CER
N-T
HES
IS-2
004-
082
09/1
2/20
04
THÈSE NO 3132 (2004)
ÉCOLE POLYTECHNIQUE FÉDÉRALE DE LAUSANNE
PRÉSENTÉE À LA FACULTÉ SCIENCE DE BASE
Institut de radiophysique appliquée
SECTION DE PHYSIQUE
POUR L'OBTENTION DU GRADE DE DOCTEUR ÈS SCIENCES
PAR
diplôme en physique, Université d'Athènes, Grèceet de nationalité hellénique
acceptée sur proposition du jury:
Dr J.-F. Valley, directeur de thèseDr A. Aroua, rapporteurProf. A. Bay, rapporteurDr M. Silari, rapporteur
Lausanne, EPFL2004
MEASUREMENTS AND MONTE CARLO CALCULATIONS WITHTHE EXTENDED-RANGE BONNER SPHERE SPECTROMETER
AT HIGH-ENERGY MIXED FIELDS
Evangelia DIMOVASILI
To Professor Pavlos D. Ioannou
i
CONTENTS CHAPTER 1- Introduction to Bonner Sphere Spectrometry ..... 1
1.1 Multi sphere spectrometry ......................................................................1
1.2 Synoptic historical review and developments of multi- sphere
spectrometry..........................................................................................3
1.3 The CERN Bonner Sphere Spectrometer...............................................7
1.4 Electronics............................................................................................13
1.5 Unfolding methods ...............................................................................13
1.5.1 Introduction ....................................................................................13
1.6 Unfolding codes....................................................................................16
1.6.1 MAXED ..........................................................................................16
1.6.2 GRAVEL ........................................................................................17
1.7 MONTE CARLO method ......................................................................18
1.7.1 Introduction ....................................................................................18
1.7.2 Major components of a Monte Carlo algorithm ..............................19
1.8 The FLUKA Monte Carlo code .............................................................20
CHAPTER 2- Calibration facilities .................................................... 22
2.1 Introduction...........................................................................................22
2.2 The PTB accelerator facility.................................................................23
2.2.1 Description .....................................................................................23
2.3 Calibration of the BSS with Quasi- Monoenergetic Neutron Fields
at UCL ..................................................................................................26
2.3.1 Introduction ...................................................................................26
2.3.2 The neutron facility and the experimental set- up .........................27
2.4 The CERN-EU Reference Field (CERF)...............................................29
2.4.1 Description of the facility ................................................................29
2.4.2 Beam monitoring at CERF .............................................................32
CHAPTER 3- Calibration at reference neutron fields ................ 33
3.1 Calibration of the BSS in monoenergetic neutron fields .......................33
ii
3.1.1 Introduction ....................................................................................33
3.1.2 Calibration at PTB- Materials and methods ..................................33
3.1.3 Data analysis ...............................................................................39
3.1.4 Results .........................................................................................43
3.1.5 Comparison of experimental to MC calculated responses .............44
3.2 Calibration at Quasi- Monoenergetic Neutron Fields...........................50
3.2.1 Materials and methods...................................................................50
3.2.2 Experiment and data analysis ........................................................54
3.2.3 Results and discussion ..................................................................56
3.3 Combined results of the 2001 and 2002 calibration measurements....66
CHAPTER 4- The response of the extended-range Bonner Sphere Spectrometer to charged hadrons.................................... 74
4.1 Introduction...........................................................................................74
4.2 BSS measurements at high-energy mixed fields..................................74
4.2.a Calculation of spectral fluences of hadrons (first MC study) ..........77
4.2.b Response functions of the BSS to charged hadrons
(second MC study)..................................................................................80
4.3 Comparison of the MC studies and estimation of correction factor ....83
4.4 Experimental test with 120 GeV/c hadrons at CERF...........................85
4.5 Conclusions.........................................................................................86
CHAPTER 5- Neutron spectral measurements with a Bonner sphere spectrometer............................................................................. 87
5.1 Introduction...........................................................................................87
5.2 Experiment ...........................................................................................88
5.3 Monte Carlo simulations and spectrum unfolding.................................92
5.4 Data analysis........................................................................................93
5.4.1 Beam normalization factor .............................................................93
5.4.2 BSS response to charged hadrons ................................................94
5.5 Results and discussion.........................................................................96
5. 6. Conclusions......................................................................................104
iii
CHAPTER 6 - Upgrade and development of the CERN-EU Reference Field (CERF) .....................................................................105 PART A: The beam monitoring at CERF .....................................................105
6.1 Monitoring system ..............................................................................105
6.1.1 Introduction ..................................................................................105
6.1.2 Measurements of beam profile with a Multi-Wire Proportional
Chamber ...............................................................................................107
6.2 Performance tests of the BIG PIC .....................................................108
6.2.1 Stability tests...............................................................................109
6.2.2 Linearity test ................................................................................111
6.2.3 Region of ion saturation ...............................................................112
6.2.4 Investigation of leakage current existence ...................................113
6.3 Inter-comparison of PIC and BIG PIC monitors at CERF ...................114
6.3.1 Recombination effects for the BIG PIC .......................................116
6.4 Tests of the Triggers 4, 5, 6 in the H6 beam line................................117
6.4.1 Efficiency measurements of Trigger4...........................................118
6.4.2 Measurements of Trigger5 and Trigger6......................................121
6.5 Conclusions........................................................................................123
PART B: Monte Carlo studies ......................................................................124 6.6 Introduction.........................................................................................124
6.7 Space studies for the CERF facility ....................................................127
6.8 Conclusions........................................................................................133
CHAPTER 7- Conclusions.................................................................134 APPENDICES.........................................................................................138
iv
RESUME
L’utilisation de la spectrométrie comme moyen de renseignement dans le cadre de la protection contre le rayonnement neutronique est devenue une activité de plus en plus importante durant ces dernières années. Le besoin d’avoir des données spectrales est apparu, car ni les instruments d’inspection des zones ni les dosimètres personnels ne donnent l’équivalent de dose de maniére correcte pour toutes les énergies de neutrons. Il est donc important de connaître le spectre des champs dans lesquels ces appareils sont utilisés. L’un des appareils les plus souvent utilisés dans la spectrométrie des neutrons et dans la dosimétrie est le spectromètre de sphère de Bonner (‘Bonner Sphere Spectrometer’ or BSS). La gamme du spectromètre utilisé au cours de ce travail consiste en 7 sphères et offre une réponse aux neutrons allant jusqu’à 2 GeV. Un détecteur à 3He est utilisé comme compteur thermal au centre de chaque sphère. Dans le contexte de cette thèse le ‘BSS’ a été calibré dans un champ monoénergétique de neutrons à des énergies basses et intermédiaires. Il a été également employé pour des mesures dans différents champs mixtes à haute énergie. Ces mesures ont amené aux calculs de champs de neutrons et de fluences spectrales pour des cibles non protégées. De telles données sont très utiles pour les calculs des blindages. De plus, la réponse des ‘BSS’ aux hadrons chargés a été améliorée. Parmi les champs dans lesquels le ‘BSS’ a été testé, le complexe CERF s’est révélé d’un intérêt particulier. Il est installé dans l’un des faisceaux secondaires issus du ‘SPS’ (Super Proton Synchroton) situé sur la partie françaize du CERN. La composition en particules et les fluences spectrales au-delà du blindage de CERF simulent celles rencontrées dans les champs cosmiques situés aux altitudes des vols commerciaux. Durant la dernière décennie les instituts de recherche travaillant dans le domaine de la recherche spatiale ont montré un intérêt grandissant pour l’installation CERF. Il y a donc un besoin de mettre en place une nouvelle zone d’exposition aux radiations reproduisant le champ de radiation situé au-delà de l’atmosphère ou à l’intérieur d’un vaisseau spatial. Les simulations ‘Monte-Carlo’ menées dans cet objectif concordent avec les résultats expérimentaux pris à l’intérieur de la station spatiale MIR. Ainsi, l’installation CERF pourrait s’avérer un outil très utile pour de futurs tests avec les ‘BSS’, en vue de simulation des champs de rayonnement cosmique.
v
ABSTRACT
The use of spectrometry to provide information for neutron radiation
protection has become an increasingly important activity over recent years. The need
for spectral data arises because neither area survey instruments nor personal
dosimeters give the correct dose equivalent results at all neutron energies. It is
important therefore to know the spectra of the fields in which these devices are used.
One of the systems most commonly employed in neutron spectrometry and
dosimetry is the Bonner Sphere Spectrometers (BSS). The extended- range BSS that
was used for this work, consists of 7 spheres with an overall response to neutrons up
to 2 GeV. A 3He detector is used as a thermal counter in the centre of each sphere. In
the context of this thesis the BSS was calibrated in monoenergetic neutron fields at
low and intermediate energies. It was also used for measurements in several high
energy mixed fields. These measurements have led to the calculation of neutron yields
and spectral fluences from unshielded targets. Such data are very useful for shielding
calculations and other radiation protection measurements. Furthermore, an upgrade of
the response of the BSS to charged hadrons was achieved.
Among the fields were the BSS has been tested, the CERF facility is of
particular interest. It is installed in one of the secondary beam lines from the Super
Proton Synchrotron (SPS) in the French site of CERN. The particle composition and
spectral fluences outside the shielding of CERF simulate those in the cosmic radiation
field at commercial flight altitudes. In the last decade interest has arisen in CERF
from research institutions working in the space programme. There is therefore a need
for setting-up a new exposure area reproducing the radiation environment found
outside the atmosphere or inside a space vessel. The results from Monte Carlo studies
that have been carried out for this purpose show a good agreement with experimental
data taken inside the MIR space station. Therefore, CERF may prove to be a very
useful tool for further tests with the BSS system, towards its upgrade for
measurements in cosmic ray fields.
vi
INTRODUCTION
The aim of this doctoral thesis is to upgrade and develop the Extended- Range
Bonner Sphere Spectrometer (BSS) in high-energy mixed radiation fields. The correct
use of the BSS must be certified by calibration measurements. Three calibration
campaigns have taken place between 2001 and 2003, in order to validate the Monte
Carlo calculated response matrix of the extended BSS. The overall response of the
spectrometer has been extended to charged hadrons and a full response matrix has
been calculated. It has also been possible to calibrate the BSS in a mixed charged
hadron beam at the CERN-EU Reference Field facility.
In Chapter 1 the reader can find an overview of the history of the multi-sphere
spectrometry and a description of the instruments and methods that were used for the
purposes of this work. Chapter 2 describes the calibration facilities that have been
used in the framework of this dissertation. Calibration campaigns have been
performed at PTB (Braunschweig, Germany) and at UCL (Louvain- La Neuve,
Belgium) at monoenergetic neutrons with energies from 0.565 MeV up to 60 MeV.
The aim of the calibration was the verification of the theoretical response functions
calculated with the FLUKA Monte Carlo code and the estimation of the calibration
factor of the specific 3He detector. The data analysis and the results of the calibration
are presented in Chapter 3.
A BSS can be used to measure neutron spectra both outside accelerator
shielding and from unshielded targets. Measurements that were performed in the
NA-57/ALICE experimental area at CERN have provided evidence that the BSS
under certain conditions can show a significant response to charged hadrons. A
complete response matrix of the BSS to charged pions and protons was calculated
with FLUKA. An experimental verification was carried out with a 120 GeV/c hadron
beam at the CERF facility at CERN. The analysis and the results of these studies are
given in detail in Chapter 4.
The aim of the experiments at the NA-57/ALICE area was the calculation
of the neutron yields and spectral fluences from unshielded, semi-thick targets, in
order to provide source term data for neutron production from high-energy hadrons.
vii
This information is necessary for shielding calculations and other radiation protection
purposes. The work is analytically presented in Chapter 5.
The calibration of the BSS at the charged hadron beam of CERF provoked the
need for the further development and improvement of this facility. The studies
performed are presented in Chapter 6 that consists of two parts. In the first part, the
extensive performance tests of the CERF back-up monitor are described. This is an
ionization chamber of similar design to the reference CERF beam monitor that was
submitted to a number of tests in the SC/RP calibration laboratory. It was then tested
in real experimental conditions at the CERF facility. The second part of Chapter 6
discusses the FLUKA Monte Carlo studies that have been performed for the upgrade
of CERF. The new challenging objective was to design a different shielding
configuration inside the CERF cave that could be of interest for measurements in the
framework of the space programme but also for the future use of the BSS at high
energy fields such as cosmic ray fields.
In Chapter 7 the conclusions of the overall work are summarised and an
outlook on further development of the Extended –Range BSS and of the CERF
facility is proposed. Appendices include part of the calculations that have been done
in the context of this thesis.
1
CHAPTER 1
Introduction to Bonner Sphere Spectrometry
1.1 Multi sphere spectrometry
The term ‘radiation spectrometry’ can be used to describe the measurement of
the intensity of a radiation field with respect to energy, wavelength, momentum, mass,
angle of incidence or any other related quantity. The distribution of the intensity with
one of these parameters is commonly referred to as the spectrum. Neutron energy
spectra are frequently measured by indirect methods because of the experimental
limitations in detecting fast neutrons. The need for spectrometry stems mainly from
the fact that both area survey instruments and personal neutron dosimeters have a dose
equivalent* response which is a function of energy. Spectra thus need to be known in
order to determine precisely the dose equivalent values in fields where individuals are
exposed to neutrons, e.g. in workplaces in the nuclear industry, around accelerators, or
at aircraft flight altitudes. Spectra also need to be determined to characterize
calibration fields.
A multi-sphere spectrometer uses a thermal neutron detector at the centre of
moderating spheres of different diameters, usually made of polyethylene. A schematic
drawing of such a sphere is shown in Fig. 1.1. The fast neutrons are slowed down in
the moderator and reach the detector thermalized, while the thermal neutrons initially
present in the field are mostly captured in the moderator. Therefore, when the
moderating sphere's diameter increases, the maximum sensitivity of the system moves
to higher energies. The multi-sphere spectrometer shows several characteristics useful
for radiation protection measurements. Amongst its main advantages one can include
its functional simplicity, its wide energy domain (from thermal energies to several
MeV), the high neutron sensitivity allowing the measurement of low dose equivalent
rates encountered in radiation protection, the good discrimination of electronic noise
* See Appendix A
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
2
and photon counting by a suitable choice of the counter and adjustment of the
associated electronics. The electronics modules needed to operate the detector are
relatively simple, adding to the benefits of using this spectrometer.
Fig. 1.1. Schematic representation of a polyethylene Bonner Sphere.
One drawback of this method is the low energy resolution. This is partially
due to the fact that the statistical fluctuations in the number of collisions in the
neutron slowing down processes are large, and the capture reactions are completely
indistinguishable from one another. This results in loss of information about the
primary neutron energy and, consequently, low resolution. However, the energy
resolution of the system being low can be judged as satisfactory for the evaluation of
the dosimetric quantities used in radiation protection.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
3
1.2 Synoptic historical review and developments of multi- sphere spectrometry
Spectrometers for measuring photon and neutron energy spectra have been
used for almost as long as there has been an awareness of the existence of these
radiations. The devices have however improved dramatically over the years. In
parallel, the original detection mechanisms have been augmented by more modern
techniques. During this period of research and advancement of technology,
spectrometry measurements have played a vital role in understanding the nature and
origins of radiation. They have subsequently been an invaluable source of information
in atomic and nuclear physics research.
Bramblett, Ewing and Bonner, working at Rice University, developed and
tested the first multi-sphere detector in 1960, known widely since then as the Bonner
Sphere Spectrometer (BSS) [1]. The first BSS consisted of a small cylindrical (4 mm
high by 4 mm diameter) 6LiI(Eu) scintillator optically coupled to a Photo-Multiplier
placed at the centre of a series of polyethylene neutron moderating spheres (Fig. 1.2).
Fig. 1.2. Schematic drawing of the Bonner [1] experimental set –up.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
4
Count rates with the LiI detector placed inside 5.08 cm, 7.62 cm, 12.7 cm,
20.32 cm and 30.48 cm diameter spheres were used to obtain information about
neutron spectra. The size of the crystal was chosen to be small so as to allow good
γ-ray discrimination. The responses of the first BSS were measured for incident
neutrons in the energy range 0.06-15 MeV and are shown in Fig. 1.3.
Fig. 1.3. The energy dependence of the relative detection efficiencies of Bonner sphere neutron detectors of various diameters up to 30.48 cm [1].
Since the first use of the LiI crystal in moderating spheres many other researchers
have shown preference in this type of detector [2-4]. In cases where higher sensitivity
is required larger scintillation crystals can be used. The main disadvantage of the use
of LiI crystals is the subtraction of γ events which requires an analysis of the light-
output spectrum and it is generally a difficult procedure. 10BF3 proportional counters have been used as an alternative to LiI crystals.
BSS sets based on such counters have not been very common, however systems using
small diameter cylindrical counters have been built and used extensively [5- 8].
During the 1970s and early 1980s many papers were published describing the
use of 9 mm diameter cylindrical 3He proportional counters (type 0.5NH10) as
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
5
thermal sensors (see for example [9, 10]). The use of this type of counter which
contained about 8 atm of 3He provided a system with a much higher sensitivity.
Although it is less sensitive to γ- rays than a LiI one, the low gas amplification may
restrict the discrimination of all neutron induced events, while excluding noise.
Nevertheless, this counter has been used by many groups [11-14] and it provides a
useful lower-response alternative to the more sensitive type SP9 3He counter, for use
in intense neutron fields.
In the 1980s and early 1990s, several research teams [15-19] investigated the
use of the SP9 spherical 3He proportional counter, produced and commercially
available by Centronic Ltd., UK. This counter has a lower gas pressure of about 2 atm
but a bigger diameter, resulting in a larger geometrical cross-section. As a result, the
overall fluence responses of Bonner spheres incorporating this counter are of the order
of a factor of 10 higher than for the 4 mm by 4 mm LiI system, depending on sphere
size and neutron energy. Additionally, the discrimination with respect to γ- rays and
noise is excellent, except in the highest-intensity γ- ray fields where pile-up becomes
a problem. The 3He counters are fairly insensitive to radiations other than neutrons
and their efficiency proved to be stable with time. The characteristics of BSS using
this type of counter are nowadays very well established [13, 20-26].
Although the SP9 counter provides reasonably high sensitivity, there are
situations where a higher efficiency is desirable, e.g. for the investigation of cosmic-
ray-induced neutrons at ground and flight altitudes with very low fluence rates. For
such purposes, BSS systems based on large 3He counters have been developed
[27-28].
On the other hand, there are situations where low efficiency is preferred. BSS
systems which use activation detectors as the thermal sensor have been built [27, 29-
31]. The activation material is usually gold or indium. These systems do not present
problems, such as dead-time losses, that can be of great concern for active devices in
intense fields, particularly in pulsed ones. Activation foils have proved to be more
efficient in other environments, because of their very low sensitivity to γ-rays.
However, γ-ray induced neutron production in the material of the sphere needs to be
considered in intense γ-ray fields, for example in the photon beam from a medical
electron accelerator [29].
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
6
The development of a lower-efficiency system based on the SP9 counter has
been of great interest. There have been few attempts made towards this direction [24],
by either covering this counter with a close-fitting thin cadmium shell, or by reducing
the gas pressure.
It is commonly known that the resolution of a conventional BS at low and
intermediate energies is particularly poor. For radiation protection the resolution
below about 10 keV, where dose equivalent conversion coefficients are roughly
constant, is not very important. However, there are applications, e.g. the
characterization of epithermal fields for boron neutron capture therapy (BNCT),
where better resolution is needed. Including an outer shell of a thermal neutron
absorber, such as boron or lithium, around the smaller spheres, enhances the response
function up in energy with increasing absorber thickness. The disadvantage of the
method is that the overall sensitivity is reduced. Nevertheless, by using absorber
layers of different thicknesses, spheres with response functions covering the low- and
intermediate-energy range can be produced. Several groups have investigated this
approach [32-34].
BSS have recently started being adapted to measure spectra up to neutron
energies in the GeV range [26, 28, 35, 36]. An interesting approach among new
designs that have evolved from the BSS is to use a single block of moderator
containing either several extended, position-sensitive thermal neutron detectors [37]
or a number of small thermal neutron detectors [38] mounted at different positions.
Passive detectors have also been used in Bonner spheres, in order to measure
very intense pulsed neutron fields such as those encountered around particle
accelerators [39-41], or in the case where a low intensity neutron field requires a very
long integration time such as in some environmental measurements. The types of
passive detectors employed include activation detectors sensitive to thermal neutrons,
pairs of 6Li and 7Li fluoride thermo-luminescent detectors, and track detectors with
radiators made of 10B, 6Li or 235U.
Interest in measurements around high-energy accelerators, and at high
altitudes in the atmosphere, have stimulated the development of spheres with shells of
lead, iron or copper within the polyethylene [26, 28, 42-44]. Neutron multiplication,
which for high-energy neutrons occurs within the metal shell, increases the high-
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
7
energy response. Variations in the response function shapes can be obtained by
varying the configuration. The alternative of obtaining data at higher energies is to
include in the BS analysis data from ‘threshold’ reactions like neutron induced fission
reactions, e.g. 232Th(n,f) and 209Bi(n,f) (where ‘f’ stands for ‘fission’) [41].
In summary, the Bonner Sphere spectroscopy, one of the most widely
employed methods, is basically the same today as it was when initially developed.
Different detector types are nowadays used and computers have been improved to
such a point that they can perform complex unfolding procedures very quickly.
Moreover, the new systems employ portable personal computers that perform data
analysis and display of results quickly after measurement. New developments in
technology are expected to decrease the dimensions of data analysis equipment but it
is not likely that the large moderating spheres can be reduced in size.
Bonner Sphere spectroscopy is a very useful tool to provide information that
can reduce uncertainties in dosimetric measurements. It has been the system that
researchers have shown preference at, in order to provide solutions to measurement
problems. Despite innovations, energy resolution is always going to be poor. It should
however be possible to combine BSS data with high resolution measurements to
produce spectra with good resolution in the energy region where it is important.
1.3 The CERN Bonner Sphere Spectrometer
The Radiation Protection (RP) group at CERN is in possession of an Extended
range BSS (Fig.1.4), a recently built descendent of a conventional BSS. The
conventional BSS consists of a set of five moderating spheres made of polyethylene,
having outer diameters of 81 mm, 108 mm, 133 mm, 178 mm and 233 mm. In order
to better absorb the thermal neutron component, the sphere with the smallest diameter
can be surrounded by a cadmium (Cd) shell of thickness 1 mm, therefore this
configuration is called 81cd. The active part of the spectrometer is a spherical 3He
proportional counter with a diameter of 32 mm located in the centre of each sphere.
Two 3He proportional counters are used by the RP group, with gas pressures of 2 atm
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
8
and 4 atm. They both use a gas mixture of He- Kr inside their effective volume. The
Kr is used in order to reduce the range of the reaction products or in other words to
increase the stopping power. A more detailed description of the BSS and of the
specific 3He counters is given in [45]. For the purposes of the present study only the
2- atm 3He detector has been used.
In order to be able to study the neutron spectra around hadron accelerators,
where the high-energy neutron component is important, it was desirable to extend the
response of the BSS to substantially higher energies. To extend the range of the
conventional Bonner Sphere Spectrometer by means of two new spheres, nineteen
configurations of different sizes and materials were thoroughly investigated by means
of Monte Carlo simulations. The reader can find the details about the design studies of
the spheres or about other influencing factors (material, density, geometry, etc.)
elsewhere [45]. The convention of labeling each sphere by its diameter in centimeters
has been adopted in the present study. The two new spheres were named Ollio (for the
255 mm diameter sphere) and Stanlio (for the 119.5 mm diameter sphere).
Fig. 1.4. The extended range Bonner Sphere Spectrometer of the RP group at CERN.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
9
Ollio consists of moderator shells of 3 cm polyethylene, 1 mm cadmium, 1 cm
lead and 7 cm polyethylene (from the central 3He proportional counter outwards). The
characteristic of this configuration is that it suppresses the response to incident
neutrons with energies lower than 100 keV and increases it for energies above
10 MeV and up to 1 GeV, as compared to the 233 mm sphere of the conventional
BSS. The response function shows the peak at about 10 MeV as it is typical for all
large detectors of a BSS. The second newly built sphere, Stanlio, consists of
moderator shells of 2 cm polyethylene, 1 mm cadmium and 2 cm lead (Fig.1.5). Its
response function does not show the peak at 10 MeV. This feature makes Stanlio a
useful complement to the other detectors. At low energies it behaves like a small
Bonner sphere, but at high energies the response is increased compared to the 233 mm
sphere. The response functions of the various detectors are shown in Figs.1.6- 1.7.
The nominal response matrix can be found in Appendix B.
Fig. 1.5. Cut through the moderator of the Bonner Sphere Stanlio.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
10
Fig.1.6. The calculated absolute neutron fluence response functions of the CERN
BSS at full energy scale. For clarity, the statistical errors have been omitted in the
lower graph.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
11
Fig. 1.7. The calculated absolute neutron fluence response functions of the CERN
BSS at low energies (upper plot) and at high energies (lower plot).
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
12
The accurate evaluation of the response matrix is essential for a correct use of
the spectrometric system. The response functions of the BSS were calculated by
Monte Carlo simulations performed with FLUKA98 [46]. The neutron response of
each detector was calculated for 78 incident neutron energies. For the neutron
energies of 0.05, 0.1, 0.25, 0.5, 1 and 2 GeV, a broad parallel beam having a slightly
larger extension than the sphere was assumed. In most experimental conditions low
energy neutrons are generated by down-scattering and they are undirectional.
Therefore, an isotropic distribution of the incident neutrons was chosen for the
72 low-energy groups, between E=19.6 x 106 eV and 1x 10-5 eV [45].
As Fig.1.6 shows, for each detector the response peaks at a given energy, depending
on the moderator’s size. The response of the various polyethylene spheres decreases
rapidly for energies above a few MeV. This is due to decreasing (n, p) cross section
with increasing neutron energy. For a small sphere the degree of moderation is small,
as is the capture of thermal neutrons in the moderator. Low-energy neutrons thus have
a reasonable probability of arriving at the thermal sensor and being detected, whereas
fast neutrons tend to escape. For larger spheres there is considerably more
moderation. There is also more capture which means low-energy neutrons tend to get
absorbed in the polyethylene. It is the high energy neutrons which thus have the
greatest probability of being detected in the sensor, and the response function peaks in
the high-energy region. In order to reproduce correctly the measured responses by
calculation and supplement them in energy ranges not measured, detailed information
concerning the geometry of the moderator and the 3He counter is required.
The difference in the shapes and position of the maxima in these response
curves serves as the basis for using the set of spheres as a simple neutron
spectrometer. By measuring the count rate with each sphere individually, an unfolding
process can provide some information about the energy distribution of the incident
neutrons.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
13
1.4 Electronics
The electronics used with the CERN BSS system consist of a preamplifier, a
Serial Micro Channel SMC 2100 box (a module housing an amplifier, high voltage
and multi-channel analyser) produced by MAB [47] and called the mab box, a 3H
proportional counter and power supply. The Serial Micro Channel SMC 2100
modular concept design is capable of being connected to a serial interface of a pc. The
SMC 2100 is supported by a range of modules that are necessary in certain
applications for the measurement of radiation pulses. The multi-channel analyser
operates with 2048 channels. Minimum is channel zero and maximum is channel
2047. All pulses arriving from the spectroscopy amplifier are passing the window of
an existing single-channel analyser. The analog pulses of the amplifier are digitized
by a fast 12-bit analog-to-digital converter (ADC). The recognition of the peak
maximum is made in the MCA. The ADC values are stored in the MCA and are
normalized upon 2048 channels.
A program package (AM-SMCA01) is used with the electronics for the data
acquisition and processing. The program is available with different options depending
on the purpose of application. For the purposes of this dissertation the option ‘pulse
height analysis’ has always been used. The software is compatible with Windows
95/98 and more recently it has been upgraded to Windows 2000 and NT. More details
on the software can be found in [48].
1.5 Unfolding methods
1.5.1 Introduction
One of the drawbacks of the multi-sphere spectrometer is the mathematical
problem of unfolding the neutron spectrum. For the CERN BSS applications, eight
detector/moderator configurations are used, resulting in eight values of count rate
recorded from the detector. These eight data points are then used in the unfolding
code with the response function matrix to determine the neutron flux in the 78
energy bins covering the entire range of neutron energies from thermal up to
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
14
2 GeV. The derivation of the neutron spectrum from the experimental data is done
as follows. If sphere i has response function Ri(E) and is exposed in a neutron field
with spectral fluence Φ(E) then the sphere reading Mi is obtained mathematically
by folding Ri(E) with Φ(E), i.e.:
dEEERii )()( Φ=Μ ∫ (1.1)
Equation (1.1), formally known as a Fredholm integral of the first kind,
extends over the range of neutron energies present in the field. This equation does not
provide a unique solution because the continuous function Φ(E) cannot be defined by
a set of n discrete measurements. It can however be approximated by equation (1.2)
where Φj is the fluence in group j extending from energy Ej to Ej+1 and Rij represents
Ri(E) averaged over group j.
j
n
jiji R Φ=Μ ∑
=1 i=1, …, m (1.2)
The degree of approximation decreases as the number of groups n increases. If
there are m spheres, equation (1.2) represents a set of m linear equations. In the case
that m ≥n, equation (1.2) can be solved by using special mathematical methods.
However, because m is usually small, of the order of 10, the solution may provide a
poor representation of the spectrum. Spectra are therefore usually represented by an
array with n> m, which means that equation (1.2) can only be solved for Φj by using
smoothing functions or so-called additional a priori information. A clear distinction
has to be made between input guess spectrum and a priori information. Knowledge of
the spectral fluence distribution of the investigated neutron field, derived either from
calculations or previous measurements, is usually called specific a priori information.
In other words, adding of specific a priori information during unfolding is equivalent
to increasing the number of measurements. This means that specific a priori
information controls the results.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
15
Even in the case that specific a priori information is available, it is reasonable
to require for the output spectrum a series of criteria based on the general physical
representation of the investigated neutron field. As an example, such criteria can be:
• No negative values for the differential fluence can be accepted
• The solution spectrum should be smooth. Even if structures are present in
the real spectrum, reliable information on these structures can only come from
specific a priori information. A Bonner Sphere system provides a few channel
unfolding and it is not capable of reproducing finer structures, although it can
indicate broad peaked fluence distributions. In a case that fine structures or
oscillations that are not included in the guess spectrum appear in the solution
spectrum, one should make sure that the unfolding code has not been
improperly used.
A solution can be called exact, approximate, or appropriate [49]. Exact
solutions may have zero errors, and might look reasonable. However, they may have
unphysical characteristics, such as oscillations. Usually, the unfolded data should not
be expected to have too good a fit, at least not better than the error of the input data. In
general, the trial vectors should contain the features of the neutron spectra one can
expect from the physics of the problem. Similarly, the smoothing functions, if any,
have to be properly chosen. Appropriate solutions can be obtained from good
measurements, and considerable experience is needed to judge just when a reasonable
spectral solution is reached.
Good approximations to Ri(Ej) can be obtained from simulation calculations
supported by measurements with well characterized monoenergetic and radionuclide
source neutrons. The increased computing power available nowadays makes the
Monte Carlo method the most appropriate approach to calculating response functions.
Monte Carlo codes have the advantage of allowing the sphere plus detector to be
geometrically modeled in detail, including non-spherically symmetric features.
Knowledge of the energy-dependent response, with the sphere diameter as a
parameter, is of fundamental importance for the spectrometry using unfolding
procedures. Using these data, measurements with a BSS set in an unknown field will
allow information on Φ(E) to be extracted. Many unfolding codes employing
different mathematical techniques have been used to perform spectrum unfolding for
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
16
multi sphere systems. Due to the non-uniqueness of the deconvolution, there are many
different methods that are based on different mathematical principles. Some of the
mathematical methods used are: maximum entropy, least-squares iteration etc. The
unfolding codes employed for the purposes of the present work are briefly described
below.
1.6 Unfolding codes
1.6.1 MAXED
The FORTRAN code MAXED (MAXimum Entropy Deconvolution) was
developed at EML (Environmental Measurements Laboratory, DOE, USA) [50]
specifically for the deconvolution (unfolding) of multi-sphere neutron spectrometer.
The maximum entropy deconvolution algorithm used in MAXED is a modification of
the one in Wilczek and Drapatz [51]. In data analysis, the maximum entropy principle
is widely used as a general and powerful technique for reconstructing positive
distributions in situations where only incomplete information is available. It requires
the maximization of entropy S given by equation (1.3)
])ln([ iDEF
iDEFi
ii fff
ffS −+−= ∑ (1.3)
where fi is the determined fluence in group i and fiDEF is the (discretized) default
spectrum.
For this purpose the MAXED code implements a special algorithm (annealing
global optimization algorithm) described in [52]. In the case of the deconvolution of
neutron spectra, the distribution to be determined is the neutron energy spectrum f(E)
also called solution spectrum and the constraints are the measurements and the
experimental errors associated with them.
The default spectrum contains all a priori information, based on which the
code is using the maximum entropy method to derive a new probability distribution
that takes into account the new information provided by the measurements. A formal
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
17
argument due to Shore and Johnson [53] shows that the maximum entropy method is
the only general method of choosing the new distribution that does not lead to
inconsistencies.
The approach followed in MAXED has several features that make it attractive:
it permits inclusion of a priori information in a well-defined and mathematically
consistent way, the algorithm used to derive the solution spectrum is not ad hoc (it
can be justified on the basis of arguments that originate in information theory), and
the solution spectrum is a non-negative function that can be written in closed form.
For more information on MAXED, the reader should refer to [50].
1.6.2 GRAVEL
The GRAVEL code is based on the formalism described by M. Matzke
[54-56]. It is adapted to utilize the capabilities of the MATLAB [57] language for the
I/O of data and for the graphical representation of other relevant quantities (e.g. χ2,
total flux, etc.) during the iterations, thus allowing a better understanding of the
evolution of the procedure. This code performs a nonlinear least-square adjustment.
The constraint of non-negative particle fluences is essential for the operation of the
program. To take into account the condition of non-negative fluence, the logarithms
of fluence Φ are used instead of the fluence Φ. They are determined by a special
gradient method [58] which minimizes the χ2 value obtained by the comparison
between experimental and reconstructed data.
An initial input spectrum (guess spectrum) is needed. A solution always exists,
but the solution spectrum depends on this input spectrum. The initial spectrum has
usually the shape of 1/E. In case there exists a more precise initial spectrum provided
for example by Monte Carlo calculations, the unfolding solution is a higher resolution
spectrum. These iterations are done in a way that is not quite transparent, so that a
uncertainty propagation cannot be easily performed. For this reason, a lot of
experience is needed for a successful unfolding of neutron spectra employing the
GRAVEL code. The unfolding in the context of the present work has been performed
with GRAVEL by experts in the Polytechnic of Milan [59].
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
18
1.7 MONTE CARLO method
1.7.1 Introduction
Numerical methods that are known as Monte Carlo methods can be described as
statistical simulation methods, where statistical simulation is defined in quite general
terms to be any method that utilizes sequences of random numbers. Credit for
inventing the Monte Carlo method often goes to Stanislaw Ulam, a Polish born
mathematician who worked for John von Neumann on the United States’ Manhattan
Project during World War II. Ulam is primarily known for designing the hydrogen
bomb with Edward Teller in 1951. He invented the Monte Carlo method in 1946
while pondering the probabilities of winning a card game of poker.
Ulam did not invent statistical sampling. This had been employed to solve
quantitative problems before, with physical processes such as dice tosses or card
draws being used to generate samples. Ulam’s contribution was to recognize the
potential for the newly invented electronic computer to automate such sampling.
Working with John von Neuman and Nicholas Metropolis, he developed algorithms
for computer implementations, as well as exploring means of transforming non-
random problems into random forms that would facilitate their solution via statistical
sampling. This work transformed statistical sampling from a mathematical curiosity to
a formal methodology applicable to a wide variety of problems. It was Metropolis
who named the new methodology after the casinos of Monte Carlo.
