: phosphorescent light for slow decays (τ ≥ 1µs) and fluorescent light for fastdecays (τ ∼ 1ns). Both time scales can be correlated to different electron excitation-states,singlet or triplet. Fluorescence corresponds to photo-emissions from de-excitation of an ex-cited singlet state to a lower state of the same spin multiplicity. Singlet-singlet transitionsare quantum mechanically favored and are therefore fast. Phosphorescence corresponds tophoto-emissions from the de-excitation of an excited triplet state to a lower singlet state.Triplet-singlet transitions are quantum mechanically forbidden and the decays are thus de-layed.

Scintillation in most materials is an inefficient process and scintillators convert only asmall fraction of the energy deposited by ionizing radiation into scintillation light. Most ofthis energy is lost to heat and vibrational excitation and not available for light production.The fraction of the deposited energy that becomes scintillation light, the scintillation effi-ciency S, depends for most scintillator materials on both the type of ionizing radiation and

8

the incident particle energy. This is explained by a reduction of the scintillation efficiency(quenching) of (organic) molecules damaged by the high ionization density of protons andheavy ions. If S is independent of the particle energy, the light yield L, the amount of scin-tillation light produced, is linear with respect to the incident particle energy. This is the casefor electrons in both inorganic and organic (above 100 keV) scintillator materials.

The amount of scintillation light dLdx created by a particle per unit path length is given by

[24]

dLdx

= SdEdx

, (6)

where dEdx is the energy deposition per unit path length. The density of damaged molecules

along the path of a particle is to good approximation directly proportional to the ionizationdensity of the particle. The damaged molecule density can therefore be estimated to be B dE

dx ,where B is a proportionality constant. If quenching occurs in only a fraction k of these dam-aged molecules, the light yield L of a proton and heavy ion can be estimated by [25, 24]

dLdx

=S dE

dx

1 + kB dEdx

(7)

The different light yields of a NE-213 liquid scitillator detector for electrons, protons,alpha-particles and 12C are shown in Fig. 7. A more detailed description of light productionin scintillators can be found in [24, 25, 26].

ELJEN TECHNOLOGYSWEETWATER, TEXAS

RESPONSE OF EJ-301 LIQUID SCINTILLATOR

PARTICLE ENERGY (MeV)

NU

MBE

R O

F SC

INTI

LLAT

ION

PH

OTO

NS

PRO

DU

CED

SCINTILLATION LIGHT PRODUCED VS. PARTICLE ENERGY

ELECTRONS

PROTONS

ALPHAS

CARBONS

106

105

104

103

102

100.1 1 10 100

Data Source: V.V. Verbinski et al, Nucl. Instrum. & Meth. 65 (1968) 8-25

Figure 7: The light yield of NE-213 for electrons, protons, alpha-particles and 12C. The non-linear light production and the quenching of the light yield from protons and ions can beclearly identified. Figure from [27].

1.6 Scintillator Materials

Scintillators come in all states of matter and the fundamental interactions leading to scintilla-tion light are diverse. It is often convenient to divide scintillator materials into three groups:

9

Figure 8: A schematic representation of the excitation states of organic scintillators. Figurefrom [24].

organic scintillators, inorganic scintillators, and noble-gas scintillators. Organic scintillatorsare mixtures of aromatic molecules which are in liquid solutions, plastics or organic crys-tals. Inorganic scintillators come in a wide variety of states such as single crystals, glasses,ceramics and gases. Noble-gas scintillators can be used as gases or liquids. In this studentlaboratory, a liquid organic scintillator (NE-213 based on xylene) and an inorganic single-crystal (YAP) are employed.

