IAEA-TECDOC-1681 n NEuTrON TrANsmuTATION DOpINg Of sIlICON AT rEsEArCh rEACTOrs

IAEA-TECDOC-1681

Neutron Transmutation Doping of

Silicon at Research Reactors

INTERNATIONAL ATOMIC ENERGY AGENCY

VIENNA
ISBN 978–92–0–130010–2
ISSN 1011–4289
Neutron Transmutation Doping
of Silicon at Research Reactors
 The following States are Members of the International Atomic Energy Agency:

AFGHANISTAN GHANA NIGERIA

ALBANIA GREECE NORWAY
ALGERIA GUATEMALA OMAN
ANGOLA HAITI PAKISTAN
ARGENTINA HOLY SEE PALAU
ARMENIA HONDURAS PANAMA
AUSTRALIA HUNGARY PAPUA NEW GUINEA
AUSTRIA ICELAND PARAGUAY
AZERBAIJAN INDIA PERU
BAHRAIN INDONESIA PHILIPPINES
BANGLADESH IRAN, ISLAMIC REPUBLIC OF POLAND
BELARUS IRAQ PORTUGAL
BELGIUM IRELAND
QATAR
BELIZE ISRAEL
REPUBLIC OF MOLDOVA
BENIN ITALY
ROMANIA
BOLIVIA JAMAICA
RUSSIAN FEDERATION
BOSNIA AND HERZEGOVINA JAPAN
SAUDI ARABIA
BOTSWANA JORDAN
SENEGAL
BRAZIL KAZAKHSTAN
SERBIA
BULGARIA KENYA
SEYCHELLES
BURKINA FASO KOREA, REPUBLIC OF
KUWAIT SIERRA LEONE
BURUNDI
KYRGYZSTAN SINGAPORE
CAMBODIA
LAO PEOPLE’S DEMOCRATIC SLOVAKIA
CAMEROON
CANADA REPUBLIC SLOVENIA
CENTRAL AFRICAN LATVIA SOUTH AFRICA
REPUBLIC LEBANON SPAIN
CHAD LESOTHO SRI LANKA
CHILE LIBERIA SUDAN
CHINA LIBYA SWEDEN
COLOMBIA LIECHTENSTEIN SWITZERLAND
CONGO LITHUANIA SYRIAN ARAB REPUBLIC
COSTA RICA LUXEMBOURG TAJIKISTAN
CÔTE D’IVOIRE MADAGASCAR THAILAND
CROATIA MALAWI THE FORMER YUGOSLAV
CUBA MALAYSIA REPUBLIC OF MACEDONIA
CYPRUS MALI TUNISIA
CZECH REPUBLIC MALTA TURKEY
DEMOCRATIC REPUBLIC MARSHALL ISLANDS UGANDA
OF THE CONGO MAURITANIA UKRAINE
DENMARK MAURITIUS UNITED ARAB EMIRATES
DOMINICA MEXICO UNITED KINGDOM OF
DOMINICAN REPUBLIC MONACO GREAT BRITAIN AND
ECUADOR MONGOLIA NORTHERN IRELAND
EGYPT MONTENEGRO UNITED REPUBLIC
EL SALVADOR MOROCCO OF TANZANIA
ERITREA MOZAMBIQUE UNITED STATES OF AMERICA
ESTONIA MYANMAR URUGUAY
ETHIOPIA NAMIBIA UZBEKISTAN
FINLAND NEPAL VENEZUELA
FRANCE NETHERLANDS VIETNAM
GABON NEW ZEALAND YEMEN
GEORGIA NICARAGUA ZAMBIA
GERMANY NIGER ZIMBABWE

The Agency’s Statute was approved on 23 October 1956 by the Conference on the Statute of the IAEA held at
United Nations Headquarters, New York; it entered into force on 29 July 1957. The Headquarters of the Agency are
situated in Vienna. Its principal objective is “to accelerate and enlarge the contribution of atomic energy to peace,
health and prosperity throughout the world’’.
 IAEA-TECDOC-1681

NEUTRON TRANSMUTATION
DOPING OF SILICON
AT RESEARCH REACTORS

INTERNATIONAL ATOMIC ENERGY AGENCY

VIENNA, 2012
 COPYRIGHT NOTICE

All IAEA scientific and technical publications are protected by the terms of the Universal
Copyright Convention as adopted in 1952 (Berne) and as revised in 1972 (Paris). The
copyright has since been extended by the World Intellectual Property Organization (Geneva)
to include electronic and virtual intellectual property. Permission to use whole or parts of
texts contained in IAEA publications in printed or electronic form must be obtained and is
usually subject to royalty agreements. Proposals for non-commercial reproductions and
translations are welcomed and considered on a case-by-case basis. Enquiries should be
addressed to the IAEA Publishing Section at:

Marketing and Sales Unit, Publishing Section

International Atomic Energy Agency
Vienna International Centre
PO Box 100
1400 Vienna, Austria
fax: +43 1 2600 29302
tel.: +43 1 2600 22417
email: [email protected]
http://www.iaea.org/books

For further information on this publication, please contact:

Physics Section
International Atomic Energy Agency
Vienna International Centre
PO Box 100
1400 Vienna, Austria
email: [email protected]

© IAEA, 2012
Printed by the IAEA in Austria
May 2012

IAEA Library Cataloguing in Publication Data

Neutron transmutation doping of silicon at research reactors. –

Vienna : International Atomic Energy Agency, 2012.
p. ; 30 cm. – (IAEA-TECDOC series, ISSN
1011-4289 ; no. 1681)
ISBN 978-92-0-130010-2
Includes bibliographical references.

1. Nuclear reactors – Research. 2. Neutron transmutation

doping of semiconductors. 3. Silicon. I. International Atomic
Energy Agency. II. Series.

IAEAL 12-00743
 FOREWORD
The IAEA has for many years been conducting several missions to its Member States in order
to assist them in ensuring and enhancing the performance of their neutron irradiation facilities.
One area that can serve as a significant source for commerce and cooperation for these
facilities is the irradiation of materials for added value, including the production of
semiconductor material and gemstone colouration through neutron bombardment and
transmutation, a process in which research reactors (RRs) possess prevalence over chemical
techniques with regard to uniformity and enhanced electro-conductivity. In the Asia-Pacific
region in particular, its research reactors are uniquely positioned to gain substantially through
implementing demanding standards and promoting their burgeoning irradiation and
processing capacity to the robust electronics industry and gemstone dealers. In order to gather
the region’s leading producers and interested research reactor staff, the IAEA sponsored under
a Regional Cooperative Agreement (RCA) the project RAS4026, ‘Adding Value to Materials
through Irradiation with Neutrons,’ which concluded in 2010. The participating Member
States were Bangladesh, China, India, Indonesia, the Republic of Korea, Pakistan, Thailand
and Vietnam, and the project’s objectives were:

— To expand utilization of RRs in the region with a specific focus on neutron irradiation
of materials to increase their material value and thereby generate additional revenue for
the utility;
— To gather valuable information to improve the design and capability of such irradiation
activities in new research reactors being planned by some Member States in the region.

Within the eight participating Member States are eight operational research reactors already
practicing neutron transmutation techniques on various materials, and 33 research reactors
overall of a wide spectrum in terms of design, power rating and age. Also, each is at a
different stage in the management of its irradiation facilities, making the exchange of
experiences very important to achieving safe and effective practices as well as optimized
production in collaboration with end users and industrial partners. Among the several research
reactor utilization fields in the region that need information exchange, silicon doping, gem
colouration and membrane production activities have not been collaboratively organized so
far, but Member States assert high interest because such technologies are growing more
important, especially regarding the semiconductors necessary for a rapidly advancing
electronics sector.

This report is the culmination of the project, and its purpose is to relate guidance on the
preparation, design, execution and quality control of neutron transmutation doping (NTD) for
silicon in a single document. The publication also contains a detailed history of techniques
used at research reactors over the past 50 years and guidelines for quality design and
operation of a Member State’s irradiation facilities. In addition, an extensive database
containing information on experiences world-wide and existing capabilities in using neutron
irradiation for the increase of material value has been included for assessing reactor capacity
for providing irradiation services. Ultimately, the report is intended to improve research
reactor utilization by promulgating the best practices in use today and the quality control
demanded by industrial partnership in the field of neutron irradiation services.