Ulam and Metropolis published the first paper on the Monte Carlo method in
1949 [60]. Monte Carlo is now used routinely in many diverse fields, from the
simulation of complex physical phenomena such as radiation transport in the earth's
atmosphere to the simulation of the esoteric sub-nuclear processes in high-energy
physics experiments. Its name does not mean to imply that the method is either a
‘gamble’ or ‘risky’. It simply refers to the manner in which individual numbers are
selected from valid ‘representative collections of input data’ so they can be used in an
iterative calculation process. These representative collections of data are some sort of
a Frequency Distribution that is converted to a Probability Distribution.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
19
Monte Carlo Simulation methods are primarily used in situations where:
• The system being studied can be mathematically described by a metric,
which can be either parametric or analytic.
• The Input Data can be written as some sort of a frequency distribution.
• The calculated distribution histogram of the ‘answer’, or Output, must
accurately reflect the Input data.
• The calculated uncertainty in the ‘answer’, or Output, must be an accurate
measure of the validity of the model.
Monte Carlo Simulations could be regarded as ‘True Stochastic Simulations’ in
that they describe the final state of a model by just knowing the frequency
distributions of the parameters describing the beginning state and the appropriate
metric that maps or transforms the beginning state to the final state.
1.7.2 Major components of a Monte Carlo algorithm
The primary components of a Monte Carlo simulation method include the following:
• Probability distribution functions (pdf's)- the physical (or mathematical)
system must be described by a set of pdf's.
• Random number generator- a source of random numbers uniformly distributed
on the unit interval must be available.
• Sampling rule - a prescription for sampling from the specified pdf's, assuming
the availability of random numbers on the unit interval, must be given.
• Scoring (or tallying) - the outcomes must be accumulated into overall tallies or
scores for the quantities of interest.
• Error estimation - an estimate of the statistical error (variance) as a function of
the number of trials and other quantities must be determined.
• Variance reduction techniques - methods for reducing the variance in the
estimated solution to reduce the computational time.
• Parallelization and vectorization - algorithms to allow Monte Carlo methods
to be implemented efficiently on advanced computer architectures.
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
20
1.8 The FLUKA Monte Carlo code
FLUKA is a FORTRAN simulation code generally used to calculate particle
transport and interactions with matter. The development of FLUKA since its first
appearance in the sixties has seen many advances that cover an extended range of
applications. These applications include proton and electron accelerator shielding,
calorimetry, dosimetry, detector design, cosmic rays, neutrino physics, radiotherapy,
etc. A more thorough and detailed historic review of FLUKA can be found in [46].
This simulation code can perform high accuracy simulations for the interaction
and propagation in matter of about 60 different particles: photons and electrons with
energy from 1 keV to TeV, muons and hadrons with energy up to 20 TeV (to be
extended soon to 10 PeV), neutrinos, and all the corresponding antiparticles. In
particular neutrons are simulated down to thermal energies as well as heavy ions.
FLUKA can employ very large-scale and complex geometries. For this
purpose it is using an updated version of the well-known Combinatorial Geometry.
This improved version of the Combinatorial Geometry is fast, flexible and user-
friendly at a much higher degree than the older versions, allowing the users to track
correctly charged particles even in the presence of magnetic or electric fields. New
bodies have been introduced, resulting in increased rounding accuracy, speed and
even easier input preparation. Various visualization and debugging tools have also
been integrated into FLUKA thus making it a very powerful tool with many different
capabilities according to the user’s needs.
Several models are employed in FLUKA for the transport and propagation of
the different groups of particles in different energy ranges. The hadron-nucleon
interaction models are based on resonance production and decay below a few GeV,
and on the Dual Parton model at higher energies. Two models are also used in hadron-
nucleus interactions. At momenta below 3 GeV/c the PEANUT package includes a
very detailed Generalized Intra-Nuclear Cascade (GINC) model. At high energies the
Gribov-Glauber multiple collision mechanism is included in a less refined GINC. One
can also simulate photonuclear interactions with FLUKA. These are described by
Vector Meson Dominance and Delta, Quasi-Deuteron and Giant Dipole Resonance.
Multiple Coulomb scattering and ionization fluctuation are also integrated in the code,
CHAPTER 1. Introduction to Bonner Sphere Spectrometry
21
allowing it to handle problems such as electron backscattering even in the few keV
energy range. FLUKA can also simulate synchrotron radiation and optical photons.
The FLUKA physical models are described in detail in several journal and conference
papers [61-63].
A very interesting and important feature of FLUKA, probably unique in
comparison with other Monte Carlo codes, is its capability to be used both in a biased
mode and in a fully analog mode. This means that on the one hand it can be used to
predict fluctuations, signal coincidences and other correlated events and on the other
hand it can be used to investigate other rare events because of its wide choice of
available statistical techniques.
22
CHAPTER 2
Calibration facilities
2.1 Introduction
Calibration is a set of operations that establish, under specific conditions, the
relationship between values indicated by a detector (dosemeter), and the
corresponding known (i.e. conventionally true [64]) values of the quantity to be
measured [65,66]. This relationship can be established by determining the response of
a device for the full range of radiation energies and angles of incidence for its
intended use. Laboratory calibrations determine the calibration factor for each
individual instrument under standard test conditions [66,67]. The calibration factor f
is the factor by which the reading of the device is multiplied to obtain the value of the
quantity to be measured.
The calibration procedure is linked to the quality assurance of an instrument.
Assuming that the available instrument is the suitable one for the specific need, its
performance depends strongly on the quality of the response matrix as well as on
other parameters (unfolding technique, accuracy of data, etc.). Therefore, the
calibration of the instrument for the verification of the calculated (usually by Monte
Carlo) response matrix is essential.
In order to calibrate the Bonner sphere detectors and to determine their
response functions, measurements with monoenergetic neutrons are required. Typical
energies used for this purpose are those recommended by the International
Organization of Standardization (ISO) [68]. An extensive calibration campaign has
been carried out to evaluate the response of the BSS to monoenergetic neutrons.
Calibrations with monoenergetic neutron beams were carried out at the Physikalisch-
Technische Bundesanstalt (PTB) in Braunschweig (Germany) in two consecutive
years. In 2001 the BSS was calibrated at energies 0.144 MeV, 1.2 MeV, 5 MeV and
14.8 MeV [45] and in 2002 the calibration was done at energies 0.565 MeV, 2.5 MeV,
CHAPTER 2. Calibration facilities
23
8 MeV and 19 MeV. For the full characterization of the BSS another calibration
campaign was performed at UCL (Louvain–la Neuve, Belgium) in 2003.
The calibration of the BSS in the hadron beam of the CERN-EU Reference
Field (CERF) was later performed in the context of the upgrade of its response to
charged hadrons (see chapter 4). A short description of all irradiation facilities that
have been used for calibration and tests of the BSS system, is given below.
2.2 The PTB accelerator facility
2.2.1 Description
The Physikalisch- Techmsche Bundesanstalt (PTB) accelerator facility [69-71]
for fast neutron research, was constructed at Braunschweig between 1971 and 1974. A
schematic drawing of the facility is shown in Fig. 2.1.
Fig. 2.1. Neutron research facility at PTB. (1) the cyclotron, (2) the quadrupole magnet, (3) the neutron producing target.
CHAPTER 2. Calibration facilities
24
The cyclotron is located in the basement of the building. It is mounted on a
swivel arm, about 5 m in radius. The external beam can thus be moved in a horizontal
plane such that it is always directed to the vertical axis of rotation. The scattering
probe S (Fig 2.2) is located on the axis of rotation. The neutron producing target T, set
up at a distance of 15-25 cm from the scattering probe, can be turned around the probe
together with the whole cyclotron, thus varying the scattering angle.
For experiments involving neutron sources in the low backscatter experimental
hall, the external cyclotron beam can be transported via two 90 ° deflecting magnets
from the cyclotron room to one of the entrance ports of the switching magnet on the
ground floor. The beam then can be directed into one of the four beam lines within the
experimental hall. These beam lines are terminated by the targets, i.e. the
monoenergetic neutron sources.
Fig. 2.2. Layout of the neutron scattering experiment. C: cyclotron, Q: quadrupole magnet, T: neutron producing target, S: scattering probe (= pivot of cyclotron movement), P: polyethylene shields, W: water tank, D: neutron detector, B: concrete shield.
CHAPTER 2. Calibration facilities
25
The low backscatter experimental hall has an area of 24 m x 30 m and a height
of 14 m. In order to protect the environment against radiation, the walls of the
experimental hall are 1.2 m thick, the ceiling is 0.4 m thick concrete and an additional
0.3 m water layer on the roof may be provided in case additional shielding is needed.
The four beam lines are installed on a plane at 6.25 m above ground. A low scattering
grid floor of aluminum 4.5 m above ground, allows access to the targets (Fig 2.3).
Fig. 2.3. Scheme of the experimental set-up at PTB (not to scale).
CHAPTER 2. Calibration facilities
26
The reference monitors used to measure the neutron fluence (Fig. 2.3) are the
following :
• The New monitor (NM): it consists of a long cylindrical 3He proportional
counter surrounded by a cylindrical polyethylene layer 15 cm thick. It was
mounted at a distance of 5.5 m from the target at an angle of 19 degrees.
• The 3He proportional counter: A 3He proportional counter embedded inside a
polyethylene moderator (50 mm in diameter and 233 mm long) was placed on
the right hand side to the beam line, just beside it and upstream of the target.
• The Geiger-Mϋller counter (GM): In order to measure the photon contribution
a Geiger-Mϋller counter was mounted on top of the production target.
• The charge monitor: it is used for monitoring the charge and thus the intensity
of the beam impinging on the target. It was placed in the beam line but it is not
shown in Fig. 2.3.
• Precision long counter (PLC): it consists of a long BF3 proportional counter
placed inside a large, specially shaped, moderating cylinder made of
polyethylene, boron loaded polyethylene, aluminum and cadmium. It was
placed at a distance of about 5.5 m from the target, at an angle of 100 degrees.
2.3 Calibration of the BSS with Quasi- Monoenergetic Neutron Fields at UCL
2.3.1 Introduction
For the full characterization of the BSS the calibration at higher energies than
those used in PTB was considered as essential. For that purpose the calibration
campaign of the BSS was completed at the cyclotron of the Université Catholique de
Louvaine La Neuve (UCL, Belgium) at energies of 33 MeV to 60 MeV [72-76].
CHAPTER 2. Calibration facilities
27
2.3.2 The neutron facility and the experimental set- up
The CYClotron of LOuvaine la NEuve (CYCLONE) is a multiparticle,
variable energy isochronous cyclotron capable of accelerating protons up to 80 MeV,
deuterons up to 55 MeV, alpha particles up to 110 MeV and heavier ions up to an
energy of 110 Q²/M MeV (where Q is the charge state and M the mass of the ion).
The UCL neutron beam facility enables the production of quasi-monoenergetic
neutron beams. Monoenergetic protons with energies between about 25 and 70 MeV,
impinging on a 3 mm or 5 mm thick Li target of natural isotope composition, are used
for neutron production. The layout of the neutron production and collimation area is
shown in Fig. 2.4. The pulsing system (D in Fig. 2.4) consists of a pair of deflecting
plates, inserted in the beam line between the cyclotron beam exit port and the
switching magnet.
Fig. 2.4. Schematic drawing of the cyclotron (left) and the neutron beam facility at UCL. D: deflector, S: switching magnet, T: neutron production target, C: collimator. The scale indicates the distance from the target in metres.
The deflecting voltage is provided by a resonance amplifier, phase-locked to
the main cyclotron oscillator. Proton bursts passing in between the plates during the
time of zero deflecting voltage arrive unaffected at the neutron target while all other
bursts observe a deflecting voltage and are stopped by a slit system in the beam line.
After passing through the target the protons are deflected with a dipole magnet to a
graphite beam dump. A collimator restricts the neutron beam to a diameter of 30 mm
at its exit. A cleaning magnet serves to remove charged particles contaminating the
neutron beam. The neutron beam is dumped in a cave downstream of the
measurement cave.
CHAPTER 2. Calibration facilities
28
The calibration measurements were performed in Cave Q, a heavily shielded
concrete room (Fig. 2.5).
Fig. 2.5. The cave Q at UCL.
CHAPTER 2. Calibration facilities
29
2.4 The CERN-EU Reference Field (CERF)
2.4.1 Description of the facility
The CERF facility is installed in one of the secondary beam lines (H6) from
the Super Proton Synchrotron (SPS), in the North Experimental Area on the Prevessin
(French) site of CERN (Figs 2.6- 2.8). A positive hadron beam with momentum of
usually 120 GeV/c is stopped in a copper target, 7 cm in diameter and 50 cm in length
which can be installed in two different positions inside an irradiation cave.
Fig. 2.6. Axonometric view of the CERF facility. The reference positions are also marked (iron and concrete roof and side shield). The side shielding on the Salève side is removed to show the inside of the irradiation cave with the copper target set-up.
CHAPTER 2. Calibration facilities
30
Fig. 2.7. Axonometric view of the CERF facility. The reference positions are also shown. The side shielding on the Jura side is removed to show the external set- up of the facility.
The secondary particles produced in the target traverse a shielding on top of
these two positions. The shielding is made up of either 80 cm concrete or 40 cm iron
and in both cases uniform radiation fields are produced. The fields extend over two
areas of 2 x 2 m2 located at approximately 90° with respect to the incoming beam
direction. Each of them is divided into 16 squares of 50 x 50 cm2, with each element
of these ‘grids’ representing a reference exposure location. Additional measurement
positions are available behind the lateral shielding of the irradiation cave, at the same
angles with respect to the target as for the two roof positions. Shielding is either 80
cm or 160 cm concrete, and at both positions 8 additional exposure locations
(arranged in 2 x 4 grids made up of the same 50 x 50 cm2 elements) are provided. The
nominal measurement locations (the reference field) are at the centre of each square at
25 cm height above floor, i.e. at the centre of a 50 x 50 x 50 cm3 air volume, where
the radiation field is calculated.
The beam is slow-extracted from the SPS over a few seconds. During the
pulse the beam intensity is constant, thus producing a constant radiation field at the
exposure locations. Typical values of dose equivalent rates are 1–2 nSv per PIC-count
CHAPTER 2. Calibration facilities
31
(see section 2.4.2) on top of the 40 cm iron roof-shield and about 0.3 nSv per PIC-
count outside the 80 cm concrete shields (roof and side). The dose equivalent rate at
the reference positions can be varied by proper adjustments of the beam intensity on
the target and it typically ranges from 25 µSv/h to 1 mSv/h on the iron roof-shield and
from 5 µSv/h to 600 µSv/h on the 80 cm concrete roof or lateral shield. The energy
distributions of the various particles (mainly neutrons, but also photons, electrons,
muons, pions and protons) at the various exposure locations, have been obtained by
Monte Carlo simulations performed with the FLUKA code [61-63]. Details of the
latest simulations are given elsewhere [77]. A more thorough description of the CERF
facility is given in [78].
Fig. 2.8. Plan and sectional views of the CERN-EC facility.
CHAPTER 2. Calibration facilities
32
2.4.2 Beam monitoring at CERF
The intensity of the primary beam at CERF is monitored by a Precision
Ionization Chamber (PIC), which is installed about 405 m downstream of the T4
production target. It has been the primary monitor at the CERF facility since many
years. It is placed in the beam just upstream of the copper target, connected to a
current digitizing circuit. Its design is described elsewhere [79-80].
The PIC is an open-air ionization chamber with cylindrical shape. Its sensitive
volume is 0.86 litres (diameter: 185 mm, active length: 32 mm). The charge produced
by ionization by the beam in this volume is collected at a capacitor that is discharged
whenever the charge attains a predefined threshold. It then issues one count that is a
measure of the number of beam particles that have produced this charge. The PIC
serves to normalize the experimental data to the number of particles in the H6 beam.
One PIC-count corresponds (within 10%) to 2.2x104 particles impinging on the target.
In the first years of the CERF operation a check of the beam shape and
position just before the copper target was performed by taking images of the beam,
originally with a radiographic film (for example, ref. [81]) and in the recent past with
a Polaroid film that has the advantage of immediate development. X-ray films
required development in the laboratory of the individual dosimetry service, so that the
results were often available only after the run. This method was time consuming and
only provided rough information on the ‘beam spot’, mainly because of the resolution
of the films. The use of X-ray films allowed a reconstruction of the beam profile, but
only by an off-line analysis [82]. That method was good enough for the purpose of
checking the correct alignment of the beam in the H6 line, which usually did not
present any problem as the beam set-up was done by the operator from the
Experimental Areas control room.
In 2002 two new instruments were installed in the irradiation cave. The
advances in the beam and reference field monitoring at the CERF facility since 2002
are described in chapter 6.
33
CHAPTER 3
Calibration at reference neutron fields
3.1 Calibration of the BSS in monoenergetic neutron fields
3.1.1 Introduction
In order to calibrate the Bonner sphere detectors and to determine their
response functions, measurements with monoenergetic neutrons were performed at
Physikalisch-Technische Bundesanstalt (PTB) in Braunschweig (Germany) in 2001
and 2002. The calibration campaign was completed with the measurements at UCL
(Louvain–la Neuve, Belgium), at energies 33 MeV and 60 MeV, in 2003. The 2001
and 2002 results are presented in detail and in the following the results of all
calibrations are summarised and graphically illustrated at the end of this chapter
(Figs 3.17- 3.20).
3.1.2 Calibration at PTB- Materials and methods
Monoenergetic neutron beams were produced at the 4 MV Van de Graaff
accelerator of the PTB [69-71]. The production reactions are presented in Table 3.1.
Table 3.1. Neutron production reactions for the calibration at PTB.
Neutron Energy [MeV]
Ion energy [MeV] Reaction
0.565 2.304 7Li(p,n)7Be 2.5 3.36 3H(p,n)3He 8.0 5.2 2H(d,n)3He 19.0 2.664 3H(d,n)4He
CHAPTER 3. Calibration of the BSS at reference neutron fields
34
The widths of the neutron distributions are sufficiently narrow to allow the
neutron fields to be considered as monoenergetic. The width of the distribution is
governed mainly by the energy loss of the beam in the target and the energy
distribution thus approximates more to a rectangular shape rather than to a Gaussian.
The measurements were performed in a low scatter hall of dimensions 24 x 30
x 14 m3. The irradiation location was on a grid floor in the centre of the hall (Fig. 3.1).
The irradiations were performed in open geometry, i.e. the neutron producing target
was situated in the centre of the hall while the fluence monitors and the BSS were set
up without any shielding. The BSS support stands were light to minimize neutron
scatter and were designed to allow spheres to be changed without disconnecting the
electronics. Each sphere had its own specific stand and all stands fit on a common
base.
Fig. 3.1. The irradiation hall at PTB.
CHAPTER 3. Calibration of the BSS at reference neutron fields
35
Measurements were performed by exposing each Bonner sphere to neutrons
emitted in the forward direction at a distance of 2 m from the production target. Solid
targets were used consisting of thin layers of metallic LiOH and Ti, loaded with
deuterium or tritium and evaporated on Ta, Ag or Al backings. For the 0.565 MeV
measurement the target consisted of Li while for the 8 MeV a gas target was used.
The target is air-cooled and rotated around the beam spot during the experiment.
For the calibration of the BSS the 2-atm 3He detector was used. The spheres
were positioned for irradiation on a support that could be moved around the target on
a circle of a maximum radius of 5.2 m. The irradiation of the spheres was always
performed in such a geometry that allowed a perpendicular orientation of the 3He
detector with respect to the beam axis. From the previous calibration in the same hall
[45] it was shown that the results from a perpendicular irradiation of the 3He detector
agree with the results of a parallel irradiation within the statistical uncertainties. In the
case of an accidental displacement of the spheres during the measurement, it was
estimated that the angle subtended was less than 5° with respect to the beam axis and
thus the anisotropy effect was judged to be negligible [83]. The charge monitor which
integrates the beam current was used for monitoring the beam.
The radiation components that contribute to the BSS reading are schematically
shown in Fig. 3.2. Various corrections have to be applied in order to derive the
response to monoenergetic neutrons.
Fig. 3.2. Radiation components contributing to the BSS reading. T: direct neutrons, B: target- scattered neutrons reaching the detector, Ain: neutrons scattered by air in the detector, Aout: neutrons scatter by air away from the detector, S: neutrons scattered by other sources (wall, etc) in the detector.
CHAPTER 3. Calibration of the BSS at reference neutron fields
36
The contribution to the counter reading from room and air scatter neutrons was
taken into account by measurements made with a shadow cone. The shadow cone was
made of 20 cm iron and 30 cm polyethylene and was interposed between the target
and the monitor (Fig. 3.3). In the previous calibration of the BSS different shadow
cones were used, depending on the size of the sphere. For the present measurements
only one shadow cone was used. The use of a single shadow cone did not increase the
risk of overshadowing for the smaller spheres or of insufficient shadowing for the
larger ones [83]. It was thus decided to use one shadow cone that would allow a good
compromise between precision and time saving. The overall uncertainty due to this
effect did not exceed 1%.
Fig. 3.3. The shadow cone technique.
CHAPTER 3. Calibration of the BSS at reference neutron fields
37
Neutrons were emitted from a target at all angles. Those emitted at laboratory
angles around 90o are scattered in the target backing and target mounting. The target
assembly is basically a light construction, however the mass of the tube of the vacuum
chamber, the mounting ring of the backing, and the target backing itself all give rise to
neutron scattering or (n,2n) reactions. Some of these neutrons can reach the Bonner
sphere undergoing calibration at 0o. This contribution cannot be subtracted using
shadow- cone measurements and must be evaluated separately. The number of target
scattered neutrons and their spectral distribution can be obtained from Monte Carlo
calculations that consider the detailed description of the target geometry and the
reaction kinematics [84].
For the purposes of this study, the spectral distributions of the target-scattered
neutrons were provided by Schlegel [85] and are shown in Fig. 3.4. Obviously no
scattering occurs in the gas target used for the calibration at 8 MeV.
Especially for the calibration at 19 MeV an additional correction had to be
made, the blank target correction [86]. At this energy, reactions with components of
the target layer such as other isotopes or elements can produce unwanted neutrons that
induce a considerable contribution to the reading of the BSS. To correct for this effect
another measurement was performed with a background target known as blank target.
The blank target is almost identical to the main target disc, except that the material for
the production of the primary neutrons is omitted. For instance, the blank target used
in the measurements with Tritium targets consists of a titanium layer of the same
thickness on an identical backing disc. All beam parameters were kept constant for the
subsequent measurements, always with the beam current integrator as monitor.
CHAPTER 3. Calibration of the BSS at reference neutron fields
38
Fig. 3.4. Uncollided and target-scattered neutron spectral distributions calculated with the TARGET code [85] (upper left: 0.565 MeV, upper right: 2.5 MeV, lower left: 8 MeV, lower right: 19 MeV).
CHAPTER 3. Calibration of the BSS at reference neutron fields
39
3.1.3 Data analysis
For the measurements with the BSS detectors, the MAB (Münchner Apparate
Bau, Germany) electronics [47] was used. The electronics consist of an amplifier,
high voltage and a multi-channel analyser housed in a box. In addition, a pre-amplifier
is connected to the box. This system allows the acquisition of a pulse height spectrum
of the 3He proportional counter. The shape of a typical pulse height spectrum of the 3He proportional counter is shown in Fig. 2.5. Such a spectrum was acquired for every
Bonner sphere. The mean count rates varied between 25 cts/sec and 1200 cts/sec for
measurements without shadow cone, 20 cts/sec and 150 cts/sec for measurements
with shadow cone and between 80 cts/sec and 325 cts/sec for measurements with the
blank target. The count rates were kept low with frequent adjustments of the beam
current by the accelerator operator. This way neither pile –up occurred or dead time
corrections had to be applied. The duration of each measurement varied between 300
sec and 600 sec. The integral detector counts above threshold (Fig. 3.5) were in all
cases above 10000 so the statistical uncertainties were kept low.
Fig. 3.5. A typical pulse height spectrum acquired with the electronics of the 3He proportional counter. A simple threshold placed at about channel 100 allows the separation of the neutron-induced events from noise and gamma ray induced events.
CHAPTER 3. Calibration of the BSS at reference neutron fields
40
At the calibration energy of 8 MeV measurements done without gas filling are
important in order to evaluate the contribution of neutrons that are produced in the
entrance window. Because of technical reasons, these measurements could not be
done. In the calibration in 2001 the percentage of the detector’s reading from neutrons
produced in a gas target without filling to one with gas filling, was evaluated for each
of the Bonner Spheres and it was found to be of the order of 1-2 % for 5 MeV [45].
Therefore, for the calibration in 2002 it was considered that the corresponding
percentage is 1.5% for the larger spheres and 2% for the smaller ones, given the fact
that the production reactions for 5 MeV (in 2001) and for 8 MeV (in 2002) have
similar cross sections.
The only monitor used for the measurement of the spectral fluence was the
charge monitor. The fluence values during the time intervals of the measurements
were communicated by PTB, already corrected for the attenuation in air as well as for
the dead time of the fluence monitor.
The absolute fluence response of the BSS is expressed as the number of counts per
incident neutron fluence. The response RT(E) of each Bonner sphere to direct neutrons
with spectral fluence ΦT(E) is associated with the reading MT of the sphere in the
following expression:
dEEER TTT )()( Φ=Μ ∫ (3.1)
As explained in the Section 3.1.2, there are other contributions to the reading
of the sphere that have to be subtracted, either as background contributions or as
‘noise’ to the real measurement. Thus the reading of the Bonner sphere due to direct
neutrons can be expressed in the following way:
TSSCST Μ−Μ−Μ=Μ (3.2)
where MS is the reading from measurements without shadow cone, MSC is the reading
of the Bonner spheres from measurements with shadow cone and MTS is the reading
of the Bonner spheres due to target scattered neutrons. In the case of calibration of the
BSS with 19 MeV neutrons, equation (3.2) is written as:
CHAPTER 3. Calibration of the BSS at reference neutron fields
41
BTTSSCST Μ−Μ−Μ−Μ=Μ (3.3)
where MBT is the reading from measurements with the blank target. Each of the terms
MSC, MTS, MBT expresses a fraction of the response of the BSS to unwanted neutrons,
hence they can be written as follows:
dEEER SCSCSC )()( Φ=Μ ∫ (3.4)
dEEER TSTSTS )()( Φ=Μ ∫ (3.5) dEEER BTBTBT )()( Φ=Μ ∫ (3.6)
where ΦSC(E), ΦTS(E), ΦBT(E) are the corresponding spectral fluences. For the
definition of the target-scattered fluence ΦTS(E), Monte Carlo calculations were
performed by the PTB [85,86].
To determine the fraction of the reading of each Bonner sphere due to target
scattered neutrons, the spectral fluence ΦTS(E) provided by PTB was folded with the
Monte Carlo calculated response functions according to equation (3.5). Then this
number of theoretical counts MTS was subtracted from the reading obtained during the
measurement without shadow cone, and normalized to the fluence given for the same
time interval. The magnitude of the target- scattering correction depends strongly on
the neutron energy and on the diameter of the sphere. For the present analysis it was
found that the target- scattering effect is not contributing essentially to the reading of
the BSS, however the corrections were done for a more precise determination of the
response to direct neutrons [87- 88].
Large percentages occur only where the absolute value of the response is very
small. Most of the corrections lie between 0.02% and 3.6% as shown in Table 3.2.
CHAPTER 3. Calibration of the BSS at reference neutron fields
42
Table 3.2. Percent contribution of the target- scattered neutrons to the total reading of the BSS.
Energy [MeV] Sphere
0.565 2.5 19 81 0.02 % 2.24 % 0.05 %
81cd 0.02 % 2.63 % 0.05 % 108 0.05 % 2.11 % 0.11 % 133 0.08 % 1.72 % 0.11 % 178 0.05 % 1.70 % 0.11 % 233 0.02 % 1.36 % 0.11 %
Ollio 0.02 % 3.59 % 0.10 % Stanlio 0.02 % 2.81 % 0.06 %
The effect of scattered neutrons depends on how the response of the
instrument in the energy region where the scattered neutrons occur compares with the
response to the primary neutrons. A contribution of 2-3% due to scattered neutrons
whose spectral distribution is rather flat and extends to very low energies, can induce
an important contribution (up to 40%) in the response of a small Bonner sphere when
calibrated at high energies. This can be observed in Table 3.3, which shows the
contributions to the BSS readings due to scattered radiation. Table 3.4 shows the same
effect but in terms of fluence responses.
Table 3.3. Percent contribution of scattered radiation
to the BSS reading.
Energy [MeV] Sphere 0.565 2.5 8 19
81 18% 40% 29% 28% 81cd 18% 36% 24% 24% 108 26% 20% 9% 19% 133 32% 10% 8% 16% 178 17% 7% 5% 13% 233 5% 6% 17% 9%
Ollio 5% 9% 4% 6% Stanlio 16% 37% 20% 19%
CHAPTER 3. Calibration of the BSS at reference neutron fields
43
Table 3.4. Ratio of the response of the Bonner spheres to direct neutrons over the response to direct plus scattered neutrons.
Energy [MeV] Sphere 0.565 2.5 8 19
81 0.85 ± 0.06 0.72 ± 0.07 0.78 ±0.06 0.04 ± 0.01
81Cd 0.87 ± 0.06 0.75 ± 0.07 0.81 ± 0.06 0.08 ± 0.01
108 0.90 ± 0.06 0.84 ± 0.07 0.89 ± 0.06 0.15 ± 0.01
133 0.92 ± 0.06 0.89 ± 0.07 0.92 ± 0.06 0.20 ± 0.04
178 0.94 ± 0.06 0.93 ± 0.07 0.95 ± 0.06 0.31 ± 0.04
233 0.95 ± 0.07 0.95 ± 0.07 0.96 ± 0.07 0.47 ± 0.07
Ollio 0.96 ± 0.06 0.96 ± 0.07 0.97 ± 0.06 0.64 ± 0.07
Stanlio 0.87 ± 0.06 0.76 ± 0.07 0.88 ± 0.06 0.31 ± 0.03
For the error analysis of the absolute fluence responses the following errors
were considered and propagated with the error propagation formula:
• Statistical uncertainty in the counting rates of the BSS (raw data). This is between
1% and 3% depending on the total counts above the threshold.
• Systematic uncertainty of the fluence monitors and statistical uncertainty in their
counting rate. The first uncertainty varied between 3.8% for the energies
0.565 MeV and 8 MeV, and 5.6% for the energies 2.5 MeV and 19 MeV.
• Especially for the measurements at 8 MeV, the contribution of neutrons produced
in the entrance window was taken into account. This was estimated to be 1.5% for
the larger spheres and 2% for the smaller spheres, since the latter are more sensitive
to scattered radiation.
3.1.4 Results
Following the procedure described above and correcting for room and in-
scatter contribution as well as for target- scattering, the experimental absolute fluence
responses of the BSS were determined. These are listed in Table 3.5 together with
their total experimental uncertainties.
CHAPTER 3. Calibration of the BSS at reference neutron fields
44
Table 3.5. Experimental absolute fluence responses to monoenergetic neutrons with energies 0.565 MeV, 2.5 MeV, 8 MeV and 19 MeV.
Energy [MeV]
Sphere 0.565 2.5 8 19
81 0.735±0.033 0.264±0.020 0.092±0.005 0.028±0.003
81Cd 0.741±0.033 0.270±0.020 0.097±0.005 0.049±0.003
108 1.965±0.084 0.936±0.059 0.381±0.016 0.171±0.016
133 2.797±0.116 1.788±0.105 0.792±0.032 0.414±0.075
178 2.899±0.118 2.768±0.155 1.510±0.059 0.777±0.080
233 1.819±0.104 2.789±0.152 1.914±0.104 1.066±0.139
Ollio 0.591±0.024 1.237±0.067 0.974±0.038 0.786±0.071
Stanlio 0.661±0.029 0.253±0.018 0.159±0.007 0.207±0.020
3.1.5 Comparison of experimental to MC calculated responses
The calculated responses taken as references in this work are those reported as
‘mean response functions’ in reference [45]. The responses at the energies of
calibration were derived by linear interpolation. These are presented in Table 3.6.
Table 3.6. The calculated absolute fluence responses for the calibration energies (linear interpolation between the mean response functions of the BSS).
Energy [MeV]
Sphere 0.565 2.5 8 19
81 0.886 ± 0.089 0.302 ± 0.030 0.083 ± 0.008 0.036 ± 0.004
81Cd 0.860 ± 0.086 0.281 ± 0.028 0.089 ± 0.009 0.050 ± 0.005
108 2.157 ± 0.216 1.036 ± 0.104 0.352 ± 0.035 0.154 ± 0.015
133 3.017 ± 0.302 1.859 ± 0.186 0.737 ± 0.074 0.339 ± 0.034
178 3.013 ± 0.301 2.800 ± 0.280 1.489 ± 0.149 0.734 ± 0.073
233 1.877 ± 0.188 2.890 ± 0.289 2.006 ± 0.201 1.139 ± 0.114
Ollio 0.605 ± 0.061 1.301 ± 0.130 1.069 ± 0.107 0.878 ± 0.088
Stanlio 0.715 ± 0.072 0.258 ± 0.026 0.146 ± 0.015 0.209 ± 0.021
CHAPTER 3. Calibration of the BSS at reference neutron fields
45
Any new calculation should be validated against measurements to avoid errors
due to, for example, mistakes in the computational model. To check the consistency
of the calculated responses with the measured ones, the ratio Rcalc/Rexp was
determined, where Rcalc is the calculated response of a given sphere and Rexp its
experimentally derived response. The results are presented in Table 3.7 and plotted in
Figs. 3.6- 3.8.
This ratio should be close to unity, within the uncertainties, for all spheres.
The overall ratio that is calculated as the weighted mean of all ratios, i.e. <Rcalc/ Rexp>
is the calibration factor of the specific 3He proportional counter. These ratios are
shown in Table 3.8. At the end of this chapter the reader can find the final calibration
factor for both calibration campaigns.
Table 3.7. Ratios of calculated over measured absolute fluence responses for all energies.
Energy [MeV] Sphere
0.565 2.5 8 19
81 1.205±0.132 1.143±0.144 0.906±0.101 1.295±0.192
81Cd 1.160±0.127 1.041±0.129 0.922±0.102 1.019±0.119
108 1.098±0.119 1.107±0.131 0.924±0.100 0.901±0.123
133 1.079±0.117 1.040±0.121 0.931±0.100 0.819±0.170
178 1.039±0.112 1.012±0.116 0.986±0.106 0.945±0.136
233 1.032±0.119 1.036±0.118 1.048±0.119 1.068±0.175
Ollio 1.023±0.110 1.052±0.120 1.098±0.118 1.117±0.151
Stanlio 1.082±0.118 1.022±0.126 0.920±0.100 1.010±0.141
CHAPTER 3. Calibration of the BSS at reference neutron fields
46
Fig. 3.6. Ratios Rcalc/ Rexp of calculated over measured responses for all spheres at a given calibration energy.
Table 3.8. Weighted mean of ratio of calculated over measured responses for all spheres.
Sphere <Rcalc /Rexp>
81 1.076 ± 0.066 81cd 1.022 ± 0.059 108 0.997 ± 0.058 133 0.983 ± 0.060 178 0.999 ± 0.058 233 1.043 ± 0.064
Ollio 1.066 ± 0.061 Stanlio 0.999 ± 0.059
<Rcalc /Rexp> 1.021 ± 0.021
CHAPTER 3. Calibration of the BSS at reference neutron fields
47
In order to illustrate the behavior of each sphere at the energies of the present study,
the ratios (calculated/ experimental response) are plotted versus sphere in Fig. 3.7.
Fig. 3.7. Mean ratios <Rcalc/Rexp> of calculated over measured responses if all calibration energies for a given sphere are considered. The bars represent the total uncertainty. The ratio is the calibration factor of the 3He proportional counter as derived from the four calibration energies.
CHAPTER 3. Calibration of the BSS at reference neutron fields
48
Fig. 3.8. Ratios Rcalc/Rexp (calculated over measured responses) for all calibration energies for each Bonner Sphere. The average ratio <Rcalc/Rexp> over all energies is also given.