1.6.1 Organic Scintillators

Organic scintillators consist of aromatic organic molecules, hydrocarbons with benzene likecyclo-hexal structures.Carbon atoms have four available valence electrons and form withhydrogen either saturated hydrocarbons or molecules with double or triple electron bonds.In aromatic molecules, carbon atoms form double bonds and one s-orbital and two p-orbitalscombine to three equivalent orbitals (sp2 hybridization). Such σ-electrons are strongly boundand the orbitals do not produce luminescence. However, the last p-orbital is symmetric withrespect to the molecular plane and the associated electron is called a π-electron. Such π-electrons are weakly bound (see Fig. 8) and form in aromatic molecules one de-localized

10

orbit. In organic scintillators, scintillation is caused by transitions from the excited statesof these π-orbitals to the ground state. An electron excitation diagram of the π-molecularorbitals is shown in Fig. 8 (b) [24]. The different energy levels are a result of the relative spinorientation of the excited electron, compared to the spin orientation of the unpaired electronin the ground state, in either parallel (triplet Ti) or anti-parallel alignment (singlet Si). Thevibrational excitation states (Sij) arise from the collective motion of atoms in a molecule.The de-excitation from the singlet state S1 to S0 (fluorescence) by the emission of light is afast (∼ ns) process. The excited triplet state T1 de-excites by slower processes. The highlyforbidden transition process T1 → S0 (phosphorescence) has a time scale of µs or longer. Themost common way to de-excite the T1 states is indirect delayed-luminescence via S1 → S0.This process is considerably faster than phosphorescence and responsible for the delayedpart of scintillator light pulses in organic scintillators. T1 can be excited by inter-molecularinteraction to singlet states. For a detailed description of scintillation in organic scintillators,see [24, 25]

1.6.2 Inorganic Scintillators

Scintillation-light production in inorganic scintillators depends on the energy states of thecrystal lattices of the material. In crystals, electrons have only discrete energy levels availablewhich form clearly separated energy bands. Electrons in the the valence band, the lowerenergetic band, are strongly bound at the lattice site. Electrons in the conduction band havesufficient energy to freely migrate through the crystal. In the band gap of a pure crystal, theenergy levels between the valence band and the conduction band, no electrons are found.If energy is absorbed by the crystal, an electron from the valence band can be excited to theconduction band.

The hole left by such a process can then be filled again if the electron returns to the valenceband resulting in the emission of light. It is important to remember that in pure crystals thisprocess has two major drawbacks for detectors. First, the excited electron state has a longlife time and the light emission process is often delayed (∼ 100 ms). Second, the band gapin typical scintillator materials is large (∼ 4 eV) and the emitted (UV) light is outside thesensitivity range of photo-multiplier tubes (PMTs) (see Sec. 1.8).

To increase the amount of visible scintillation light, small amounts of impurities are reg-ularly added to inorganic scintillators. These activator atoms create locations in the crystallattice where the energy-band structure of the pure crystal is modified. The activators arenormally chosen to create new energy states in the otherwise forbidden energy-band gap.Electrons can now de-excite in a two-step process through the intermediate activator state.The light emitted by these transitions will be shifted to lower energies compared to the fullband gap and can be detected with PMTs. For a more detailed explanation of the scintillationprocess in inorganic scintillators, see [24, 25, 26].

1.7 Scintillator Detectors

The general detector concept is the same for both scintillator types employed in this thesis.The scintillation material is enclosed in a light-tight housing which can be thin aluminum

11

layers and black tape for plastic scintillators or several millimeter thick metal cells for gasscintillators. The inside of this housing is often covered with a reflective material, polishedmetal or reflective paint. If a particle interacts in the detector medium, it creates scintillationlight. This light is focused onto a light detector which is optically coupled to the scintillla-tor material with a window and a non-scintillating light guide. For solid materials, directcoupling between the scintillator material and the light detector is also possible. Figure 9illustrates this concept for a liquid-scintillator detector interacting with a neutron. The mostcommon light sensitive devices used to register scintillation light are photo-multiplier tubes(PMTs).

n Scintillator

Light GuidePMT

n’

p

Figure 9: The basic concept behind a scintillator detector. Scintillation light is produced bya neutron-induced recoiling proton. The light is guided from the detector cell through thelight guide to the photo-cathode.