The IAEA officers responsible for this publication were D. Ridikas of the Division of
Physical and Chemical Sciences and N.D. Peld of the Division of Nuclear Fuel Cycle and
Waste Technology.
 EDITORIAL NOTE

The use of particular designations of countries or territories does not imply any judgement by the
publisher, the IAEA, as to the legal status of such countries or territories, of their authorities and
institutions or of the delimitation of their boundaries.
The mention of names of specific companies or products (whether or not indicated as registered) does
not imply any intention to infringe proprietary rights, nor should it be construed as an endorsement
or recommendation on the part of the IAEA.
 CONTENTS
1. INTRODUCTION ......................................................................................................... 1
1.1. Background ......................................................................................................... 1
1.2. Purpose and scope ............................................................................................... 1
1.3. Structure .............................................................................................................. 2
2. WHAT IS NTD? ........................................................................................................... 2
2.1. Definition of NTD ............................................................................................... 2
2.2. Targets of NTD ................................................................................................... 2
3. WHY IS NTD NEEDED? ............................................................................................. 4
3.1. Doping of a semiconductor ................................................................................. 4
3.2. Crystal growth ..................................................................................................... 8
4. HISTORY, PRESENT STATUS AND FUTURE PERSPECTIVES
OF NTD-SI .................................................................................................................. 12
4.1. History of NTD-Si............................................................................................. 12
4.2. Parameters for the future of NTD-Si ................................................................. 17
5. NEUTRON IRRADIATION AND RESISTIVITY .................................................... 19
5.1. Resistivity in semiconductors ........................................................................... 19
5.2. Doping concentration for NTD-Si .................................................................... 21
5.3. Neutron flux and cross-section.......................................................................... 23
5.4. Estimation of the K value .................................................................................. 25
6. TECHNICAL REQUIREMENTS FOR NTD ............................................................ 27
6.1. Radial uniformity .............................................................................................. 28
6.2. Axial irradiation uniformity .............................................................................. 29
6.3. Irradiation accuracy ........................................................................................... 29
6.4. Interfering radiation and NTD .......................................................................... 30
6.5. Residual radioactivity........................................................................................ 31
6.6. Interference with other facilities ....................................................................... 32
6.7. Handling of silicon ingots ................................................................................. 33
6.8. Neutron fluence monitoring .............................................................................. 33
6.9. Washing, decontamination and radioactivity monitoring ................................. 33
7. METHODS FOR UNIFORM IRRADIATION .......................................................... 34
7.1. Radially uniform irradiation .............................................................................. 34
7.2. Axially uniform irradiation ............................................................................... 35
7.3. Effect of initial resistivity.................................................................................. 37
8. NEUTRON FLUENCE MONITORING .................................................................... 37
8.1. Self-powered neutron detector .......................................................................... 38
8.2. Flux monitoring using activation ...................................................................... 39
9. PROCEDURES OF NTD FOR SILICON .................................................................. 42
9.1. Overall procedure of NTD-Si wafer production ............................................... 42
9.2. NTD procedures at the reactor site.................................................................... 44
9.3. Warnings and irradiation control ...................................................................... 49
10. DOCUMENTATION ................................................................................................ 51
11. POSSIBLE EMERGENCIES ................................................................................... 52
REFERENCES ................................................................................................................ 53
ANNEX I. DATABASE OF RESEARCH REACTORS INVOLVED IN NTD ........... 57
GLOSSARY AND ABBREVIATIONS ......................................................................... 93
CONTRIBUTORS TO DRAFTING AND REVIEW .................................................... 95
 1. INTRODUCTION

1.1. Background

For many years, the IAEA has conducted several missions to promote the research and
development activities of research reactors. Among various areas of research reactor
utilization, the neutron transmutation doping of silicon is one well established technology
desired by industry. Silicon semiconductors doped via the NTD method have been widely
used in various industrial fields, especially for high quality semiconductor power devices.

NTD can provide a direct commercial income to research reactors, and many research reactor
operators desire to implement NTD at their facilities. However, because of its commercial
features, detailed up to date techniques have been shared rarely since the last Consultants’
Meeting on the topic organized by the IAEA in 1988. This may have given a negative effect
to industry’s motive for adopting NTD technology for their products. Industry is basically
reluctant to use nuclear reactors as a part of their production process based on worries about
the availability of NTD, since many NTD reactors across the world are aged, some of them
operate with reduced capacity or may even be closed. Therefore, enhancing NTD availability
by involving more research reactor facilities would reduce the burden of potential industrial
partners developing another technology or material that is costly and needs further R&D
efforts. Benefitting not only industry but also the research reactor community, collaboration in
NTD is needed.

The worldwide demand for NTD increased rapidly during the 1980s but decreased in the
subsequent decade due to the development of gas doping technology, and then fluctuated
between 2000 and 2005. Since then, NTD has faced another new circumstance. The demand
for high power semiconductors was increasing constantly given the rapid increase of green
energy technologies such as alternative electricity generation by wind mills and solar cells as
well as hybrid and electric vehicles. The NTD method has several advantages in the
production of high power semiconductors in comparison with other conventional doping
technologies such as ion implantation or the diffusion method.

The IAEA conducted a regional TC project on ‘Adding Value to Materials through Irradiation
with Neutrons’ during 2007 and 2008. This project implemented various activities in Asia-
Pacific region to increase the Members States’ ability in NTD, initiated and contributed
significantly to the preparation of this document. Certainly, this publication will be useful for
other Member States interested in NTD technology.

1.2. Purpose and scope

The general purpose of this publication is to provide comprehensive guidelines in order to

start NTD based Si production or modernize/upgrade an existing NTD facility. This document
is composed of three major components: basic knowledge of NTD, technical requirements
and quality control. The guidance on basic knowledge constitutes a standardization of the
concept and theory. In fact, a large part of the theory in NTD and related neutron irradiation is
well known. However, standardization of the entire process still needs to be documented and
made available in all interested parties. Technical requirements for NTD are a key part in the
design of an irradiation facility. As a matter of fact, these requirements will vary depending
on the characteristics of each irradiation site of a research reactor. Therefore, in this document,
several typical conditions had to be assumed in order to provide more unified description.
Finally, the guidance for quality control is intended to define an efficient and constant control

1
of material irradiation process through all its stages and to ensure a high quality of the final
product.

The present publication is applicable to the following types of work which are necessary to
initiate and implement NTD:

— Obtaining detailed theoretical knowledge for NTD;

— Designing a new irradiation facility for NTD;
— Maintaining good quality in irradiation results;
— Improving existing facilities to increase the available quantity and irradiation quality;
— Obtaining information on the present status and future prospects of the NTD market.

1.3. Structure

The present report consists of 11 sections including references. Sections 2, 3 and 4 present the
definition, necessity, history and future prospects of NTD. The fifth section presents the basic
and theoretical explanation for NTD and neutron irradiation. Sections 6 and 7 present detailed
guidelines for aligning a neutron irradiation facility for NTD. Section 8 presents the
methodology for examining the performance of the facility and neutron irradiation. In section
9, a detailed actual procedure of neutron irradiation for NTD is described. Section 10 presents
the documentation procedure which is essential to ensure the quality of the overall process of
NTD technology.

2. WHAT IS NTD?
2.1. Definition of NTD

Nuclear transmutation means the change of a nucleus to another or multiple other nuclides
through a nuclear reaction. The transmutation is usually instigated by interaction with
neutrons, photons or high energy charged particles. If a nucleus absorbs a neutron, the
resulting compound nucleus may be in an unstable state and undergo some processes to
become stable. If its atomic number is changed in this process, the original atom is changed
into a new element and therefore will have different properties.

Doping is a process of intentionally adding a small amount of an impurity into a material to

improve its properties for dedicated purposes. Particularly, doping in semiconductor
technology is the insertion or creation of impurities in an intrinsic or extrinsic semiconductor
material to improve its electrical properties. An intrinsic semiconductor is a pure
semiconductor without any significant intentionally introduced dopant species present. Thus,
its electrical properties are determined by the properties of the material itself. An extrinsic
semiconductor, on the other hand, is a doped semiconductor.