CHAPTER 3. Calibration of the BSS at reference neutron fields
49
The spread of the data points in Fig. 3.6- 3.8 does not show any systematic
tendency and is generally well described by their uncertainties. The somewhat larger
deviations for the smallest spheres at 19 MeV can be attributed to additional
experimental uncertainties on the fluence determinations which could not be corrected
and thus were not reflected in the evaluated uncertainties. This could not be a
systematic error because it would have appeared in the results for all spheres. It is also
likely to be an uncertainty linked to the behavior of the small spheres at high energies.
As known, the small Bonner spheres are mostly sensitive to thermal neutrons. In the
case of the calibration at 19 MeV the thermalization is incomplete for high-energy
incident neutrons. However, this results only in a small energy dependence of the
response ratio, as reported elsewhere [88].
There are other factors that could influence the response of the Bonner
spheres, thus also the ratio of the calculated over the measured response. A very
important factor is the density of the polyethylene. As it is reported in ref. [88], a
small variation in the density can lead to a large change in the response. The
magnitude of this effect varies with energy. Another influencing factor could be the
geometry of the 3He detector, or the gas pressure adopted in the MC calculations.
These characteristics are not exactly known and a slight deviation from their true
value may provoke a large uncertainty in the estimation of the response function of
the sphere. Figs 3.6- 3.8 show that in general the calculated responses, Rcalc(E), are in
very good agreement with the measurements, Rexp(E).
CHAPTER 3. Calibration of the BSS at reference neutron fields
50
3.2 Calibration at Quasi- Monoenergetic Neutron Fields
3.2.1 Materials and methods The measurements were performed in cave Q that is shown in Fig. 3.9. The beam
radius was 2.78 cm as measured at a distance of 6.1 m from the target. The neutron
production reactions are given in Table 3.9.
Fig. 3.9. Experimental set up of the BSS in Cave Q at UCL.
Table 3.9. Main characteristics of the neutron production reactions of the quasi- monoenergetic beams.
Reaction 7Li(p,n)7Be
Proton energy Ep [MeV] 36.4 62.9
Energy loss of protons in target ∆Ep MeV] 3.7 1.4
Nominal neutron energy Eo [MeV] 32.9 60.6
CHAPTER 3. Calibration of the BSS at reference neutron fields
51
A set of different neutron detectors was used for the characterization of the
neutron beam. The properties of these detectors can be found in ref. [75]. A short
overview is given below.
• A proton recoil telescope (PRT) was used for the measurement of the peak
fluence, i.e. the fluence of neutrons within the high-energy peak. This
instrument is based on the detection of recoil protons produced by neutrons
that are scattered off hydrogen nuclei at center- of- mass scattering angles
greater than 148o.
• A U-238 fission chamber (FC), consisting of a stack of ionization chambers.
Due to their large energy deposition, the fission products can be easily
discriminated from other events by setting a threshold in the pulse-height
spectrum.
• A NE102 transmission detector was permanently installed in the beam. This
detector has a fast (NE102_f) and a slow (NE102_s) component.
• A beam charge (Q) monitor. It was not suited for use as a high-precision beam
monitor because of a large leakage current varying in time.
All spectral neutron fluences were measured with a NE213 liquid scintillation
detector 102 mm in length and 51 mm in diameter, with the time of flight method
[76]. They are shown in Fig. 3.10. Of the five monitors, only the fast component of
NE102 was used for the normalization of the present data. The charge monitor and
the slow component of NE102 presented linearity problems, while the fission
chamber had poor statistics. Improvement of the charge measurement was not
attempted because of lack of access to the beam dump and its high radiation level.
The measurement of the spectral fluence at each calibration energy was essential
because the neutron beam is not monoenergetic.
CHAPTER 3. Calibration of the BSS at reference neutron fields
52
Fig. 3.10. Spectral fluence measured by the NE213 scintillator for the 33 MeV and 60 MeV neutron beams.
To determine the response function of the BSS, each detector has to be
exposed to a broad parallel field, uniform over the whole sensitive area of the
instrument. The area of the BSS to be irradiated was larger than the cross section of
the beam, therefore a scanning procedure had to be applied in order to simulate a
uniform irradiation by a broad beam. For the purposes of the present work the PTB
scanning system which is based on a cross scanning movement was used (Fig. 3.11).
CHAPTER 3. Calibration of the BSS at reference neutron fields
53
Fig. 3.11. The PTB scanning system used for the irradiation of the BSS at UCL.
The scan was achieved by continuously moving the monitor in front of the
beam, according to a law that is the superposition of two motions, perpendicular to
each other, as in the case of a Lissajous figure. A Lissajous figure is obtained by
combining two harmonic motions, orthogonal to each other and having different
frequencies, non-multiple and out of phase (Fig. 3.12). More details on the scanning
procedure are given in ref. [89].
Fig. 3.12. Lissajous figure.
CHAPTER 3. Calibration of the BSS at reference neutron fields
54
Apart from the complete set of the BSS, the LINUS (Long Interval Neutron
Survey-meter) detector was also used in this calibration experiment. LINUS is a
neutron rem counter (a detector with high efficiency to thermal neutrons) of the
Andersson-Braun (A-B) type. The structure of the moderator-attenuator is modified in
such a way that the instrument has a response comparable to conventional monitors in
the low and medium energy range, i.e. from thermal to 7- 8 MeV, and much increased
at the higher energies. The specific moderator for LINUS consists of polyethylene,
borated plastic and a shell of Pb. The counter is a spherical 3He proportional counter
(3.2 cm active diameter, filled with 304 kPpa 3He and 101 kPa Kr). For more details
on the LINUS the reader can see references [90-93]. LINUS is a fully characterized
instrument and its configuration is very similar to the configuration of Ollio. For these
reasons and in order to inter-compare the responses of the two detectors,
measurements were performed with LINUS during the experiments at UCL.
3.2.2 Experiment and data analysis
The procedure followed for the calibration at UCL is different from the one
followed at PTB (see section 3.1.2). No shadow cone measurements were needed
because the contributions from neutrons scattered in the target, in the collimators and
in the air, are negligible. However, background measurements were performed for
each sphere. Another contribution of scattered radiation can be produced during the
scanning procedure from the sphere itself. For this purpose special measurements
were performed and it was found that this contribution does not exceed 2%.
The mean count rates varied between 35 cts/sec and 325 cts/sec for the
calibration measurement and between 7 counts/s and 45 counts/s for the background
measurements. No pile –up or dead time corrections had to be applied. Measurements
lasted between 150 sec and 1600 sec. The total number of counts for each
measurement provided good statistics.
The fluence and spectral fluence measurements were performed by the PTB
metrology team [94]. In contrast to the PTB monoenergetic neutron fields,
the 33 MeV and 60 MeV fields are quasi-monoenergetic fields [95]. The spectral
CHAPTER 3. Calibration of the BSS at reference neutron fields
55
distributions of the neutron beams used are characterized by a high-energy peak and a
continuum that extends to low energies. The peak results from transitions to the
ground state and the first excited state in 7Be (Ex = 0.429 MeV), which are stable
against particle emission. The continuum is caused by breakup reactions and by
interactions of neutrons with the collimator, etc.
The low-energy cut-off of the spectral fluence measured with the scintillation
detector and the fission chamber is about 3 MeV. Below this energy the spectral
fluence is unknown. For the determination of the absolute fluence response of the
BSS the knowledge of this information is essential because the BSS responds to this
part of the spectra as well. Especially for the smaller Bonner spheres this response is
assumed to be higher in comparison to their response to the peak neutrons. A flat
extrapolation to thermal energies could be a good approximation to the real case
because of the absence of scattered neutrons in the cave, however a 1/E extrapolation
should also be considered.
In order to be able to extrapolate, an unfolding with the experimental data
taken with the BSS at the energies of 33 MeV and 60 MeV was performed. The
unfolding provided the spectral fluence at all energy intervals above zero. The
solution spectrum indicated that a constant extrapolation to lower energies seemed to
be justified. For the analysis of the irradiations of the Bonner spheres the spectral
fluence measured with the reference monitor NE102 should be used. The number of
peak neutrons N0 per unit monitor count Mf of the fast NE102 detector can be
determined with the following formulas:
(N0/Mf) = 2.78(10)⋅102 at 33 MeV (3.7)
and
(N0/Mf) = 1.72(14)⋅102 at 60 MeV (3.8)
The reference distance from the Li-target is 6 m. The fluence monitor data
communicated by PTB provide only the fluence in the peak. Therefore, the total
fluence was ‘reconstructed’ by using the ratio ‘peak/total’ from the spectra provided
by PTB (Fig. 3.10).
CHAPTER 3. Calibration of the BSS at reference neutron fields
56
3.2.3 Results and discussion
For the calculation of the Bonner sphere absolute fluence response, two
methods can be used.
The first method has already been used in the past in similar cases [93]. It
involves Monte Carlo simulations for the determination of the absolute fluence
response of each Bonner sphere to the total spectrum and to the peak of the spectrum
only. Then the experimental absolute fluence response to monoenergetic neutrons of
33 MeV and 60 MeV can be derived by equation (3.9):
QQQ
QRR
R
RRR FLUKA
peakFLUKA
totalFLUKAtotal
totaltotalpeak
ΦΦΦ
Φ−×−= ΦΦ
Φ
ΦΦΦ
0
0,,
,
exp,exp
,exp
, )]([ (3.9)
where exp, peakRΦ is the experimental absolute fluence response to the high- energy peak
of the spectrum, exp,totalRΦ is the experimental absolute fluence response to the total
spectrum, FLUKApeakR ,Φ is the FLUKA [61-63] Monte Carlo calculated response to the high-
energy peak and FLUKAtotalR ,Φ is the FLUKA Monte Carlo calculated response to the total
spectrum. The coefficients Φ0/Q and Φ/Q are the neutron fluence in the high-energy
peak per unit charge in the primary beam and the total neutron fluence per unit charge
in the primary beam, respectively.
The second method is less direct than the first one: if the Monte Carlo
calculated response functions of the BSS are folded with the experimental spectral
fluences, they should reproduce the experimental reading (i.e, the counts) for each
detector. Then, if the ratios of the experimental counts over the theoretical ones is
close to unity within the statistical uncertainties, this means that the shape of the MC
calculated response functions is correct and therefore the response of the BSS at the
calibration energies should be close to the MC calculated values.
The first method is generally accepted as more precise in comparison to the
second one but the latter is less time-consuming. Therefore, in the context of this
thesis and because of a big delay in the communication of the monitor data from PTB,
it was decided to use the second method.
CHAPTER 3. Calibration of the BSS at reference neutron fields
57
The first step was to re-adjust (‘re-bin’) the width of the energy bins of the
fluence spectra from PTB, so as to obtain the energy binning used in the FLUKA
Monte Carlo code. This was necessary in order to fold the spectra with the MC
calculated response functions of the BSS, since they were also calculated with the
FLUKA code [96]. The response functions involved in the study presented in this
chapter are the ones reported as ‘nominal’ in [45]. The reader can address to this
reference for more details and can find the response matrix in the Appendix B.
It was also necessary to extrapolate the fluence spectra to energies lower than
3.5 MeV and 5 MeV for the spectra of 33 MeV and 60 MeV respectively. As
mentioned in section 3.2.2, a justified extrapolation would be a flat one for both
energies. However it was decided to use several types of extrapolations for each
spectrum. This allowed a number of conclusions to be drawn that are presented later
in this chapter. The extrapolations that were used are summarised in Table 3.10.
Table 3.10. Extrapolations used for the fluence spectra of 33 and 60 MeV in order to use them in the unfolding program.
Extrapolation type Peak
Energy
[MeV] Flat 1/E
33 3.5 MeV 0 1 keV 0 10 keV 0 100 keV 0 500 keV 0 1 MeV 0
60 5 MeV 0 1 keV 0 100 keV 0 150 keV 0 220 keV 0 300 keV 0 500 keV 0 1 MeV 0
As seen from Fig. 3.10 (section 3.2.1), the cutoff energy E1 for the peak region
is 29.0 MeV for the 33 MeV while for the 60 MeV it is 54.5 MeV. Knowing the
partial contribution of each part, i.e. of the tail and of the high-energy peak to the total
and the number of neutrons in either of the two parts, one can easily reconstruct the
total number of neutrons in the whole spectrum. For both energies it was calculated
that the tail contributes about 50% to the total spectrum. It is obvious that for every
type of extrapolation, the relative contribution of the tail to the total spectrum
changes. The percentage of neutrons included in the extrapolated part over the number
of neutrons in the total spectrum, varies between 0.03% (for the 1/E extrapolation
between 1 keV and zero, for the 33 MeV) and 32% (for the 1/E extrapolation between
1 MeV and zero, for the 60 MeV).
CHAPTER 3. Calibration of the BSS at reference neutron fields
58
In the next step, the different spectra were folded with the MC calculated
responses and the sum of the products for all energy bins was divided by the total
number of neutrons in the full spectrum. The results are presented in Tables 3.11 and
3.12 for all types of extrapolation.
Table 3.11. Results of folding the MC responses with different types of spectra, for the 33 MeV calibration energy. The values given are counts/neutron/cm2.
Type of extrapolation Sphere Flat
< 3.5 MeV 1/E
< 1 MeV 1/E
< 0.5 MeV 1/E
< 0.1 MeV 1/E
<10 keV 1/E
< 1 keV 81 0.093 0.799 0.503 0.182 0.101 0.094
81cd 0.110 0.751 0.481 0.190 0.117 0.110 108 0.324 1.018 0.719 0.406 0.331 0.324 133 0.594 1.078 0.863 0.647 0.598 0.594 178 1.033 1.097 1.060 1.035 1.033 1.033 233 1.338 1.084 1.187 1.305 1.335 1.337
Stanlio 0.264 0.740 0.541 0.325 0.270 0.265 Ollio 0.874 0.627 0.732 0.844 0.872 0.874
LINUS 0.377 0.279 0.320 0.365 0.376 0.377
Table 3.12 Results of folding the MC responses with different types of spectra, for the 60 MeV calibration energy. The values given are counts/neutron/cm2.
Type of extrapolation Sphere Flat
<5 MeV 1/E
< 1 MeV 1/E
<500 keV 1/E
<300 keV 1/E
<220 keV 1/E
<150 keV 1/E
<100 keV 1/E
<1 keV 81 0.051 0.476 0.280 0.193 0.155 0.122 0.097 0.051
81cd 0.068 0.455 0.276 0.197 0.163 0.132 0.110 0.068 108 0.192 0.626 0.422 0.332 0.294 0.261 0.236 0.192 133 0.366 0.693 0.535 0.468 0.439 0.415 0.397 0.366 178 0.677 0.778 0.725 0.704 0.696 0.689 0.684 0.677 233 0.933 0.854 0.888 0.905 0.912 0.919 0.924 0.933
Stanlio 0.309 0.583 0.457 0.401 0.377 0.355 0.339 0.309 Ollio 0.886 0.738 0.807 0.838 0.851 0.862 0.871 0.886
LINUS 0.346 0.293 0.317 0.328 0.333 0.337 0.340 0.346
CHAPTER 3. Calibration of the BSS at reference neutron fields
59
The ratios of the experimental total counts over the peak fluence were
calculated for each BSS detector. The peak fluence was calculated by dividing the
peak number of neutrons with the field size used at each measurement. Then the total
fluence was calculated by the peak fluence multiplied by the factor (Total/Peak)
derived from the spectra. The reader should keep in mind that the factor (Total/Peak)
varies because the total fluence varies according to the type of extrapolation used.
From these ratios one can calculate the ratios Exp.counts/total fluence using the
following formula (3.10)
total
peak
peaktotal
countsExpcountsExpΦ
Φ⋅
Φ=
Φ.. (3.10)
The results are shown in Tables 3.13 and 3.14 for 33 MeV and 60 MeV
respectively.
Table 3.13. Ratios (Exp.counts/ Total fluence) for 33 MeV calculated according to formula (3.10). The values given are counts/neutron/cm2.
Exp. Total (33 MeV) Sphere Flat
< 3.5 MeV 1/E
<1 MeV 1/E
<0.5 MeV 1/E
<0.1 MeV 1/E
< 10 keV 1/E
< 1 keV 81 0.043 0.030 0.035 0.041 0.042 0.043
81cd 0.040 0.028 0.034 0.039 0.040 0.040 108 0.138 0.097 0.114 0.133 0.137 0.138 133 0.250 0.176 0.208 0.242 0.250 0.250 178 0.443 0.311 0.368 0.428 0.442 0.443 233 0.577 0.405 0.479 0.557 0.575 0.557
Stanlio 0.118 0.082 0.098 0.113 0.117 0.118 Ollio 0.422 0.296 0.351 0.407 0.421 0.422
LINUS 0.115 0.081 0.096 0.111 0.115 0.115
CHAPTER 3. Calibration of the BSS at reference neutron fields
60
Table 3.14. Ratios (Exp.counts/ Total fluence) for 60 MeV calculated according to formula (3.10). The values given are counts/neutron/cm2.
Exp. Total (60 MeV) Sphere Flat
< 5 MeV 1/E
<1 MeV 1/E
<500 keV 1/E
<300 keV 1/E
< 220 keV 1/E
< 150 keV 1/E
< 100 keV 1/E
< 1 keV 81 0.100 0.082 0.090 0.094 0.096 0.097 0.098 0.100
81cd 0.100 0.082 0.091 0.094 0.096 0.097 0.098 0.100 108 0.156 0.129 0.142 0.148 0.150 0.152 0.154 0.156 133 0.260 0.214 0.236 0.245 0.249 0.253 0.256 0.260 178 0.407 0.335 0.369 0.384 0.390 0.396 0.400 0.407 233 0.539 0.444 0.489 0.509 0.517 0.525 0.530 0.539
Stanlio 0.228 0.188 0.207 0.215 0.219 0.222 0.224 0.228 Ollio 0.583 0.480 0.529 0.550 0.559 0.567 0.572 0.583
LINUS 0.146 0.120 0.132 0.138 0.140 0.142 0.143 0.146
The aim is to obtain the ratios of the experimental counts over the counts
calculated by the folding. These results are presented in Tables 3.15 and 3.16 for 33
MeV and 60 MeV respectively.
Table 3.15. Experimental counts / calculated counts for 33 MeV.
Type of extrapolation Sphere Flat
< 3.5 MeV 1/E
<1 MeV1/E
<0.5 MeV1/E
<0.1 MeV1/E
< 10 keV 1/E
< 1 keV 81 0.462 0.038 0.070 0.225 0.416 0.457
81cd 0.364 0.037 0.071 0.205 0.342 0.364 108 0.426 0.095 0.159 0.328 0.414 0.426 133 0.421 0.163 0.241 0.374 0.418 0.421 178 0.429 0.284 0.347 0.414 0.428 0.429 233 0.431 0.374 0.404 0.427 0.431 0.417
Stanlio 0.447 0.111 0.181 0.348 0.433 0.445 Ollio 0.483 0.472 0.480 0.482 0.483 0.483
LINUS 0.305 0.290 0.300 0.304 0.306 0.305
CHAPTER 3. Calibration of the BSS at reference neutron fields
61
Table 3.16. Experimental counts / calculated counts for 60 MeV.
Type of extrapolation Sphere Flat
< 5 MeV1/E
<1 MeV 1/E
<500 keV1/E
<300 keV 1/E
<220 keV1/E
<150 keV 1/E
< 100 keV1/E
< 1 keV81 1.961 0.172 0.321 0.487 0.619 0.795 1.010 1.961
81cd 1.471 0.180 0.330 0.477 0.589 0.735 0.891 1.471 108 0.813 0.206 0.336 0.446 0.510 0.582 0.653 0.813 133 0.710 0.309 0.441 0.524 0.567 0.610 0.645 0.710 178 0.601 0.431 0.509 0.545 0.560 0.575 0.585 0.601 233 0.578 0.520 0.551 0.562 0.567 0.571 0.574 0.578
Stanlio 0.738 0.322 0.453 0.536 0.581 0.625 0.661 0.738 Ollio 0.658 0.650 0.656 0.656 0.657 0.658 0.657 0.658
LINUS 0.422 0.410 0.416 0.421 0.420 0.421 0.421 0.422
The data presented in Tables 3.15 and 3.16 are the final results of the method
that was followed for the analysis of the calibration experiment of the BSS. They are
plotted in Figs 3.13 and 3.14 respectively. The horizontal axis represents the upper
limit of the energy interval in which the extrapolation 1/E is done.
Fig. 3.13. Ratios Experimental counts / calculated counts for 33 MeV.
CHAPTER 3. Calibration of the BSS at reference neutron fields
62
Fig. 3.14. Ratios Experimental counts / calculated counts for 60 MeV.
As shown in Tables 3.15 and 3.16 the ratios (experimental/calculated counts)
are far from unity. This big discrepancy was first attributed to a possible error in the
definition of the field size that was used in each measurement for the determination of
the total neutron fluence. This argument was discarded after a thorough investigation
of the experimental data. First, any change of the field size or of any other detail in the
set-up was recorded in the logbook, and second, data obtained for Ollio for different
field sizes gave consistent experimental results. In Table 3.17 the experimental data
for Ollio along with the results are shown.
Table 3.17. Extract from the logbook of the experimental data obtained for Ollio. Energy (MeV)
Field size (cm2)
Net Ollio counts
Peak fluence (1/ cm2)
Counts/peak fluence (cm2)
31 x 30 421592 425552 0.991 33 42 x40 192421 190489 1.010 31 x 30 502605 292573 1.718 60 42 x40 235042 135512 1.734
CHAPTER 3. Calibration of the BSS at reference neutron fields
63
As seen from Table 3.17 the results for Ollio for both field sizes are consistent
within a statistical uncertainty of maximum 2%. This provides evidence of the correct
recording of the field size during the experiment. Therefore it was decided to
investigate other possible sources of error.
The possibility that the MC calculated response matrix of the BSS system is
not correct can be excluded. The reason is that the MC responses have been verified
by the two previous calibrations of the BSS at PTB and the agreement between the
experimental and the MC response functions is very good (see Figs. 3.17-3.20 at the
end of this chapter). There is no reason to believe that the MC simulations provide
correct results up to 20 MeV but not at 33 MeV and 60 MeV.
Taking into account that the experimental procedure did not include any other
source of error and that the monitor data provided by PTB are correct, it is clear that
the discrepancies observed are due to a factor that had not been noticed and hence not
taken into account up to now. The only possible source for the big discrepancies thus
seems to be a malfunction of the electronics and data acquisition system of the BSS.
The task of the electronics is to ensure the correct determination of the number
of neutron events produced in the 3He proportional counter. For that purpose, it is
recommended that the pulse height -spectrum (the spectrum of the proton recoils in
the 3He counter) obtained from the detector system be recorded. This has always been
done for all the calibration campaigns of the BSS (at PTB 2001, 2002 and at UCL
2003) since the same electronics is used. By comparing the shapes of pulse height
spectra obtained by the same electronics one can see that the UCL spectra seem
distorted. There is no definite plateau region while the thermal peak is also not well
defined. This may provide an indication of a malfunction of the data acquisition
system and may therefore explain the big discrepancies observed in the experimental
results. Examples of the spectra acquired in the several calibration experiments of the
BSS are shown in Fig. 3.15.
CHAPTER 3. Calibration of the BSS at reference neutron fields
64
Fig. 3.15. Spectra from the PTB 2001 calibration (upper left), the PTB 2002 calibration (upper right), and the UCL 2003 calibration (lower left: a normal measurement, lower right: a background measurement).
A typical 3He pulse height spectrum is shown in Fig 3.5 (section 3.1.3). By the
comparison of that spectrum with the ones shown in Fig. 3.15 it was concluded that
the MAB data acquisition electronics box was not operating correctly during the UCL
calibration. To verify the problem with the MAB electronics, measurements with a
second set of electronics were performed long after the calibration. It was not possible
to make this test immediately after the calibration at UCL, because the reference
fluence data was communicated with a big delay and until the moment that this data
arrived and the analysis could be done, there was no evidence that the electronics was
not functioning properly. As soon as the analysis of the experimental data was
finished, it was obvious that there was a source of error. The mab box was then sent
for repair and only very recently it could be tested again in the calibration hall of the
SC-RP group.
CHAPTER 3. Calibration of the BSS at reference neutron fields
65
The electronics used for this purpose, is the so called LINUS electronics. It is
based on NIM standard and it consists of an amplifier (Ortec 570), a single channel
analyser (Ortec 550A), a counter (Ortec 994) and the power supply assembled in a
portable NIM crate. This electronics are used with the LINUS that includes the same
type of 3He proportional counter as the Bonner spheres. The only difference is the
operating high voltage of the detector. Normally the comparison of the results
obtained with the two different electronics systems, should give the ratio unity. This
must be the case when the two systems are operating correctly. From the
measurements in the calibration hall, it was found that the ratio of the two systems is
approximately 0.3. This means that the mab box counted about three times less that it
would, if it worked properly. This test can only prove the initial hypothesis about the
malfunction of the electronics and cannot provide a correction factor to be applied on
the experimental data from UCL. This might have been possible if the tests had been
done before the shipping of the electronics to the MAB company or immediately after
the calibration.
From Figs 3.13 and 3.14 one can draw some additional interesting conclusions.
• The ratios of the experimental /calculated (convoluted) counts for each
Bonner Sphere vary as a function of the extrapolation used for the part of the
spectrum where the shape was not precisely defined. It is obvious that the
type of extrapolation affects seriously the sphere count-rate, especially for the
smaller ones. This is due to the fact that these spheres are more sensitive to
the thermal and low- energy neutrons and this is the energy region where the
extrapolation is done. From this behavior one can conclude that the Bonner
spheres respond correctly to the energy spectrum, but for reasons mentioned
above (failure of electronics) their correct experimental responses could not
be derived.
• From the same graphs it can be seen that the sensitivity curves of LINUS and
Ollio are parallel but not identical. This is expected because the 3He
proportional counters used inside them as active sensors, are not the same.
This fact can be considered as an additional proof of the correct behavior of
the spheres.
• The ratios calculated for the flat extrapolation at 33 MeV for each sphere are
more ‘centered’ around a certain value (0.45). This is not the case for all other
CHAPTER 3. Calibration of the BSS at reference neutron fields
66
extrapolations where the ratios fluctuate randomly. This agreement indicates
that these ratios would be close to unity if the electronics had not failed. It
also provides evidence that the spectrum is flat below 3.5 MeV. Although the
poor energy resolution of the BSS does not allow a detailed description of the
spectral fluence to be given, it has been shown [97] that a well characterized
set of spheres allows the integral neutron fluence to be determined within 4%.
• For the same reason, for the 60 MeV the 1/E extrapolation between zero and
220 keV seems more correct than the flat extrapolation given in [95].
3.3 Combined results of the 2001 and 2002 calibration measurements - conclusions
The BSS was calibrated for the first time at PTB in 2001 with monoenergetic
neutrons with energy 0.144 MeV, 1.2 MeV, 5 MeV and 14.8 MeV. The full
description of the data analysis and the presentation of the experimental results are
given in [45]. Apart from the verification of the MC calculated response functions of
the BSS, the second task of the calibration experiments is the determination of the
calibration factor of the 2-atm 3He proportional counter used with the BSS. In Table
3.18 the experimental absolute fluence responses of both calibration experiments of
2001 and 2002 are listed. The MC calculated absolute fluence responses are given in
Table 3.19.
For the determination of the calibration factor of the 3He proportional counter
with the given set of spheres, the weighted ratios Rcalc/Rexp are needed. First, the ratios
of the calculated over the experimental fluence responses were calculated by simply
dividing the data given in Tables 3.18 and 3.19. The results are shown in Table 3.20.
Then, the mean weighted ratios over all energies for every sphere were calculated
(Table 3.21) and plotted in Fig. 3.16.
From Table 3.21 and Fig. 3.16 one can see that the weighted ratios for all
spheres are slightly higher than unity. This may indicate a systematic uncertainty that
is due to factors influencing the MC calculated responses. For example, the density of
the polyethylene used in the FLUKA MC simulations is a factor that can introduce an
uncertainty in the calculations. It has already been stressed [45] that the polyethylene
density is not the same for all spheres and for all the filler pieces used to match the
CHAPTER 3. Calibration of the BSS at reference neutron fields
67
3He shape to the central hole in each sphere. It has been reported by other researchers
[87] that small uncertainties in the determination of the density can result in large
discrepancies in the MC calculations.
Another important parameter is the 3He detector’s geometry that was used in
the MC simulations. The geometry used in the calculations was not the precise one.
This was verified in 2003 when the technical drawings of the 3He proportional counter
were confidentially communicated by Centronics Ltd [98].
CHAPTER 3. Calibration of the BSS at reference neutron fields
68
Table 3.18. Experimental absolute mean fluence responses from the calibration experiments in PTB (2001 & 2002).
Counts per unit neutron fluence [cm2] Sphere
0.144 MeV 0.565 MeV 1.2 MeV 2.5 MeV 5 MeV 8 MeV 14.8 MeV 19 MeV
81 1.199±0.033 0.735±0.033 0.421±0.008 0.264±0.020 0.119±0.008 0.092±0.005 0.041±0.002 0.028±0.003
81Cd 1.197±0.028 0.741±0.033 0.404±0.008 0.270±0.020 0.113±0.002 0.097±0.005 0.051±0.002 0.049±0.003
108 2.386±0.052 1.965±0.084 1.137±0.022 0.936±0.059 0.476±0.037 0.381±0.016 0.178±0.005 0.171±0.016
133 2.848±0.102 2.797±0.116 2.201±0.030 1.788±0.105 1.008±0.052 0.792±0.032 0.396±0.007 0.414±0.075
178 2.182±0.051 2.899±0.118 2.808±0.039 2.768±0.155 1.808±0.086 1.510±0.059 0.835±0.013 0.777±0.080
233 0.984±0.030 1.819±0.104 2.263±0.031 2.789±0.152 2.226±0.113 1.914±0.104 1.241±0.016 1.066±0.139
Stanlio 0.951±0.020 0.661±0.029 0.369±0.008 0.253±0.018 0.159±0.005 0.159±0.007 0.179±0.004 0.207±0.020
Ollio 0.192±0.006 0.591±0.024 0.909±0.012 1.237±0.067 1.085±0.039 0.974±0.038 0.816±0.010 0.786±0.071
Table 3.19. Calculated absolute fluence responses of the BSS at monoenergetic neutrons for the calibration energies.
Counts per unit neutron fluence [cm2] Sphere 0.144 MeV 0.565 MeV 1.2 MeV 2.5 MeV 5 MeV 8 MeV 14.8 MeV 19 MeV
81 1.306±0.149 0.886±0.089 0.524±0.094 0.302±0.030 0.144±0.029 0.083±0.008 0.043±0.008 0.036±0.004
81Cd 1.260±0.104 0.860±0.086 0.515±0.069 0.281±0.028 0.148±0.023 0.089±0.009 0.053±0.008 0.050±0.005
108 2.588±0.200 2.157±0.216 1.539±0.115 1.036±0.104 0.544±0.042 0.352±0.035 0.182±0.014 0.154±0.015
133 3.005±0.199 3.017±0.302 2.519±0.149 1.859±0.186 1.114±0.064 0.737±0.074 0.403±0.022 0.339±0.034
178 2.359±0.181 3.013±0.301 3.220±0.233 2.800±0.280 2.042±0.131 1.489±0.149 0.858±0.053 0.734±0.073
233 1.109±0.120 1.877±0.188 2.617±0.222 2.890±0.289 2.533±0.186 2.006±0.201 1.291±0.087 1.139±0.114
Stanlio 1.065±0.145 0.715±0.072 0.442±0.076 0.258±0.026 0.185±0.038 0.146±0.015 0.199±0.037 0.209±0.021
Ollio 0.214±0.014 0.605±0.061 1.085±0.065 1.301±0.130 1.263±0.076 1.069±0.107 0.896±0.057 0.878±0.088
Table 3.20. Ratios of calculated over measured (experimental) absolute fluence responses for all energies.
Rcalc/Rexp Sphere
0.144 MeV 0.565 MeV 1.2 MeV 2.5 MeV 5 MeV 8 MeV 14.8 MeV 19 MeV
81 1.089±0.108 1.205±0.132 1.245±0.145 1.144±0.144 1.210±0.175 0.902±0.101 1.049±0.183 1.286±0.192
81Cd 1.053±0.082 1.161±0.127 1.275±0.106 1.041±0.129 1.310±0.119 0.918±0.102 1.039±0.150 1.020±0.119
108 1.085±0.074 1.098±0.119 1.354±0.057 1.107±0.131 1.143±0.096 0.924±0.100 1.022±0.080 0.901±0.123
133 1.055±0.071 1.079±0.117 1.144±0.053 1.040±0.121 1.105±0.070 0.931±0.100 1.018±0.056 0.819±0.170
178 1.081±0.074 1.039±0.112 1.147±0.064 1.012±0.116 1.129±0.071 0.986±0.106 1.028±0.062 0.945±0.136
233 1.127±0.100 1.032±0.119 1.156±0.074 1.036±0.118 1.138±0.078 1.048±0.119 1.040±0.066 1.068±0.175
Stanlio 1.120±0.123 1.082±0.118 1.198±0.145 1.020±0.126 1.164±0.179 0.918±0.100 1.112±0.168 1.010±0.141
Ollio 1.115±0.065 1.024±0.110 1.194±0.051 1.052±0.120 1.164±0.060 1.098±0.118 1.098±0.059 1.117±0.151
CHAPTER 3. Calibration of the BSS at reference neutron fields
69
Table 3.21. Weighted mean of ratios of calculated over experimental responses for all spheres.
Sphere <Rcalc /Rexp>
81 1.106±0.048 81cd 1.097±0.039 108 1.136±0.031 133 1.063±0.027 178 1.070±0.029 233 1.090±0.033
Ollio 1.132±0.026 Stanlio 1.056±0.046
Fig. 3.16. The weighted average ratios (Calculated response / Experimental response) for all energies and all spheres.
The calibration factor of the 2- atm 3He proportional counter is calculated as
the mean weighted factor for all spheres, as derived by the data of Table 3.21, and it
has the value:
The uncertainty 0.012 is the weighted average of all uncertainties shown in Table
3.20. The value of the χ2 indicates that the errors were not overestimated.
fc= 1.096 ± 0.012 with χ2=0.811
CHAPTER 3. Calibration of the BSS at reference neutron fields
70
Fig. 3.17. MC calculated and experimental response functions for the spheres 81 mm and 81Cd.
CHAPTER 3. Calibration of the BSS at reference neutron fields
71
Fig. 3.18. MC calculated and experimental response functions for the spheres 108 mm and 133 mm.
CHAPTER 3. Calibration of the BSS at reference neutron fields
72
Fig. 3.19. MC calculated and experimental response functions for 178 mm and 233 mm.
CHAPTER 3. Calibration of the BSS at reference neutron fields
73
Fig. 3.20. MC calculated and experimental response functions for Stanlio and Ollio.
74
CHAPTER 4
The response of the extended-range Bonner Sphere Spectrometer to charged hadrons
4.1 Introduction
A BSS can be used to measure neutron spectra both outside accelerator
shielding and from an unshielded target. In the former case the contamination of the
neutron field with other types of hadrons is usually comparatively small. On the
contrary, when measuring the neutron emission from an unshielded target bombarded
by a high-energy hadron beam, a large contribution of hadrons other than neutrons
may be present. These secondary hadrons may interact with the moderator and
generate neutrons, which are in turn detected by the 3He proportional counter. This
study investigates the importance of this effect when using an extended-range BSS in
a mixed high-energy radiation field. It also provides a response matrix to charged
hadrons along with an experimental verification.
4.2 BSS measurements at high-energy mixed fields
Measurements with the extended range BSS were performed at SPS (Super
Proton Synchrotron) at CERN of the neutron yield and spectral fluence from 50 mm
thick copper, silver and lead targets bombarded by a hadron beam (composed of about
80% positive pions and 20% protons) of 40 GeV/c. The neutron emission per incident
hadron on target was measured at 60 cm from the target in the angular range from 30º
to 135º with respect to the beam direction [99]. This experiment is described in
chapter 5.