1.8 PMT

Figure 10 shows a schematic drawing of a PMT. An evacuated glass tube is the outer enve-lope containing the photo-cathode, a photo-sensitive metal layer, and the electron-multiplyingdynode assembly. The photocathode transforms ∼ 25% of the incident light quanta intophoto-electrons [24]. Focusing electrodes guide the primary electrons to the first multiplyingdynode. The electron number increases throughout the dynode system in an avalanche-likeprocess. The amplified pulse eventually strikes the anode and the resulting anode currentcan be measured.

1.9 Fast-Neutron Detection with Scintillator Detectors

Neutrons are electrically uncharged particles which are detected by reactions with the de-tector medium. The preferred detection reactions are energy dependent. Neutrons can becaptured on high cross section target isotopes such as 10B, 6Li, 155,157Gd and 3He. This is thepredominant detection concept for cold (∼ 2.2 meV), thermal (∼ 25 meV) and epithermal(∼ 1 eV) neutrons. For fast neutrons (>0.5 MeV) neutron-nucleus scattering is employed

12

Figure 10: The basic elements and functions of a PMT. Light is converted on the photo-cathode into electrons. The focusing electrodes and electron multiplier dynodes are alsoshown. Figure from [28].

as the more common detection technique. Figure 11 shows the neutron-nucleus scatteringprocess in the laboratory frame and in the center-of-mass frame. The energy of the recoilnucleus in the lab system can be calculated for non-relativistic neutrons by:

ER =4A

(1 + A)2 cos2 (θ)En, (8)

where A is the mass number of the target nucleus, ER is the kinetic energy of the recoilnucleus, θ represents the scattering angle of the scattered nucleus in the lab frame and En isthe kinetic energy of the neutron [24]. Equation 8 shows that the energy transfer is inverselyproportional to the mass number of the target nucleus and hence light targets are preferred.The maximum energy transfer to some of the typical target materials in fast-neutron detec-tors are listed in Table 1.

Table 1 shows that hydrogen and helium are the preferred target materials for neutronscattering. In NE-213 liquid scintillator made from hydrogen and 12C, neutrons are detectedmainly by neutron-proton scattering.

13

En

ER

A

A

neutron

neutron

target nucleus

recoiling nucleus

E’n

Lab System

θ

En

ER

A

A

neutron

neutron

target nucleus

recoiling nucleus

E’n

CM System

Θ

θ

φ

Figure 11: Scattering of an incoming neutron of the energy En and the target nucleus withrecoil energy ER in the center-of-mass frame on the left side and in the lab fram on the rightside. Figure from [24]

2 Setup and Experimental Techniques

In this section, an overview of the measurement setup and the experimental techniques isgiven. The neutron sources, the laboratory environment of the Source Testing Facility (STF),the shielding, the detectors and the data-acquisition systems are described. The detectorcalibration of the liquid-scintillator detectors, the pulse-shape discrimination technique andthe time-of-flight technique for the detection of energy-tagged fast neutrons are presented.

2.1 Neutron Sources

Two different neutron sources can be used in this laboratory and some of their propertiesare listed in Table 2. Both actinide/Be sources are housed in robust X.3 type cylinders of tig-welded, double-layered stainless steel, and are 31 mm in height and 22.4 mm in diameter.The alpha-particles from the decay of the actinide isotopes are therefore completely stoppedin the capsules.

14

Target A Maximum Energy Transfer1H 1 1.0002H 2 0.889

3He 3 0.7504He 4 0.64012C 12 0.28416O 16 0.211

Table 1: Maximum neutron–energy transfer for different target isotopes.

Location STF STFLaboratory

241Am/9Be 238Pu/9Be3

Supplier High Technology The Radiochemical CentreSources (HTS) [29] Amersham [30]

year of production 2015 1972neutron yield 1.1 · 106 n

s –nominal activity 18.5 GBq –

unwanted radiation gamma-ray n/a∼ 60 keV

Table 2: An overview of the properties of the neutron sources employed in this laboratory.