Therefore, NTD is defined as the process by which neutron irradiation creates the impurity in
an intrinsic or extrinsic semiconductor to increase its value for various uses [1, 2, 3, 4].

2.2. Targets of NTD

The most prominent target or candidate materials for NTD are Si, Ge, GaAs, GaN, GaP [5],
InP [6], InSe [7] and HgCdTe [8]. Among them, silicon and germanium belong to the carbon
group, which is group 14 in IUPAC classification in the periodic table.

2
Each of the elements in this group has four valence electrons in its outermost orbital. The
valence electrons of Si or Ge enter into covalent bonds with the valence electrons of
neighbouring atoms resulting in the formation of a diamond lattice, i.e. two penetrating face
centred cubic Bravais lattices, displaced along the body diagonal of the cubic unit cell by one
quarter the length of the diagonal. The covalent bonds in the diamond lattice make Si and Ge
a semiconductor or in other words an insulator at low temperatures.

Natural silicon atoms are composed of three isotopes, 28Si (abundance: 92.23%), 29Si
(abundance: 4.67%) and 30Si (abundance: 3.10%). Among them, when 28Si or 29Si atoms
absorb a thermal neutron, they are changed into other stable silicon atoms. The absorption of a
fast neutron leads to the direct or indirect (via decay) production of Al or Mg isotopes. The
probability of this type of absorption, however, is lower by about two orders of magnitude.
The absorption of fast neutrons disturbs the goal of Si doping. It is therefore desirable to
suppress it by the provision of a well thermalized neutron spectrum for doping purposes.

In the case of 30Si, thermal neutron capture leads to the unstable isotope 31Si, which undergoes
beta decay. The product of this process is a phosphorous atom, 31P, resulting in the n type
impurity doping in silicon material. The process is as follows without accounting for the
associated antineutrino emission:
30 31
Si + n → Si → (β-) 31P. (2.1)
The 31P atom is a stable pentavalent impurity from group 15 in the periodic table. All of the
component elements from this group have five electrons in their outermost shell, i.e. n type
impurity doping is achieved through this process. The fact that the irradiation of Si with
thermal neutrons results only in a single nuclear reaction and the short half-life of 31Si of only
2.62 h are parameters of crucial importance with respect to the use of the NTD doping
technique on an industrial scale.
When natural germanium is irradiated by thermal neutrons, gallium-31, arsenic-33 and
selenium-34 atoms are formed by electron capture and some β- decay. Among these products,
the yield of gallium generated by the following reaction is the largest:
70 71
Ge + n → Ge → (EC) 71Ga. (2.2)
Since 71Ga is a stable trivalent impurity from group 13 in the periodic table, p type impurity
doping is possible by the neutron irradiation of germanium [9]. The NTD method for
germanium doping demonstrates its superiority in the microdistribution of resistivity (0.7%)
compared to reference resistances prepared by metallurgical techniques (4.3%) [10].
Germanium doped by the NTD method is used for the far-infrared p-Ge laser and several
sensors [11], including extremely low temperature measurement devices like germanium
cryogenic thermistors [12].

Gallium arsenide (GaAs) is an important semiconductor material in the manufacture of

devices such as microwave frequency integrated circuits, infrared light-emitting diodes, laser
diodes and solar cells. GaAs has some superior electronic properties in comparison with
silicon. When GaAs is irradiated by neutrons, germanium or selenium atoms are generated,
and the amount of generated selenium is much larger than that of germanium. Thus, n type
impurity doping in GaAs is possible by neutron irradiation [13, 14, 15].

Gallium nitride (GaN) can be used to produce an n type semiconductor through two doping
processes:

3
 69
Ga + n → 70Ga → (β-) 70Ge, (2.3)
71 72 - 72
Ga + n → Ga → (β ) Ge. (2.4)
HgCdTeSe is a very useful sensor material, and it can be doped into a p type semiconductor
by neutron irradiation using the process:
74 75
Se + n → Se → (β-) 75As. (2.5)
While many significant commodities are produced from germanium, the main target of NTD
is yet silicon. A large commercial amount of NTD is applicable only for silicon because of the
following reasons:

— Silicon is the most widely used semiconductor material;

— The demand for an extremely high quality silicon semiconductor useful for power
devices and sensors has been significant and is still increasing;
— Si ingots are available in the form of extremely pure perfect single crystals;
— Unlike a pure silicon base, compound semiconductors in certain cases are more difficult
to manufacture, primarily in the introduction of the dopant into the correct sub-lattice;
— NTD technology is far superior for the production of a semiconductor with an extremely
uniform dopant concentration.

3. WHY IS NTD NEEDED?

3.1. Doping of a semiconductor
3.1.1. Intrinsic semiconductors

The electrical conductivity of a semiconductor can be controlled over a wide range either
permanently or dynamically. Semiconductor devices are essential in modern technology.
Silicon is instrumental as a matrix material to be doped in a majority of semiconductor
devices, while other materials, although at lesser extent, are used as well.

An intrinsic semiconductor [16], also called an undoped semiconductor, is a pure material

without any significant dopant species present. Elements from the group 14 in the periodic
table such as silicon and germanium have four valence electrons, and they are called
tetravalent elements. In their crystalline structure (diamond lattice), they form covalent bonds
with four adjacent atoms. The periodic lattice of a semiconductor crystalline structure
establishes certain energy bands for electrons. The energy of an electron is confined to one of
these energy bands. Figure 1 shows a simplified representation of the two uppermost bands in
silicon. The lower lying energy bands are of no importance for the electrical conduction

The lower band is called the valence band, in which outer shell electrons are bound to specific
lattice sites by the covalent bonding that constitutes the interatomic forces within the crystal.
The next higher band is called the conduction band, in which electrons are free to migrate
through the crystal. Only these electrons and the corresponding holes (see below) contribute
to the electrical conductivity of the material. The two bands are separated by the band-gap.
Electrons in the valence band must cross the band-gap to reach the conduction band. This
scenario is similar to insulators, but the band-gap for an insulator is usually 5 eV or more,
which is much higher than that of semiconductors.

4
 Conduction band

Electron
energy Eg=1.115 eV

Valence band

FIG. 1. Simplified representation of the electron energy bands of silicon.

In intrinsic semiconductors at room temperature, just as in insulators, very few electrons gain
enough thermal energy to leap over the band-gap. Therefore, intrinsic semiconductors and
insulators have roughly the same conductivity at ambient temperature.

If an electron in the valence band leaps over the band-gap, the electron can leave the specific
bonding site and drift in the conduction band throughout the crystal. The excitation process
not only creates an electron in the conduction band, but also leaves a vacancy called a hole in
the valence band, creating what is called an electron–hole pair. The hole can also move under
the influence of an applied electric field in the opposite direction of the electron. The mobility
of both of these charges contributes to the observed conductivity of the material.

After their formation, both the electron and the hole take part in random thermal motion that
results in their diffusion away from their point of origin. The diffusion coefficient D can be
estimated using the following relationship [16]:

kT
D=μ , (3.1)
e

where μ is the mobility of the charge carrier, k is the Boltzmann constant and T is the absolute
temperature.

If an electric field is applied, the electrons and holes undergo a net migration. In the presence
of a low-to-moderate electric field, the drift velocity ν is proportional to the applied field E.
The mobility for both electrons and holes is defined by:

vh = μ h E , (3.2)
ve = μ e E . (3.3)

In a completely pure intrinsic semiconductor, the number of electrons in the conduction band
must exactly equal the number of holes in the valence band. However, this condition is
practically impossible to achieve due to the very small levels of residual impurities. However,
neglecting the residual impurities, the concentration of electrons and holes in an intrinsic
semiconductor under thermal equilibrium is equal, i.e.

5
 n = p = ni , (3.4)
where n represents the electron concentration in the conduction band, p represents the
concentration of holes in the valence band and ni is the material’s intrinsic carrier
concentration. The intrinsic carrier concentration is dependent on the material and
temperature. Table 1 shows several properties of intrinsic silicon and germanium, including
their hole and electron densities.