CHAPTER 4. The response of the BSS to charged hadrons
75
The unfolding of the data obtained with the set of eight Bonner spheres was
performed with a code based on the GRAVEL formalism [54], using as pre-
information the neutron energy distribution calculated with the FLUKA Monte Carlo
(MC) code [61-63]. For the MC simulations a pencil beam of 40 GeV/c protons/pions
impinging on a 50 mm thick Ag, Cu and Pb target and the spectral fluence of the
generated secondary neutrons was scored in void spheres with diameter of 133 mm
(corresponding to that of the medium- size Bonner sphere). The irradiation locations
for the simulations were the same as in the experiment, i.e. at angles of 30°, 45°, 60°,
75°, 90°, 105°, 120°, 135°, with respect to the direction of the incoming proton/pion
beam. The full results of the experiment are given in chapter 5. In the present chapter
only the response of the BSS to charged hadrons is discussed.
As an example, the neutron spectral fluences at 30° resulting from unfolding of the
experimental data and from the simulations are shown in Fig. 4.1 for a silver target.
The experimental spectrum shows the two peaks predicted by the MC guess spectrum;
an isotropic evaporation component centred at 3 MeV and a high-energy peak situated
around 100- 150 MeV. However, the unfolded spectrum shows a much more
pronounced high-energy peak and a comparatively smaller evaporation peak with
respect to the distribution predicted by the MC simulations. Precise background
corrections were done but the repeated unfolding showed that the discrepancy
remained large.
CHAPTER 4. The response of the BSS to charged hadrons
76
10-6 1x10-5 1x10-4 10-3 10-2 10-1 100 101
0.0
1.0x10-5
2.0x10-5
3.0x10-5
4.0x10-5
5.0x10-5
6.0x10-5
7.0x10-5
8.0x10-5
9.0x10-5
E*M
(E) p
er p
rimar
y pa
rticl
e (c
m-2)
Guess Spectrum Experimental (uncorrected) Experimental (corrected)
Neutron energy (GeV)
Fig. 4.1. Neutron spectral fluences from 40 GeV/c positive hadrons (80%
pions + 20% protons) on a 50 mm thick silver target, at 30° with respect to beam
direction, resulting from the MC simulations and from the uncorrected and corrected
experimental data unfolding (see text).
This discrepancy was eventually attributed to the contribution of secondary
neutrons produced in the BSS detectors by charged hadrons coming from the target.
This contribution cannot be suppressed by background subtraction. The charged
hadrons (mainly charged pions and protons) undergo inelastic interactions in the
moderator of the spheres, especially in the lead shell that is a part of Stanlio’s and
Ollio’s configuration. They produce secondary neutrons which are subsequently
slowed-down and thermalized by the polyethylene. In this way the 3He detector shows
an enhanced count- rate without of course being possible to discriminate the unwanted
counts due to charged hadrons. The determination of the response functions of the
BSS to charged hadrons is discussed in this chapter. In order to verify this hypothesis,
two studies involving FLUKA Monte Carlo simulations were performed.
CHAPTER 4. The response of the BSS to charged hadrons
77
4.2.a Calculation of spectral fluences of the hadrons (first Monte Carlo study)
The main task of the first study was to calculate the spectral fluence of all
secondary hadrons. These calculations were performed using the same geometry as
for the estimation of the energy distribution of neutrons emitted from the target.
As an example the spectral fluences of neutrons and charged hadrons at 30°
are shown in Fig. 4.2. To save computing time, the calculations were only made for a
primary pencil beam of 40 GeV/c positive pions that was the dominant component of
the primary beam. It was found that the most intense components of the spectral
fluence at all angles are protons and charged pions. Positive and negative pions have
very similar energy distributions, with a peak at almost 1 GeV. Protons show a two-
peak distribution centred at about 200- 300 MeV and 500- 600 MeV.
The ratios of each of the response to charged hadrons and neutrons over the
response to the total (all charged hadrons + neutrons) are listed in Table 4.1 for the
silver target. Similar tables for the copper and the lead target are given in Table C13
in the Appendix C. The ratios of the neutron response to the total represent the
correction factors that have to be applied to the experimental data in order to suppress
the charged hadron contribution.
As shown in Table 4.1 the response of Stanlio and Ollio to charged hadrons is
quite similar and contributes for more than 50% of the 3He detector counts at 30°. The
correction for the contribution of charged hadrons was also applied at large angles
because as shown from Table 4.1, these particles are responsible for more than 6% of
the counts even at 135°.
CHAPTER 4. The response of the BSS to charged hadrons
78
Fig. 4.2. Spectral fluences of neutrons and charged hadrons at 30° from a 40 GeV/c beam of positive pions impinging on a 50 mm thick Ag target.
The correction factors of Table 4.1 (section 4.2) were applied to the
experimental data and a new unfolding was performed with the corrected data. The
corrected neutron spectral fluence at 30° is shown in Fig. 4.1. The agreement with the
simulated spectrum is now satisfactory.
The procedure adopted for estimating these correction factors is very time-consuming.
It consisted of ninety- six sets of simulations (8 angles × 4 particles × 3 number of
spheres) per each target material, with each set consisting of five individual runs, and
it is obviously specific to the present experimental conditions. On the other hand, if
the response matrix of the BSS to charged hadrons is known, the correction factors
can be determined directly by folding the response functions with the spectral fluence
of charged hadrons coming from the target. Ideally one should be able to determine
the spectral fluence or at least the yield of neutrons and charged hadrons from the
experimental data, the response matrices of the BSS to the various particles and some
pre- information on the relative importance of the various contributions. The
calculation of the response matrix of the BSS to charged hadrons is discussed in the
next section.
CHAPTER 4. The response of the BSS to charged hadrons
79
Table 4.1. Ratios of the response of three Bonner spheres to p, n, π+ and π- to the total, for a 50 mm thick silver target (first MC study).
Angle Particle 233 mm Stanlio Ollio
neutron 0.902±0.002 0.439±0.001 0.456±0.001
π+ (2.572±0.011)x10-2 0.204±0.001 0.194±0.001
π- (5.688±0.019)x10-2 0.262±0.001 0.263±0.001
proton (1.537±0.004)x10-2 (9.427±0.005)x10-2 (8.700±0.014)x10-2
30°
π++π-+ proton (9.797±0.031)x10-2 0.561±0.001 0.544±0.001
neutron 0.951±0.001 0.638±0.001 0.653±0.001
π+ (1.023±0.005)x10-2 0.117±0.001 0.109±0.001
π- (3.085±0.006)x10-2 0.182±0.001 0.183±0.001
proton (7.879±0.063)x10-3 (6.255±0.005)x10-2 (5.410±0.010)x10-2
45°
π++π-+ proton (4.897±0.011)x10-2 0.362±0.001 0.347±0.001
neutron 0.970±0.001 0.766±0.001 0.778±0.001
π+ (5.049±0.017)x10-3 (6.570±0.006)x10-2 (6.042±0.011)x10-2
π- (1.994±0.003)x10-2 0.124±0.001 0.125±0.002
proton (4.909±0.023)x10-3 (4.449±0.004)x10-2 (3.674±0.010)x10-2
60°
π++π-+ proton (2.991±0.005)x10-2 0.234±0.001 0.222±0.001
neutron 0.979±0.002 0.840±0.001 0.848±0.002
π+ (2.807±0.014)x10-3 (3.919±0.003)x10-2 (3.560±0.006)x10-2
π- (1.438±0.003)x10-2 (8.829±0.007)x10-2 (8.999±0.017)x10-2
proton (3.442±0.017)x10-3 (3.262±0.004)x10-2 (2.630±0.007)x10-2
75°
π++π-+ proton (2.061±0.004)x10-2 0.160±0.001 0.152±0.002
neutron 0.985±0.002 0.885±0.001 0.890±0.001
π+ (1.766±0.008)x10-3 (2.609±0.002)x10-2 (2.359±0.003)x10-2
π- (1.114±0.002)x10-2 (6.771±0.008)x10-2 (6.941±0.008)x10-2
proton (2.214±0.011)x10-3 (2.188±0.002)x10-2 (1.731±0.004)x10-2
90°
π++π-+ proton (1.512±0.003)x10-2 0.115±0.001 0.110±0.001
neutron 0.988±0.002 0.910±0.001 0.914±0.002
π+ (1.275±0.007)x10-3 (1.927±0.002)x10-2 (1.743±0.003)x10-2
π- (9.589±0.018)x10-3 (5.640±0.004)x10-2 (5.820±0.010)x10-2
proton (1.396±0.008)x10-3 (1.377±0.002)x10-2 (1.042±0.004)x10-2
105°
π++π-+ proton (1.226±0.002)x10-2 (8.944±0.009)x10-2 (8.605±0.014)x10-2
neutron 0.990±0.002 0.927±0.001 0.929±0.002
π+ (9.328±0.041)x10-4 (1.480±0.001)x10-2 (1.314±0.003)x10-2
π- (8.492±0.018)x10-3 (4.955±0.004)x10-2 (5.126±0.007)x10-2
proton (8.972±0.062)x10-4 (8.697±0.001)x10-3 (6.469±0.018)x10-3
120°
π++π-+ proton (1.032±0.002)x10-2 (7.305±0.006)x10-2 (7.086±0.009)x10-2
neutron 0.991±0.002 0.936±0.001 0.938±0.002
π+ (7.991±0.038)x10-4 (1.287±0.001)x10-2 (1.139±0.002)x10-2
π- (7.591±0.015)x10-3 (4.468±0.004)x10-2 (4.643±0.007)x10-2
proton (6.055±0.043)x10-4 (6.094±0.008)x10-3 (4.448±0.018)x10-3
135°
π++π-+ proton (8.996±0.019)x10-3 (6.364±0.006)x10-2 (6.227±0.009)x10-2
CHAPTER 4. The response of the BSS to charged hadrons
80
4.2.b Response functions of the BSS to charged hadrons (second Monte Carlo study)
A second set of simulations was performed to estimate the contribution to the 3He count-rate of the secondary neutrons generated by the charged hadrons in the
moderator. Stanlio and Ollio were first investigated because they include a shell of
lead in their design. The two spheres were assumed to be irradiated by parallel
monoenergetic beams of neutrons, protons and charged pions, with energies between
50 MeV and 150 GeV (see Tables in the Appendix C). For each energy, five
independent simulations were performed in order to achieve good statistics. The
statistical error in all cases did not exceed 5%. The results are shown in Figs. 4.3-4.5.
The response functions to charged hadrons were calculated with the FLUKA
code for the complete BSS, for monoenergetic broad parallel beams of protons,
positive and negative pions with energy in the range 50 MeV– 150 GeV. For each
particle the neutron spectral fluence was scored in the 3He detector and subsequently
folded with the 3He(n,p)3H cross sections [100] to evaluate the response (see section
4.3). The same calculations were repeated for the 233 mm sphere. This sphere showed
a definitely smaller response to charged hadrons (about 10% at 30° and below 1% at
135°), confirming that this effect is far less important for polyethylene. Following this
result it was decided not to investigate the response to charged hadrons of the other
polyethylene spheres of smaller diameter, since their response would be negligible.
The ratios of the response to charged hadrons and neutrons to the total (all charged
hadrons + neutrons) are listed in Table 4.1 (section 4.2.a) for the silver target.
For the various detectors, the response functions to protons and positive pions
is quite similar, with a rapid increase up to about 100 MeV followed by a flat trend.
This is due to the fact that above this energy the inelastic cross sections tend to the
geometrical value. As expected, Ollio and Stanlio have a much higher response with
respect to the polyethylene spheres. Negative pions show for most spheres an
increased response below about 100 MeV. This can be because some π− are slowed
down to rest inside the sphere, they are captured and produce spallation reactions.
This effect is expected to be larger for larger spheres. As it can also be noted from the
graphs, the addition of a 1 mm cadmium shell to the 81 mm sphere increases its
response to charged hadrons by a factor of 5 to 10.
CHAPTER 4. The response of the BSS to charged hadrons
81
102 103 104 105
10-3
10-2
10-1
100
101
81 81cd 108 133 178 233 stanlio ollio
Res
pons
e [c
m2 ]
Energy [MeV]
Fig. 4.3. Fluence response of the extended BSS to protons.
These response functions were folded with the spectral fluences of neutrons
and charged hadrons calculated for the 40 GeV/c positive pion beam on a silver target
(Section 4.2.a), to verify the consistency with the former procedure used for
estimating the percent contribution of neutrons and charged hadrons to the total 3He
count-rate. For this purpose the response functions of the BSS to neutrons, previously
calculated up to 2 GeV had to be extended to 40 GeV. The results of such a
comparison are given in a next section.
CHAPTER 4. The response of the BSS to charged hadrons
82
102 103 104 105
10-3
10-2
10-1
100
101
81 81cd 108 133 178 233 stanlio ollio
Res
pons
e [c
m2 ]
Energy [MeV]
Fig. 4.4. Fluence response of the extended BSS to positive pions.
102 103 104 10510-3
10-2
10-1
100
101
81 81cd 108 133 178 233 stanlio ollio
Res
pons
e [c
m2 ]
Energy [MeV]
Fig. 4.5 Fluence response of the extended BSS to negative pions.
CHAPTER 4. The response of the BSS to charged hadrons
83
4.3 Comparison of the Monte Carlo studies and estimation of correction factor
Two methods were used to determine the correction factor to be applied to the
experimental results. They can be summarised as follows:
• Simulation of parallel beam irradiation by neutrons, protons, π+ and π− with
source spectra obtained by scoring yields at various angles in a previous
simulation of a target yield experiment. The correction factor for each angular
interval is obtained as the ratio of the neutron response to the sum of the four
responses to neutrons, protons, π+ and π− (Table 4.1, section 4.2.a).
• Calculation of energy response functions for parallel monoenergetic beams of
protons, π+ and π− and when not already done, neutrons (see section 4.2.b).
Folding these response functions with the spectra obtained at various angles in
the same previous target yield simulation as before, one obtains the
contribution of each component. The correction factor for each angular interval
is obtained as the ratio of the neutron contribution to the sum of the four
contributions, neutrons, protons, π+ and π- (Table 4.2).
The ratios of the results obtained with the two methods are listed in Table 4.2.
The agreement is satisfactory for the 233 mm sphere and for Ollio. For Stanlio the
agreement is satisfactory for neutrons, whilst data for charged hadrons show
discrepancies up to a factor of 2. Although one is essentially interested in the
correction factors for neutrons, this discrepancy is not completely negligible. One
should also note that there are systematic trends in the discrepancy also for the other
two detectors (233 and Ollio), which cannot be explained as random fluctuations but
may rather point to a systematic error in one of the two procedures, possibly under-
sampling of particles in the tail of the spectra. It should also be mentioned that the two
procedures are based on the same calculated spectra, and therefore they are not
completely independent of each other.
CHAPTER 4. The response of the BSS to charged hadrons
84
Table 4.2. Ratios of the percent contribution by neutrons, π+, π- and p to the total 3He count-rate determined by the full MC simulations of the experimental set-up (section 4.2.a ) to that determined by folding the detector response function with the particle spectra (section 4.2.b) .
Angle Particle 233 mm Stanlio Ollio neutron 0.974 ± 0.061 0.713 ± 0.080 1.026 ± 0.058
π+ 1.342 ± 0.046 1.483 ± 0.043 0.982 ± 0.060 π- 1.319 ± 0.047 1.446 ± 0.043 0.987 ± 0.060
30°
proton 1.320 ± 0.047 1.440 ± 0.044 0.950 ± 0.062 neutron 0.988 ± 0.060 0.815 ± 0.071 1.020 ± 0.058
π+ 1.360 ± 0.046 1.699 ± 0.038 0.974 ± 0.060 π- 1.281 ± 0.048 1.659 ± 0.039 0.970 ± 0.061
45°
proton 1.358 ± 0.046 1.620 ± 0.040 0.916 ± 0.064 neutron 0.993 ± 0.060 0.881 ± 0.066 1.019 ± 0.058
π+ 1.390 ± 0.045 1.833 ± 0.036 0.966 ± 0.061 π- 1.252 ± 0.049 1.788 ± 0.037 0.944 ± 0.062
60°
proton 1.324 ± 0.047 1.728 ± 0.038 0.891 ± 0.065 neutron 0.995 ± 0.059 0.920 ± 0.064 1.017 ± 0.058
π+ 1.385 ± 0.045 1.897 ± 0.035 0.946 ± 0.062 π- 1.214 ± 0.050 1.848 ± 0.036 0.917 ± 0.064
75°
proton 1.350 ± 0.046 1.784 ± 0.037 0.871 ± 0.067 neutron 0.997 ± 0.059 0.943 ± 0.062 1.015 ± 0.058
π+ 1.375 ± 0.045 1.941 ± 0.034 0.939 ± 0.062 π- 1.184 ± 0.051 1.888 ± 0.035 0.893 ± 0.065
90°
proton 1.346 ± 0.046 1.809 ± 0.036 0.854 ± 0.068 neutron 0.998 ± 0.059 0.955 ± 0.062 1.014 ± 0.058
π+ 1.385 ± 0.045 1.959 ± 0.034 0.932 ± 0.063 π- 1.161 ± 0.052 1.900 ± 0.035 0.868 ± 0.067
105°
proton 1.364 ± 0.046 1.824 ± 0.036 0.824 ± 0.070 neutron 0.999 ± 0.059 0.964 ± 0.061 1.012 ± 0.059
π+ 1.350 ± 0.046 1.969 ± 0.034 0.911 ± 0.064 π- 1.153 ± 0.052 1.908 ± 0.035 0.855 ± 0.068
120°
proton 1.379 ± 0.045 1.817 ± 0.036 0.812 ± 0.071 neutron 0.999 ± 0.059 0.968 ± 0.061 1.011 ± 0.059
π+ 1.356 ± 0.046 1.977 ± 0.034 0.909 ± 0.064 π- 1.141 ± 0.053 1.915 ± 0.035 0.853 ± 0.068
135°
proton 1.348 ± 0.046 1.819 ± 0.036 0.801 ± 0.072
CHAPTER 4. The response of the BSS to charged hadrons
85
4.4 Experimental test with 120 GeV/c hadrons at CERF
A real ‘calibration’ of the BSS response to charged hadrons would be ideal for
the verification of the response matrix to charged hadrons but it would require the
availability of broad beams of protons and charged pions of several defined energies
up to tens of GeV. Since this was not feasible, an experimental verification of the
response of the BSS to a monoenergetic beam of high-energy hadrons was performed
at the CERF facility at CERN [78].
Each detector of the BSS was exposed to a 120 GeV/c positive hadron beam.
The beam is composed of 1/3 protons and 2/3 pions and it shows a Gaussian profile
with FWHM of 30.5 mm and 31.7 mm in the horizontal and vertical planes,
respectively (measured with a multi-wire proportional chamber). The beam impinged
on each sphere at 25 mm from its centre because the irradiation of the spheres at their
centre provoked a very high count-rate that could damage the detector or result in
pile- up effects. MC simulations with the FLUKA code were performed reproducing
the exact experimental conditions.
The experimental data (counts per beam particle) are compared with the MC
results in Table 4.3. The agreement is rather good except for the two smaller spheres
where the experimental value is almost twice than the MC value. This may be due to
the fact that, given the dimension of the sphere and the beam size, the sphere was not
fully intercepting the beam. Even a small variation in the beam position and/or
dimension would cause a significant variation in the count-rate.
The good agreement between the MC predictions and the experimental data
give confidence in the Monte Carlo calculations of the BSS response functions.
CHAPTER 4. The response of the BSS to charged hadrons
86
Table 4.3. Counts per beam particle of the BSS detectors exposed to a narrow beam of 120 GeV/c protons/positive pions. The beam impinged on each sphere at 25 mm from its centre. The MC results are the weighted sum of the response to protons (1/3) and pions (2/3).
Sphere Monte Carlo result Experimental result
81 (4.95±0.04) x10-5 (7.81±0.78) x10-5
108 (2.24±0.02) x10-4 (3.41±0.34) x10-4
133 (4.30±0.04) x10-4 (3.88±0.39) x10-4
178 (8.97±0.10) x10-4 (1.16±0.12) x10-3
233 (1.31±0.01) x10-3 (1.69±0.17) x10-3
Stanlio (2.59±0.03) x10-2 (2.72±0.27) x10-2
Ollio (4.47±0.01) x10-2 (3.60±0.36) x10-2
4.5 Conclusions
The studies presented in this chapter show that the extended range BSS can be
used to measure around unshielded targets. An important condition that has to be met
is that the response functions of the lead-enriched detectors must be corrected for the
contribution due to charged hadrons. This fact has been verified by neutron spectral
measurements performed at CERN in a high-energy radiation field. Stanlio and Ollio
were found to have a significant response to the charged hadron component
accompanying the neutrons emitted from the target. The other detectors of the
extended-range BSS, that are composed only of polyethylene, showed a similar
behavior although of lesser importance.
The complete response matrix of the extended BSS to charged pions and
protons that was calculated with FLUKA, can be generally used whenever the BSS is
used to measure at fields where a large contribution of charged hadrons is present.
Some discrepancies for Stanlio have to be thoroughly studied, although they are not
associated with the corrections that need to be made for its response to hadrons.
The Monte Carlo calculated response functions were experimentally verified,
so this agreement can additionally be considered as another successful benchmark test
of the FLUKA code itself.
87
CHAPTER 5
Neutron spectral measurements with a Bonner sphere spectrometer
5.1 Introduction
This chapter presents experimental measurements of the neutron yield and energy
distribution from comparatively thin copper, silver and lead targets bombarded by high-
energy hadrons. The experiments performed at CERN were intended to provide source
term data for neutron production from high-energy hadrons, necessary for shielding
calculations and other radiation protection purposes.
Measurements of the spectral fluence and of the ambient dose equivalent of
secondary neutrons produced by 250 GeV/c protons and 158 GeV/c per nucleon lead ions
were performed around a thick beam dump a few years ago [101]. The results showed
that the spectral fluence of the secondary neutrons outside a thick shield is similar for
light (protons and pions) and heavy (lead) hadrons of comparable energy per nucleon
stopped in a massive target. It was also shown that the approach of considering a high-
energy lead ion as a group of independent protons is sufficiently accurate for the purpose
of evaluating the ambient dose equivalent of secondary neutrons outside a thick shield.
While in a massive dump the development of the hadronic cascade results in a
neutron yield Y which varies with energy E as 8.0EY ∝ , with a thin target the high-energy
cross-sections vary much slower and are roughly equal to their geometrical values [102].
Measurements of the neutron emission from comparatively thin copper and lead targets
bombarded by beams of high-energy protons/pions and lead ions have been discussed in
[103,104]. A comparison between a Monte Carlo simulation for protons and the
experimental results for lead ions has shown that the neutron yield for lead ions can be
reasonably predicted by scaling the result of a Monte Carlo calculation for protons by the
projectile mass number of the ion to the power of 0.85 to 0.95 for a lead target and 0.88 to
1.03 for a copper target.
CHAPTER 5. Neutron spectral measurements with the BSS
88
Following the above results, further measurements were conducted to extend these
data to different energies and target materials over a wider angular range. The present
chapter discusses measurements that were performed with the extended range Bonner
Sphere Spectrometer (BSS) and the experimental apparatus used also in [68].
5.2 Experiment
The neutron yield and spectral fluence from 50 mm thick copper, silver and lead
targets bombarded by a mixed beam of protons and pions (about 75% pions and 25%
protons) of 40 GeV/c momentum was measured at CERN in one of the secondary beam
lines of the Super Proton Synchrotron (SPS). The target thickness is about 1/3 of
interaction length, as the interaction length of 40 GeV/c protons and pions is 15 cm in Cu,
14.6 cm in Ag and 17 cm in Pb [105]. The neutron emission per incident particle on target
was measured in the angular range from 30º to 135º with respect to the beam direction.
The experimental data were compared with the results of Monte Carlo simulations
performed with the FLUKA code [61-63]. The simulations were also used as a priori
information for the unfolding of the experimental data.
The measurements were performed in the experimental area used by the NA57
[106,107] and ALICE [108] collaborations. The experimental apparatus used for the latest
measurements with lead ions was also used in this experiment. The equipment is provided
with remotely- controlled multiple target-holder and detector support, so that the
measurements could be conducted in a purely parasitic mode during the normal running
of NA57. Access to the experimental area was thus only needed to change the Bonner
detector every 32 measurements (8 angles x 4 target positions). The system was installed
in the same position as in the previous experiments, i.e. a few metres downstream of the
NA57 apparatus (see Fig. 5.1), on a concrete platform normally employed by the ALICE
collaboration. The zone is laterally and top shielded by 80 cm concrete.
CHAPTER 5. Neutron spectral measurements with the BSS
89
Fig. 5.1. The NA-57 set –up and the exit pillars of the GOLIATH magnet.
The experimental device (Fig. 5.2) consists of a supporting frame (A) over which
a fan-shaped platform (B) slides horizontally, perpendicularly to the beam axis, to
position the target holder in and out of the beam. An arm, with the detector support at its
end, moves on the platform swinging around the target holder. The target to detector
distance is 600 mm, which permits a good angular resolution of the measurement yet
keeping the device compact. Four targets with diameter of up to 25 mm can be mounted
on the holder (Fig. 5.3), but one position is left empty for background measurements. The
holder moves vertically to position the selected target in the beam. The target holder has a
vertical excursion of 126 mm and a positioning accuracy of better than 0.1 mm. Its
position is determined by a rotative transducer fitted to a worm screw driven by a
stepping motor. A CsI scintillator coupled to a photodiode is mounted at the centre of the
holder. This scintillator is used as a reference to correctly align the targets in the beam.
The 150 mm horizontal excursion of the platform B is driven by a stepping motor coupled
to a worm screw, with a resolution of better than 0.1 mm. The speed can be adjusted
between 0.1 and 4 mm s-1.
CHAPTER 5. Neutron spectral measurements with the BSS
90
Fig. 5.2. The experimental device used for the neutron spectrometry measurements, allowing the semiautomatic positioning of target and detector.
The detector arm moves over an angular range of 120º in 15º steps, from 15º to
135º with respect to the beam direction. The arm is driven by a stepping motor fitted
directly to a wheel (Fig. 5.3); the positioning accuracy is 0.05º.
The entire system is remotely controlled; the correct positioning of the detector
arm at the various angular locations is sensed by micro-switches and indicated on the
control panel by leds. The stepping motors for the three independent movements
(horizontal of the platform, vertical of the target holder and angular of the detector arm)
are driven by four-phase unipolar control boards of Eurocard format, operating at low
voltage and installed in an Europac standard rack close to the device. The transducer
signals are processed by boards installed in the same rack and sent to the control panel.
The actual positions of the platform, of the detector arm and of the target holder are
displayed on the control panel, from which the three movements are also controlled. The
correct operation of the equipment is additionally surveyed by a video camera.
CHAPTER 5. Neutron spectral measurements with the BSS
91
Fig. 5.3. Left: detail of the driving mechanism of the swinging arm for detector positioning. Right: sketch of the multiple target holder.
The intensity of the primary beam was monitored by an air- filled Precision
Ionization Chamber (PIC) at atmospheric pressure, placed in the beam just upstream of
the target, connected to a current digitising circuit. One PIC-count corresponds to
2.45x104 ± 675 particles impinging on the target [109]. The extended-range neutron rem
counter LINUS [90-93] was placed in a fixed position on the measurement platform and
used as a reference monitor to verify the stability of the experimental conditions.
The spectrometer uses a Centronics 3He proportional counter connected via a
preamplifier to a MAB [47] acquisition system (amplifier, high voltage, multi channel
analyser housed in a single box). More details for the electronics are given in chapter 1 as
well as in [45].
For each target, measurements were performed in the angular range from 30º to
135º in steps of 15º. The count rate was sufficiently high, hence the statistical uncertainty
on the single measurements was always below a few percent. Measurements without
target were performed with each detector for background subtraction. The background is
generated by interaction of the beam with upstream components (although the overall
thickness of the material present in the beam path is small) and by radiation scattered
from the local shielding and the dump (situated approximately 20 m downstream of the
CHAPTER 5. Neutron spectral measurements with the BSS
92
target). Interactions with air are negligible, as the interaction length for protons in air is
about 750 m (and about 2/3 of this value for pions [104]) and the air path upstream of the
target is about 20 m.
5.3 Monte Carlo simulations and spectrum unfolding
The Monte Carlo simulations were performed with the FLUKA code [61-63]. The
spectral fluence of secondary neutrons generated by 40 GeV/c protons striking 50 mm
thick copper, silver and lead targets was scored in void spheres with diameter 133 mm
(corresponding to that of one of the Bonner spheres used for the measurements) placed at
the same positions used in the experiment (i.e., at angles of 30°, 45°, 60°, 75°, 90°, 105°,
120°, 135°).
Separate simulations were performed considering only the target and reproducing
strictly the geometry of the concrete structure which shields locally the experimental set-
up. This allowed a part of the contribution of the scattered neutrons to be estimated. A
detailed treatment of the diffused component would have requested the simulation of all
the interactions of the primary beam upstream of the target, the contribution of other
experiments running in parallel and of the beam dump placed about 20 m downstream of
the target. The complexity of the geometry of some structures upstream of the target
together with their large distance from the scoring positions are difficult to simulate and
make almost impossible to converge to sufficiently accurate results without using
variance reduction techniques. Some structures (such as the beam dump) were considered
in a set of simulations done in the past [103], showing that, as expected the local concrete
shield is responsible for most of the scattered neutrons contributing to the spectral fluence
at the scoring positions.
The calculated spectra were used as pre-information for unfolding the
experimental data. The statistical uncertainty of the simulations is well below 1%.
As in previous experiments [103,104], the unfolding of the data obtained with the
set of eight Bonner spheres was performed with a code based on the GRAVEL formalism
[54]. The code employs the MATLAB™ capabilities for the I/O of data and results and
for the graphical presentation of other relevant quantities (e.g. chi-square, total fluence,
dose, etc.) during the iterations.
CHAPTER 5. Neutron spectral measurements with the BSS
93
The sensitivity of the unfolded spectra against the statistical fluctuations of the
experimental data was checked by randomly varying the experimental counts of the
Bonner spheres according to a normal distribution with a standard deviation of 3%.
5.4 Data analysis
The uncertainty on the experimental data is in the range 6- 9%. However, two
important corrections had to be applied to the experimental data: a normalization factor
required by the fact that the beam size was larger than the target cross-section, and a
correction factor on the counts of Ollio and Stanlio to account for the neutron production
in the lead shell of these two spheres by the charged hadrons coming from the target.
5.4.1 Beam normalization factor
Since the beam was larger than the neutron-producing target, the fraction of the
beam hitting the target, the surrounding material or not interacting, had to be estimated. In
order to correctly assess these values, a program was written [110] which calculates the
beam fractions for given beam profiles and offsets. The beam was considered as Gaussian
shaped with σhorizontal = 9.3 mm and σvertical = 7.8 mm. The cylindrical target of 10 mm
radius was mounted inside a plastic ring with a lateral thickness of 2 mm. This ring was in
turn embedded in a second ring made of aluminum, which had a lateral thickness of
3 mm.
As mentioned above, the beam was centred on target with the help of a
scintillator. Assuming that the beam centre can be aligned with the target axis with an
uncertainty of 2 mm, the beam distribution on the target and target holder is given in
Table 5.1. The maximum neutron contribution coming from the target holder can be
estimated by taking into account the nuclear interaction lengths of the various materials
constituting the targets and holder, the fractions of the beam hitting the various parts and
the thickness of the plastic and aluminum rings (3.9 mm) as compared to the target
(50 mm). For all targets, the neutron production in the holder is less than 1.5% of the
CHAPTER 5. Neutron spectral measurements with the BSS
94
neutron production in the target and can therefore be neglected. The experimental results
were thus multiplied by a factor of 1/0.487 as deduced from Table 5.1.
It must be noted that the uncertainly in the calculation of the σhorizontal and σvertical is
of the order of 10 – 20%. A measurement of the beam size 20 m upstream of the
experimental position confirmed the computed values within 10%. In addition, the beam
distribution is not exactly Gaussian. By decreasing both σhorizontal and σvertical by 10% to
20%, the fraction of the beam hitting the target increases approximately in the same
proportion. This uncertainty in the correction to be applied may explain the systematic
error found in the comparison of the experimental results with the Monte Carlo
predictions (see section 5.5).
Table 5.1. Fraction of beam hitting the target, the components of the target holder or missing the system, assuming a precision of 2 mm for aligning the beam on-target [courtesy of Helmut Vincke, CERN].
5.4.2 BSS response to charged hadrons
The second important correction to be applied became manifest while performing
the unfolding of the experimental data. The experimental spectra showed the two peaks
predicted by the Monte Carlo guess spectrum, an isotropic evaporation component
centred at 3 MeV and a high-energy peak sitting around 100-150 MeV (Fig. 5.4).
However, the unfolded spectrum showed a much more pronounced high-energy peak and
a comparatively smaller evaporation peak with respect to the distribution predicted by the
MC simulations.
Component Beam fraction (%)
Target 48.68 ± 0.91
Plastic ring 12.94 ± 0.06
Aluminum ring 15.86 ± 0.07
Lost particles 22.52 ± 0.91
CHAPTER 5. Neutron spectral measurements with the BSS
95
Fig. 5.4. Neutron spectral fluences from 40 GeV/c p/π+ on a 50 mm thick silver target, at 30° with respect to beam direction (from MC simulations and uncorrected experimental data unfolding).
This discrepancy can be explained by the contribution of secondary neutrons
produced in the BSS detectors by charged hadrons coming from the target, a contribution
which cannot be suppressed by background subtraction. These charged hadrons (mainly
charged pions and protons) undergo inelastic interactions in the moderator (mainly in the
lead shell of Stanlio and Ollio), producing secondary neutrons which are subsequently
slowed-down and thermalized by the polyethylene giving rise to additional counts in the 3He detector. The importance of this effect when using an extended-range BSS in a mixed
high-energy radiation field is thoroughly discussed in chapter 4 (see also ref. [111]).
A set of FLUKA simulations was performed to determine the correction factor.
First the spectral fluence of all secondary hadrons was calculated using the same
geometry that was implemented for estimating the energy distribution of neutrons and
charged hadrons emitted from the target. A second set of simulations was then performed
to estimate the contribution to the 3He count-rate of the secondary neutrons generated by
these charged hadrons in the moderator. The reader can find a table with the correction
factors for the response of the BSS to charged hadrons and a full description of this part
of the analysis in chapter 4.
CHAPTER 5. Neutron spectral measurements with the BSS
96
5.5 Results and discussion
The neutron spectral fluences for the three targets are shown in Fig. 5.5- 5.7
at angles of 30°, 60°, 90° and 120°. The spectra show the two peaks predicted by the
Monte Carlo a priori information: an isotropic evaporation component at 3 MeV and a
high-energy peak around 100- 150 MeV. The valley between the evaporation and the
high-energy components is due to a minimum in the hadron-nucleon cross sections at
intermediate energies. It should be noted that the relative importance of the high-energy
peak over the evaporation peak is larger for the copper target. The same effect was
observed with a lead ion beam [104].
Fig. 5.5. Neutron spectral fluence [EΦ(E)] per primary hadron from 40 GeV/c p/π+ on a 50 mm thick copper target, at emission angles of 30°, 60°, 90° and 120°.
CHAPTER 5. Neutron spectral measurements with the BSS
97
Fig. 5.6. Neutron spectral fluence [EΦ(E)] per primary hadron from 40 GeV/c p/π+ on a 50 mm thick silver target, at emission angles of 30°, 60°, 90° and 120°.
Fig. 5.7. Neutron spectral fluence [EΦ(E)] per primary hadron from 40 GeV/c p/π+ on a 50 mm thick lead target, at emission angles of 30°, 60°, 90° and 120°.
CHAPTER 5. Neutron spectral measurements with the BSS
98
The neutron yields resulting from the unfolding of the experimental data are listed
in Tables 5.2 – 5.4. The results are given in four energy bins and as total yield, in the
angular range 30° – 135°. The total yield is compared with the results of the Monte Carlo
simulations. The simulations included the shielding around the target, so that
experimental and simulation data can be directly compared. The ratio between the
experimental results and the calculations is given in the last column of Tables 5.2 – 5.4.