2.2 The Source Testing Facility

The Source Testing Facility has been established at Lund University in a joint effort betweenthe Division of Nuclear Physics and the Detector Group of the European Spallation SourceERIC (ESS). Envisioned as a user-facility, the STF not only provides access to a wide array ofneutron and gamma-ray sources but also to all infrastructure necessary for neutron-detectordevelopment and shielding studies. It is located next to the Lund Ion-Beam Analysis Facility(LIBAF) [31].

A schematic 3-D drawing of the STF is shown in Fig. 12. The interlocked irradiation areais shown on the left side of the figure. This is the space where irradiations are performed.The interlocked area is separated from the main part of the STF by an access-controlled gate.Patch-panels route signals, high voltage and gas connections to the DAQ area on the rightside of the gate. Here, the measurements can be remotely controlled and data-acquisitionperformed.

2.3 Source Tank and Shielding

A major component of the apparatus located at the STF is a water-filled tank (see Fig. 13),known as the Aquarium. It is located in the interlocked area of the STF and is used to de-fine beams from the mixed neutron/gamma-ray radiation field that is emitted by a neutron

15

Storage & StagingArea

Interlocked Area

DAQArea

LIBAF

Figure 12: A schematic drawing of the STF. The interlocked area, the DAQ area and thedetector storage and staging area are identified by the boxed labels. The position of theLIBAF is indicated.

Inner Tank

Lower Tank

Upper Tank

Figure 13: The Aquarium. (a) 3-D schematic drawing of the Aquarium, made up of the lowertank, the upper tank and the inner tank. The air-pad frame is also shown. (b) A photographof the inner chamber. The 4 YAP detectors and the sleeves holding them in position areclearly visible.

16

source. The Aquarium is made from poly-methyl-methacrylate (PMMA), also known asplexiglass, and was produced in a modular design of lower tank, upper tank and inner tank(see Fig. 13). The Aquarium has outer dimensions of 1400 mm × 1400 mm × 1400 mm.

The inner chamber (400 mm× 400 mm× 280 mm) houses the neutron source (see Fig. 13)and up to 4 YAP detectors (see Section 2.4.1). PMMA sleeves (see Fig. 13) are used to setthe YAP detectors in well-defined positions. These sleeves are inserted through designatedopenings A fifth slightly smaller vertical penetration for the source is located in the geo-metrical center of the Aquarium. The Aquarium is filled with ∼ 2650 liters of ultra pure,de-ionized water. Four beamports 172 mm (diameter) and 500 mm (length) allow for foursimultaneous, independent irradiations.

2.4 Detectors

Two different kinds of radiation, gamma-rays and neutrons, are investigated with the detec-tors presented here.

2.4.1 The YAP Scintillator Detector

YAP:Ce detectors (Ce+ doped yttrium aluminum perovskit) were used to measure the 4.44 MeVgamma-rays from the first excited state of 12C. YAP is an excellent choice for a gamma-raydetector operated near an intense neutron source. The material is radiation hard and is quite(but not completely) insensitive to neutrons. YAP is a mechanically and chemically resistant,single-crystal scintillator with a relative low effective Z ∼ 31.4. The material has a fast rise-time (∼ 5 ns) and a short decaytime (∼ 27 ns) which allows for fast timing information andhigh count rates. The 1 inch high and 1 inch diameter single crystal is coupled to a 1 inchPMT. The PMTs are operated at about −800 V. The energy resolution was determined to be∼ 10% at 662 keV determined using 137Cs. The YAP detectors are not used for spectroscopybut rather as trigger detectors.