TABLE 1: PROPERTIES OF INTRINSIC SILICON AND GERMANIUM [16]

Si Ge
Atomic number 14 32
Atomic weight 28.09 72.60
Stable isotope nucleon numbers 28-29-30 70-72-73-74-76
Density (300 K), g/cm3 2.33 5.32
Atomic density, cm-3 4.96×1022 4.41×1022
Dielectric constant (relative to vacuum) 12 16
Forbidden energy gap (300 K), eV 1.115 0.665
Forbidden energy gap (0 K), eV 1.165 0.746
Intrinsic carrier density (300 K), cm-3 1.5×1010 2.4×1013
Intrinsic resistivity (300 K), Ω·cm 2.3×105 47
Electron mobility (300 K), cm2/V·s 1350 3900
Hole mobility (300 K), cm2/V·s 480 1900
Electron mobility (77 K), cm2/V·s 2.1×104 3.6×104
Hole mobility (77 K), cm2/V·s 1.1×104 4.2×104
Energy per electron-hole pair (300 K), eV 3.62
Energy per electron-hole pair (77 K), eV 3.76 2.96

3.1.2. Doped semiconductors [16]

A semiconductor’s intrinsic electrical properties are often permanently modified by doping.

The amount of dopant added to an intrinsic semiconductor determines the level of
conductivity. Silicon is tetravalent and its normal crystalline structure forms covalent bonds
with four adjacent atoms as shown in Figure 2(a). In this figure, each dot represents a normal
valence electron involved in the covalent bond. One of these covalent electrons can be lost by
thermal excitation, leaving behind an unsaturated bond or hole. In silicon doped by an
element in group 15 of the periodic table, i.e. phosphorus, arsenic or antimony, some silicon
atoms in the lattice are substituted by the dopant atoms as shown in Figure 2(b). Because
there are five valence electrons surrounding a dopant atom, one electron is left over after all
covalent bonds have been formed. This extra electron is only weakly bound to the atom at the
original site. From the perspective of the energy bands, it populates a level within the band
gap but is separated by only a comparatively small energy (i.e. 44 meV for P in Si [17]) from
the conduction band and can be excited at moderate temperature easily into the conduction
band to form a conduction electron without a corresponding hole. At normal temperature,
virtually all such electrons are excited into the conduction band. Since excitation of these
electrons does not result in the formation of holes, the number of electrons far exceeds the
number of holes. In this case, the electrons are the majority carriers, and the holes are the

6
minorityy carriers. As
A the dopaant atoms hhave extra electrons
e to donate, theey are calleed donor
atoms. SSemiconducctor materiaals doped w with donors are called n type semicconductors from the
negativee charge off the majorrity charge carrier eleectrons. Notte that eachh movable electron
within tthe semiconnductor has always beeen donated by an imm mobile positiive dopant ion, and
the dopeed material has a net ellectric chargge of zero.

(a) (b) (c)

FIG. 2. Schematicc representa ation of thee covalent bond of the intrinsic silicon cryystal (a),
doped ssilicon crysttal with n typ
ype impurityy (b) and p type
t impuritty (c).

In nearlly all cases of n type semiconduct

s tors, the do
onor concenntration ND iis much larrger than
the elecctron conceentration in the conduuction band of the intrrinsic materrial. Thereffore, the
numberr of conducttion electron ns is dominnated by thee contributio
on from thee donor, and
d we can
write:

n = ND . (3.5)

As the nnumber of electrons

e in
n the conducction band is i much mo ore than thee intrinsic value, the
numberr of holes becomes mu uch less duee to recomb bination. Att equilibrium
m, the prodduct of n
and holee concentraation p is thee same as foor the intrinsic materiall:

np = ni pi . (3.6)

Intrinsicc silicon haas 1.5×1010 cm-3 electroons and holles (Table 1). Thereforee, as an exaample, if
it is dopped by 10177 cm-3 (abou ut 2 parts pper million)) donor atoms, the dennsity of con nduction
17 -3
electronns n will be 10 cm and a the conccentration of o holes p will
w be in thehe order of 101 3 cm-3.
Becausee the total number
n of charge
c carriiers is now much greatter than in iintrinsic siliicon, the
electricaal conductiivity of thee doped seemiconductor is much h larger thhan the onee of the
correspoonding puree (un-doped d) material.

The chaarge neutrallity in the doped

d materrial is mainttained by a balance beetween the electrons
e
and thee immobilee ionized donor
d impur
urities in th
he n type semiconduc
s ctor. The electrical
e
conducttivity is determined almost excluusively by the flow of o electronss. The resistivity of
doped m material caan be calcu ulated with the dopan nt concentraation and ththe mobility
y of the
majorityy carriers.

A p typpe semicondductor, with

h positive hholes as thee dominant charge carr rrier, is obtaained by
doping a trivalent element in group 13 oof the perio odic table, typically booron, alumiinium or
gallium
m. This type of doping agent
a is als o called an acceptor. If a silicon aatom is replaced by
an accepptor atom as
a shown inn Figure 2(cc)., the misssing valence electron lleaves one covalent
bond ammong the suurrounding silicon atom
ms unsaturaated, i.e. one electron iis missing from
f one

7
of the four covalent bonds. Analogous to the description given above, for donors the energetic
levels of acceptors are located within the band gap but in this case closely above the valence
band edge. This vacancy represents a hole similar to that left behind when a normal valence
electron is excited to the conduction band. If an electron is captured to fill this vacancy, it
participates in a covalent bond, albeit one not identical to the bulk of the crystal because of
the trivalent atom. Consequently, the electron is slightly less firmly attached than a typical
valence electron. Some electrons are always available to fill the holes created by acceptors by
virtue of thermal excitation. These electrons come from other normal covalent bonds
throughout the crystal and therefore leave holes behind in the valence band, meaning that an
extra hole is created in the valence band for every acceptor atom.

If the concentration NA of acceptors is sufficiently large compared to the intrinsic

concentration of holes pi, then the number of holes is completely dominated by the
concentration of acceptors, or:

p = NA. (3.7)

By the equilibrium constant discussed above, the equilibrium number of conduction electrons
becomes much smaller. In this material, holes are the majority carrier and dominate the
electrical conductivity. The filled acceptor sites represent fixed negative charges that balance
the positive charges of the majority holes.

Semiconductors doped with donor impurities are called n type, while those doped with
acceptor impurities are known as p type. The n- and p type designations indicate which charge
carrier acts as majority carrier. The opposite carrier is called the minority carrier, which exists
due to thermal excitation at a much lower concentration than the majority carrier.

Degenerately, or very highly, doped semiconductors have conductivity levels comparable to

metals. From the point of view of the energy bands, a degeneracy means that the majority
carrier levels interact and form a band on their own which merges with the conduction band in
case of n type semiconductors or the valence band in case of p type semiconductors. Like in
metals, in degenerate semiconductors occupied and un- occupied energy levels are no longer
separated by a band-gap. Degenerate semiconductors are often used in modern integrated
circuits as a replacement for metal. Superscript plus and minus symbols are often used to
denote relative doping concentration in semiconductors. For example, n+ denotes an n type
semiconductor with a high, often degenerate, doping concentration. Similarly, p− indicates a
very lightly doped p type material.

Even at degenerate levels of doping, the concentration of dopant with respect to the base
material is much lower. The doping concentration for silicon semiconductors may be in the
range from 1013 cm-3 to 1018 cm-3. Doping concentrations above 1018 cm-3 is considered
degenerate at room temperature. Degenerately doped silicon contains dopant atoms in the
order of parts per thousand of silicon.

3.2. Crystal growth

3.2.1. Czochralski process

The Czochralski (CZ) process is a method of crystal growth using a melted material [18]. The
most important application of this process may be the growth of large cylindrical ingots of
single crystal silicon (CZ-Si). High purity, semiconductor grade silicon, i.e. having only a few

8
parts peer million of impurities is melteed in a cruucible, whicch is usuallly made off quartz.
Subsequuently, a seeed crystal is
i touched tto the molteen silicon and
a a crystaal ingot is pulled
p by
slowly wwithdrawingg the seed as
a shown in Figure 3 annd Figure 4..