Table 5.2. Neutron yield (per energy group and total yield) for 40 GeV/c p/ π+ on a 50 mm thick copper target. Data are neutrons per primary particle on target.
Angle
(o)
<100
keV
0.1-20
MeV
20-500
MeV
0.5-2
GeV
Total Yield
(experimental)
Total Yield
(Monte Carlo)
Exp./Calc.
Yield
30 0.056 0.42 0.26 4.7E-02 0.78 0.57 1.37
45 0.054 0.39 0.29 1.5E-02 0.74 0.50 1.48
60 0.041 0.37 0.27 3.4E-03 0.68 0.46 1.48
75 0.041 0.37 0.19 7.5E-04 0.60 0.41 1.46
90 0.038 0.35 0.18 3.9E-04 0.57 0.37 1.54
105 0.050 0.34 0.12 1.5E-04 0.51 0.34 1.50
120 0.048 0.33 0.12 1.4E-04 0.50 0.31 1.61
135 0.046 0.30 0.15 3.1E-04 0.50 0.29 1.72
Table 5.3. Neutron yield (per energy group and total yield) for 40 GeV/c p/ π+ on a 50 mm thick silver target. Data are neutrons per primary particle on target.
Angle
(o)
<100
keV
0.1-20
MeV
20-500
MeV
0.5-2
GeV
Total Yield
(experimental)
Total Yield
(Monte Carlo)
Exp./Calc.
Yield
30 0.091 0.67 0.41 7.4E-02 1.24 0.87 1.43
45 0.089 0.70 0.29 1.2E-02 1.09 0.81 1.35
60 0.078 0.69 0.27 2.3E-03 1.04 0.77 1.35
75 0.083 0.67 0.25 6.5E-04 1.0 0.72 1.39
90 0.086 0.64 0.19 2.4E-04 0.92 0.67 1.37
105 0.082 0.64 0.14 9.6E-05 0.87 0.62 1.40
120 0.080 0.62 0.16 1.5E-04 0.86 0.58 1.48
135 0.085 0.58 0.19 1.9E-04 0.85 0.53 1.60
CHAPTER 5. Neutron spectral measurements with the BSS
99
Table 5.4. Neutron yield (per energy group and total yield) for 40 GeV/c p/ π+ on a 50 mm thick lead target. Data are neutrons per primary particle on target.
Angle
(o)
<100
keV
0.1-20
MeV
20-500
MeV
0.5-2
GeV
Total Yield
(experimental)
Total Yield
(Monte Carlo)
Exp./Calc.
Yield
30 0.14 1.29 0.50 8.7E-02 2.01 1.31 1.53
45 0.15 1.27 0.44 1.6E-02 1.87 1.25 1.50
60 0.14 1.26 0.39 2.8E-03 1.79 1.21 1.48
75 0.13 1.24 0.33 6.2E-04 1.70 1.16 1.47
90 0.14 1.20 0.30 2.2E-04 1.64 1.10 1.49
105 0.13 1.18 0.24 1.1E-04 1.56 1.04 1.50
120 0.13 1.14 0.26 1.4E-04 1.54 0.98 1.57
135 0.13 1.10 0.27 2.0E-04 1.50 0.92 1.63
From Tables 5.2- 5.4 one can observe a systematic discrepancy in the range 30% to
40% between experimental and simulation data. The discrepancy may be possibly due to
an uncertainty in the beam normalization factor as discussed in section 5.4.1. The ratios of
the experimental to the MC calculated yields are plotted in Fig. 5.8 for all targets.
Fig. 5.8. Ratio of experimental to simulation neutron yield data for the three targets in the angular range 30° – 135°.
CHAPTER 5. Neutron spectral measurements with the BSS
100
For the copper target (Table 5.2 and Fig. 5.5) one notices that the high-energy peak (the
20 – 500 MeV bin) is larger at 60° than at 30°, contrary to what one would expect. A
similar trend is observed for all targets at angles of 105° to 135° (not shown in Figs. 5.5 –
5.7). On the other hand, for all targets the total yields regularly decrease with increasing
emission angle as expected (Tables 5.2 – 5.4). This unexpected behavior can be explained
by the influence of the statistical fluctuations of the experimental data on the unfolding
process.
The sensitivity of the unfolded spectra against the statistical fluctuations of the data
was checked by randomly varying the experimental counts of the Bonner spheres
according to a normal distribution with a standard deviation of 3%, as done in a previous
work [104]. It was found that the neutron fluences in large energy groups vary from 3%
up to 12% for each group. These variations may be considered as a lower limit of the total
uncertainties, which can be estimated as lower than 20% in each of these energy groups,
if all other sources of errors are taken into account. The total fluences are obviously more
stable, as shown in Tables 5.2 – 5.4. The variation of the yield per energy group versus
angle is shown in Fig. 5.9- 5.11.
Fig. 5.9. Neutron yield per energy group as a function of emission angle for 40 GeV/c p/ π+ on a 50 mm thick copper target.
CHAPTER 5. Neutron spectral measurements with the BSS
101
Fig. 5.10. Neutron yield per energy group as a function of emission angle for 40 GeV/c p/ π+ on a 50 mm thick silver target.
Fig. 5.11. Neutron yield per energy group as a function of emission angle for 40 GeV/c p/ π+ on a 50 mm thick lead target.
CHAPTER 5. Neutron spectral measurements with the BSS
102
The variation of the integral yield versus angle is shown in Fig. 5.12 for all targets.
Fig. 5.12. Neutron yield as a function of emission angle for 40 GeV/c p/ π+ on 50 mm thick copper, silver and lead targets. The lines are fits to the data according to the law Y =Y0 + α·exp(-θ/t) (expression (5.1)).
The contribution of the high-energy component at the forward angles is obvious
and results in an increase of the yield although it is not as pronounced as in the case of
heavy ions [104]. The neutron yield Y versus angle can be fitted by the expression:
Y = Y0 + α·exp (– θ/t) (5.1)
in which θ is the emission angle. The parameters of the fits Y0, α and t are given in
Table 5.5 for the three targets.
Table 5.5 Parameters of the fit (expression 5.1) for the total neutron yield as a function of emission angle for the three targets (Fig. 5.12).
Target Fit parameter
Cu Ag Pb Y0 0.35 ± 0.05 0.78 ± 0.05 1.36 ± 0.12 α 0.69 ± 0.02 0.79 ± 0.07 1.00 ± 0.06 t 77.11 ± 12.02 53.11 ± 12.79 69.26 ± 24.39
CHAPTER 5. Neutron spectral measurements with the BSS
103
The variation of the integral neutron yield Y versus target mass number A is shown in Fig. 5.13.
Fig. 5.13. Neutron yield as a function of target mass number for 40 GeV/c protons/pions in the angular range 30° – 135°. The lines are fits to the data according to the law Y = a·Ab (expression (5.2)).
Fig. 5.13 shows that the yields increase by a factor in the range 2.5 – 3 from copper
to lead. This trend is explained by the partial development of the hadronic cascade in the
targets, both in the radial and in the longitudinal direction. The data at the eight angles
approximately follow the same trend and can be fitted by the expression:
Y = a·Ab (5.2)
The parameters of the fits are given in Table 5.6. They take into account a maximum 10%
uncertainty in the experimental data. The exponential b varies from 0.77 to 0.92, which is
not too far from the value one would theoretically expect from the expression of the
geometrical cross-section of a proton striking a target with mass number Atarget :
3/2arg0arg tettp Arπσ =→ (5.3)
where r0 = 1.3- 1.5x10-15 m.
CHAPTER 5. Neutron spectral measurements with the BSS
104
Table 5.6. Parameters of the fit (expression 5.2) for the total neutron yield as a function of target mass number in the angular range 30o – 135o (Fig. 5.13).
Emission angle (o)
30 45 60 75 90 105 120 135
a 0.033±0.006 0.028±0.003 0.023±0.001 0.018±0.004 0.015±0.001 0.012±0.002 0.011±0.002 0.013±0.002
b 0.769±0.037 0.789±0.021 0.816±0.005 0.848±0.045 0.876±0.014 0.915±0.031 0.924±0.043 0.897±0.039
5. 6. Conclusions
The neutron yields and spectral fluences from 50 mm thick copper, silver and lead
targets bombarded by a mixed beam of protons and pions with momentum of 40 GeV/c
were measured in the angular range from 30º to 135º with respect to the beam direction.
These experiments are intended to provide source term data for neutron production from
high-energy heavy ions, necessary for shielding calculations and other radiation
protection purposes. Two peaks can be observed in the neutron spectral fluence,
according to the different steps of the intranuclear cascade model. The high- energy peak
at around 100- 150 MeV is due to direct hadron-nucleon reactions and pre- equilibrium
emission, while the peak at about 3 MeV comes from evaporation neutrons. It was found
that the relative importance of the high- energy peak over the evaporation peak is larger
for the copper target.
The experimental results are in good agreement with predictions obtained with the
FLUKA Monte Carlo code. The 30% to 40% discrepancy between experiment and
simulations can most likely be imputed to the uncertainty in the beam normalization
factor. It can be stated that the values of total yields given in Tables 5.2 – 5.4 lie in
between the experimental and the Monte Carlo data.The results complement previous
measurements of the neutron emission from comparatively thin targets bombarded by
beams of high-energy protons/pions and lead ions [103,104], as well as measurements of
neutron production from a massive dump [101].
105
CHAPTER 6
Upgrade and development of the CERN-EU Reference Field (CERF)
PART A: The beam monitoring at CERF
6.1 Monitoring system
6.1.1 Introduction
Two new instruments were installed in the beam line in the CERF irradiation
cave before the start of the June 2002 run at CERF. The first is a Multi-Wire
Proportional Chamber (MWPC) for accurate measurements of the profile of the
120 GeV/c hadron beam impinging on the copper target (Fig. 6.1). The second is a
5- litre ionization chamber of similar design as the reference CERF beam monitor (the
PIC), put in place as a back-up instrument to the PIC (Fig. 6.2). This chamber was
first submitted to extensive performance tests with 137Cs sources in a calibration
laboratory at CERN and later tested and inter-compared with the PIC in the CERF
hadron beam.
CHAPTER 6. Upgrade and development of CERF
106
Fig. 6.1. The wire chamber installed in the CERF cave.
Fig. 6.2. The BIG PIC (foreground) and the PIC (background) monitors in the CERF cave.
CHAPTER 6. Upgrade and development of CERF
107
6.1.2 Measurements of beam profile with a Multi-Wire Proportional Chamber
A more reliable control of the beam position and profile had always been a
general claim by the CERF users. A solution was identified in a MWPC, which is one
of the standard monitors used to measure the beam profile in the SPS secondary beam
lines.
The MWPC was installed between the two CERF target positions,
approximately one metre upstream of the one below the concrete roof-shield. Its effect
on the beam is negligible, so it was decided that the MWPC be mounted on a static
support and stay in the beam during the whole CERF run period. To measure the
beam profile, the CERF copper target must obviously be removed from the upstream
support (the one under the iron roof-shield). It was verified (by tests during the beam
set-up) that backscattering from the target installed under the concrete roof-shield
does not affect the profile measurement, so in this case one can measure the beam
profile on-line while taking data. An example of the horizontal and vertical beam
profiles measured with the MWPC is shown in Fig. 6.3. Changing collimators C3 and
C5 from ± 11 mm to ± 14 mm modifies the rms width of the beam in the two
transverse dimensions from 10.4 mm (horizontally) and 9.1 mm (vertically) to
11.5 mm (horizontally) and 9 mm (vertically).
Fig. 6.3. Horizontal and vertical beam profiles measured with the MWPC installed one meter upstream of the copper target under the concrete roof-shield. The beam intensity was 3x107 particles per SPS pulse.
CHAPTER 6. Upgrade and development of CERF
108
6.2 Performance tests of the BIG PIC
The Precision Ionization Chamber (PIC) [80], the primary beam monitor at
CERF, had until now no back-up instrument. Since all measurements at CERF are
normalized to unit beam particle incident on the copper target, the importance of the
PIC is apparent. Therefore, a second beam monitor was tested, installed in the facility
and inter-compared to the PIC, in order to have it characterized and ready for use in
case of a PIC failure in any future run.
This second device was built at CERN about 25 years ago, at the same time as
the standard PIC. It is a 4.9-litre effective volume open- air ionization chamber of the
same design as the PIC (i.e., of cylindrical shape), with identical external cross-
sectional area. Knowing the effective volume and the diameter of the chamber one
can calculate its effective length, which is 99.8 mm. This is 3.1 times the effective
length of the PIC, which is 32 mm [80]. For this reason this device is called here the
‘BIG PIC’. From this result it was expected that the sensitivity of the BIG PIC is
approximately three times the sensitivity of the PIC. This was later verified by the
experimental results discussed in section 6.3. The technical characteristics of the BIG
PIC are summarised in Table D1 in the Appendix D.
The device is connected to a high-voltage power supply and digitizer housed
in a small metal box. In this box, a small electronic device manufactured by
HAMAMATSU and referenced C4960-1 provides the high-voltage supply with a very
good stability and linearity. The voltage can be adjusted from 0 Volts to –1250 Volts
(in the following, all voltage values mentioned are negative polarity). Under normal
conditions the chamber is operated at 600 Volts. Two S-HVS connectors are used for
the power supply of the chamber and for the control of the output voltage. A digitizer,
with a sensitivity of 1 pC/digit, converts the input current to an output frequency.
There are 2 connectors LEMO 00, one for connecting the chamber and the other for
checking and recording the output frequency.
The BIG PIC was submitted to a number of performance tests similar to those
made in the past on the PIC [112,113]. Measurements were performed in the SC/RP
calibration laboratory to verify the region of ion saturation, the linearity of the
response versus intensity of the radiation field, the stability and the leakage current of
the chamber. The BIG PIC was then tested in the hadron beam at the CERF facility
CHAPTER 6. Upgrade and development of CERF
109
during the CERF run, and its response was compared to that of the standard PIC.
Measurements of the instrument stability were repeated in the calibration laboratory
after the CERF runs.
All measurements in the SC/RP calibration laboratory were performed at the
nominal voltage of 600 V, except for the determination of ion saturation for which the
voltage was varied between 10 V and 1000 V. Air pressure and ambient temperature
were sufficiently stable throughout the measurements, hence no corrections had to be
applied. There were some variations in the humidity, but this parameter has only a
minor influence on the operation of the chamber [114].
6.2.1 Stability tests
To investigate the warming up time of the electronics, i.e. the time needed to
reach stable operating conditions, measurements were performed for seven hours after
switching on the power. Measurements were performed with 137Cs sources of
different activities providing air kerma rates varying between 10 µGy/h and
30 mGy/h. For the air kerma rates of 10 µGy/h up to 300 µGy/h the measuring time
was 1000 s for each measurement. The measurements with air kerma rates between
500 µGy/h and 3 mGy/h lasted 300 s each, while measurements at the highest air
kerma rates lasted 100 s each. The sources used, the values of air kerma rates and the
duration of the measurements are summarised in Table D2 in the Appendix D. The air
pressure, temperature and humidity in the calibration laboratory were recorded before
each measurement. The pressure and the temperature were fairly stable,
P = (968.7 ± 1.0) hPa and T = (19.8 ± 1.0) 0C, respectively. The humidity varied in
the range 43.3% to 73%, which translates in a variation of 2% on the reading of the
chamber [114].
This first stability test indicated that after several hours the chamber might still
not have reached stable operating conditions. The stability test was therefore repeated
over a longer period (72 hours). For this series of measurements only one source was
used, providing an air kerma rate of 3 mGy/h. The measuring time for the first four
measurements was 1000, 900, 300 and 360 s, respectively, and 300 s for all the other
measurements. The air pressure and temperature in the calibration laboratory were
CHAPTER 6. Upgrade and development of CERF
110
recorded before each measurement and were found again stable, P = (960.3 ± 1.0) hPa
and T = (19.4 ± 1.0) 0C, respectively. The humidity varied in the range 53.4% to
72.3%. The results of the different series of measurements are plotted together in
Fig. 6.4. The slight difference between the values of air pressure in the two periods
corresponds to a variation in the sensitivity of the chamber of approximately 1%.
Since this is a minor correction, it was not accounted for. The results of the second
series of measurements alone are plotted in Fig. 6.5. The results of the two series of
measurements indicate that the BIG PIC needs almost ten hours to stabilize. For
comparison, the standard PIC needs about two hours to reach conditions of operation
[112].
Fig. 6.4. Stability test of the BIG PIC: sensitivity of the chamber as a function of time after switching it on. The error bars include the statistical uncertainties of the measurement, the 2% variation with humidity and the 3% uncertainty on the source output.
CHAPTER 6. Upgrade and development of CERF
111
Fig. 6.5. Stability test of the BIG PIC: sensitivity of the chamber as a function of time after switching it on. The error bars include the statistical uncertainties of the measurement, the 2% variation with humidity and the 3% uncertainty on the source output. The line is only to guide the eye.
6.2.2 Linearity test
To test the linearity of the response of the BIG PIC, the chamber was
irradiated with 137Cs sources providing air kerma rates in the range 10 µGy/h to
30 mGy/h. Table D2 in the Appendix D lists the relevant parameters of the
irradiations (source identification, air kerma rates, number and duration of
measurements). The results, shown in Fig. 6.6, indicate a good linearity of the
response in the entire interval investigated. The parameters of the linear fit are given
on the plot. The chi-square (Χ2) divided by the degrees of freedom N-m (where N is
the number of data points and m is the number of parameters estimated by the fit) is
the reduced chi-square. The values of the reduced chi-square corresponding to the
probability Px(x2; N-m) of exceeding x2 versus the number of degrees of freedom are
tabulated in [115].
CHAPTER 6. Upgrade and development of CERF
112
Fig. 6.6. Linearity test of the BIG PIC: measured count-rate (counts per second) versus air kerma rate. The line is a linear fit Y =A +B X to the experimental data, with Y = log10·y and X = log10·x.
6.2.3 Region of ion saturation
The region of ion saturation (i.e., the region in which there is no ion
recombination in the chamber), which determines the operating region of the
instrument, was evaluated with a series of measurements made with a 137Cs source.
The chamber was exposed to an air kerma rate of 30 mGy/h at a distance of 1.56 m
from the source. The voltage applied to the chamber was varied from 10 V to 1000 V
in steps of 50 V (except for the first two measurements, where the step was 20 V). For
each voltage setting three measurements were made of duration 20 to 60 s. The results
are listed in Table D3 in the Appendix D and are plotted in Fig. 6.7. The figure shows
that the nominal operating voltage of 600 V lies well within the region of ion
saturation for photon air kerma rates of up to 30 mGy/h, which is the maximum one
available in the calibration laboratory.
CHAPTER 6. Upgrade and development of CERF
113
Fig. 6.7. Count-rate versus applied voltage, showing the region of ion saturation of the BIG PIC. The line is only to guide the eye.
6.2.4 Investigation of leakage current existence
Another test made on the BIG PIC was the investigation of leakage current.
A leakage in the capacity of the digitizer would modify the response of the chamber to
a pulsed radiation field like at CERF, where the beam comes in pulses lasting a few
seconds and spaced by about 13 seconds. To investigate this effect, the data from the
linearity test were divided by the air kerma rate and plotted as a function of kerma rate
in Fig. 6.8. The data and the fit seem to indicate the absence of leakage current in the
interval of air kerma rate investigated, in agreement with the results of the linearity
test shown in Fig. 6.6. However, in the region of very low air kerma rates (below
10 µGy/h) no conclusions can actually be drawn. To verify the absence of a small
leakage current, which would cause a deviation from linearity, additional
measurements at low kerma rates (< 10 µGy/h) would be needed, to simulate the
behavior of the chamber for very low intensities of the CERF hadron beam.
CHAPTER 6. Upgrade and development of CERF
114
Fig. 6.8. Sensitivity versus air kerma rate of the BIG PIC. The line is a fit y = A, to the experimental data.
Long-term measurements also showed that the chamber did not give any
spurious counting in the course of the test. This means that the electronics of the
chamber is not affected by any noise.
6.3 Inter-comparison of PIC and BIG PIC monitors at CERF
The BIG PIC was tested at CERF during the June and July 2002 runs. The
chamber was installed in the beam about one metre downstream of the standard PIC.
Preliminary tests were carried out in June and measurements were performed in July,
to inter-compare the response of the two beam monitors and to verify the presence of
any recombination effect while operating the BIG PIC in a hadron beam.
In order to inter-compare the response of the BIG PIC to that of the PIC, the
readings of the two instruments were recorded over a number of SPS pulses and for
several beam intensities. The results are given in Table D4 in the Appendix D
and are plotted in Fig. 6.9. Since the distance traversed by the particles in the active
CHAPTER 6. Upgrade and development of CERF
115
volume of the BIG PIC is 3.1 times the distance traversed in the standard PIC (see
section 6.2), one expects that the ratio of the readings of the two instruments is about
3, within an uncertainty that is ± 5% for the PIC [109] and can reasonably be assumed
of the same order for the BIG PIC.
From Table D4 (Appendix D) and Fig. 6.9 one sees that the ratio between the
readings of the BIG PIC and the PIC ranges from 3.09 at low beam intensities (200
PIC-counts per pulse) down to 2.95 for very high intensities. These variations are well
within the experimental uncertainties and in agreement with the above factor.
Nonetheless, the ratio seems to show a slightly decreasing trend with increasing
intensity of the hadron beam starting at about 6000 PIC/ pulse (Fig. 6.9).
Another inter-comparison of the two PIC monitors was performed in August
2003. For another verification of their ratio, new measurements were performed at
CERF with intensities varying from about 70 PIC-counts/pulse to 4,000 PIC/ pulse.
The raw data are shown in Table D5 in Appendix D. From this table one can see that
the ratio between the readings of the BIG PIC and the PIC ranges from 2.98 at low
beam intensity (75 PIC-counts/pulse) up to 3.02 at ~500 PIC-counts/pulse. Then it
slightly decreases again for higher beam intensities. These variations are well within
the experimental uncertainties and in agreement with the factor found in the previous
year.
Fig. 6.9. Ratio of the readings BIG PIC and PIC as a function of beam intensity. The lines are fits Y = A and Y = B + C·x to the experimental data.
CHAPTER 6. Upgrade and development of CERF
116
6.3.1 Recombination effects for the BIG PIC
It has previously been demonstrated that the standard PIC is not subjected to
charge recombination when exposed to the CERF beam (120 GeV/c positive hadrons)
and that its response remains constant with increasing beam intensity [113]. The
above results may therefore suggest a small recombination effect at high intensity in
the BIG PIC. To verify this hypothesis, a series of measurements were performed to
obtain saturation curves of the chamber (i.e, voltage characteristics curves) for
different beam intensities.
The beam intensity was varied by adjusting collimators C3 and C5; all the
other collimators were left at their nominal CERF settings (i.e., collimators C1 and C2
set at ± 20, and C8 to C11 fully open). The beam intensity was varied between 200
and 14000 PIC-counts (of the standard PIC monitor) per pulse. These intensities
correspond to three times higher count- rates of the BIG PIC. Beam intensities and
fluctuations are summarised in Table D6 in the Appendix D.
For each setting of the collimators, data were taken by varying the voltage
applied to the BIG PIC from 10 V to 1000 V. The reading of the chamber was
corrected for slight beam intensity fluctuations by recording the number of primary
protons impinging on the T4 production target according to the expression:
collsetTVT
pulseppulsep
pulsecountsPICBIG //
/4
4⋅
− (6.1)
For each voltage setting, five readings of the BIG PIC per pulse and the
corresponding proton intensities on the T4 target were recorded. Expression (6.1)
represents the average of the count-rates (BIG PIC-counts/pulse) normalized to the
proton rate on T4 (pT4/pulse) for one voltage setting, multiplied by the mean proton
rate for each collimator setting.
The above quantity is plotted in Fig. 6.10 as a function of the voltage applied
to the BIG PIC, for various beam intensities. The uncertainties associated to the data
points range from 0.14% to 2.13% and are too small to appear in the plot. The graph
seems to confirm the existence of a slight ion recombination effect in the chamber for
beam intensities above about 6000 counts/pulse.
CHAPTER 6. Upgrade and development of CERF
117
0 200 400 600 800 1000
0.0
5.0x103
1.0x104
1.5x104
2.0x104
2.5x104
3.0x104
3.5x104
4.0x104
4.5x104 14000 counts/pulse 12000 counts/pulse 10000 counts/pulse 8000 counts/pulse 6000 counts/pulse 4000 counts/pulse 2000 counts/pulse 1000 counts/pulse 450 counts/pulse 200 counts/pulse
BIG
PIC
cou
nts
corr
ecte
d fo
r T4
read
ing
Voltage (V)
Fig. 6.10. Voltage characteristic curves of the BIG PIC for different beam intensities at
CERF. The count-rates are corrected for slight beam intensity fluctuations. The lines are
sigmoidal fits to the data to guide the eye.
6.4 Tests of the Triggers 4, 5, 6 in the H6 beam line
The Trigger4 is one of the scintillators installed in the H6 beam line very close to the
PIC. It is routinely used for the verification of the calibration factor of the PIC and
subsequently of the BIG PIC. More precise calibration methods for the PIC were used
in the years 1998 and 1999 [109,116], but a routine check is done before each CERF
run. The characteristics of Trigger4 are given in [117].
The Trigger5 and Trigger6 are two scintillation counters located 15 m
downstream of the iron beam dump installed at the end of the CERF area. These
counters measure the muons produced by the pion component of the beam, and thus
provide an indirect measurement of the beam intensity. Therefore they can be useful
for a more precise estimation of the PIC response.
CHAPTER 6. Upgrade and development of CERF
118
6.4.1 Efficiency measurements of Trigger4
A routine check of the PIC calibration factor is performed according to the
procedure described briefly in the following. The PIC-counts are read out online via a
LabView program running on a PC. A certain number of cycles are preset and the
PIC-counts are accumulated during the same time interval. The reading of Trigger4 is
received directly by the SPS beam-control program. The beam cycle lasts 16.8 s while
the pulse extraction lasts 4.8 s (August 2003). After selecting a number of beam
cycles, the Trigger4 counts the beam particles and after the last cycle the beam-
control program gives the average number of particles. An estimation of the
calibration factor is obtained from the ratio of the average number of particles to the
average PIC-counts.
First measurements showed a significant deviation from the known calibration
factor, i.e. 31,000 instead of the expected ~23,000 particles per PIC-count. The reason
of such discrepancy was an inappropriate high voltage (HV) of the photomultiplier of
Trigger4 (-1.85 kV). By varying the HV of the photomultiplier and recording the
counts of the scintillator, it was found that a good operating HV value (i.e., in the
middle of the so-called “plateau”) is -1.73 keV (Fig. 6.11).
1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0
0.0
0.5
1.0
1.5
2.0
2.5
Even
ts p
er p
rimar
y pa
rticl
e [x
10-6]
HV of PMT of Trigger 4 [-kV] Fig. 6.11. Characteristic curve of the Trigger4 scintillator. Events on Trigger4, normalized to the protons on T4 production target, as a function of high-voltage (HV) of the photomultiplier (PMT).
CHAPTER 6. Upgrade and development of CERF
119
With the HV set to the new value of -1.73 kV, the counts per pulse of Trigger4
were recorded for different apertures of collimators C3 and C5, and compared to the
counts per pulse of the PIC-counter (Table 6.1).
Table 6.1 Ratio of Trigger4-counts over PIC-counts for different beam intensities.
PIC-counts /SPS
pulse
(mean over 5 cycles)
Trigger4 response
per SPS pulse
(mean over 5 cycles)
Ratio
(Trigger4-counts
/ PIC-counts)
73.6 1,697,103 23,058
165.4 3,742,178 22,625
293.8 6,521,542 22,197
458 9,953,216 21,732
652.8 13,853,153 21,221
1,160.4 23,141,498 19,943
2,132.2 33,055,128 15,503
As noted in [117], Trigger4 is not reliable above 1700 PIC-counts per second
because in this region the limit of the photomultiplier is reached. Results in Table 6.1
show that at intensities around 300 PIC-counts per pulse it is already important to
correct the response of the scintillator for dead time losses.
In order to check if these data are consistent with the usual calibration factor
of 23,000, the Trigger4 response was corrected for dead time losses. When calculating
the effects of dead time, the entire detector system must be taken into account. There
are two models for dead time behavior: paralyzable and nonparalyzable response
[118]. The expressions of the true interaction rate n as a function of the recorded
count rate m and the system dead time τ are:
τnenm −= , paralyzable model (6.2)
and τn
nm+
=1
, non- paralyzable model (6.3)
CHAPTER 6. Upgrade and development of CERF
120
Each element of a detector system usually has its own dead time, which can be
extendable (paralyzable model) or non extendable (non-paralyzable model) [119]. The
two models predict the same first-order losses and significantly differ for high true
event rates (i.e., higher than 1/τ, where τ is the system dead time). Since no
information about the detector components and their dead times was available, data
analysis was done with both models. It was assumed that they represented the two
extremes of the experimental set- up and the real set- up lies in between.
The data presented in Table 6.1 have been normalized to a pulse length of
4.8 s in order to obtain the recorded count rate m for Trigger4, the PIC-counts per
second and their ratio Γ. The dead time of the system and the ratio Γ have been
estimated by fitting the experimental data with expressions (6.2) and (6.3), as shown
in Fig. 6.12.
The mean value of the calibration factors obtained from the two fitting models,
results in the value Γ=23,640. This value is consistent within 3% with the calibration
factor in use of 23,000 particles per PIC-count.
CHAPTER 6. Upgrade and development of CERF
121
0 50 100 150 200 250
0
1x106
2x106
3x106
4x106
5x106 Γ 23607.9 ±38.9τ 29 ns ±0.2
Res
pons
e Tr
igge
r4 (p
artic
les/
s)
Intensity (PIC-counts/s)
Paralyzable model
0 50 100 150 200 250
0
1x106
2x106
3x106
4x106
5x106 Γ 23672.1 ±66.8τ 32 ns ±0.6
Res
pons
e Tr
igge
r4 (p
artic
les/
s)
Intensity (PIC-counts/s)
Nonparalyzable model
Fig. 6.12. Measured count rate by Trigger4 as a function of the true count rate by the PIC (symbols). The solid line is a fit according to the paralyzable model (upper plot) and the nonparalyzable model (lower plot).
6.4.2 Measurements of Trigger5 and Trigger6
The response of the PIC was compared with the reading of the scintillation
counters Trigger5 and Trigger6, installed downstream of the CERF area. The Trigger5
and Trigger6 measure the muons produced by the pion component of the beam that is
roughly proportional to the beam intensity in the H6. The muon flux is about three
orders of magnitude lower than the beam intensity, thus it allows monitoring over a
wider intensity range than Trigger4 [116]. The response of the scintillation counters
versus PIC-counts is shown in Fig. 6.13 for different beam intensities.
CHAPTER 6. Upgrade and development of CERF
122
0 500 1000 1500 2000 2500 3000 3500 40000
20
40
60
80
100
120
140
160
180
200
Trig
ger5
[x1,
000
muo
ns p
er s
pill]
PIC [counts per pulse]
Trigger5 Linear fit: y = 0.04111 x + 6.24887
0 500 1000 1500 2000 2500 3000 3500 40000
20
40
60
80
100
120
140
160
180
200
Trig
ger6
[X1,
000
muo
ns p
er s
pill]
PIC [counts per pulse]
Trigger6 Linear fit: y = 0.0373 x + 4.99083
Fig. 6.13. Response of Trigger5 (upper plot) and Trigger6 (lower plot) versus PIC-counts. The straight line is the fit in the range of linearity (squares).
The experimental points in Fig. 6.13 were fitted by a linear function in the non-
saturated range. The data points above 2,000 PIC-counts (about 105 muons) were
excluded from the fit because at these intensities the Trigger5 and Trigger6 start to
saturate. The calculation of the average of the linear fits shown in Fig. 6.13, results in
the expression:
T5/6 = (0.039205 ±0.0011) · P + 5.61985 ± 1.25. (6.4)
This expression gives the expected number of Trigger5/6-counts (T5/6) as a
function of the PIC-counts (P). In 1998 a similar expression was calculated in order to
CHAPTER 6. Upgrade and development of CERF
123
correct the response of Trigger4 up to 2000 PIC-counts per pulse and may be
compared to the expression that was calculated in July 1998 [116] for correcting the
response of Trigger4 up to 2,000 PIC-counts per pulse:
T5/6 = 0.04078 · P + 4.57536 (6.5)
The equations (6.4) and (6.5) agree well within the experimental uncertainties.
6.5 Conclusions
The beam set- up at CERF has been made much easier by the MWPC installed
in the irradiation cave close to the copper target. The beam profile in the horizontal
and vertical planes can thus be measured on- line.
The BIG PIC seems slightly less performing than the standard CERF beam
monitor (the PIC): it shows a comparatively long warm- up time and a slight
recombination at high beam intensities. However, the present results seem to indicate
that its characteristics and performance are sufficient to establish it as a back-up
instrument to the primary CERF beam monitor, but some corrections for ion
recombination might have to be applied at high beam intensity.
Trigger4 has proved to be a reliable instrument to perform calibration tests for
intensities below 2,000 PIC-counts per pulse, if the appropriate corrections for dead-
time losses are made. Measurements taken during the CERF run are consistent with
the calibration factor in use (23,000 Trigger4-counts/PIC-counts), the correlation
factor between the PIC and BIG PIC (~3) and the expected Trigger5/6-counts per
PIC-counts within the experimental uncertainties.
CHAPTER 6. Upgrade and development of CERF
124
PART B: Monte Carlo studies
6.6 Introduction
Under the term cosmic radiation one usually means the flux of energetic
particles that enter the Earth’s atmosphere from outer space [120]. Extra solar cosmic
rays, usually called Galactic Cosmic Rays (GCR), represent the dominant component
of the cosmic rays flux in the Earth vicinity for energies above ~ 10 MeV. These are
particles accelerated at distant sources, which propagate in the galaxy through
relatively weak magnetic fields and experience electromagnetic and nuclear
interactions with photons and nuclear matter in the interstellar medium. Approaching
the heliosphere they interact with the magnetic fields carried by the solar wind, that
effectively modify their flux up to the energy region of several GeV.
Over the past 10 years, there has been increasing concern about the exposure
of air crews to atmospheric cosmic radiation [121-123]. At aviation altitudes, the
neutron component of the secondary cosmic radiation contributes about half of the
dose equivalent, but until recently it has been difficult to accurately calculate or
measure the cosmic- ray neutron spectrum in the atmosphere to determine accurate
dosimetry [124-127]. Dose rates from atmospheric cosmic radiation at commercial
aviation altitudes are such that crews working on present day jet aircraft are an
occupationally exposed group with a relatively high average effective dose
[123,125,128]. Crews of future high-speed commercial aircraft flying at higher
altitudes would be more exposed [126].
The CERF particle composition and spectral fluences are similar to those in
the cosmic radiation field at commercial flight altitudes between 10 and 20 km
(Fig.6.14). A large fraction of the exposure of aircrew is due to neutrons with a
comparable energy spectrum to that found around high- energy hadron accelerators.
The CERF facility provides a radiation field for the inter-comparison and the
calibration of instrumentation that is used for in- flight measurements. Moreover, the
fast neutron component at CERF is considered to be more similar to that in space.
Therefore, interest has arisen from several International Institutes to study whether a
CHAPTER 6. Upgrade and development of CERF
125
different shielding configuration at CERF could produce a radiation field, rich in
high-energy particles, as the one found inside a space station.
Fig. 6.14. Comparison of the energy spectrum at commercial flight altitude to the one on the concrete roof-shield at CERF [45].