2.4.2 The Liquid Scintillator Detector

The liquid scintillator detector is sensitive to both gamma-rays and fast-neutrons. An as-sembled fast-neutron detector is shown in Fig. 15. The detector consistes of a 3 mm thickaluminum cell with a 94 mm inner diameter and a 62 mm inner depth filled with ∼ 430 mlof NE-213 [32]. NE-213 is a commonly used liquid organic scintillator for fast-neutron de-tection based on xylene4, an aromatic molecule of the formula C6H4(CH3)2. The scintillatoremits light at 425 nm and has a bulk light-attenuation length of ∼ 3 m. The filled cell wascoupled to a 72.5 mm (diameter)× 34 mm (length) PMMA lightguide. Acrylic PMMA trans-mits light down to wavelengths of 300 nm [33] and has a refractive index of ∼ 1.491. Thewalls of the lightguide were painted with water-soluble EJ-510 [34] reflective paint. The light-guides were pressure-coupled to spring-loaded, magnetically shielded 3 inch ET Enterprises9821KB PMT assemblies [28] operated at about -2000 V.

4Newer versions are based on pseudocumene which is both less toxic and less flammable than xylene.

17

Figure 14: Photograph of a YAP detector. The PMT and the PMT assembly (base and mag-netic shielding) are located on the polished side to the left. The YAP crystal is located underthe matt surface to the right.

Figure 15: A photograph of an assembled liquid-scintillator detector. The black tube attachedto the detector cell is a magnetically shielded 3 inch ET Enterprises 9821KB PMT assembly.

Type Module Name Manufacture StandardCFD PS 715 Phillips Scientific NIMCFD Ortec 935 Ortec NIM

SG QDC V792 CAEN VMELG QDC V792 CAEN VME

TDC V1190B-2eSST CAEN VMEScaler V830 CAEN VME

PCI-VME adapter SIS3100 Struck Optical Fiber ConnectionCBD CES 8210 Creative Electronic Systems VME

Table 3: An overview of the electronic modules used in the laboratory.

18

2.5 Data Acquisition and Electronics

The data-acquisition system used in this laboratory is written in C++ and based on the data-analysis framework ROOT [35] to provide online data monitoring and data storage. Signalsare recorded on an event-by-event basis for offline analysis using a LINUX PC DAQ system.The analog signals from the liquid-scintillator detector and the YAP detectors are passed tolong-gated (LG) charge-to-digital converters (QDCs) with integration gates of 500 ns andto short-gated (SG) QDCs with integration gates of 60 ns. The gates are opened ∼ 25 nsbefore the analog pulses arrived. The analog signals are also passed to constant fraction(timing) discriminators (CFDs). The logic signals from the discriminators are passed to digi-tal scalers and time-to-digital converters (TDCs). The VME crate containing the QDCs,TDCsand scalers is connected to the PC by a PCI-VME bus adapter. The DAQ is triggered by thedistant detector, NE-213. The modules used for this lab are listed in Table 3.

Figure 16: A simplified overview of the tagged-neutron setup (not to scale). The Am/Besource, the Pb sleeve, a single YAP detector, and the NE-213 detector are all shown togetherwith a block electronics diagram.

2.6 Experimental Configuration

A block diagram of the tagged-neutron configuration is shown in Fig. 16. The actinide/Besources are placed in the inner chamber of the Aquarium so that their cylindrical-symmetryaxes correspond to the vertical direction in the lab. The YAP detectors are placed with their

19

crystals approximately 10 cm from the actinide/Be sources. The crystal orientation of theYAPs is such that their cylindrical symmetry axis also corresponds to the vertical directionin the lab. These detectors trigger overwhelmingly on the 4.44 MeV gamma-rays radiatingfrom the source which come from the decay of the first excited state of 12C. The NE-213 cellis placed 0.5 m - 2.5 m from the actinide/Be sources at source height and is free-running. TheNE-213 detector detects both neutrons and gamma-rays.

2.7 Pulse Shape Discrimination

The different shapes of the light pulses from gamma-ray-induced recoiling electrons andfrom neutron-induced recoiling protons were measured using the “Tail-to-Total” ChargeComparison Method [36, 37, 38] (see Fig.17). This method compares the delayed light (tail) ofa scintillation pulse to the total light amount of a scintillation pulse produced by the incidentradiation.