For hom
mogeneity, the
t seed cry ystal and thee crucible arre rotated. The
T small siingle crystaal seed is
mountedd on a rotatting axis and
d put in conntact with th
he surface of the liquid.. The rotatio
on speed
can varyy from 2 too 50 rpm (rotations perr minute). TheT temperature gradieents are adjjusted in
such a w
way that cryystal growthh occurs at tthe surface of the seed. Crystals w
with diameteers up to
50 cm aare obtainabble by this method
m [20] .

FIG. 3. The
T phases of
o crystal grrowth durin
ng the Czoch
hralski proccess [19].

F
FIG. 4. A 20
00 mm Czocchralski cryystal pulled from
f a meltt. [19].

9
As the crystal is grown, the impurity concentration including the dopants will increase in the
melt due to segregation, and therefore some portion of the impurities is incorporated into the
crystal. The temperature profile of the whole system will also change. All these factors
influence the homogeneity and the radial and lateral doping level of the crystal. In this process,
an especially important impurity is interstitial oxygen. In general, the concentration of oxygen
increases from ‘head’ to ‘tail’.

Semiconductors made through this method are mainly used in low power semiconductor
devices such as personal computers. In this method, the crucible is a major source of pollution.
At the melting point of silicon, 1418ºC, the crucible always chemically attacks and pollutes
the melt. Due to impurities from the crucible and dissolved oxygen, the resulting
semiconductor has relatively low purity and quality. Also, it is difficult to obtain a
semiconductor with a resistivity larger than several tens of Ωcm by this method.

Impurity atoms can be added to the molten intrinsic silicon in precise amounts for the purpose
of doping. Since the dopant distribution in the axial direction varies, a proper part of the
grown crystal is chosen and cut. In the radial direction of the crystal, the semiconductor
grown by this method shows variations in doping concentration in a rather regular pattern
known as striation. This is because the temperature of the crystal surface varies by convection
of the molten silicon in the crucible. Since the temperature gradients in the crucible cannot be
avoided, there will be some convection in the liquid silicon.

In order to suppress the natural convection and pollution within the crucible, the
semiconductor crystal can also be grown under a strong magnetic field resulting in the
product called Magnetic-Field-Applied Czochralski Silicon (MCZ-Si) [21]. Since the molten
silicon has metallic properties, it undergoes convective motion under the magnetic field. The
intensity of the applied magnetic field is about 50–4000 G. In this case, it is possible to
compensate for the original convection of the CZ method. Therefore, non-uniformity of
resistivity in the radial direction can be much smaller than with the CZ-Si, although the
dopant distribution in the axial direction is as large as in the CZ type. Since a high purity
silicon crystal with about several thousand Ω•cm can be obtained by this method, if NTD is
applied to MCZ-Si, a silicon semiconductor with a very uniform dopant distribution is
obtainable. Floating zone silicon (FZ-Si) in several power devices could be replaced by MCZ-
Si. MCZ-Si is also used for charge coupled devices (CCDs) and highly integrated circuits.

3.2.2. Floating zone method

The floating zone method is a crystal growing method that ensures a very high purity. When
the molten liquid silicon is in contact with solid silicon, the impurity atoms will diffuse in a
process called segregation to the liquid region since they prefer to sit in the liquid phase. Thus,
by passing a boule of silicon through a thin section of a furnace very slowly such that only a
small region of the boule is molten at any time, the impurities will collect in the molten region
and migrate with it as the boule is moved through the heat zone. This is called the zone
refining method developed by William Gardner Pfann, and the concept is shown in Figure 5.

10
 FIG. 5. Concept of the floating zone method [22].

The narrow part of the ingot is heated by a coil attached to a high frequency generator. The
local induction heating forms a narrow molten zone. By slowly moving the heating coil
upward, a slow displacement of the molten zone takes place followed by solidification. Once
the whole polycrystalline ingot has been passed through by the molten zone, a single crystal
ingot is obtained. By repeating this process one can obtain an extremely pure crystal.

The ratio of the impurity concentration in the liquid phase cL to that in the solid phase cs is
represented by the segregation coefficient K, i.e.:

cL
K= . (3.8)
cs

In general, K is much larger than unity because impurities are more soluble in the liquid than
in the solid phase. When the impurity concentration in the starting ingot is c0, the initial
molten zone also has impurity concentration c0, but the first solid segment obtained after the
passage of the molten zone has the concentration given by:

c′=c0/K. (3.9)

Since K>1, the solid segment has a lower impurity concentration than at its initial state. As the
molten zone progresses, it becomes enriched with impurities until its impurity concentration
reaches the product of K and c0. Further motion of the molten zone does not give additional
purification.

11
After the first passage of the molten zone is completed, the ingot is cooled down. A second
passage can be started from the bottom of the ingot with the lowered impurity concentration.
Many passages can be made, starting always at the bottom of the ingot. Eventually the
impurity concentration profile reaches a steady state which cannot be improved upon by
further passages [20].

Since the melt never comes into contact with anything but vacuum or inert gases, there is no
incorporation of impurities in the melt. This is especially vital considering oxygen impurities,
which cannot be avoided with CZ crystals. FZ crystals therefore are always used when a very
low oxygen concentration is important, but FZ crystal growth is limited in terms of crystal
diameter.

Silicon crystals are basically grown by one of the two methods; CZ-Si is by the Czochralski
method and FZ-Si is by the floating zone method. While FZ-Si has much a higher crystal
quality than CZ-Si, CZ-Si occupies more than 90% of the silicon market. The diameters of
these crystals have become larger as the process has been refined. The largest crystal diameter
of the major CZ-Si market recently jumped from 8 to 12 inches. For the case of FZ-Si, the
major size had been 5 and 6 inches for several years, but recently 8 inch FZ-Si has been seen.

For the case of CZ-Si, the dopant is mixed with silicon before crystal growing. Then, the
dopant distribution in the silicon crystal is less uniform than in FZ-Si due to the
aforementioned striation. Also, the longitudinal variation of dopant concentration gradually
increases or decreases over the axial length of CZ-Si, which limits the usage of the entire
ingot for applications with a tight resistivity tolerance. The incorporation of oxygen in CZ-Si
is much higher than in FZ-Si as well.

FZ-Si shows much higher quality than CZ-Si, and gas diffusion doping for FZ-Si (DW-Si)
results in much more uniform resistivity than for CZ-Si as well. Currently, NTD is applied
mainly to FZ-Si (NTD-Si) in order to attain more uniform resistivity than DW-Si. As the
technology developed, resistivity uniformity of DW-Si was enhanced, which was the reason
that NTD demand decreased during the 1990s. NTD-Si holds about 20% of the current FZ-Si
market.

4. HISTORY, PRESENT STATUS AND FUTURE PERSPECTIVES

OF NTD-SI
4.1. History of NTD-Si

4.1.1. The beginning stages of NTD

The possibility of producing silicon semiconductors with perfectly uniform phosphorus

dopant distribution by neutron transmutation was first identified and described by Karl Lark-
Horovitz in 1951 [23]. In 1961, Morris Tanenbaum and A.D. Mills at Bell Telephone
Laboratories carried out the first NTD experiment [24], by irradiating small pieces of silicon
with neutrons while measuring the phosphorus distribution.

While the NTD technique had been known and practiced for a decade in research laboratories,
it was not commercially used to any significant extent until the mid-1970s. In 1973, a German
high power device manufacturer used NTD-Si for the production of thyristors, which was a
practical commercial breakthrough [25]. Due to the superiority of the resulting device

12
characteristics, between 1974 and 1976, NTD was offered by a number of research reactors in
the USA, England and Denmark, and quickly the amount of irradiated silicon reached several
tonnes.

At the Danish reactor DR2 of RISØ National Laboratory, the first industrial production of a 2
inch diameter silicon ingot began in 1974 in cooperation with the Danish company Topsil of
Frederikssund [26]. They were called T-silicon, which was the registered trademark of
commercial transmutation doped silicon. Two inch silicon ingots were irradiated in 53 mm
diameter aluminium cans in a horizontal graphite stringer in the reactor’s thermal column.
Upon the final shutdown of DR2 in November 1975, about 100 kg silicon had been
irradiated. Afterwards an irradiation facility for a 3 inch silicon crystal was constructed for the
DR3 reactor, and its commercial service started in 1976. Additional irradiation facilities were
constructed in 1977 and installed in the graphite reflector surrounding the heavy water tank.
In 1997, they installed an additional 5 inch irradiation facility [27]. DR3 continued
irradiations of 3, 4 and 5 inch ingots and produced a large amount of the NTD-Si supply until
it was permanently shut down in 2000 because of a small leak of heavy water.