The radiation field in low-Earth orbit consists of three main components:
1. GCR components trapped in the Earth’s magnetic field;
2. Neutrons, as secondary particles created in GCR interactions with spacecraft
materials, which could scatter back from the Earth (albedo neutrons);
3. Radiation coming from Solar Particle Events (SPEs).
The relative intensities of the components and the fluence rates of the different
particle types depend on the orbit altitude and inclination, on solar activity, and on the
occurrence of SPEs. Both inside and outside a spacecraft, the radiation field at a point
in tissue or in free air is affected by the shielding, scattering and generation of
secondary particles by surrounding material including an astronaut’s and/or
colleagues’ bodies. For the altitude range and inclination of a manned space station,
secondary neutrons can be a major contributor to effective dose, up to a few tens of
percent of effective dose rates. The exact proportion is very dependent on the amount
CHAPTER 6. Upgrade and development of CERF
126
of shielding. Most of the neutrons inside the station result from nuclear interactions of
charged particles (mainly galactic protons and protons trapped by the earth's magnetic
field) with the wall material of the vessel.
Preliminary Monte-Carlo simulations were performed [45] with FLUKA in
order to investigate this possibility. In order to keep computing time reasonable,
simulations were based on a simplified spherical geometry (Fig. 6.15) rather than
modeling the complete facility. The aim was to understand whether a given
target/shielding combination and angular scoring region would indicate a promising
situation which could subsequently be further investigated. The preliminary studies
revealed that the real geometry of the facility has to be slightly modified and
implemented in the Monte Carlo calculations.
Fig. 6.15. Simplified geometry for the preliminary CERF studies.
The results of the simulations confirmed that there is no need of changing the
‘standard’ copper target or the present beam momentum of 120 GeV/c at CERF. The
new exposure area will be a high-intensity one, hence an additional back shield will
be needed; the combination of iron backed by concrete may be able to reduce the low-
energy neutron component. It was also shown that a 40 cm thick aluminum slab
CHAPTER 6. Upgrade and development of CERF
127
placed at the forward direction fulfills satisfactorily the requirement for high
intensities, high energy particles and high neutron fluence.
6.7 Space studies for the CERF facility The most recent geometrical set- up of the CERF facility was implemented in
Monte Carlo FLUKA simulations with main focus on the new exposure area in the
forward direction (only the area under the concrete roof was modeled). The exact
shape of the external shielding was not implemented since it does not affect the
radiation field inside the cave. The geometry used for the simulations is shown in Figs
6.16- 6.18. An aluminum slab of 40 cm thickness 75 cm width and 2.3 m height was
added downstream of the Cu target. The slab was placed at the forward direction (30
degrees from the Cu target, with respect to the beam axis) at a distance 3.5 m from the
target. In addition, an iron slab of 40 cm thickness and of 240 cm height was added
behind the aluminum slab, attached to the existing concrete wall as a back shield for
the low energy neutron component that contributes significantly to the increase of the
dose equivalent. The fluence of neutrons, protons, pions, muons, photons and
electrons was scored behind the aluminum slab. The calculated spectra of all particles
are shown in Fig. 6.19.
CHAPTER 6. Upgrade and development of CERF
128
Fig. 6.16. Top view of the modified CERF facility. The position of the Al and the Fe slab are indicated. The
external concrete walls on the left (with respect to the beam axis) and the roof have been excluded for clarity.
Fig. 6.17. Side view of the simulated CERF facility. The external concrete wall on the right and the roof
are not shown.
CHAPTER 6. Upgrade and development of CERF
129
Fig. 6.18. Side view of the simulated CERF facility. The external concrete wall on the left and the roof
are not shown.
Fig. 6.19. Monte Carlo FLUKA calculated energy spectra of different particles at the modified CERF facility.
CHAPTER 6. Upgrade and development of CERF
130
In Fig 6.19 the E·M(E)/dE known as fluence per unit lethargy is plotted versus
log(E). The lethargy representation is often the most appropriate, since spectra, when
plotted with a linear energy scale, very often show a steep negative slope at lower
energies. The proton spectrum in the forward direction (from 0o to 30o) is similar to
the energy spectrum of the galactic cosmic protons inside the MIR space station [129]
(see also Fig 6.20). In order to emphasize the similarity of experimental and
calculated spectra, a polynomial fit was done for the calculated spectra, that leads to
the comparison shown in Fig. 6.21. The uncertainty of the MC calculated data (up to
20%) and the uncertainties of the polynomial fit (10%) have been omitted for reasons
of clarity.
In order to compare computed results with experimental data many parameters
have to be taken into account, such as solar modulation, the inclination of orbit of the
space vessel, the type of the space vessel etc. For the present comparison, data taken
inside the module ‘Spektr’ of the MIR space station [130] during a solar minimum
and a solar maximum were used. The module which was added at the MIR complex
in 1995, contained equipment that was used for atmospheric research and surface
studies.
Fig. 6.20. FLUKA calculated and experimental proton spectra for minimum and maximum solar activity.
CHAPTER 6. Upgrade and development of CERF
131
Fig. 6.21. Comparison of experimental proton spectra for minimum and maximum solar activity with a polynomial fit of the FLUKA calculated spectrum.
The FLUKA calculated proton spectrum at CERF, as shown in Fig. 6.21, is
more similar to the GCR spectrum as measured at MIR under conditions of minimum
solar activity, particularly for the energy range below 1 GeV. For higher energies the
curves converge for both conditions of minimum and maximum solar activity. The
explanation for this phenomenon may be that the low energy portion of the cosmic ray
spectrum is subject to strong variability due to solar modulation, following the 11 year
solar cycle. The modulation effects decrease with increasing particle rigidity 1 and
become negligible for particles with rigidities of a few GV, or approximately 5 GeV
per nucleon. During high solar activity the solar wind is stronger and so are the
magnetic fields transported by it, resulting in a decrease of the cosmic ray flux. The
contrary is true during low solar activity.
It must be stressed that the FLUKA calculated spectra had to be scaled in
order to be compared with the experimental data. The scaling factor corresponds to
1 The penetrating ability of a cosmic ray is quantified by the rigidity that it is expressed as the GCR’s momentum divided by its charge.
CHAPTER 6. Upgrade and development of CERF
132
intensities at the CERF facility in the range of 600- 1100 PIC/ spill. The agreement
can be considered as satisfactory, judging especially from the shape of the two spectra
that extend over the same energy range.
It is also possible to find the measured neutron spectrum in the MIR station
[129] for minimum solar activity and compare it with the calculated neutron spectrum
at CERF. The comparison shown in Fig 6.22 indicates that there is a good agreement
between experimental and calculated spectra. This is even more important if one takes
into account the lack of statistical uncertainties in the experimental data.
Fig. 6.22. Comparison of experimental neutron spectra for minimum solar activity.
CHAPTER 6. Upgrade and development of CERF
133
6.8 Conclusions
The modified geometry of CERF can provide a radiation field whose proton
and neutron components are very similar to the ones found inside a space station. For
the other radiation components no conclusions can be drawn since it was not possible
to find experimental data in the same energy range. The hadron beam at CERF
consists of protons, pions and kaons and it is known that the behavior of the pions is
similar to the protons’ at high energies. For this reason it can be concluded that the
CERF facility can be widely used for tests also for the pion component.
Aside from protons, neutrons and pions, the radiation environment inside a
space station is also composed of heavy ions which primarily originate from the GCR
and secondly produced by interactions of the GCR with the walls of the space vessel
or with other materials. At the CERF facility heavy ions are not present.
It can be concluded that the optional new exposure area at the CERF facility
may be very useful for the tests and inter-comparison of detectors used in space
stations or for other measurements in the framework of the space program.
134
CHAPTER 7
Conclusions
The present dissertation focused on the upgrade and development of the
extended- range Bonner Sphere Spectrometer (BSS). The BSS has gone under an
extensive calibration campaign that started in 2001 and was completed in 2003. It has
always been highlighted by many researchers that significant errors may be
introduced in spectroscopy performed by Bonner Spheres, by using incorrect response
functions. It is therefore vital to obtain the most appropriate response matrix. The only
way to accomplish this task is to validate the Monte Carlo calculated response
functions against calibration measurements.
The BSS has also been successfully used in measurements of neutron yields
and spectral fluences from unshielded targets. In addition, it has been calibrated at the
CERN- EU Reference Field facility (CERF), in the context of the upgrade of its
response to charged hadrons. CERF is a special irradiation facility that simulates the
radiation environment at commercial flight altitudes. The BSS has already been tested
in that field, several times in the past. A continued need for tests at CERF has been
expressed by many scientists worldwide. Hence, the improvement of the facility is
essential for its future existence and use. In consideration of this fact, extensive
performance tests were applied to an ionization chamber in order to certify its
capacity to serve as back- up to the reference monitor used at CERF. Moreover, the
potential development of the BSS towards its deployment for fluence measurements at
cosmic fields has arisen the need for Monte Carlo studies for the modification of the
facility.
The following conclusions can be drawn.
CHAPTER 7. Conclusions
135
Studies with the Extended– Range Bonner Sphere Spectrometer
1. The fluence response functions obtained from the analysis of the calibration
of the BSS at PTB (Germany) are in a very good agreement (within 20%)
with the Monte Carlo calculated fluence response functions. The calibration at
UCL (Belgium) verified that the Bonner spheres respond correctly to the
energy spectra (33 MeV and 60 MeV), but due to the electronics’ failure, the
correct experimental responses could not be derived.
2. The ratio of the calculated fluence response over the measured one is defined
as the calibration factor for the specific 2- atm 3He proportional counter and
its value is : fc= 1.096 ± 0.012.
3. The BSS can be used to measure around unshielded targets but the response
functions of the lead-enriched detectors must be corrected for the contribution
due to charged hadrons. This fact has been verified by neutron spectral
measurements performed at CERN in a high- energy radiation field. The other
detectors of the extended-range BSS that are composed only of polyethylene,
showed a similar behavior but of lesser importance.
4. The complete response matrix of the extended BSS to charged pions and
protons was calculated with FLUKA. It can be applied whenever the BSS is
used to measure at fields where a large contribution of charged hadrons is
present.
5. The neutron yields and spectral fluences from 50 mm thick copper, silver and
lead targets bombarded by a mixed beam of protons and pions with
momentum of 40 GeV/c were measured. Two peaks can be observed in the
neutron spectral fluence, according to the different steps of the intranuclear
cascade model. A high-energy peak at around 100 – 150 MeV, that is due to
direct hadron-nucleon reactions and pre-equilibrium emission, and a peak at
about 3 MeV that comes from evaporation neutrons. The experimental results
are in good agreement with predictions obtained with the FLUKA Monte
Carlo code.
CHAPTER 7. Conclusions
136
The upgrade and development of the CERF facility
1. The beam set-up at CERF has been made much easier by the Multi Wire
Proportional Counter (MWPC) that is installed in the irradiation cave close to
the copper target. The beam profile in the horizontal and vertical planes can
thus be measured on-line. The ionization chamber (BIG PIC) reached a
sufficient level of performance so as to be established as a back-up instrument
to the primary CERF beam monitor. However, some corrections for ion
recombination might have to be applied at high beam intensity.
2. Trigger4 has proved to be reliable for performing calibration tests at
intensities below 2,000 PIC-counts per pulse, with the appropriate corrections
for dead-time losses.
3. The modified geometry of CERF can provide a radiation field whose proton
and neutron components are very similar to the ones found inside a space
station. Therefore, the facility can be very useful for future tests and inter-
comparison of detectors used in space stations.
Outlook
In the framework of this thesis it was shown that there have been significant
improvements to the extended BSS, in particular to the knowledge of its
characteristics and performance capabilities. The BSS is a reliable system that has
been successfully used in mixed radiation fields. This fact becomes more important if
one takes into account that one of the major difficulties in neutron spectrometry for
protection purposes is the very wide range of energies which need to be covered.
Hence, the ability of the BSS to operate reliably can be employed in future
measurements at experimental areas or for the control of the radiation monitors
around the Large Hadron Collider (LHC).
The Monte Carlo studies for the design of a new exposure area at the CERF
facility are very promising. The first results showed a good agreement between
calculated and experimental data. Since there is no other known facility where to
perform measurements on Earth with detectors that will be used in space, it might be
CHAPTER 7. Conclusions
137
worth to upgrade the CERF facility by constructing this new area. In order to achieve
higher reliability further investigations will be needed. The new calculations will have
to be based on inputs that consider spectra of primary cosmic rays, other parameters
such as solar modulation or geomagnetic fields and atmospheric models. It might be
more interesting to include different atmospheric profiles, in order to reproduce the
expected conditions at different latitudes. Although these studies may not be easy to
implement, they could be valuable for the future development of the CERF facility.
Towards the extension of the use of the BSS in the region of cosmic ray energies, the
CERF facility may prove to be very useful for future tests and measurements.
APPENDIX A – Physical and Dosimetric quantities
138
A.1 Introduction
In 1990, the International Commission on Radiological Protection (ICRP)
[131] published the definition of two new quantities to be used for protection
purposes. These are the protection quantities equivalent dose, HT,R, and effective
dose, E. Since equivalent dose and effective dose are not directly measurable, the
International Commission on Radiation Units and Measurements (ICRU) [132]
defined the quantities ambient dose equivalent, H*(d), and directional dose
equivalent, H'(d, W), for area monitoring, and personal dose equivalent, HP(d), for
determining the dose to the individual. These quantities are measurable and are
termed operational quantities. ICRP introduced new energy and radiation type
dependent radiation weighting factors, wR, to account for differences in the biological
response associated with different radiation qualities (photons, neutrons, electrons and
heavy charged particles).
A.2 BASIC PHYSICAL QUANTITIES
Energy fluence, Φ the quotient of dN by da, where dN is the number of particles incident on an
elementary sphere of cross-sectional area da.
dadN=Φ
The most frequently used unit is the reciprocal square centimeter [cm-2].
Energy fluence rate The quotient of dΦ by dt, where dΦ is the increment of energy fluence in the time
interval dt.
dtdad
dtd
⋅Ν=Φ=Ψ
2
The unit is [cm-2·s-1].
APPENDIX A – Physical and Dosimetric quantities
139
Absorbed dose D is the energy absorbed per unit mass . It is defined as the quotient of the mean energy
imparted, εd , to matter of mass dm,
dmdD ε=
Its unit is the Joule per kilogram (J·kg-1) which is given the special name Gray (Gy).
Kerma The quantity kerma relates to kinetic energy of the charged particles, released in
matter by uncharged particles. It is the quotient of dEir by dm, where dEir is the sum of
the initial kinetic energies of all the charged particles liberated by uncharged particles
in the mass dm of material, thus:
dmirdΕ
=Κ
A.3. DOSIMETRIC QUANTITIES
A.3.1. Protection quantities The protection quantities give a measure of the risk of damage due to ionising
radiation and are suitable for defining exposure limits.
Organ absorbed dose
It is defined as the mean absorbed dose, DT, in a specified tissue or organ of the
human body, T, given by
∫ ⋅= dmDTmTD 1
where mT is the mass of the tissue or organ, and D is the absorbed dose in the mass
element dm. Unit: Gray [Gy].
APPENDIX A – Physical and Dosimetric quantities
140
Equivalent dose, HT, of an organ or tissue
It is defined as the absorbed dose averaged, DT,R , over the tissue or organ, T,
due to radiation R, multiplied by the relevant radiation weighting factor, wR , for
radiation, R. When the radiation field is composed of radiations with different values
wR , the absorbed dose is subdivided into blocks, each multiplied by its own value of
wR and summed to determine the total equivalent dose, i.e.,
RTDR RwTH ,∑=
Unit: Sievert [Sv], 1 Sv = 1 Jkg-1
The radiation weighting factor, wR, reflects the different radiobiological
effectiveness of the various types and energies of radiation. Table A1 provides the
values of radiation weighting factor used for radiological protection purposes as now
recommended by ICRU [133]. Table A.1. Radiation weighting factors according to ICRU 60.
Types and energy range of radiation Radiation weighting factor wR
Photons (all energies) 1
Electrons and muons (all energies) 1
Neutrons
< 10 keV 5
10-100 keV 10
> 100 keV- 2 MeV 20
>2 -20 MeV 10
>20 MeV 5
Protons (not recoil), energy > 2 MeV 5
Alpha particles, fission fragments, heavy nuclei 20
APPENDIX A – Physical and Dosimetric quantities
141
Effective dose, E, is a summation of the equivalent doses in tissue or organs, each multiplied by the
appropriate tissue weighting factor. It is given by the expression
RTDR Rw
T TwTHT TwE ,∑∑ ⋅=∑=
where HT is the equivalent dose in tissue or organ, T, and wT is the tissue weighting
factor for tissue, T, and accounts for the different susceptibilities of different organs to
radiation damage.
Unit: Sievert [Sv], 1 Sv = 1 J·kg-1
A.3.2. Operational quantities
Operational quantities were designed by the ICRU to provide appropriate
estimates of the protection quantities and to serve as calibration quantities for
dosimetric devices. As stated by the ICRP : ‘The probability of stochastic effects is
found to depend not only on the absorbed dose but also on the type and energy of the
radiation causing the dose’. Operational quantities are based on the dose equivalent
concept which combines the absorbed dose at the point of interest, D, and Q, the
quality factor at that point. Q is related to the type and energy of the radiation via the
unrestricted linear energy transfer, L, of charged particles in water.
Dose equivalent, H derives from the relation
∫ ⋅⋅=⋅= dLLDLQDQH )()( where DL is the distribution of dose D in linear energy L and Q(L) is the quality factor
as a function of L in water. Unit: Sievert [Sv], 1 Sv = 1 Jkg-1
APPENDIX A – Physical and Dosimetric quantities
142
Area monitoring
For area monitoring the phantom of definition is the ICRU sphere, a 30 cm
diameter sphere made of a four- element tissue- like material, with a mass density of
1 g/cm3 and a mass composition of 76.2% oxygen, 11.1% carbon, 10.1% hydrogen
and 2.6% nitrogen. In addition, certain mathematic conventions, namely expansion
and alignment of the actual radiation field, are included in the definitions of quantities
to be used for area monitoring.
Different dose equivalent quantities are defined for strongly and weakly
penetrating, radiation and for area and personal monitoring. Radiation is strongly
penetrating if the dose equivalent received by the terminative layer of the skin
(0.07 mm) at normal incidence to a broad radiation beam is lower than ten times the
effective dose. Radiation is weakly penetrating if for normal incidence the skin dose is
higher than ten times the effective dose. The area dose provides an estimate of the
effective dose that a person would receive if he stayed at a particular location while
the personal dose is a measure of the exposure of an individual to external radiation.
The human body influences the radiation field so they are not in general equivalent.
Area dose The area dose is the dose equivalent of soft tissue measured at a specific point
(Unit: [Sv]). For strongly penetrating radiation the ambient dose equivalent H*(10) is
used while for weakly penetrating radiation the relevant quantity is the directional
dose equivalent H’(0.07 Ω).
Ambient dose equivalent, H*(10), at a point of interest in a real radiation field is the dose equivalent that would be
produced by the corresponding expanded and aligned radiation field at a depth
of 10 mm in the ICRU sphere in the opposing direction to the aligned field.
APPENDIX A – Physical and Dosimetric quantities
143
The directional dose equivalent, H’(0.07 Ω), at a point of interest in a real radiation field is the dose equivalent that would be
produced by the corresponding expanded radiation field in the ICRU sphere at a depth
of 0.07 mm on a radius in a specified direction Ω. Often, the maximum value is used
as it is not known a priori what orientation a person will have in the radiation field.
Personal dose is the dose equivalent in soft tissue measured at a point on the body surface
representative of the radiation conditions prevailing (Unit: [Sv]). Again different
quantities are used for strongly and weakly penetrating radiation. The personal depth
dose equivalent, Hp(10), is the dose equivalent in ICRU soft tissue at a depth
of 10 mm in the body at the location where the personal dosimeter is worn.
Personal surface dose equivalent, Hp(0.07), is the dose equivalent in ICRU soft tissue at a depth of 0.07 mm in the body in the
location where the personal dosimeter is worn. These values may vary between
individuals and depend on the part of the body to which the dosimeter is attached.
APPENDIX B – Nominal Response functions of the BSS to neutrons
144
Nominal response functions of the extended range BSS. The group response is given.
E (MeV) 81 81cd 108 133 178 233 Stanlio Ollio 1.00E-11 4.14E-07 1.24E+00 4.32E-02 8.38E-01 5.56E-01 2.47E-01 8.47E-02 3.89E-02 3.22E-03 6.83E-07 2.50E+00 1.00E+00 1.71E+00 1.15E+00 5.05E-01 1.75E-01 9.92E-01 5.38E-03 1.13E-06 2.75E+00 2.46E+00 2.01E+00 1.34E+00 6.14E-01 1.97E-01 2.13E+00 6.33E-03 1.64E-06 3.05E+00 2.77E+00 2.23E+00 1.50E+00 6.67E-01 2.25E-01 2.52E+00 6.38E-03 2.38E-06 3.13E+00 3.15E+00 2.35E+00 1.58E+00 7.42E-01 2.47E-01 2.72E+00 6.55E-03 3.47E-06 3.19E+00 3.06E+00 2.46E+00 1.68E+00 7.76E-01 2.59E-01 2.91E+00 8.32E-03 5.04E-06 3.31E+00 3.10E+00 2.61E+00 1.87E+00 8.21E-01 2.83E-01 2.96E+00 6.58E-03 7.34E-06 3.31E+00 3.38E+00 2.72E+00 1.92E+00 8.53E-01 2.88E-01 2.89E+00 6.25E-03 1.07E-05 3.24E+00 3.24E+00 2.83E+00 1.98E+00 9.22E-01 3.05E-01 2.92E+00 8.23E-03 1.55E-05 3.28E+00 3.13E+00 2.93E+00 2.05E+00 8.99E-01 3.30E-01 2.56E+00 1.26E-02 2.26E-05 3.17E+00 3.06E+00 2.97E+00 2.08E+00 9.93E-01 3.45E-01 2.89E+00 1.25E-02 3.73E-05 3.19E+00 2.75E+00 2.97E+00 2.14E+00 1.04E+00 3.54E-01 2.70E+00 1.17E-02 6.14E-05 3.08E+00 3.13E+00 2.94E+00 2.26E+00 1.08E+00 3.75E-01 2.46E+00 1.12E-02 1.01E-04 3.19E+00 2.64E+00 3.00E+00 2.29E+00 1.15E+00 3.87E-01 2.22E+00 1.50E-02 1.67E-04 2.93E+00 2.70E+00 3.05E+00 2.37E+00 1.20E+00 4.08E-01 2.09E+00 1.54E-02 2.75E-04 2.84E+00 2.67E+00 3.11E+00 2.39E+00 1.17E+00 4.33E-01 2.30E+00 1.88E-02 4.54E-04 2.82E+00 2.58E+00 2.91E+00 2.35E+00 1.24E+00 4.37E-01 2.06E+00 2.52E-02 6.89E-04 2.62E+00 2.73E+00 3.02E+00 2.45E+00 1.32E+00 4.60E-01 2.22E+00 2.73E-02 1.04E-03 2.45E+00 2.40E+00 2.92E+00 2.53E+00 1.32E+00 4.80E-01 2.11E+00 3.08E-02 1.58E-03 2.36E+00 2.42E+00 2.95E+00 2.56E+00 1.35E+00 4.88E-01 1.92E+00 3.73E-02 2.31E-03 2.42E+00 2.16E+00 2.99E+00 2.54E+00 1.43E+00 5.22E-01 1.89E+00 3.81E-02 3.35E-03 2.33E+00 2.17E+00 2.88E+00 2.52E+00 1.42E+00 5.31E-01 1.88E+00 4.68E-02 4.88E-03 2.21E+00 2.04E+00 2.84E+00 2.62E+00 1.44E+00 5.39E-01 1.70E+00 5.03E-02 7.10E-03 2.16E+00 2.00E+00 2.81E+00 2.51E+00 1.47E+00 5.59E-01 1.87E+00 5.58E-02 1.03E-02 2.07E+00 1.94E+00 2.75E+00 2.49E+00 1.55E+00 5.83E-01 1.63E+00 5.43E-02 1.50E-02 1.96E+00 1.89E+00 2.73E+00 2.61E+00 1.56E+00 6.05E-01 1.48E+00 6.39E-02 2.19E-02 1.87E+00 1.82E+00 2.73E+00 2.57E+00 1.62E+00 6.26E-01 1.53E+00 7.68E-02 3.18E-02 1.85E+00 1.70E+00 2.69E+00 2.62E+00 1.63E+00 6.80E-01 1.42E+00 8.28E-02 5.25E-02 1.72E+00 1.61E+00 2.74E+00 2.66E+00 1.76E+00 7.54E-01 1.40E+00 9.64E-02 8.65E-02 1.70E+00 1.46E+00 2.64E+00 2.82E+00 1.85E+00 7.60E-01 1.26E+00 1.31E-01 1.23E-01 1.54E+00 1.45E+00 2.56E+00 2.86E+00 2.04E+00 8.96E-01 1.20E+00 1.65E-01 1.50E-01 1.47E+00 1.42E+00 2.51E+00 2.89E+00 2.14E+00 9.43E-01 1.26E+00 1.96E-01 1.83E-01 1.42E+00 1.30E+00 2.54E+00 2.90E+00 2.27E+00 1.03E+00 1.17E+00 2.14E-01 2.24E-01 1.33E+00 1.27E+00 2.51E+00 2.94E+00 2.29E+00 1.11E+00 1.15E+00 2.88E-01 2.73E-01 1.34E+00 1.21E+00 2.50E+00 2.95E+00 2.39E+00 1.22E+00 1.06E+00 3.27E-01 3.34E-01 1.21E+00 1.10E+00 2.32E+00 2.88E+00 2.64E+00 1.33E+00 9.68E-01 3.86E-01 4.08E-01 1.15E+00 1.09E+00 2.34E+00 2.98E+00 2.69E+00 1.48E+00 9.80E-01 4.58E-01 4.98E-01 1.07E+00 1.02E+00 2.23E+00 3.01E+00 2.76E+00 1.73E+00 8.96E-01 5.39E-01 6.08E-01 9.85E-01 9.06E-01 2.11E+00 2.93E+00 2.92E+00 1.78E+00 7.66E-01 6.25E-01 7.43E-01 8.91E-01 8.19E-01 1.97E+00 2.77E+00 2.90E+00 1.97E+00 6.59E-01 7.72E-01 8.21E-01 8.05E-01 7.38E-01 1.87E+00 2.77E+00 2.98E+00 2.17E+00 6.72E-01 8.43E-01 9.07E-01 7.56E-01 7.28E-01 1.85E+00 2.72E+00 3.06E+00 2.29E+00 5.90E-01 8.77E-01 1.00E+00 7.05E-01 7.18E-01 1.82E+00 2.73E+00 3.01E+00 2.35E+00 5.37E-01 9.24E-01
APPENDIX B – Nominal Response functions of the BSS to neutrons
145
Nominal response functions of the extended range BSS (continued).