The pulse-shape function PS is defined to measure the ratio between delayed scintillationlight and the total light pulse and uses the difference of the charge measured by the LG QDC(500 ns) and the SG QDC (60 ns) normalized to the LG QDC charge

PS =LG QDC− SG QDC

LG QDC. (9)

The Figure-of-Merit (FOM), a value indicating the discrimination strength of a PSD tech-nique, is a useful concept to measure the gamma-ray rejection strength of different scintilla-tor materials and decide which is best. The FOM is defined as

FOM =|mn −mγ|

δn + δγ. (10)

An integrated PS spectrum (L ≥ 0.58 MeVee) is shown in Fig.18. Two peaks are clearlyvisible. Events below PS = 0.2 correspond to gamma-ray induced recoiling electrons. Eventsabove PS = 0.2 correspond to neutron-induced recoiling protons. The FHWMs of the Gaus-sian fits δn,γ and the positions of the means mn,γ are used to calculate the FOM of the detectorfor a given pulse-height threshold cut. The FOM of the detector is a number strongly corre-lated to the integration threshold. The higher the threshold value, the better the discrimina-tion and the larger the FOM. Figure 19 shows a contour plot of PS plotted against the lightyield L for the NE-213 detector.

The excellent discrimination between neutron and gamma-ray events is apparent for L >0.5 MeVee. A small overlap exists between the two distributions below this threshold inthe vicinity of PS = 0.2. gamma-rays events can be discriminated against by setting a PSthreshold to about 0.2 so that only neutron like events will be processed further.

20

Figure 17: An illustration of the difference in the pulse shape of a neutron-induced recoilingproton (red) and a gamma-ray induced recoiling electron (blue) .

Figure 18: An integrated PS plot for all light yields larger than 0.58 MeVee is shown as anexample of a FOM.

21

]ee

L [MeV0 1 2 3 4 5

PS

[arb

. u

nit

s]

0.2−

0

0.2

0.4

0

2000

4000

6000

8000

10000

Figure 19: Contour plot of PS versus light yield L in MeVee.

2.8 Energy-Tagged Fast-Neutrons and Time-of-Flight

Neutron energies in this laboratory are measured indirect by establishing a coincidence be-tween the products of the nuclear reaction 9Be(α,ngamma)12C in actinide/Be sources. Neu-trons detected in these types of coincidence measurements have a well-defined reference tothe time of the reaction T0. They are thus tagged and their energy can be determined fromtheir time-of-flight TOFn and their flight path s.

The tagging setup used to measure TOFn is illustrated in Fig. 20. The flight-path from thesource to the NE-213 detector s, the point were the nuclear reaction occurs at the time T0, andthe time-of-flight of the gamma-rays to the NE-213 detector (TOFγ), of the neutrons to theNE-213 detector (TOFn) and of the gamma-rays to a YAP detector (tYAP,γ) are all illustrated.The CFDs, the delay of the YAP signals and the TDC are also shown. The TDC measures thetime difference between the arrival of a start signal and the arrival of a stop signal.

The experiment shown in Fig. 20 does not measure the TOF direct. Instead, the timedifference between the arrival of the start signal from the liquid-scintillator detector and thearrival of the stop signal from the YAP detector is measured. The absolute TDC values of agamma-ray/gamma-ray coincidence (t’γ,γ) and a neutron/gamma-ray (t’n,γ) coincidence arethus influenced by the experimental setup and will change for example if the length of thecable connecting one of the detector to the DAQ is changed. However, the time differencebetween the two events |t′γ,γ − t′n,γ| will not change. This fixed difference is used to calculatethe neutron TOFn:

TOFn = |t′γ,γ − t′n,γ|+ TOFγ, (11)

where the TOFγ, the gamma-ray TOF, can be calculated from the speed of light c and theflight-path s. The actinide/Be sources produce two prompt regions in the TDC spectrum:one from gamma-ray/gamma-ray coincidences and the other from neutron/gamma-ray co-incidences. The neutron/gamma-ray region is the result of a gamma-ray detected in the YAPdetector and a neutron detected in the liquid-scintillator detector. The prompt neutron/gamma-ray distribution is wide and structured, since neutrons corresponding to the first-excited

22

Am/Bes

NE-213

T0TOFn,γ

γ

n

YAP

γ

T0

TOFYAP,γ

TDCCFD

CFD

start delaystop

Figure 20: A schematic illustration of a TOF measurement.

TDC [arb. units]1500 1550 1600 1650 1700 1750 1800 1850 1900

co

un

ts

0

1000

2000

3000

4000

5000

6000

7000

γn,t’ 0Tγ,γ

t’nTOF ­flashγ

neutrons

Figure 21: The measured TDC coincidence spectrum between particles detected in the NE-213 and particles detected in a YAP detector. A TDC entry represents the time between thearrival of the start signal from the NE-213 detector and the arrival of the stop signal from theYAP detector. The gamma-flash and the neutron region are indicated with a blue and a redarrow respectively.

23

state in 12C are emitted with a wide range of energies. The gamma-ray/gamma-ray region isproduced by the simultaneous detection of a gamma-ray in the NE-213 detector and a corre-lated gamma-ray in the YAP detector. The gamma-ray/gamma-ray coincidence produces anarrow peak called the gamma-flash. The gamma-flash position relative to T0 (when the re-action occurred) in the coincidence spectrum can be calculated from the flight-path s and theTDC time resolution (250 ps

CH ). The neutron region populates lower coincidence TDC chan-nels, while the gamma-ray/gamma-ray region is at higher TDC channels. This is due to thedelay of the YAP signal. The time between a start signal coming from a neutron detected inthe NE-213 detector and the stop signal from a gamma-ray detected by the YAP detector isindicated by t’n,γ. The time between a start signal arising from a gamma-ray in the NE-213and a stop signal from a gamma-ray detected in the YAP is shown as t’γ,γ.

The neutron TOF can be transformed on an event-by-event base into neutron energy. Thekinetic energy is calculated according to:

Tn =12

mnc2 s2

TOF2n · c2

(12)

where mn is the neutron mass, s is the distance from the source to the detector, TOFn is theneutron TOF and c is the speed of light. Using the classical energy formula is acceptable inthis work due to the low energies (<10 MeV) involved.

References

[1] https://commons.wikimedia.org/wiki/File:Quark_structure_neutron.svg.

[2] CERN Maximilien Brice. View of the lhc tunnel sector 3-4., 5th October 2009.

[3] Tennessee Valley Authority. Watts bar nuclear plant, n.d.

[4] https://europeanspallationsource.se/.

[5] Andrew Taylor, Mike Dunne, Steve Bennington, Stuart Ansell, Ian Gardner, Peter Nor-reys, Tim Broome, David Findlay, and Richard Nelmes. A route to the brightest possibleneutron source? Science, 315(5815):1092–1095, 2007.

[6] H.L. van der Horst. "VIIIc Neutron Generators". Gas-Discharge Tubes. Philips TechnicalLibrary, 1964. OCLC 10391645. UDC No. 621.387.

[7] D. Rezaei Ochbelagh, H. Miri Hakimabadb, and R. Izadi Najafabadib. Nucl. Inst. andMeth. A, 794:102, 2015.

[8] http://www3.nd.edu/ wzech/LabManual_0907c.pdf.

[9] A.A Naqvi et al. Nucl. Inst. and Meth. A, 353:156, 1994.

[10] C.M. Frankle and G.E. Dale. Nucl. Inst. and Meth. A, 723:24, 2013.

24

[11] J. Chadwick. Nature, 129:312, 1932.

[12] I. El Agib and J. Csikai. Nucl. Inst. and Meth. A, 435:456, 1999.