In the United Kingdom, NTD was first undertaken by the Materials Test Reactors DIDO and
PLUTO at the Harwell site of the United Kingdom Atomic Energy Authority (UKAEA) in
1975 [28]. Following the first successful trial for a silicon irradiation in 1975, the amount of
NTD production increased from 2 t during 1976 to a level of around 30 t in 1985. According
to the Harwell reactor’s report in 1987, the demand for NTD had rapidly risen, and that the
application fields of NTD Si had been extended. In order to satisfy an even higher demand in
the future, the Harwell reactors had prepared for a capacity of about 35 t per annum [29].
Then in early 1990, the DIDO and PLUTO reactors at the Harwell site were shut down. At
that time, about 206 t of the silicon had been doped in these reactors over the course of 15
years [30]. To compensate for the shutdown of the Harwell reactors, a joint venture was set up
between AEA Technology (the trading name of the UKAEA) and the Centre for the Study of
Nuclear Energy (SCK/CEN) involving silicon irradiation in the Centre’s BR2 reactor in Mol,
Belgium. A silicon irradiation facility named Silicon Doping by Neutron Irradiation
Experiment (SIDONIE) with an annual capacity of 28 t was designed and installed for this
reactor in early 1992 [31, 32].

In Switzerland, the first irradiations in the SAPHIR reactor were carried out as early as 1975.
The total annual quantities remained modest at less than 200 kg until about 1990, but in 1992,
NTD activity was stepped up to about 2 t of silicon [33].

MURR, a 10 MW open pool reactor at the University of Missouri, USA, has performed NTD
actively, starting with several test runs in 1975. It was reported that the reactor’s capacity in
1978 was on the order of 15 t with 50 Ω•cm resistivity, and that it could be enhanced to 60 t
with some modification [34].

In Japan, NTD tests began in 1975 in compliance with a local semiconductor company’s
request, and commercial service began in 1977. Irradiation rigs in the JRR-4 reactor were
developed for 2.5, 4 and 5 inch silicon ingots of 45 cm length. JRR-2 had irradiated silicon up
to 3 inch ingots from 1983 till its shutdown in 1996.

In the Russian Federation, NTD semiconductor technology was first realized in the early
1970s based at the WWR-C nuclear reactor developed at the Obninsk Branch of the Karpov
Scientific Research Institute of Physical Chemistry. In 2003, several tonnes of NTD-Si and
other semiconductor materials were processed at this facility along with simultaneous

13
development and testing of different NTD procedures, while several other institutes were also
participating in NTD [35].

HIFAR in Australia had irradiated a relatively large amount of silicon ingots for many years
[36] until its final shutdown in 2007.

In Seibersdorf, Austria, the ASTRA, a 10 MW swimming pool type reactor was in parallel to
scientific applications intensively used for Si doping until its final shut down in 1999.
According to an Austrian press release in 1991 about 5 % of the global demand of NTD Si
was produced in Seibersdorf .In the former German Democratic Republic the RFR, a 10 MW
WWR-SM reactor of the former USSR design was used for isotope production but also for Si
doping. The reactor was finally shut-down in 1991 and has been decommissioned meanwhile.

4.1.2. Major reactors for NTD at present

BR2 in Belgium has been an active reactor for NTD irradiation. This reactor underwent a
refurbishment shutdown for about two years and then resumed operation in April 1997 [37].
The irradiation facility for silicon installed in 1992 was for ingots of 4 and 5 inches in
diameter. In anticipation for the shifting of the crystal diameter to larger sizes, they recently
realized a larger irradiation facility [32, 38].

JRR-3M in Japan has a 17 cm diameter irradiation hole for NTD. The irradiation of 5 inch
ingots began in 1990, and it was reconstructed to irradiate 6 inch ingots throughout 1994 and
1995. It can irradiate ingots up to 60 cm in length. The amount of NTD-Si produced was
about 3.7 t at JRR-3 and about 0.7 t at JRR-4 in 2006 [39].

In the Republic of Korea, the HANARO reactor at the Korea Atomic Energy Research
Institute (KAERI) has two vertical holes in the heavy water reflector region for NTD. The
development of an NTD irradiation facility began in 2001, and commercial service at a
smaller hole began at the end of 2002 using 5 inch ingots. In 2005, service for 6 inch ingots
also began at the same hole. The irradiation record in 2007 totalled about 15 tonnes in spite of
the reduced operation time of the reactor. From 2007, 6 and 8 inch irradiation facilities have
been studied at a larger hole [40].

OPAL, a new reactor in Australia has six vertical irradiation holes in the heavy water reflector
vessel for 5 inch (one hole), 6 inch (three holes) and 8 inch (two holes) ingot irradiations. The
possible ingot diameters are 4-8 inches. It uses a combination of water bearing and drive
rotation utilizing reactor pool water. OPAL achieved its full power rating of 20 MW at the
end of 2006 [41].

South Africa’s SAFARI-1 research reactor has a silicon irradiation facility installed in the
pool side region next to the reactor core. The commissioning of the facility took place in 1992,
and it has been upgraded and improved to meet the growing international demand for NTD-Si.
An initial test irradiation was performed for 4 inch ingots. In 2008, it was redesigned to
accommodate a maximum diameter of 6 inches and a maximum length of 600 mm [42, 43].

FRM-II in Germany was successfully commissioned in 2004 and 2005. An automatic doping
system in FRM-II was designed to accommodate up to 8 inch silicon ingots exhibiting a total
stack length of 500 mm. FRM II offers a single irradiation channel. In order to allow 5 inch
and 6 inch ingots to be irradiated suitable Al-spacers are used to minimize the gap within the
standard 8 inch irradiation basket. Irradiation service began in 2007 [44], and that year, a total

14
of 3.8 t was irradiated by the semiautomatic silicon doping facility, which increased its output
to 9.8 t in 2008 [45] and about 15 t in 2010.

4.1.3. Other reactors related to NTD

Other reactors that have some activities related to NTD or plan to carry out similar irradiation
services are as follows:

— MITR-II (5 MW): Massachusetts Institute of Technology in US [46]

o Two horizontal through ports
o 4TH1-3: 4 inch crystals, 6TH1-2: 4, 5, 6 inch crystals
o 400 mm long magnesium cans
— MURR (10 MW): Missouri University Research Reactor in US
— IRT-T (6 MW): Tomsk Polytechnic University in the Russian Federation [47]
o 150 mm in diameter, horizontal GÉK-4 experimental channel
o Reciprocal motion at a velocity of 270 mm/min passing through Be reflector
o In 1995, capacity of ingots with 12.7 cm in diameter: 4.5 t/yr
— JEEP II (2 MW) in Norway [48] [Annex]
— HFR (45 MW) in the Netherlands [48] [Annex]
— MARIA (30 MW) in Poland [Annex]
— SPR IAE (3.5 MW), HWRR-II (15 MW) and CARR (60 MW): the Institute of Atomic
Energy (CIAE) in China [Annex]
— SPRR-300 (3 MW): Tsinghua University in China
— HFETR (125 MW) [Annex], MJTR (5 MW): the Nuclear Power Institute of China
— CIRUS (NTD test, 40 MW) in India
— ETRR-2 (22 MW) in Egypt [49]
o Conception for 3 and 5 inch crystals with resistivity lower than 50 Ωcm, using two
vertical holes in the thermal column
— LVR-15 (10 MW) in the Czech Republic [Annex]
— TRR (5 MW) in the Islamic Republic of Iran
o Conception for ingots up to 10 cm in diameter [50]
— IEA-R1 (5 MW) in Brazil [Annex]
o Suitable for ingots of 5 inches up to 500 mm in length, used for low resistivity Si of
30 Ωcm, capacity 1.2 t/year [51, 52]
— OSIRIS (70 MW) and ORPHEE (14 MW) in France [Annex]
o 4 irradiation channels for ingots of 4 and 6 inches [31]

The reader should note that the above list should be considered as not fully complete, i.e.
other research reactors involved in NTD of Si technology do exist according to the existing
information available through the IAEA RRDB [48], i.e. in Kazakhstan, the Russian
Federation, Ukraine, USA, and Uzbekistan.