E (MeV) 81 81cd 108 133 178 233 Stanlio Ollio
1.11E+00 6.78E-01 5.98E-01 1.69E+00 2.58E+00 3.15E+00 2.47E+00 5.66E-01 1.00E+00 1.22E+00 6.25E-01 6.12E-01 1.59E+00 2.53E+00 3.07E+00 2.42E+00 5.15E-01 1.05E+00 1.35E+00 5.87E-01 5.11E-01 1.44E+00 2.35E+00 3.07E+00 2.50E+00 5.03E-01 1.14E+00 1.50E+00 5.31E-01 4.91E-01 1.45E+00 2.32E+00 3.09E+00 2.68E+00 4.14E-01 1.19E+00 1.65E+00 5.07E-01 4.59E-01 1.38E+00 2.20E+00 3.09E+00 2.76E+00 4.26E-01 1.25E+00 1.83E+00 4.66E-01 4.16E-01 1.30E+00 2.22E+00 2.86E+00 2.66E+00 3.85E-01 1.27E+00 2.02E+00 4.22E-01 3.77E-01 1.21E+00 2.05E+00 2.85E+00 2.79E+00 3.63E-01 1.27E+00 2.23E+00 3.80E-01 3.59E-01 1.12E+00 1.90E+00 2.92E+00 2.71E+00 3.20E-01 1.28E+00 2.47E+00 3.67E-01 3.08E-01 1.10E+00 1.87E+00 2.68E+00 2.74E+00 3.01E-01 1.28E+00 2.73E+00 3.05E-01 3.06E-01 9.42E-01 1.65E+00 2.62E+00 2.82E+00 3.15E-01 1.31E+00 3.01E+00 3.04E-01 2.68E-01 9.00E-01 1.60E+00 2.41E+00 2.53E+00 2.63E-01 1.28E+00 3.33E+00 2.69E-01 2.44E-01 7.89E-01 1.50E+00 2.35E+00 2.59E+00 2.60E-01 1.26E+00 3.68E+00 2.32E-01 2.18E-01 7.51E-01 1.39E+00 2.25E+00 2.41E+00 2.50E-01 1.20E+00 4.07E+00 2.10E-01 1.98E-01 6.84E-01 1.28E+00 2.17E+00 2.38E+00 2.38E-01 1.21E+00 4.49E+00 1.98E-01 1.75E-01 6.76E-01 1.18E+00 2.08E+00 2.45E+00 3.69E-01 1.29E+00 4.97E+00 1.77E-01 1.75E-01 5.68E-01 1.12E+00 2.02E+00 2.43E+00 2.14E-01 1.27E+00 5.49E+00 1.62E-01 1.55E-01 5.20E-01 1.06E+00 1.88E+00 2.37E+00 2.31E-01 1.24E+00 6.07E+00 1.50E-01 1.44E-01 4.99E-01 9.74E-01 1.75E+00 2.35E+00 1.85E-01 1.20E+00 6.70E+00 1.34E-01 1.31E-01 4.44E-01 9.10E-01 1.70E+00 2.30E+00 1.94E-01 1.17E+00 7.41E+00 1.05E-01 1.04E-01 3.91E-01 8.17E-01 1.58E+00 2.11E+00 1.59E-01 1.17E+00 8.19E+00 9.76E-02 1.02E-01 3.61E-01 7.16E-01 1.42E+00 1.86E+00 1.72E-01 1.04E+00 9.05E+00 8.23E-02 8.65E-02 3.06E-01 6.35E-01 1.27E+00 1.76E+00 2.29E-01 1.00E+00 1.00E+01 7.06E-02 7.46E-02 2.63E-01 5.62E-01 1.08E+00 1.62E+00 2.14E-01 9.56E-01 1.11E+01 6.67E-02 7.28E-02 2.33E-01 5.00E-01 1.08E+00 1.57E+00 2.04E-01 9.54E-01 1.22E+01 6.20E-02 7.30E-02 2.23E-01 4.87E-01 1.00E+00 1.47E+00 2.11E-01 9.61E-01 1.35E+01 5.05E-02 6.30E-02 2.03E-01 4.18E-01 8.79E-01 1.28E+00 2.28E-01 9.02E-01 1.49E+01 5.12E-02 5.95E-02 1.93E-01 4.09E-01 8.32E-01 1.26E+00 2.26E-01 9.10E-01 1.75E+01 5.00E-02 6.17E-02 1.77E-01 3.96E-01 8.31E-01 1.21E+00 2.38E-01 9.01E-01 1.96E+01 3.83E-02 5.23E-02 1.47E-01 3.05E-01 6.44E-01 1.05E+00 2.35E-01 8.81E-01 2.10E+01 2.90E-02 4.66E-02 1.44E-01 3.06E-01 6.54E-01 1.03E+00 2.49E-01 9.18E-01 2.50E+01 1.77E-02 3.79E-02 1.28E-01 2.77E-01 6.04E-01 9.31E-01 3.01E-01 9.55E-01 3.50E+01 1.20E-02 3.01E-02 8.84E-02 1.96E-01 4.41E-01 7.02E-01 3.64E-01 9.56E-01 5.00E+01 7.91E-03 2.51E-02 5.38E-02 1.24E-01 2.89E-01 4.79E-01 4.01E-01 9.61E-01 7.50E+01 7.03E-03 2.34E-02 4.12E-02 9.31E-02 2.16E-01 3.60E-01 4.21E-01 9.72E-01 8.50E+01 6.46E-03 2.32E-02 3.59E-02 8.02E-02 1.83E-01 3.05E-01 4.30E-01 9.90E-01 1.00E+02 6.15E-03 2.29E-02 3.32E-02 7.36E-02 1.68E-01 2.86E-01 4.38E-01 1.02E+00 1.35E+02 5.72E-03 2.12E-02 2.99E-02 6.43E-02 1.52E-01 2.58E-01 4.54E-01 1.05E+00 1.75E+02 5.18E-03 2.06E-02 2.54E-02 5.78E-02 1.36E-01 2.33E-01 4.78E-01 1.10E+00 2.50E+02 4.94E-03 2.04E-02 2.29E-02 5.40E-02 1.24E-01 2.11E-01 5.15E-01 1.17E+00 3.00E+02 4.93E-03 2.04E-02 2.25E-02 5.04E-02 1.15E-01 1.95E-01 5.59E-01 1.25E+00 4.00E+02 4.67E-03 2.06E-02 2.18E-02 4.86E-02 1.14E-01 1.96E-01 6.18E-01 1.37E+00 5.00E+02 4.27E-03 2.10E-02 2.02E-02 4.73E-02 1.13E-01 1.96E-01 6.99E-01 1.54E+00 6.50E+02 4.18E-03 2.20E-02 1.91E-02 4.48E-02 1.09E-01 1.90E-01 7.85E-01 1.71E+00 7.50E+02 4.15E-03 2.31E-02 1.89E-02 4.29E-02 1.06E-01 1.87E-01 8.61E-01 1.87E+00 1.00E+03 4.02E-03 2.34E-02 1.87E-02 4.17E-02 1.03E-01 1.83E-01 9.38E-01 2.04E+00 1.50E+03 3.95E-03 2.42E-02 1.89E-02 4.14E-02 1.01E-01 1.80E-01 1.09E+00 2.36E+00 2.00E+03 3.98E-03 2.53E-02 1.96E-02 4.26E-02 1.01E-01 1.79E-01 1.25E+00 2.69E+00
APPENDIX C - Response functions of BSS to positive pions
146
TABLE C1
Sphere 81 mm
Sphere 81Cd Energy
[MeV] Response
[cm2] Uncertainty Response [cm2]
Uncertainty
50 4.43E-04 7.81E-06 2.61E-03 2.86E-05
70 6.90E-04 1.16E-05 4.19E-03 4.30E-05
80 8.59E-04 1.25E-05 5.44E-03 5.21E-05
100 1.57E-03 1.84E-05 5.98E-03 4.63E-05
120 1.69E-03 1.84E-05 7.27E-03 4.92E-05
150 1.94E-03 1.87E-05 7.57E-03 4.85E-05
180 1.84E-03 1.74E-05 7.58E-03 5.52E-05
200 1.96E-03 1.99E-05 7.40E-03 5.07E-05
220 1.79E-03 1.88E-05 7.60E-03 5.00E-05
250 1.86E-03 2.00E-05 7.24E-03 4.95E-05
280 1.55E-03 1.83E-05 6.78E-03 4.85E-05
300 1.54E-03 2.47E-05 6.57E-03 4.72E-05
400 1.34E-03 1.84E-05 6.27E-03 5.83E-05
500 1.10E-03 1.38E-05 5.59E-03 4.91E-05
700 1.09E-03 1.33E-05 5.66E-03 5.34E-05
1000 1.16E-03 1.29E-05 6.16E-03 4.98E-05
2.00E+03 1.07E-03 1.17E-05 6.75E-03 4.55E-05
3.00E+03 1.16E-03 1.44E-05 6.17E-03 4.41E-05
4.00E+03 1.45E-03 1.70E-05 7.01E-03 4.90E-05
5.00E+03 1.51E-03 1.74E-05 7.29E-03 4.82E-05
7.00E+03 1.49E-03 1.59E-05 8.05E-03 6.91E-05
1.00E+04 1.74E-03 2.06E-05 7.93E-03 6.30E-05
5.00E+04 1.65E-03 2.02E-05 8.30E-03 6.35E-05
1.00E+05 1.70E-03 1.89E-05 8.88E-03 8.19E-05
1.50E+05 1.77E-03 1.92E-05 8.71E-03 7.45E-05
APPENDIX C - Response functions of BSS to positive pions
147
TABLE C2
Sphere 108 mm
Sphere 133 mm Energy
[MeV] Response
[cm2]
Uncertainty Response
[cm2]
Uncertainty
50 2.97E-03 5.96E-05 5.89E-03 1.20E-04
70 3.67E-03 5.38E-05 9.77E-03 1.75E-04
80 4.66E-03 7.43E-05 1.14E-02 1.70E-04
100 7.11E-03 7.93E-05 1.68E-02 1.79E-04
120 9.92E-03 9.67E-05 2.38E-02 2.33E-04
150 1.19E-02 1.14E-04 2.81E-02 2.67E-04
180 1.20E-02 1.11E-04 2.94E-02 2.79E-04
200 1.07E-02 9.54E-05 3.12E-02 3.10E-04
220 1.18E-02 1.04E-04 2.84E-02 2.26E-04
250 1.03E-02 8.73E-05 2.75E-02 2.41E-04
280 9.22E-03 8.66E-05 2.45E-02 2.65E-04
300 9.70E-03 1.10E-04 2.41E-02 2.27E-04
400 7.65E-03 9.04E-05 2.00E-02 2.50E-04
500 7.49E-03 9.31E-05 1.83E-02 2.27E-04
700 7.43E-03 7.09E-05 1.83E-02 1.81E-04
1000 7.76E-03 9.60E-05 1.95E-02 1.81E-04
2.00E+03 7.67E-03 1.02E-04 2.06E-02 2.13E-04
3.00E+03 7.09E-03 8.51E-05 2.01E-02 2.29E-04
4.00E+03 8.60E-03 9.86E-05 2.13E-02 2.56E-04
5.00E+03 9.51E-03 1.07E-04 2.58E-02 2.83E-04
7.00E+03 1.01E-02 1.10E-04 2.77E-02 2.46E-04
1.00E+04 1.06E-02 1.02E-04 2.93E-02 2.79E-04
5.00E+04 1.15E-02 1.37E-04 3.00E-02 3.46E-04
1.00E+05 1.10E-02 1.34E-04 3.37E-02 3.33E-04
1.50E+05 1.22E-02 1.23E-04 3.22E-02 2.51E-04
APPENDIX C - Response functions of BSS to positive pions
148
TABLE C3
Sphere 178 mm
Sphere 233 mm Energy
[MeV] Response
[cm2] Uncertainty
Energy
[MeV] Response
[cm2] Uncertainty
50 1.06E-02 1.89E-04 50 1.43E-02 5.64E-04
70 2.20E-02 3.32E-04 65 2.78E-02 6.42E-04
80 2.78E-02 4.05E-04 80 4.49E-02 8.95E-04
100 4.11E-02 3.60E-04 100 6.55E-02 1.10E-03
120 5.58E-02 4.58E-04 110 8.44E-02 9.41E-04
150 6.89E-02 6.60E-04 120 9.17E-02 1.18E-03
180 7.53E-02 5.43E-04 130 1.09E-01 1.57E-03
200 7.88E-02 5.75E-04 150 1.25E-01 1.77E-03
220 7.42E-02 5.03E-04 200 1.47E-01 1.50E-03
250 7.22E-02 5.89E-04 250 1.40E-01 2.01E-03
280 6.73E-02 6.71E-04 300 1.22E-01 1.25E-03
300 6.57E-02 4.99E-04 400 1.02E-01 9.01E-04
400 5.54E-02 5.99E-04 500 9.46E-02 1.87E-03
500 5.09E-02 4.52E-04 700 9.87E-02 1.96E-03
700 5.36E-02 5.51E-04 1000 1.15E-01 2.63E-03
1000 5.81E-02 5.26E-04 1.50E+03 1.18E-01 2.04E-03
2.00E+03 5.95E-02 6.02E-04 2.00E+03 1.15E-01 1.50E-03
3.00E+03 5.80E-02 6.32E-04 3.00E+03 1.13E-01 1.49E-03
4.00E+03 6.52E-02 7.16E-04 4.00E+03 1.29E-01 1.39E-03
5.00E+03 7.09E-02 5.88E-04 5.00E+03 1.41E-01 1.47E-03
7.00E+03 8.00E-02 6.43E-04 7.00E+03 1.58E-01 1.72E-03
1.00E+04 8.32E-02 9.60E-04 9.00E+03 1.61E-01 1.68E-03
5.00E+04 8.89E-02 8.40E-04 1.00E+04 1.66E-01 3.38E-03
1.00E+05 9.87E-02 8.95E-04 5.00E+04 1.99E-01 1.74E-03
1.50E+05 1.10E-01 7.85E-04 1.00E+05 2.02E-01 1.85E-03
1.50E+05 2.25E-01 2.23E-03
APPENDIX C - Response functions of BSS to positive pions
149
TABLE C4
STANLIO
OLLIO Energy
[MeV] Response
[cm2] Uncertainty
Response
[cm2] Uncertainty
50 1.54E-02 1.59E-04 3.67E-02 7.84E-04
100 1.37E-01 5.13E-04 6.92E-01 4.65E-03
150 2.69E-01 6.63E-04 1.35E+00 5.45E-03
250 3.03E-01 7.64E-04 1.62E+00 6.41E-03
300 3.05E-01 7.14E-04 1.64E+00 5.15E-03
500 3.15E-01 6.42E-04 1.68E+00 7.39E-03
700 3.43E-01 7.03E-04 1.88E+00 6.45E-03
1000 3.85E-01 7.58E-04 2.09E+00 6.63E-03
2.00E+03 5.00E-01 8.75E-04 2.67E+00 7.02E-03
3.00E+03 5.47E-01 9.03E-04 2.92E+00 7.57E-03
4.00E+03 6.20E-01 8.99E-04 3.30E+00 9.14E-03
5.00E+03 6.52E-01 9.72E-04 3.47E+00 7.96E-03
7.00E+03 7.56E-01 1.07E-03 3.95E+00 9.91E-03
9.00E+03 8.45E-01 1.26E-03 4.36E+00 9.28E-03
1.00E+04 8.76E-01 1.10E-03 4.53E+00 1.44E-02
5.00E+04 1.32E+00 2.57E-03 6.66E+00 2.21E-02
1.00E+05 1.59E+00 3.11E-03 8.03E+00 2.77E-02
1.50E+05 1.81E+00 3.08E-03 8.95E+00 2.41E-02
APPENDIX C - Response functions of BSS to negative pions
150
TABLE C5
Sphere 81 mm
Sphere 81Cd Energy
[MeV] Response [cm2] Uncertainty
Energy
[MeV] Response
[cm2] Uncertainty
50 2.12E-03 1.87E-05 50 8.71E-03 7.51E-05
60 2.30E-03 2.10E-05 60 8.50E-03 7.22E-05
80 2.87E-03 2.51E-05 80 9.21E-03 7.65E-05
100 3.43E-03 2.53E-05 100 1.07E-02 8.44E-05
110 3.77E-03 2.52E-05 110 1.16E-02 8.23E-05
130 3.80E-03 2.74E-05 130 1.17E-02 7.64E-05
150 3.83E-03 2.74E-05 150 1.11E-02 8.09E-05
200 3.20E-03 2.48E-05 200 1.03E-02 7.70E-05
250 2.74E-03 2.33E-05 250 9.26E-03 5.68E-05
300 2.38E-03 1.97E-05 300 8.99E-03 5.68E-05
400 2.14E-03 2.19E-05 400 7.59E-03 4.31E-05
500 1.91E-03 1.79E-05 500 7.39E-03 4.87E-05
700 1.82E-03 1.73E-05 700 7.53E-03 6.12E-05
1000 1.80E-03 2.05E-05 1000 7.38E-03 5.72E-05
2.00E+03 1.57E-03 1.66E-05 2.00E+03 6.80E-03 5.35E-05
3.00E+03 1.34E-03 1.67E-05 3.00E+03 6.81E-03 6.06E-05
5.00E+03 1.56E-03 1.80E-05 5.00E+03 7.36E-03 5.82E-05
7.00E+03 1.69E-03 1.73E-05 7.00E+03 8.01E-03 6.86E-05
1.00E+04 1.68E-03 1.71E-05 1.00E+04 8.15E-03 6.54E-05
5.00E+04 1.80E-03 2.16E-05 5.00E+04 8.03E-03 6.66E-05
1.00E+05 1.81E-03 1.94E-05 1.00E+05 8.16E-03 6.00E-05
1.50E+05 1.57E-03 1.70E-05 1.50E+05 8.60E-03 6.86E-05
APPENDIX C - Response functions of BSS to negative pions
151
TABLE C6
Sphere 108 mm Sphere 133 mm Energy
[MeV] Response
[cm2] Uncertainty
Energy
[MeV] Response
[cm2] Uncertainty
50 6.26E-02 2.99E-04 50 4.90E-01 1.18E-03
55 1.58E-02 1.21E-04 60 7.73E-02 4.60E-04
60 1.50E-02 1.46E-04 80 4.53E-02 3.61E-04
80 1.73E-02 1.32E-04 100 5.45E-02 3.11E-04
100 2.13E-02 1.39E-04 110 5.83E-02 3.91E-04
110 2.27E-02 1.45E-04 130 6.37E-02 5.17E-04
130 2.44E-02 1.60E-04 150 6.34E-02 4.15E-04
150 2.35E-02 1.46E-04 200 5.42E-02 3.69E-04
200 2.12E-02 1.61E-04 250 4.63E-02 3.84E-04
250 1.70E-02 1.13E-04 300 3.90E-02 2.97E-04
300 1.48E-02 1.16E-04 400 3.12E-02 2.77E-04
400 1.25E-02 1.03E-04 500 3.18E-02 3.30E-04
500 1.16E-02 1.14E-04 700 3.08E-02 3.44E-04
700 1.23E-02 1.32E-04 1000 3.12E-02 2.64E-04
1000 1.17E-02 1.08E-04 2.00E+03 2.67E-02 2.50E-04
2.00E+03 1.03E-02 1.23E-04 3.00E+03 2.32E-02 2.37E-04
3.00E+03 8.46E-03 8.65E-05 5.00E+03 2.86E-02 3.18E-04
5.00E+03 1.05E-02 1.20E-04 7.00E+03 3.00E-02 2.54E-04
7.00E+03 1.11E-02 1.00E-04 1.00E+04 3.02E-02 3.30E-04
1.00E+04 1.10E-02 1.01E-04 5.00E+04 3.02E-02 2.63E-04
5.00E+04 1.15E-02 1.19E-04 1.00E+05 3.26E-02 3.38E-04
1.00E+05 1.15E-02 1.08E-04 1.50E+05 3.53E-02 3.62E-04
1.50E+05 1.16E-02 1.18E-04
APPENDIX C - Response functions of BSS to negative pions
152
TABLE C7
Sphere 178 mm
Sphere 233 mm Energy
[MeV] Response
[cm2]
Uncertainty
Energy
[MeV] Response
[cm2]
Uncertainty
50 1.54E+00 3.38E-03 50 2.48E+00 5.40E-03
60 9.49E-01 2.99E-03 60 2.13E+00 5.97E-03
70 2.98E-01 1.68E-03 70 1.26E+00 4.62E-03
80 1.21E-01 1.06E-03 80 5.71E-01 3.17E-03
90 1.30E-01 1.26E-03 90 2.60E-01 2.86E-03
100 1.45E-01 1.04E-03 100 2.72E-01 2.05E-03
110 1.58E-01 8.69E-04 110 2.91E-01 3.85E-03
130 1.66E-01 9.05E-04 120 3.10E-01 2.15E-03
150 1.74E-01 1.09E-03 130 3.23E-01 1.99E-03
200 1.50E-01 8.93E-04 140 3.21E-01 2.10E-03
250 1.26E-01 1.07E-03 150 3.27E-01 2.45E-03
300 1.05E-01 7.59E-04 200 2.94E-01 1.80E-03
400 8.89E-02 8.44E-04 250 2.56E-01 2.47E-03
500 8.41E-02 6.63E-04 300 2.07E-01 1.67E-03
700 8.66E-02 7.60E-04 400 1.67E-01 2.03E-03
1000 8.96E-02 7.20E-04 500 1.71E-01 2.05E-03
2.00E+03 7.61E-02 6.70E-04 700 1.67E-01 1.35E-03
3.00E+03 6.91E-02 6.74E-04 1000 1.65E-01 4.31E-04
5.00E+03 8.03E-02 6.77E-04 2.00E+03 1.54E-01 1.35E-03
7.00E+03 8.24E-02 6.77E-04 3.00E+03 1.36E-01 1.64E-03
1.00E+04 8.89E-02 7.20E-04 5.00E+03 1.58E-01 1.67E-03
5.00E+04 9.35E-02 8.87E-04 7.00E+03 1.68E-01 2.04E-03
1.00E+05 9.98E-02 1.09E-03 1.00E+04 1.74E-01 2.67E-03
1.50E+05 1.07E-01 8.06E-04 5.00E+04 2.02E-01 1.56E-03
1.00E+05 2.14E-01 1.47E-03
1.00E+05 2.28E-01 1.86E-03
APPENDIX C - Response functions of BSS to negative pions
153
TABLE C8
STANLIO OLLIO Energy
[MeV] Response
[cm2] Uncertainty
Energy
[MeV] Response
[cm2] Uncertainty
50 9.00E-01 2.31E-03 50 6.63E+00 1.03E-02
60 9.93E-01 1.19E-03 60 6.74E+00 1.63E-02
65 8.00E-01 2.04E-03 65 5.93E+00 1.67E-02
70 7.09E-01 9.72E-04 70 4.84E+00 8.62E-03
75 7.61E-01 2.07E-03 75 4.51E+00 1.64E-02
80 8.66E-01 1.25E-03 80 4.45E+00 9.16E-03
85 9.51E-01 2.34E-03 85 4.91E+00 2.93E-02
90 9.40E-01 1.16E-03 90 5.54E+00 1.35E-02
100 8.32E-01 1.98E-03 100 5.04E+00 1.31E-02
110 7.00E-01 1.86E-03 110 3.71E+00 1.06E-02
120 5.77E-01 1.73E-03 120 3.00E+00 7.64E-03
130 5.30E-01 1.63E-03 130 2.70E+00 7.11E-03
150 4.84E-01 1.36E-03 150 2.50E+00 7.19E-03
200 4.30E-01 1.39E-03 200 2.38E+00 6.94E-03
300 3.97E-01 1.53E-03 300 2.17E+00 7.76E-03
500 3.69E-01 1.37E-03 500 2.04E+00 8.84E-03
700 4.12E-01 1.57E-03 700 2.20E+00 7.48E-03
1000 4.52E-01 1.40E-03 1000 2.40E+00 8.25E-03
2.00E+03 5.54E-01 1.50E-03 2.00E+03 2.96E+00 1.06E-02
3.00E+03 5.80E-01 1.61E-03 3.00E+03 3.12E+00 8.55E-03
5.00E+03 6.90E-01 1.99E-03 5.00E+03 3.59E+00 1.08E-02
7.00E+03 7.61E-01 1.87E-03 7.00E+03 4.07E+00 7.97E-03
1.00E+04 9.02E-01 2.18E-03 1.00E+04 4.64E+00 8.36E-03
5.00E+04 1.33E+00 2.66E-03 5.00E+04 6.69E+00 2.69E-02
1.00E+05 1.59E+00 3.12E-03 1.00E+05 7.87E+00 2.72E-02
1.50E+05 1.81E+00 3.32E-03 1.50E+05 8.97E+00 2.28E-02
APPENDIX C - Response functions of BSS to protons
154
TABLE C9
Sphere 81 mm
Sphere 81Cd Energy
[MeV] Response
[cm2] Uncertainty
Energy
[MeV] Response
[cm2] Uncertainty
50 4.13E-04 7.63E-06 50 2.14E-03 2.80E-05
60 6.06E-04 7.67E-06 60 2.60E-03 2.68E-05
80 9.43E-04 1.01E-05 80 3.15E-03 3.22E-05
100 1.07E-03 1.26E-05 100 4.43E-03 3.78E-05
150 9.91E-04 1.26E-05 150 4.61E-03 4.91E-05
200 9.60E-04 1.36E-05 200 4.34E-03 4.72E-05
300 1.04E-03 1.41E-05 300 4.90E-03 6.20E-05
500 1.18E-03 1.54E-05 500 5.26E-03 5.70E-05
700 1.20E-03 1.60E-05 700 5.71E-03 5.36E-05
1000 1.56E-03 2.01E-05 1000 6.51E-03 5.83E-05
2.00E+03 1.67E-03 2.15E-05 2.00E+03 7.22E-03 6.55E-05
5.00E+03 2.16E-03 2.43E-05 3.00E+03 7.09E-03 5.92E-05
7.00E+03 2.13E-03 2.24E-05 5.00E+03 9.08E-03 6.58E-05
1.00E+04 2.19E-03 2.13E-05 7.00E+03 9.31E-03 7.17E-05
5.00E+04 2.09E-03 1.87E-05 1.00E+04 9.68E-03 7.18E-05
1.00E+05 2.10E-03 1.95E-05 5.00E+04 1.06E-02 7.07E-05
1.50E+05 2.35E-03 2.03E-05 1.00E+05 1.08E-02 7.60E-05
1.50E+05 1.15E-02 7.23E-05
APPENDIX C - Response functions of BSS to protons
155
TABLE C10
Sphere 108 mm
Sphere 133 mm Energy
[MeV] Response
[cm2] Uncertainty
Energy
[MeV] Response
[cm2] Uncertainty
50 1.09E-03 2.75E-05 50 2.35E-03 7.69E-05
60 2.11E-03 3.65E-05 60 4.07E-03 9.91E-05
80 4.26E-03 5.67E-05 80 8.91E-03 1.27E-04
100 6.11E-03 7.57E-05 100 1.33E-02 1.43E-04
150 6.29E-03 8.52E-05 150 1.65E-02 2.28E-04
200 6.40E-03 9.18E-05 200 1.54E-02 1.75E-04
300 6.57E-03 8.14E-05 300 1.63E-02 2.74E-04
500 7.44E-03 8.32E-05 500 1.84E-02 2.12E-04
700 7.97E-03 8.64E-05 700 2.18E-02 2.52E-04
1000 9.61E-03 1.06E-04 1000 2.46E-02 2.25E-04
2.00E+03 9.45E-03 8.52E-05 2.00E+03 2.54E-02 2.38E-04
3.00E+03 9.30E-03 1.08E-04 3.00E+03 2.59E-02 3.07E-04
5.00E+03 1.20E-02 1.06E-04 5.00E+03 3.17E-02 2.82E-04
7.00E+03 1.27E-02 1.15E-04 7.00E+03 3.38E-02 2.88E-04
1.00E+04 1.37E-02 1.13E-04 1.00E+04 3.66E-02 3.38E-04
5.00E+04 1.41E-02 1.30E-04 5.00E+04 4.10E-02 2.98E-04
1.00E+05 1.58E-02 1.34E-04 1.00E+05 4.44E-02 3.32E-04
1.50E+05 1.63E-02 1.27E-04 1.50E+05 4.65E-02 3.04E-04
APPENDIX C - Response functions of BSS to protons
156
TABLE C11
Sphere 178 mm Sphere 233 mm Energy
[MeV] Response
[cm2] Uncertainty
Energy
[MeV] Response
[cm2] Uncertainty
50 4.51E-03 1.72E-04 50 6.35E-03 2.62E-04
60 7.61E-03 3.32E-04 60 9.98E-03 3.49E-04
80 1.57E-02 2.65E-04 80 2.10E-02 4.54E-04
100 2.58E-02 3.58E-04 100 3.77E-02 6.68E-04
150 4.25E-02 4.91E-04 150 7.63E-02 8.51E-04
200 4.32E-02 5.38E-04 200 8.12E-02 1.38E-03
300 4.36E-02 6.11E-04 300 8.14E-02 1.24E-03
500 5.09E-02 6.26E-04 500 9.99E-02 9.99E-04
700 5.75E-02 6.40E-04 700 1.11E-01 1.34E-03
1000 6.61E-02 6.02E-04 1000 1.22E-01 1.26E-03
2.00E+03 7.24E-02 7.46E-04 2.00E+03 1.45E-01 1.52E-03
3.00E+03 7.32E-02 6.58E-04 3.00E+03 1.46E-01 1.38E-03
5.00E+03 8.92E-02 8.00E-04 5.00E+03 1.77E-01 1.65E-03
7.00E+03 9.43E-02 7.52E-04 7.00E+03 1.88E-01 1.44E-03
1.00E+04 1.04E-01 7.88E-04 1.00E+04 2.08E-01 1.80E-03
5.00E+04 1.25E-01 1.22E-03 5.00E+04 2.62E-01 2.21E-03
1.00E+05 1.36E-01 1.36E-03 1.00E+05 3.00E-01 1.82E-03
1.50E+05 1.46E-01 8.52E-04 1.50E+05 3.25E-01 2.29E-03
APPENDIX C - Response functions of BSS to protons
157
TABLE C12
STANLIO
OLLIO
Energy
[MeV] Response
[cm2] Uncertainty
Energy
[MeV] Response
[cm2] Uncertainty
50 5.97E-03 1.90E-04 50 8.47E-03 3.27E-04
100 3.46E-02 4.37E-04 100 9.14E-02 1.29E-03
150 9.56E-02 7.27E-04 150 4.50E-01 3.52E-03
300 2.10E-01 1.25E-03 300 1.11E+00 9.47E-03
500 2.69E-01 1.10E-03 500 1.42E+00 1.62E-02
700 3.21E-01 1.21E-03 700 1.76E+00 1.11E-02
1000 3.69E-01 1.32E-03 1000 2.06E+00 1.06E-02
2.00E+03 5.24E-01 1.71E-03 2.00E+03 2.79E+00 1.45E-02
3.00E+03 6.17E-01 1.72E-03 3.00E+03 3.24E+00 1.54E-02
5.00E+03 7.61E-01 1.97E-03 5.00E+03 4.08E+00 1.89E-02
7.00E+03 8.71E-01 2.08E-03 7.00E+03 4.63E+00 1.89E-02
1.00E+04 1.01E+00 2.26E-03 1.00E+04 5.32E+00 1.86E-02
5.00E+04 1.67E+00 2.85E-03 5.00E+04 8.38E+00 2.32E-02
1.00E+05 2.07E+00 3.57E-03 1.00E+05 1.03E+01 3.27E-02
1.50E+05 2.35E+00 3.90E-03 1.50E+05 1.18E+01 2.88E-02
APPENDIX C - Response functions of the BSS to charged hadrons
158
102 103 104 105
5.0x10-4
1.0x10-3
1.5x10-3
2.0x10-3
2.5x10-3
3.0x10-3
3.5x10-3
4.0x10-3
positive pions negative pions protons
Response of 81 mm sphere to charged hadronsR
espo
nse
[cm
2 ]
Energy [MeV]
102 103 104 105
2.0x10-3
4.0x10-3
6.0x10-3
8.0x10-3
1.0x10-2
1.2x10-2
positive pions negative pions protons
Response of sphere 81cd to charged hadrons
Res
pons
e [c
m2 ]
Energy [MeV]
APPENDIX C - Response functions of the BSS to charged hadrons
159
102 103 104 105
10-3
10-2
10-1
positive pions negative pions protons
Response of 108 mm to charged hadronsR
espo
nse
[cm
2 ]
Energy [MeV]
102 103 104 10510-3
10-2
10-1
100
positive pions negative pions protons
Response of sphere 133 mm to charged hadrons
Res
pons
e [c
m2 ]
Energy [MeV]
APPENDIX C - Response functions of the BSS to charged hadrons
160
102 103 104 10510-3
10-2
10-1
100
positive pions negative pions protons
Response of sphere 178 mm to charged hadrons
Res
pons
e [c
m2 ]
Energy [MeV]
102 103 104 105
10-2
10-1
100
positive pions negative pions protons
Response of sphere 233 mm to charged hadrons
Res
pons
e [c
m2 ]
Energy [MeV]
APPENDIX C - Response function of the BSS to charged hadrons
161
102 103 104 105
0.0
5.0x10-1
1.0x100
1.5x100
positive pions negative pions protons
Response of Stanlio to charged hadrons
Res
pons
e [c
m2 ]
Energy [MeV]
102 103 104 105
0
1x100
2x100
3x100
4x100
5x100
6x100
7x100
8x100
positive pions negative pions protons
Response of Ollio to charged hadrons
Res
pons
e [c
m2 ]
Energy [MeV]
APPENDIX C - Response function of the BSS to charged hadrons
162
TABLE C13
Cu target Pb target Angle Particle Stanlio Ollio Stanlio Ollio
neutron 0.462±0.026 0.386±0.053 0.522±0.029 0.563±0.085 π+ 0.196±0.010 0.219±0.024 0.173±0.008 0.155±0.018 π- 0.249±0.010 0.294±0.039 0.228±0.010 0.214±0.027 p 0.092±0.005 0.101±0.018 0.077±0.005 0.067±0.015
30°
π++π-+ p 0.538±0.019 0.614±0.058 0.478±0.017 0.437±0.044 neutron 0.658±0.040 0.566±0.088 0.719±0.047 0.754±0.124
π+ 0.111±0.006 0.138±0.022 0.090±0.005 0.077±0.012 π- 0.169±0.008 0.225±0.027 0.147±0.008 0.134±0.016 p 0.061±0.005 0.071±0.012 0.045±0.004 0.035±0.011
45°
π++π-+ p 0.342±0.015 0.434±0.046 0.281±0.013 0.246±0.030 neutron 0.782±0.057 0.699±0.108 0.828±0.061 0.854±0.156
π+ 0.062±0.004 0.083±0.012 0.047±0.003 0.039±0.009 π- 0.115±0.007 0.168±0.022 0.094±0.006 0.084±0.012 p 0.041±0.004 0.050±0.011 0.031±0.003 0.022±0.006
60°
π++π-+ p 0.218±0.012 0.301±0.033 0.172±0.009 0.145±0.019 neutron 0.749±0.051 0.782±0.138 0.886±0.059 0.905±0.135
π+ 0.064±0.004 0.053±0.010 0.027±0.002 0.021±0.003 π- 0.140±0.010 0.130±0.017 0.064±0.004 0.058±0.007 p 0.046±0.004 0.034±0.008 0.024±0.002 0.016±0.004
75°
π++π-+ p 0.251±0.014 0.218±0.028 0.114±0.006 0.095±0.011 neutron 0.807±0.057 0.834±0.142 0.919±0.068 0.933±0.209
π+ 0.046±0.003 0.037±0.007 0.017±0.001 0.013±0.003 π- 0.117±0.006 0.108±0.017 0.047±0.003 0.042±0.007 p 0.029±0.003 0.021±0.006 0.018±0.002 0.012±0.005
90°
π++π-+ p 0.193±0.010 0.166±0.023 0.081±0.005 0.067±0.012 neutron 0.846±0.059 0.867±0.138 0.938±0.066 0.949±0.147
π+ 0.035±0.003 0.028±0.005 0.012±0.001 0.009±0.002 π- 0.102±0.006 0.093±0.013 0.037±0.002 0.033±0.005 p 0.018±0.002 0.012±0.003 0.013±0.001 0.008±0.002
105°
π++π-+ p 0.154±0.009 0.133±0.017 0.062±0.003 0.051±0.007 neutron 0.868±0.063 0.887±0.147 0.951±0.070 0.960±0.238
π+ 0.029±0.002 0.022±0.004 0.009±0.001 0.007±0.002 π- 0.092±0.006 0.084±0.011 0.031±0.002 0.028±0.005
proton 0.012±0.001 0.007±0.002 0.009±0.001 0.006±0.002
120°
π++π-+ p 0.132±0.008 0.113±0.015 0.049±0.003 0.040±0.008 neutron 0.885±0.062 0.899±0.139 0.958±0.070 0.966±0.199
π+ 0.024±0.002 0.019±0.003 0.008±0.001 0.006±0.001 π- 0.083±0.006 0.074±0.009 0.028±0.002 0.025±0.004 p 0.009±0.001 0.008±0.002 0.006±0.001 0.004±0.001
135°
π++π-+ p 0.115±0.007 0.101±0.012 0.042±0.002 0.034±0.006
APPENDIX D – Upgrade and Development of CERF
163
Table D1. Technical characteristics of the BIG PIC.
Parallel Plate Chamber for monitoring
high energetic beams of moderate intensity
Plate spacing 50 mm
Effective Diameter 250 mm
Aperture 185 mm
Effective volume 4.9 litres
Surface area 490 cm2
Electrodes Aluminized mylar foil
Electrode thickness 2.5 mg/cm2
Total Chamber thickness 35.5 mg/cm2
Table D2. 137C source identification, air kerma rate, duration and number of measurements
performed for the July 2002 stability test and the linearity test of the BIG PIC performed in the
SC/RP calibration laboratory.
Stability test Linearity test Source
Air kerma rate
(µGy/h) Duration (s) Duration (s) Repetition
10 1000 10 30
20 1000 10 20
30 1000 10 20
40 1000 10 15
Cs3739
50 1000 10 15
70 1000 10 15
80 1000 10 15
90 1000 10 15
100 1000 10 10
Cs3740
300 1000 10 10
APPENDIX D – Upgrade and Development of CERF
164
Table D2. (continued). 500 300 10 10
700 300 10 10
1000 300 10 10 Cs3609
3000 300 100 2
10000 100 100 2
20000 100 100 2 Cs2045
30000 100 100 2
Table D3. Average count rate versus applied voltage for the determination
of the region of ion saturation of the BIG PIC.
Voltage
(V)
Average count rate
(counts per s)
Voltage
(V)
Average count rate
(counts per s) 10 623.47 ± 5.58 500 1167.43 ± 4.41
30 1058.23 ± 7.27 550 1160.83 ± 4.40
50 1137.08 ± 7.54 600 1167.14 ± 4.41
100 1154.63 ± 7.60 650 1161.59 ± 4.40
150 1165.75 ± 7.63 700 1160.16 ± 4.40
200 1164.86 ± 6.23 750 1162.48 ± 4.40
250 1167.46 ± 6.24 800 1160.88 ± 4.40
300 1167.40 ± 6.24 850 1157.83 ± 4.39
350 1168.63 ± 4.41 900 1161.94 ± 4.40
400 1164.32 ± 4.41 950 1159.96 ± 4.40
450 1165.14 ± 4.41 1000 1160.80 ± 4.40
APPENDIX D – Upgrade and Development of CERF
165
Table D4. Inter-comparison of the two CERF beam monitors (the standard PIC and the BIG PIC) in the hadron
beam at CERF. The total PIC and BIG PIC values are the sum of the readings of the single pulses. The uncertainty
associated to the ratio in the last column has been computed by the usual error propagation formula.
Beam intensity
(PIC/pulse)
Average beam intensity
(PIC/pulse) BIG PIC Total PIC
Total
BIG PIC BIG PIC/PIC
11598 34259
11759 34761
11732
11696 ± 86
34661
35089 103681 2.95 ± 0.05
10507 31154
10648 31553
10293 30510
10461 31038
10661
10514 ± 151
31445
52570 155700 2.96 ± 0.07
7978 23754
7991 23802
8229 24504
8371 24887
8304
8175 ± 181
24722
40873 121669 2.98 ± 0.09
6057 18217
6238 18743
6105 18372
5889 17721
6003
6058 ± 129
18388
30292 91441 3.02 ± 0.10
3951 12060
3994 12220
4025 12314
3927 12023
4008
3981 ± 41
12275
19905 60892 3.06 ± 0.13
1991 6116
2000 6142
2002 6140
2028 6234
1987
2002 ± 16
6101
10008 30733 3.07 ± 0.18
212 656
212 654
213 657
212 655
211
212 ± 1
651
1060 3273 3.09 ± 0.55
APPENDIX D – Upgrade and Development of CERF
166
Table D5. Raw data of the PIC and BIG PIC for different beam intensities, taken at CERF in
August 2003.
Approximate beam
intensity
(PIC/pulse)
Total
PIC-counts
in 5 cycles
Total
BIG PIC-counts
in 5 cycles
BIG PIC counts
/ PIC counts
75 366 1,090 2.98
165 815 2,447 3.00
290 1,469 4,419 3.01
460 2,290 6,907 3.02
645 3,264 9,843 3.02
1,160 5,802 17,472 3.01
1,780 8,884 26,734 3.00
2,480 9,880 29,571 2.99
3,280 16,516 49,426 2.99
4,060 20,349 61,081 3.00
Table D6. Parameters during the measurements of the voltage characteristic curves of the
BIG PIC in the hadron beam at CERF.
Collimator settings Nominal beam
intensity
(PIC/pulse)
Actual beam
intensity
(PIC/pulse)
Beam fluctuation
(%) C3 C5
14,000 13,650 2.14 ± 22 ± 22
12,000 11,650 1.91 ± 19 ± 19
10,500 10,600 2.00 ± 17 ± 17
8,000 8,250 1.36 ± 14 ± 15
6,000 6,100 0.86 ± 12 ± 12
4,000 4,000 0.54 ± 9 ± 10
2,000 1,960 0.21 ± 7 ± 6
1,000 950 0.47 ± 5 ± 4
450 455 0.04 ± 3 ± 3
200 210 0.03 ± 2 ± 2
BIBLIOGRAPHY AND REFERENCES
167
1. R. L. Bramblett, R. I. Ewing and T. W. Bonner, A new type of neutron
spectrometer, Nucl. Instr. and Meth. 9, pp. 1-12 (1960).
2. G.J.H. Jacobs, R.L.P. van den Bosch, Nucl. Instr. and Meth. 175, pp. 483, (1980).
3. R.S. Sanna, US AEC Report HASL-267, (1973).
4. N.E. Hertel, J.W. Davidson, The response of bonner spheres to neutrons from
thermal energies to 17.3 MeV Nucl. Instr. and Meth. A 238, pp. 509-516 (1985).
5. D. E. Hankins, The Substitution of a BF3 Probe for the LiI Crystal in Neutron rem-
Meters, Health Phys. 14, pp. 518-520, (1968).
6. M.P. Dhairyawan, P. S. Nagarajan and G. Venkataraman, Response functions of
spherically moderated neutron detectors, Nucl. Instr. and Meth. 169, pp. 115-120
(1980).
7. J.C. Liu, F. Hajnal, C.S. Sims, and J. Kuiper, Neutron Spectral Measurements at
ORNL, Radiat. Prot. Dosim. 30, pp. 169-178 (1990).
8. W.F. Harvey, F. Hajnal, Los Alamos Report LA 1253s, (1993).
9. G.L. Watkins and G.R. Holeman, The evaluation of an iterative technique’s use in
unfolding neutron spectra data, Health Phys. 15, pp. 535-539, (1968).
10. G.R Holeman, K.W Price, L.F Friedman and R.Nath, Neutron spectral
measurements in an intense photon field associated with a high-energy x-ray
radiotherapy machine, Med. Phys. 4 (6), pp. 508-515, (1977).
11. M. Bricka, et al., Proceedings of a Symposium on Neutron Monitoring for
Radiation Protection Purposes, December 1972, IAEA, Vienna, pp. 279 (1973).
12. M. Mourgues, Proceedings of the Second Symposium on Neutron Dosimetry in
Biology and Medicine, Munchen, September–October 1974, Vol. II, CEC,
Luxembourg, pp. 907 (1975).
13. A.V. Alevra, M. Cosack, J.B. Hunt, D.J. Thomas, and H. Schraube, Experimental
Determination of the Response of Four Bonner Sphere sets to Monoenergetic
Neutrons, Radiat. Prot. Dosim. 23, pp. 293-296 (1988).
14. A. Aroua, M. Grecescu, P. Lerch, S. Prêtre, J. -F. Valley and V. Vylet, Evaluation
and test of the response matrix of a multisphere neutron spectrometer in a wide
energy range Part I. Calibration, Nucl. Instr. and Meth. A 321, pp. 298-304 (1992).
15. P. M. Thomas, K.G. Harrison and M.C. Scott, A multisphere neutron spectrometer
using a central 3He detector, Nucl. Instr. and Meth. 224, pp. 225-232 (1984).
BIBLIOGRAPHY AND REFERENCES
168
16. A. V. Alevra, M. Cosack, J. B. Hunt, D. J Thomas and H. Schraube, Experimental
determination of the response of four Bonner sphere sets to monoeergetic neutrons
(I) Radiat. Prot. Dosim. 23, pp. 249-252, (1988).