[13] Bulent Buyuk and A. Beril Tugrul. Annals of Nuclear Energy, 71:46, 2013.

[14] J. Bagi, L. Lakos, and C.T. Nguyen. Nucl. Inst. and Meth. B, 366:69, 2016.

[15] B. Milenkovic, N. Stevanovic, D. Nikezic, and D. Kosuti. Appl. Radiat. Isot., 90:225, 2014.

[16] R.B.M. Sogbadji et al. Appl. Radiat. Isot., 90:192, 2014.

[17] Lawrence Livermore National Laboratory. Actinide table, 5th June 2007.

[18] Theodore Gray. Elements: A Visual Exploration of Every Known Atom in the Universe. BlackDog & Leventhal, 2009.

[19] P. R. Fields, M. H. Studier, H. Diamond, J. F. Mech, M. G. Inghram, G. L. Pyle, C. M.Stevens, S. Fried, W. M. Manning, A. Ghiorso, S. G. Thompson, G. H. Higgins, andG. T. Seaborg. Transplutonium Elements in Thermonuclear Test Debris. Physical Review,102:180–182, April 1956.

[20] A. D. Vijaya and A. Kumar. The neutron spectrum of Am#58872Be neutron sources.Nuclear Instruments and Methods, 111:435–440, September 1973.

[21] Ali Asghar Mowlavi and Rahim Koohi-Fayegh. Determination of 4.438 mev γ-ray toneutron emission ratio from a 241am–9be neutron source. Applied Radiation and Isotopes,60(6):959 – 962, 2004.

[22] Zhenzhou Liu, Jinxiang Chen, Pei Zhu, Yongming Li, and Guohui Zhang. The 4.438mev gamma to neutron ratio for the am–be neutron source. Applied Radiation and Iso-topes, 65(12):1318 – 1321, 2007.

[23] Decay energy. In Michael F. L’Annunziata, editor, Radioactivity (Second Edition), page124. Elsevier, Boston, second edition edition, 2016.

[24] G.F. Knoll. Radiation Detection and Measurement. John Wiley and Sons, 1989.

[25] J.B. Birks. The Theorymand Practice of Scintillation Counting. Pergamon Press, 1964.

[26] W.R. Leo. Thechniques for Nuclear and Particle Physics Experiments. Springer, 1994.

[27] http://www.eljentechnology.com/images/products/data_sheets/EJ-301_EJ-309.pdf.

[28] ET Enterprise Ltd. http://www.et-enterprises.com.

[29] Supplied by High Tech Sources Limited, Unit 6, Moorbrook, Southmead, Indus-trial Estate, Didcot, Oxfordshire, Uk OX11 7HP; http:///www.hightechsource.co.uk/.For details see http:///www.hightechsource.co.uk/Legacy/Resources/Americium-Berryliu.pdf.

25

[30] The Radichemical Centre Amersham , England, is today a subsidiary of GE Healthcare.

[31] The Lund Ion Beam Analysis Facility (LIBAF) is a university laboratory atLund University for materila analysis using protons, deuterium and heliumhttp://www.nuclear.lu.se/english/research/applied-nuclear-physics/libaf/.

[32] NE-213 is no longer produced. Eljen Technologies offers EJ-301http://www.eljentechnology.com/images/products/data_sheets/EJ-301_EJ-309.pdf.

[33] http://www.eljentechnology.com/index.php/products/pmma-a-uvt-material/102-light-guides .

[34] http:///www.eljentechnology.com/index.php/products/paints/87-ej-510.

[35] https://root.cern.ch.

[36] A. Jhingan et al. Nucl. Inst. and Meth. A, 585(3):165, 2008.

[37] A. Lavagno, G. Gervino, and C. Marino. Nucl. Inst. and Meth. A, 617(1–3):492, 2010.

[38] I.A. Pawełczak et al. Nucl. Inst. and Meth. A, 711(0):21, 2013.

26