4.1.4. NTD conferences and meetings

The first formal conference associated with neutron transmutation technology was organized
in 1976 by John Cleland and Richard Wood. It had become clear that the NTD process would
be implemented on a commercial scale to prepare uniformly doped starting material for the
power device industry. The Second International Conference on Neutron Transmutation
Doping in Semiconductors was held on 23–28 April, 1978 at the University of Missouri in
Columbia, Missouri. Various aspects of NTD-Si technology were discussed at this conference
including irradiation technology, radiation induced defects and the use of NTD-Si for device

15
applications. The Third International Conference on Neutron Transmutation Doped Silicon
was held in Copenhagen, Denmark on 27–29 August, 1980. The Fourth International Neutron
Transmutation Doping Conference was held at the National Bureau of Standards in
Gaithersburg, Maryland on 1–3 June, 1982. In addition to the conferences, an IAEA
Consultants Meeting on Silicon Transmutation Doping Techniques and Practices was held at
the Institute of Atomic Energy, Otwock-Świerk, Poland, during 20–22 November 1985,
resulting in the publication of IAEA-TECDOC-456 in 1988 [4]. Unfortunately, this
publication is neither available electronically nor for additional printing.

From the related literature, we can confirm that in the early stage of NTD technology, neutron
irradiations not only for FZ-Si but also Ge, GaAs, etc., had been examined. Also, as a possible
way to overcome resistivity variations, mainly axial resistivity variations in Czochralski
grown silicon ingots, additionally controlled neutron irradiation for CZ crystals had been
proposed.

Most probably due to the growing commercial importance of Si doping to the research
reactors and to the related non-disclosure agreements between the reactors and the Si
suppliers no more topical meetings dedicated exclusively to NTD of semiconductors were
held in more recent years. However, individual contributions dealing with NTD continue to be
presented during the following conferences:

— International Conference on Isotopes (ICI);

— Research Reactor Fuel Management (RRFM);
— Meeting of the International Group of Research Reactors (IGORR);
— International. Conference on Research Reactors: Safe Management and Effective
Utilization.

4.1.5. A shift of the worldwide NTD capacity and demand

The estimated worldwide production rate of NTD-Si in 1982 was approximately 50 tonnes
per year [53]. According to future prospects at that time [54], with slow economic growth,
demand for NTD-Si in the 1990s could be 75–80 tonnes per year, whereas with higher world
economic growth and the reasonable probability of innovation in applications, demand could
be in the range 100–120 t. Since it was impossible to predict the exact demand, different rates
of market growth, 3, 5 and 7% per annum were applied. It was expected that the demand
could increase to about 150 t until 2000, at which point the 7% increase rate of the market
growth was applied. Considering the irradiation capacity of reactors, while about 50 t was
estimated in 1982, a potential capacity up to 110 t was expected. Subsequently, beyond the
early 1990s, it was expected to decrease since the primary reactors responsible for production
would be approaching an average age of more than 30 years [54].

The actual demand for NTD in 1991 reached about 160 t with a growth rate about 10% per
annum, which was higher than the above prediction, yet this increased demand was
accommodated by the research reactors. Following 1992, the demand for NTD decreased to
70 t in 1999 because of developments of MCZ- and improved FZ-Si technologies that could
substitute for the NTD process. The demand for NTD rebounded after 2000, and currently it is
more than 100 tonnes per year with a growing trend that is roughly 20% of FZ-Si production.

In 1982, since a shortage of irradiation capacity was predicted, a reactor dedicated to silicon
irradiation was examined [54]. An average annual cost of operating a medium power research
reactor, without the capital cost or depreciation of the facility, was estimated to be $6.5 M,

16
which consisted of $5 M for operation and $1.5 M for fuel. Assuming 6–7 cents/g for the
price charged for irradiation, about 100 t/year would have to be irradiated to support the
upkeep of the reactor alone. Since a tenth of this amount was the practical quantity at that
time, it was concluded that silicon irradiation must be carried out in conjunction with a large
contingent of other activities.

The current NTD market is dominated by 5 and meanwhile even more important 6 inch ingots.
2.5 inch ingots and 3 inch ingots are rarely found in the current market. The share of 4 inch
ingots has been decreasing drastically. Recently, the demand for irradiation of 8 inch ingots
has grown, and its market share could expand in the future.

At present, since the typical diameter of silicon ingots has become much larger, a large
amount of silicon can be irradiated at an irradiation site. It may not be easy for many existing
reactors to make new larger irradiation holes, but their inclusion in the design of new reactors
would be possible and will be necessary to satisfy the demand of the semiconductor industry.
As found at, i.e. OPAL, but also at FRM II and BR2 — all of them providing irradiation
facilities for ingot diameters up to 8 inch — a large scale NTD facility may be possible while
satisfying other needs of a research reactor. After all, the construction of a large irradiation
capacity for NTD could be profitable, though it is closely dependent on industrial demand for
NTD-Si.

The CZ-Si market has jumped from 8 to 12 inches. Therefore, the next wave of NTD ingots
would be 12 inches rather than in between 8 and 12 inches such as a 10 inch ingot unless
special processing lines for NTD ingots and wafers are justified. It seems that the industry is
now developing technology for 18 inch CZ-Si production but NTD for 18 inch ingots may be
difficult to satisfy not only in terms of irradiation uniformity but also accommodation at
reactors.

Besides the increasing diameter of Si ingots to be irradiated it has to be noted that an

increasing demand for high target resistivity up to >1000 Ω•cm NTD Si arose in the last few
years. On the one hand this development is connected with an increasing capacity because
high targeted resistivity is achieved by short irradiation times. On the other hand for this type
of NTD Si high quality neutron sources providing excellent thermalized neutron spectra are
required because the effect of extended irradiation defects by fast neutrons on the electrical
properties of a semiconductor is much more important at high resistivity

4.2. Parameters for the future of NTD-Si

4.2.1. Future demand and supply

As mentioned above, the worldwide demand for NTD had increased rapidly up to the early
1990s. During the 1990s, the demand decreased due to the development of other competitive
technologies, and remained stagnant until the mid-2000s at about 100 t/year. However, since
then NTD has been facing a new circumstance. The demand for high power semiconductors
may increase rapidly according to the rapid increase of alternative and much wider use of
‘green energy’ technologies.

One significant example is the quickly expanding production of hybrid electric vehicles.
Interest in electric vehicles, including plug-in hybrid cars and fuel cell engines, is also surging.
In addition to transportation, electricity generation by wind, solar and fuel cell systems is
increasing rapidly. These advances are, in principle, motivated by efforts to decrease the

17
discharge of air pollution, especially carbon dioxide, which is supposed to be the major cause
of global warming. Furthermore, various activities to increase efficiency in electricity
utilization like smart grid projects have been proposed. Each of these fields requires high
quality power devices, and NTD can supply the best base material to manufacture them.

From the aspect of supply, the ingot diameter has become larger, and recent new reactors are
apt to possess a large NTD capacity. Irradiation of large diameter ingots allows for a large
NTD capacity from the same number of irradiation holes. The 6 inch ingot has become the
major product in the NTD market, and the share of 8 inch ingots will gradually increase.
Meanwhile, shutdowns of aged research reactors serving NTD will continue. In 1999 ASTRA
in Seibersdorf, Austria, as well as in 2000 DR3 in Denmark were shut down, and then the R2
reactor in Sweden was shut down in 2005. None of them was replaced so far. Although many
research reactors currently serving NTD are aged, world-wide NTD capacity will likely
increase because replacement reactors, even at smaller number than shutdown facilities, will
have a larger and better optimized capabilities than their predecessors.