17. A. V. Alevra, M. Cosack, J. B. Hunt, D. J Thomas and H. Schraube, Experimental
determination of the response of four Bonner sphere sets to monoenergetic
neutrons (II), Radiat. Prot. Dosim. 40 (2), pp. 91-102, (1992).
18. D.J. Thomas, A. V. Alevra, J. B. Hunt and H. Schraube, Experimental
determination of the response of four Bonner sphere sets to thermal neutrons
Radiat. Prot. Dosim. 54 (1), pp. 25-31, (1994).
19. IAEA, Radiological safety aspects of the operation of electron linear
accelerators, Technical report series no. 188 (1979).
20. P. M.Thomas, K. G.Harrison and M. C.Scott, A multisphere neutron spectrometer
using a central 3He detector Nucl. Instr. and Meth. 224, pp. 225- 232 (1984).
21. A.V. Alevra, M. Cosack, J.B. Hunt, D.J. Thomas, and H. Schraube, Experimental
Determination of the Response of Four Bonner Sphere Sets to Monoenergetic
Neutrons (II), Radiat. Prot. Dosim. 40, pp. 91- 102 (1992).
22. D.J. Thomas, A.V. Alevra, J.B. Hunt, and H. Schraube, Experimental
Determination of the Response of Four Bonner Sphere Sets to Thermal Neutrons,
Radiat. Prot. Dosim. 54, pp. 25- 31 (1994).
23. V.Mares, G. Schraube and H. Schraube, Calculated neutron response of a Bonner
sphere spectrometer with 3He counter Nucl. Instr. and Meth. A 307, pp. 398- 412.
(1991).
24. B. Wiegel, A.V. Alevra and B.R.L Siebert, Calculations of the Response
Functions of Bonner Spheres with a Spherical 3He Proportional Counter Using a
Realistic Detector Model Report PTB-N-21, (1994).
25. V. Mares and H. Schraube, Improved response matrices of Bonner sphere
spectrometers with 6LiI scintillation detector and 3He proportional counter between
15 and 100 MeV Nucl. Instr. and Meth. A 366, pp. 203- 206 (1995).
26. B. Wiegel, A. V. Alevra, M. Matzke, U. J. Schrewe and J. Wittstock,
Spectrometry using the PTB neutron multisphere spectrometer (NEMUS) at flight
altitudes and at ground level , Nucl. Instr. and Meth. A 476, pp. 52- 57 (2002).
BIBLIOGRAPHY AND REFERENCES
169
27. Y.Uwamino, T. Nakamura A. Hara, Two types of multi-moderator neutron
spectrometers: Gamma-ray insensitive type and high-efficiency type, Nucl. Instr.
and Meth. A 239, pp. 299- 309 (1985).
28. P. Goldhagen, M. Reginatto, T. Kniss, J. W. Wilson, R. C. Singleterry, I. W. Jones
and W. Van Steveninck, Measurement of the energy spectrum of cosmic-ray
induced neutrons aboard an ER-2 high-altitude airplane Nucl. Instr. and Meth. A
476, pp. 42- 51 (2002).
29. E.J. Axton, A.G. Bardell, NBS Spec. Publ. 554, pp. 109 (1979).
30. A.V. Sannikov, Report CERN/TIS-RP/IR/94-05, (1994).
31. D. J. Thomas, A. G. Bardell and E. M. Macaulay, Characterization of a gold foil-
based Bonner sphere set and measurements of neutron spectra at a medical
accelerator, Nucl. Instr. and Meth. A 476, pp. 31- 35 (2002).
32. R.J. Henninger, et al., Nucl. Instr. and Meth. 96, pp. 515 (1971).
33. C.K. Wang and T.E. Blue, A neutron spectrometer for neutrons with energies
between 1 eV and 10 keV, Nucl. Instr. and Meth. A 290, pp. 237- 241 (1990).
34. A.Aroua, M. Grecescu, S. Prêtre, and J.F. Valley, Improved Neutron Spectrometer
Based on Bonner Spheres, Radiat Prot Dosimetry 70, pp. 285-289 (1997).
35. V. Vylet, Response matrix of an extended Bonner sphere system,Nucl. Instr. and
Meth. A 476, pp. 26- 30 (2002).
36. M. Grecescu, A. Aroua, M. Boschung, A. Fiechtner, K. Gmür, J. -P.Laedermann,
J. -F. Valley and C. Wernli, Neutron field measurements at the 590 MeV ring
cyclotron of the Paul Scherrer Institute, Nucl. Instr. and Meth. A476, pp. 58- 62
(2002).
37. H. Toyokawa, A. Uritani, C. Mori, N. Takeda, K. Kudo, A Spherical Neutron
Counter with an Extended Energy Response for Dosimetry, Radiat. Prot. Dosim.
70, pp. 365- 370 (1997).
38. S. Yamaguchi, A. Uritani, H. Sakai, C. Mori, T. Iguchi, H. Toyokawa, N.Takeda
and K. Kudo, Spherical neutron detector for space neutron measurement, Nucl.
Instr. and Meth. A 422, pp. 600- 605 (1999).
39. NCRP, Neutron contamination from medical electron accelerators, National
Council on Radiation Protection and Measurements, Report 79, (1984).
40. AAPM, Neutron measurements around high-energy X-ray radiotherapy machines,
Report 19, (1986).
BIBLIOGRAPHY AND REFERENCES
170
41. H. Dinter, and K. Tesch, Determination of Neutron Spectra Behind Lateral
Shielding of High Energy Proton Accelerators, Radiat. Prot Dosim. 42 , pp 5-10
(1992)
42. V.P. Kryuchkov, G.I. Semenova, IHEP Preprint 87–191, Serpukhov, 1987.
43. H.H. Hsu, K.R Alvar and D.G Vasilik, A new Bonner-sphere set for high energy
neutron measurements: Monte Carlo simulation IEEE Trans. Nucl. Sci. NS-41,
pp. 938-940 (1994).
44. V. Mares, H. Schraube, Proceedings IRPA Symposium, Prague, September 1997,
pp. 543 (1998).
45. Angela Mitaroff, Design, Calibration and Tests of an Extended Range Bonner
Sphere Spectrometer, PhD Thesis, CERN (2002).
46. A. Fassò, A. Ferrari, J. Ranft and P. Sala, The FLUKA Manual, 1998.
47. MAB (Münchner Apparate Bau), GmbH, Wallstr.122, Marl,Germany.
48. Mr. C. Hoffmann, Fundamental Software AM-SMCA01, Manual, FREIBERGER
SENSORTECHNIK, (private communication).
49. R. Gold, An Iterative Unfolding Method for Response Matrices, USAEC Report
ANL-6984, Argonne National Laboratory, Argonne, IL (1964).
50. M. Reginatto, P. Goldhagen, MAXED, a Computer Code for Maximum Entropy
Deconvolution of Multisphere Neutron Spectrometer Data, Health Phys. 77 (5),
pp. 579-583, (1999).
51. R. Wilczek and S. Drapatz, A high accuracy algorithm for maximum entropy
image restoration in the case of small data sets. Astron. Astophys.142, pp. 9-12
(1985).
52. A. Corana, M. Marchesi, C. Martini and S. Ridella, Minimizing Multimodal
Functions of Continuous Variables with the "Simulated Annealing" Algorithm,
ACM Transactions on Mathematical Software, 13 (3), pp. 262-280 (1987).
53. J.E Shore, R.W. Johnson Axiomatic derivation of the principle of maximum
entropy and the principle of minimum cross- entropy. IEEE Trans. Info. Theory:
IT-26, pp. 26-37, (1980).
54. M. Matzke, Unfolding of pulse height spectra: the HEPRO program system, PTB
report PTB-N-19 (Braunschweig: Physikalisch- Technische Bundesanstalt)
(1994).
BIBLIOGRAPHY AND REFERENCES
171
55. M. Matzke, Unfolding of particle spectra. In Proc. of Int. Conf on Neutrons in
Research and Industry, Crete 1996. Ed.G. Vourvopoulos, (Proc. SPIE 2867,
pp. 598-607, (1997).
56. M. Matzke, Propagation of uncertainties in unfolding procedures, Nucl. Instrum
Meth. A 476, pp. 230-241 (2002).
57. The MathWorks GmbH Schürmattstrasse 6+8, Gümligen, Switzerland.
58. C.C. Braga, M. S. Dias, Application of neural networks for unfolding neutron
spectra measured by means of Bonner spheres, Nucl. Instr. and Meth. A 476,
pp. 252 –255 (2002).
59. Prof. Armando Foglio Para, Polytechnic of Milan (private communication).
60. N. Metropolis and S. Ulam, The Monte Carlo method, Journal of the American
Statistical Association, 44 (247), pp. 335-341 (1949).
61. A. Fassò, A. Ferrari, J. Ranft and P. Sala: New Developments in Fluka Modelling
of Hadronic and EM Interactions. Proc. of Third Workshop on Simulating
Accelerator Radiation Environments (SARE3). KEK Tsukuba, Japan 7-9 May
1997, Ed. H. Hirayama, KEK Proceedings 97-5, pp. 32 (1997).
62. A. Fassò, A. Ferrari, J. Ranft and P. Sala, Electron-Photon Transport in FLUKA:
Status. Proc. of Advanced Monte-Carlo for Radiation Physics, Particle Transport
Simulation and Applications. Lisbon, Portugal, 23-26 October 2000. Eds. A.
Kling, F. Barao, M. Nakagawa, L. Tavora, P. Vaz, Verlag Springer, pp. 159-164
(2001).
63. A. Fassò, A. Ferrari, J. Ranft and P. Sala, FLUKA: Status and Prospective for
Hadronic Applications. Proc. of Advanced Monte-Carlo for Radiation Physics,
Particle Transport Simulation and Applications, Lisbon, Portugal, 23-26 October
2000. Eds. A. Kling, F. Barao, M. Nakagawa, L. Tavora, P. Vaz, Verlag Springer,
pp. 955-960 (2001).
64. International Organization for Standardization, Reference neutron radiations -Part
2: Calibration fundamentals of radiation protection devices related to the basic
quantities characterizing the radiation field, ISO 8529-2 (2000).
65. International Organization for Standardization. X and gamma reference radiation
for calibrating dosemeters and doserate meters and for determining their
response as a function of photon energy- Part 3: Calibration of area and
BIBLIOGRAPHY AND REFERENCES
172
personal dosemeters and the measurement of their response as a function of
energy and angle of incidence, ISO 4037-3 (1999).
66. International Organization for Standardization. Reference neutron radiations - Part
3: Calibration of area and personal dosimeters and determination of response as
a function of energy and angle of incidence, ISO 8529-3 (1998).
67. International Organization for Standardization. Reference Neutron Radiations:
Part 2: Calibration fundamentals of radiation protection devices related to the
basic quantities characterizing the radiation field. ISO 8529-2 (2000).
68. International Organization for Standardization. Reference Neutron Radiations:
Part 1: Characteristics and Methods of production. ISO 8529-1 (2001).
69. H. J. Brede, M. Cosack, G. Dietze, H. Gumpert, S. Guldbakke, R. Jahr, M.
Kutscha, D. Schlegel-Bickmann and H. Schölermann, The Braunschweig
accelerator facility for fast neutron research. I. Building design and accelerators,
Nucl. Instrum. and Meth. 169 (3), pp. 349-358 (1980).
70. H. Klein, H. J. Barrenscheen, G. Dietze and B. R. L. Siebert, The Braunschweig
accelerator facility for fast neutron research II. Data acquisition and analysis,
Nucl. Instrum. and Meth. 169, pp. 359-367 (1980).
71. S. Guldbakke, S. Dietz, H. Kluge and D. Schlegel, PTB neutron fields for the
calibration of neutron sensitive devices, Strahlenschutz: Physik und Messtechnik,
Vol. Ieds. W. Koelzer and R. Maushart, pp. 240-247 (Cologne: Verlag TUV
Pheinland) (1994).
72. A. Bol, P. Leleux, P. Lipnik, P. Macq and A. Ninane, A novel design for a fast
intense neutron beam, Nucl. Instr. and Meth 214, pp. 169-173, (1983).
73. C. Dupont, P. Leleux, P. Lipnik, P. Macq and A. Ninane, Study of a collimated
fast neutron beam, Nucl. Instr. and Meth A 256, pp. 197-206, (1987).
74. I. Slypen, V. Corcalciuc, A. Ninane and J.P. Meulders, Charged particles
produced in fast neutron induced reactions on 12C in the 45-80 MeV energy range,
Nucl. Instr. and Meth A 337, pp. 431-440, (1994).
75. H. Schuhmacher, H. J. Brede, V. Dangendorf, M. Kuhfuss, J. P. Meulders, W. D.
Newhauser and R. Nolte H., Quasi-monoenergetic neutron beams with energies
from 25 to 70 MeV, Nucl. Instr. and Meth A 421, pp. 284-295 (1999).
BIBLIOGRAPHY AND REFERENCES
173
76. H. Schuhmacher and W.G. Alberts, Reference Neutron Fields With Energies up to
70 MeV for the Calibration of Radiation Protection Instruments, Radiat. Prot.
Dosim. 42, pp. 287-290 (1992).
77. C. Birattari, A. Ferrari, M. Höfert, T. Otto, T. Rancati and M. Silari, Recent
Results at the CERN-EC High-energy Reference Field Facility. In: Proc. Satif-3.
Shielding Aspects of Accelerators, Targets and Irradiation Facilities, Sendai,
Japan, 12–13 May 1997. NEA/OECD, pp. 219–234 (1998).
78. A. Mitaroff and M. Silari, The CERN-EU high-energy Reference Field (CERF)
facility for dosimetry at commercial flight altitudes and in space, Radiat. Prot.
Dosim. 102, pp. 7-22, (2002).
79. A. Aroua, J. Bourges, R.C. Raffnsøe: Beam-Monitoring and Data-Logging for the
H6 Experiment. CERN/TIS-RP/TM/93-20 (1993).
80. R.C. Raffnsøe, Instrumentation for beam experiments in North Hall,
TIS-RP-DC/TM/92-27 (1992).
81. T. Otto and M. Silari, The July/August 1996 run at the CERN-CEC reference
facility, CERN/TIS-RP/TM/96-25 (1996).
82. T. Otto and M. Silari, The May 1996 run at the CERN-CEC reference field
facility, CERN/TIS-RP/IR/96-15 (1996).
83. S. Guldbakke, (private communication).
84. D. Schlegel, TARGET User’s Manual. Report PTB-6.41-1998-1 (Braunschweig:
Physikalisch- Technische Bundesanstalt), (1998).
85. D. Schlegel, (private communication).
86. R. Böttzer, S. Guldbakke, H. Klein, H. Schölermann and H. Schuhmacher.
Problems associated with the production of monoenergetic neutrons, Nucl.
Instrum. and Meth. A 282, pp. 358-367 (1989).
87. A.V. Alevra, M. Cosack, J.B Hunt, D.J Thomas and H. Schraube, Experimental
determination of the response of four Bonner sphere sets to monoenergetic
neutrons (II), Radiat. Prot. Dosim. 40 (2), pp. 91-102 (1992).
88. B. Wiegel, A.V. Alevra and B.R.L. Siebert: Calculations of the Response
Functions of Bonner Spheres with a Spherical 3He Proportional Counter Using a
Realistic Detector Model. PTB- N-21 (1994).
89. C. Birattari, G. Manfredi and M. Silari, Scanning procedures for calibration of
radiation detectors, Rev. Sci. Instrum. 66 (8), pp. 4198-4203 (1995).
BIBLIOGRAPHY AND REFERENCES
174
90. C. Birattari, A. Ferrari, C. Nuccetelli, M. Pelliccioni and M. Silari, An extended
range neutron rem counter, Nucl. Instrum. Meth. A 297, pp. 250 (1990).
91. C. Birattari, A. Esposito, A. Ferrari, M. Pelliccioni and M. Silari, A neutron
survey-meter with sensitivity extended up to 400 MeV, Radiat. Prot. Dosim. 44,
pp. 193. (1992).
92. C. Birattari, A. Esposito, A. Ferrari, M. Pelliccioni and M. Silari, Calibration of
the neutron rem counter LINUS in the energy range from thermal to 19 MeV,
Nucl. Instrum. Meth. A 324, pp. 232 (1993).
93. C. Birattari, A. Esposito, A. Ferrari, M. Pelliccioni, T. Rancati and M. Silari, The
extended range neutron rem counter LINUS: overview and latest developments,
Radiat. Prot. Dosim. 76, pp. 135 (1998).
94. R. Nolte (private communication).
95. R. Nolte, M. S. Allie, P. J. Binns, F. Brooks, A. Buffler, V. Dangendorf, J. P.
Meulders, F. Roos, H. Schuhmacher and B. Wiegel, High-energy neutron
reference fields for the calibration of detectors used in neutron spectrometry,
Nucl. Instr. and Meth A 476 (1-2), pp. 369-373, (2002).
96. Prof. C. Birattari (private communication).
97. A.V. Alevra, Accurate neutron fluence measurements using Bonner spheres. In:
Reactor Dosimetry, ASTM STP 1228. Eds. H. Farrar IV, E.P. Lippincott, J.G.
Williams and D. W. Wehar, pp. 290-299 (Philadelphia, PA:American Society for
Testing and Materials) (1994).
98. Mrs. Jo Carmody, Sales Account Manager, Centronic Ltd. Centronic House, King
Henry’s Drive, Croydon CR9 OBG, England (private communication).
99. S. Agosteo, C. Birattari, E. Dimovasili, A. Foglio Para, M. Silari, L. Ulrici and
H. Vincke, Neutron production from 40 GeV/c hadrons on thin copper, silver and
lead targets in the angular range 30º – 135º, (in press in NIM B).
100. http://www-nds.iaea.or.at/ndsout/endf2784.endf
101. S. Agosteo, C. Birattari, A. Foglio Para, M. Silari and L. Ulrici, Neutron
measurements around a beam dump bombarded by high energy protons and lead
ions, Nucl. Instrum. and Methods A 459, pp. 58-65 (2001).
102. D.H. Perkins, Introduction to high energy physics, Addison-Wesley, 3rd Edition,
Reading, MA, (1987).
BIBLIOGRAPHY AND REFERENCES
175
103. S. Agosteo, C. Birattari, A. Foglio Para, A. Mitaroff, M. Silari and L. Ulrici, 90º
neutron emission from high energy protons and lead ions on a thin lead target,
Nucl. Instrum. Meth. A 476, pp. 63-68 (2002).
104. S. Agosteo, C. Birattari, A. Foglio Para, L. Gini, A. Mitaroff, M. Silari and
L. Ulrici, Neutron production from 158 GeV/c per nucleon lead ions on thin
copper and lead targets in the angular range 30-135°, Nucl. Instrum. Meth. B
194, pp. 399-408 (2002).
105. Particle data group, Review of particle physics, Eur. Phys. J. C 15,
pp. 1-878 (2000).
106. V. Manzari et al., Experiment NA57 at the CERN SPS, J. Phys. G: Nucl. Part.
Phys. 25, pp. 473 (1999).
107. V. Manzari et al., Silicon pixel detectors for tracking in NA57, Nucl. Phys.
A661 (1999).
108. The ALICE collaboration, Technical Proposal, CERN/LHC95-71 (1995).
109. E. Gschwendtner, A. Mitaroff and L. Ulrici, A new calibration of the PIC
monitor in H6, CERN Internal Report TIS-RP/IR/2000-009 (2000).
110. H. Vincke, SC-RP, CERN (private communication).
111. S. Agosteo, E. Dimovasili, A. Fassò and M. Silari, The response of a Bonner
sphere spectrometer to charged hadrons, Proc. of the NEUDOS9 conference,
Delft, The Netherlands, 28 Sept - 3 Oct 2003, Radiat. Prot. Dosim. 110, pp.161-
168 (2004).
112. E. Gschwendtner, A. Mitaroff and T. Otto, Performance tests of the CERF
beam monitor, CERN/TIS-RP/TM/2000-07 (2000).
113. A. Mitaroff and T. Otto, Recombination measurements of the CERF beam
monitor for a 120 GeV/c hadron beam, CERN/TIS-RP/IR/2002-028 (2002).
114. R. G. Jaeger and W. Hübner, Dosimetrie und Strahlenschutz, 1974 (in German).
115. P.R. Bevington and D. Keith Robinson, Data Reduction and Error Analysis for
the Physical Sciences, McGraw-Hill, 2nd Edition, (1992).
116. K. Elsener, M. Heilmann and M. Silari, Verification of the calibration factor of
the CERF beam monitor, CERN/TIS-RP/TM/98-22 (1998).
117. I. Efthymiopoulos, A. Mitaroff and M. Silari, Efficiency measurements of the
Trigger4 beam monitor in H6, CERN/TIS-RP/TN/2002-027 (2002).
118. G. Knoll, Radiation detection and measurement, (John Wiley), 3rd Ed. (2000).
BIBLIOGRAPHY AND REFERENCES
176
119. W. Leo, Techniques for nuclear and particle physics experiments, Springer
(1987).
120. P. K. F Grieder, Cosmic rays at earth : researcher's reference manual and data
book, North-Holland Amsterdam 2001.
121. K. O’Brien, F. Friedberg, F.E. Duke, L. Snyder, E.B. Darden, H.H. Sauer,
The Exposure of Aircraft Crews to Radiations of Extraterrestrial Origin,
Radiat. Prot. Dosim. 45, pp. 145-162 (1992).
122. G. Reitz, K. Schnuer, K. Shaw (Eds.), Proceedings of the Workshop on
Radiation Exposure of Civil Aircrew, Luxembourg, 25–27 June 1991, Radiat.
Prot. Dosim. 48 (1993).
123. P. Goldhagen, Overview of Aircraft Radiation Exposure And Recent Er-2
Measurements, Health Physics. 79(5), pp. 526-544, (2000).
124. J.W. Wilson, J.E. Nealy, F.A. Cucinotta, J.L. Shinn, F. Hajnal, M. Reginatto,
P. Goldhagen, NASA Technical Paper 3524 (1995), National Technical
Information Service, Springfield, Virginia.
125. S. Roesler, W. Heinrich, H. Schraube, Calculation of Radiation Fields in the
Atmosphere and Comparison to Experimental Data, Radiat. Res. 149, pp.87-97,
(1998).
126. A. Kurochkin, B. Wiegel, B.R.L. Siebert, Study of the Radiation Environment
Caused by Galactic Cosmic Rays at Flight Altitudes, at the Summit of the
Zugspitze and at PTB Braunschweig, Radiat. Prot.Dosim. 83, pp. 281-291,
(1999).
127. J.W. Wilson, Overview Of Radiation Environments And Human Exposures,
Health Phys. 79 (5), pp. 470-494 (2000).
128. R. A. Nymmik, M. I. Panasyuk and A. A. Suslov, Galactic cosmic ray flux
simulation and prediction, Advances in Space Research, 17 (2), pp. 19-30
(1996).
129. N.M. Sobolevsky, Space Radiation Shielding and Environment Applications
with the SHIELD Transport Code, Geant4 Space Users Forum, 20-22 January
2003 ESTEC, Noordwijk, The Netherlands.
130. http://liftoff.msfc.nasa.gov/rsa/mir.html
131. International Commission on Radiological Protection, http://www.icrp.org
BIBLIOGRAPHY AND REFERENCES
177
132. International Commission on Radiation Units and Measurements,
http://www.icru.org
133. International Commission on Radiation Units and Measurements, Fundamental
Quantities and Units for Ionizing Radiation, ICRU Rep. 60, Bethesda,
Maryland (1998).
178
Acknowledgements
The work presented in this thesis has been carried out in the framework of the
Doctoral Student Programme at CERN in the SC/RP group in collaboration with the
University of Lausanne (UNIL) and the Swiss Federal Institute of Technology in
Lausanne (EPFL).
I wish to thank Professor Aurelio Bay in the Physics Department of EPFL for
giving me the possibility to get involved as a Doctoral student there. I also thank him
for the support and the freedom he gave me to pursue my diploma in my own way. I
would like to express my gratitude to Professor Jean-Francois Valley at the Institute
of Applied Radiophysics (IRA) in Lausanne for accepting me as his student and for
being always helpful, discrete and extremely kind. I would like to specially thank my
CERN supervisor Dr. Marco Silari for his support since I first came to CERN in 1999
as a summer student. I thank him also for offering me the privilege to spend three
years in the fascinating international environment of CERN. I have to extend my
thanks to the leader of the Radiation Protection Group, Dr. Hans Menzel and to the
Division leader Dr Helmut Schonbacher. I appreciate a lot their trust in me that was
expressed by their discrete presence, their support and help to extend my contract up
to three years.
I feel blessed for having met Professor Pavlos Ioannou in the Physics
Department of the University of Athens. I will always be indebted to him for his
unconditional support and encouragement. He has taught me much physics and an
entire philosophy of scientific thinking. I would have never arrived at this goal
without his constant assistance and confidence. My dear teacher and friend Alberto
Fassò has helped me in solving so many problems that I cannot list here. Above all, I
am grateful to him for being my spiritual guide and for reminding me many times to
keep ‘chin up’. He showed me what kind of physicist I want to become.
Special thanks are addressed to Professors Claudio Birattari and Armando
Foglio Para for providing the unfolding codes and for always making my stay in
Milan very pleasant. Many warm thanks go to Stefano Agosteo for the fruitful
collaboration and discussions. I wish to thank Alfredo Ferrari, Thomas Otto, Stefan
Roesler, and Professor Guizeppe Battistoni for their kind help and stimulating
discussions in many practical issues. I am grateful to Daniel Perrin, Michel Renou,
179
Michel Pangalo and Hubert Muller for always solving any mechanical and
electronics problem that occurred just before the experiments. I feel extremely
grateful to Matteo Magistris for his friendship and for his patience to help me with my
first steps in FLUKA.
My special thanks go to Mario Mueller for his kindness and friendship and for
spending with me many hours on FLUKA discussions. Many thanks also go to
Helmut Vincke for always finding time to answer my questions in his special way. I
would also like to thank Sabine Mayer for sharing overnight shifts with me during my
first CERF runs. I specially thank Anne-Laure Perrot for her kind help in writing the
French abstract of this thesis. I have been gifted with friendship and support from
many people during the course of this thesis. My greatest and sincere thanks to
Emmanuel Tsesmelis, Luisa Ulrici, Marta Sans Merce, Luca Reina, David
Emschermann, and John Antonakis for sharing joys and stress and for keeping always
the spirits up. I owe a HUGE ‘thank you’ to Dr Emmanuel Tsesmelis for his
friendship as well as for his unconditional and constant support through all these
years. I appreciate a lot the fact that despite his overloaded schedule he carefully read
and corrected the manuscripts of my thesis.
I gratefully acknowledge the financial support of IRA through my engagement
as University assistant. I am also indebted to Mrs Jo Carmody of Centronics Ltd,
U.K., for providing me confidential data about the 3He counter used in this work.
The last words are the most heartfelt and they are reserved for my family, for
providing me with the upbringing, opportunities and support that motivated me and
allowed me to pursue my interests. Very special thanks for their unfailing love and
constant encouragement. I would like to express my love to my fiancée Spiros
Kokolakis, for all the years together, for sharing worries and joys, for being a friend
and a companion.
180
CURRICULUM VITAE
EVANGELIA DIMOVASILI
Personal information
Nationality: Greek
Date of birth: 08.06.1974
Place of birth: Athens
Education
2001- 2004 Doctoral Thesis, EPFL, UNIL and CERN (Radiation
Protection Group). Measurements and Monte Carlo
calculations with the Extended- Range Bonner Sphere
Spectrometer at High- Energy Mixed Fields.
1998- 2001 Masters studies in Medical Physics, University of
Heraklio, Greece M. Sc. Diploma with grade
‘Distinction’.
1992- 1998 Studies in the Physics Department of University of
Athens, Greece, B. Sc. Diploma in Physics with grade
‘Very good’.
1991 Diploma of Proficiency in English, Univ. Ann Arbor,
Michigan.
1986- 1992 High School, Amfilohia, Greece
Apolitirio Diploma with grade ‘Excellent’.
181
Research and scientific activities Jan 2001- Aug 2001 Investigation of effects of TiN coating on the hydrogen
permeability of stainless steel and Ni membranes (in the
framework of a collaboration between the Departments
of Physics of the Universities of Athens and
St- Petersbourg, in the field of material science).
Oct 2000- May 2001 Masters thesis, Conceptus Radiation dose and risk from
chest screen-film radiography, performed at the
University Hospital of Heraklio, Crete.
Sept 1999- Oct 2000 •Studies on the biological effects of the E/M radiation
of 50/60 Hz.
•Quality control of Computed Tomograph, SPET
and film-developing machines.
•Medical image analysis (ANALYSE software).
July 1999- Sept 1999 Summer student at CERN with the TIS/RP group
Determination of calibration factors for gamma,
neutron and muon detectors.
May 1999- May 2001 Participation in the data analysis (using the STAPRE
code) for experiments of fission cross-section
measurements for 232Th, 235U, 238U and 239Pu (in the
framework of Greek- Russian collaboration).
Jan 2000 –today Member of the nTOF European Collaboration for High-
Resolution Measurements of Neutron Cross Sections
between 1 eV and 250 MeV.
Sept 1997- Feb 1998 Diploma thesis, Quality control of the computed
tomography and correlation studies between optical
density and Hounsfield units, performed at the
Aretaieion University Hospital of Athens.
182
Participation in Workshops/Conferences
• Ninth Symposium on Neutron Dosimetry, Delft, The Netherlands,
28 September – 3 October 2003.
• Workshop on Radiation Protection Issues Related to Radioactive Ion Beam
Facilities, CERN, Geneva (Switzerland), October 30 - November 1, 2002.
• Special Workshop of Marie Curie Fellows on Research and Training in
Physics and Technology, CERN, Geneva (Switzerland), October 3 - 4, 2002.
• Meeting of the Athens Dentists Association on the topic
‘Radiation Protection in Dental Radiology’, Athens, 30 January 1999.
• Fourth Hellenic Laser Congress under the auspices of the Greek Scientific
Association of Laser Applications in Medicine, Athens, 4 December 1998.
PUBLICATIONS
1. S. Agosteo, E. Dimovasili, A. Fassò and M. Silari
The response of a Bonner Sphere Spectrometer to charged hadrons, Radiat.
Prot. Dosimetry (110), pp.161-168 (2004).
2. S. Agosteo, C. Birattari, E. Dimovasili, A. Foglio Para, M. Silari, L. Ulrici
and H. Vincke, Neutron production from 40 GeV/c hadrons on thin copper,
silver and lead targets in the angular range 30º– 135º, Nucl. Instr. and Meth.
B, (in press).
3. C. Birattari, E. Dimovasili, A. Mitaroff and M. Silari, Calibration and latest
developments of the Extended range Bonner Sphere Spectrometer
(under preparation).
4. T.G. Kazantzeva, Yu.N. Koblik , V.P. Pikul, A.V. Hugaev, B.S. Yuldashev,
P.Ioannou, E.Dimovasili, Secondary electron emission from metal foils
formed by fission fragments (to be published).
5. J.Damilakis, K.Perisinakis, P.Prassopoulos, E.Dimovasili, H.Varveris,
N.Gourtsoyiannis Conceptus Radiation dose and risk from chest screen-film
radiography, European Radiology 13(2), pp. 406-412, (2003).
6. Abramovsky V.A.E., Dimovasili E, Ioannou P. Neutron - induced nuclear
fission cross- sections of 232Th, 235U, 238U and 239Pu from 1 MeV to 200 MeV
183
in quark- gluon model. LII Meeting on Nuclear Spectroscopy and Nuclear
structure, Nucleus 2002, MSU Moscow, pp. 239 (2002).
7. A.V. Khugaev, Yu.N. Koblik., V.P. Pikul, P. Ioannou, E. Dimovasili
About total kinetic energy distribution between fragments of binary fission
In Proceedings of the conference: II Eurasian Conference on Nuclear Science
and its Application , Almaty (Kazakhstan), September 16-19, 2002
Vol. I, pp. 159-162, (2003).
8. A.V. Khugaev, Yu.N Koblik , G.A. Mkrtchan, B.S Yuldashev, P. Ioannou, E.
Dimovasili, About some generalization of calculation algorithm of three-
dimensional imagesin emission tomography. In Proceedings of the conference:
II Eurasian Conference on Nuclear Science and its Application, Almaty
(Kazakhstan), September 16-19, 2002, Vol. III, pp. 11-16, (2003).
9. A.V. Khugaev, Yu.N. Koblik , G.A. Mkrtchan, B.S Yuldashev, P. Ioannou, E.
Dimovasili, Mathematical modeling of three-dimensional images in emission
tomography In Proceedings of the conference: II Eurasian Conference on
Nuclear Science and its Application, Almaty (Kazakhstan), September 16-19,
2002, Vol. III, pp. 119-125 (2003).
10. A. Mitaroff, E. Dimovasili, S. Mayer, C. Birattari, B. Wiegel, M. Silari, H.
Aiginger, Kalibrierung und Experiment eines Bonnerkugel Spectrometer mit
erweitertem Messbereich (Calibration and experiment of an extended range
Bonner sphere spectrometer), Proceedings of the Conference: Strahlenschutz
fur Mensch und Gesellschaft in Europa von morgen Gmuden, 17-21
September 2001.
11. I.E.Gabis, V.A.Dubrovsky. E.A.Denisov, E.Dimovasili, P.Ioannou,
T.N.Kompaniets, A.A. Kurdyumov, K. Ja.Polonsky, I.A.Khazov, Hydrogen
Permeability of Titanium Nitride, International Conference: Interaction of
Hydrogen Isotopes with Structural Materials IHISM'01, Sarov. Book of
abstracts, pp. 75-82, (2001).
12. E.Denisov, E. Dimovasili, P.Ioannou, T.Kompaniets, A.Kurdyumov
Interaction of Molecular Hydrogen with Solid Surfaces in the Lack of
Temperature Equilibrium Gas-Solid, International Conference: Interaction of
Hydrogen Isotopes with Structural Materials IHISM'01, Sarov.
Book of abstracts, pp.170-174, (2001).
13. P.Kipouros,A.Peris, P. Papagiannis , E.Dimovasili and C. Antypas
Monitoring radiotherapy beam characteristics using image processing
184
software, VI International Conference on Medical Physics,Patras, Greece
1999, Physica Medica, Volume XV, N. 3.
TECHNICAL NOTES AND INTERNAL REPORTS
1. E. Dimovasili, D. Macina and M. Oriunno, Energy Deposition in the
Window of the Roman Pot, CERN TS- LEA Note (2004),
(under preparation).
2. E. Dimovasili, A. Ferrari, M. J. Mueller and M. Silari, A proposal for
upgrading the CERF facility for space applications, CERN Technical Note
(2004).
3. E. Dimovasili, M. Magistris and M. Silari, Inter-comparison of the CERF
beam monitors, CERN Technical Note TIS-2003-016-RP-TN (2003).
4. E. Dimovasili and M. Silari, Beam and reference field monitoring during
the 2002 CERF runs, CERN Technical Note TIS-2002-033-RP-TN (2002).
5. E. Dimovasili, S. Mayer, A. Mitaroff and M. Silari, HANDI TEPC
measurements during the 1999, 2000 and 2001 CERF runs, CERN
Technical Note TIS-RP/TN/2002-020 (2002).
6. S. Mayer, T. Otto and E. Dimovasili, Further Investigations of the
Recombination Chamber REM-2 as a Mixed Field Dosimeter at CERF in
October 200, CERN Technical Note TIS-RP/TN/2002-06 (2002).
7. S. Mayer, M. Zielczynski, F. McLay, E. Dimovasili and T. Otto,
Measurements with a recombination chamber made at the CERN-EC high
energy reference field CERF, in August 2001, CERN Technical Note TIS-
RP/TN/2001-03 (2002).
8. The n_TOF Collaboration, Study of the Background in the Measuring
Station at the n_TOF Facility at CERN: Sources and Solutions,
CERN/INTC 2001-038.
9. The n_TOF Collaboration, Measurements of Fission Cross Sections for the
Isotopes relevant to the Thorium Fuel Cycle, CERN-INTC-2001-025.