A very rough but more positive projection of NTD is as follows: in 2005, the worldwide
production of new cars was about 60 million, and could be more than 100 million in 2030. If
50% of new cars will contain hybrid, electric or fuel cell engines in 2030, then the number of
new cars equipped with electric motors is assumed to be about 50 million. If one 6 inch NTD
silicon wafer is used per car, since a silicon wafer requires about 1 mm thickness including
the sawing part, NTD-Si of about 2,000 ton per annum would be needed to satisfy this
demand. It is not certain whether this future demand will be met by reactors. If the supply of
NTD-Si is not sufficient, the industry may use another, likely more expensive, technology.

At any rate, since the pressure on industry from environment groups is becoming stronger,
‘green energy’ technology could rapidly expand. Therefore, it is likely that the future demand
for NTD will increase.

4.2.2. Development of other technologies

(1) Silicon carbide (SiC)

Theoretically the most ideal semiconductor material for a power device is a diamond, but it is
expensive and would not be practical. At the moment, a wide band gap semiconductor such as
SiC is considered to be one of the most promising candidates that can replace the pure silicon
semiconductor. SiC has properties in the middle of the range between silicon and diamonds.
Power devices made from SiC could have electrical characteristics approaching those of the
ideal device. SiC is a ceramic compound of silicon and carbon that is manufactured on a large
scale for various applications from semiconductors to even nuclear fuel. In their use as a
power semiconductor, SiC Schottky diodes with a breakdown voltage of 1200 V are
commercially available, and they can be operated for high speed switching. Bipolar devices
made of SiC are being developed for higher voltages, up to 20 kV. Among its advantages, SiC
devices can be operated at higher temperatures, up to 400°C, and have a lower thermal
resistance than silicon devices, which allows for better cooling.

SiC does not have a liquid phase available for the growth of single crystals by similar silicon
crystal methods. To make the SiC crystal, a crystal growing method from gas phase to solid
state should be used, although this is much more difficult and expensive than the silicon
crystal method. In summary, even if the manufacturing method of SiC makes great strides,
and the use of devices made of SiC is extended, SiC devices will hardly encroach on the

18
territory of the silicon based devices. They will probably be used in the range of a very high
frequency or a very high power that silicon semiconductors cannot cover.

(2) Epitaxial growth

The epitaxy is a kind of interface between a thin film and a substrate, and epitaxial growth
describes a process of depositing a thin layer, typically 0.5 to 20 microns, of single crystal
material over a single crystal substrate, usually through the chemical vapour deposition
method. The epitaxy enables various applications in the semiconductor fabrication, and it can
be the source of high crystalline quality growth for use in many semiconductor materials. The
epitaxy is also used to grow layers of pre-doped silicon on the polished sides of silicon wafers
before they are processed into semiconductor devices. An epitaxial layer can be doped during
deposition by adding dopant to a source gas such as arsine, phosphine or diborane. By
growing a lightly doped epilayer over a heavily doped silicon substrate, a high breakdown
voltage across the collector–substrate junction is achieved while maintaining a low collector
resistance. Lower collector resistance allows a higher operating speed at the same current.
This epitaxial growing method is utilized for a large portion of the modern power device
fabrication process, for example, in the punch through insulated gate bipolar transistor (PT-
IGBT) fabrication procedure. However, since the thickness of an epitaxial layer is limited, the
areas in which the epitaxy is applied also remains limited.

These two competing technologies have their own weaknesses compared with the NTD
method at present. While SiC is an ideal material for power devices, the associated
technology has not matured for mass production so far. For epitaxial growth on silicon wafers,
growing an epitaxial layer with enough thickness for sufficient endurance against high
voltages is very difficult. Consequently, SiC or epitaxial wafers presently are much more
expensive than NTD wafers. Their higher cost would not allow competition in the market, in
which other lower cost devices are available, but they may establish their own market where
other substitutes are not available, similar to the role of NTD in the silicon market. However,
since in the future these technologies could have a great influence on the NTD market, their
progress or development, including new technologies, should be carefully observed.

5. NEUTRON IRRADIATION AND RESISTIVITY

5.1. Resistivity in semiconductors

While only the flow of negatively charged electrons can contribute to the electrical
conductivity of a metallic conductor, both electrons and positively charged holes contribute to
the conductivity of a semiconductor. A measure of a material's ability to conduct an electric
current is represented by either conductivity mostly denoted σ or its inverse, resistivity,
denoted by ρ. Resistivity is widely used, and it is determined by the intrinsic carrier density
and mobility. If we have a slab of a semiconductor with thickness t and surface area A, the
current I that will flow, when a voltage V is applied across the thickness is [16]:

AV AV
I= or ρ = . (5.1)
ρt It

The current is composed of the flow of holes Ih and electrons Ie. Although these two types of
charge carriers move in opposite directions, the separate currents are additive because of the

19
opposite charges of holes and electrons. Thus an external measurement of the current alone
cannot distinguish between the flow of holes or electrons. The total observed current in an
intrinsic semiconductor will be their sum, with each term given by the product of area A,
intrinsic carrier density ni, electronic charge ε and drift velocity v of the charge carrier. Thus,

I = I e + I h = Ani ε(ve + vh ) . (5.2)

From Eqs. (3.2), (3.3), (5.1) and (5.2),

V
I = Ani εE (μ e + μ h ) = Ani ε (μ e + μ h ) . (5.3)
t

Then, the resistivity becomes,

1
ρ= . (5.4)
ni ε(μ e + μ h )

For an intrinsic silicon at room temperature,

1
ρ= −19
, (5.5)
(1.6 × 10 C)(1.5 × 10 / cm 3 )(1350 + 480 cm 2 / V ⋅ s)
10

ρ = 2.3 × 105 V ⋅ s ⋅ cm/C = 230,000 Ω ⋅ cm .

This is the maximum resistivity which is theoretically obtainable. In practical silicon material,
this resistivity cannot be achieved due to residual impurities. A typical initial resistivity of Si
to be irradiated for NTD purposes ranges between 3000 and 20 000 Ωcm.

When impurity atoms are introduced into an intrinsic semiconductor material, the densities of
the charge carriers and impurities are represented by the charge neutrality [55]. Thus, most
donors and acceptors are ionized at room temperature, so the neutrality condition can be
approximated by

n + N A = p + ND , (5.6)

where n and p are the electron and hole densities, and NA and ND are the acceptor and donor
densities, respectively. Since the following product of electron and hole densities is always
independent of the added impurities, thus

np = ni pi = ni2 = pi2 (5.7)

where ni and pi are the intrinsic carrier densities. Combining above two equations, in an n type
semiconductor where ND > NA,

2
( N D − N A ) + ( N D − N A ) 2 + 4ni
n= . (5.8)
2

If N D − N A >> ni or N D >> N A ,

20
 ni2
n ≈ ND , p ≈ . (5.9)
ND

Similarly, carrier concentrations in a p type semiconductor ( N A >> N D ) is given by

ni2
p ≈ NA, n ≈ , (5.10)
NA

Therefore, the final resistivity of silicon after n type doping is as follows,

1
ρ= . (5.11)
εnμ e

As an example, assuming a silicon sample with a donor density of 1013 cm-3, then the
resistivity will be

1
ρ= , (5.12)
εN Dμ e
1
ρ= −19
,
(1.6 × 10 C)(1.0 × 10 / cm 3 )(1350 cm 2 / V ⋅ s)
13

ρ = 463 Ω ⋅ cm .

5.2. Doping concentration for NTD-Si

The resistivity of irradiated silicon is inversely proportional to the total concentration of the
produced dopants such as 31P atoms and initially existing impurities. In NTD, the added 31P
concentration is proportional to the irradiated neutron fluence, which is a product of the
neutron flux, time of irradiation with a constant neutron flux and the reaction cross-section.
As the neutron cross-section varies by neutron energy, it is influenced from the neutron
spectrum in the irradiation site. In the semiconductor industry, the resistivity rather than the
dopant concentration is usually used. Therefore, the relationship between the resistivity and
the dopant concentration should be established first.

For n type silicon doping with phosphorous, the resistivity (Ωcm) is given by

1
ρ= , (5.13)
[ P ]με

where [P] is the phosphorus atomic concentration in cm-3, ε is the electron charge,
1.602×10˗19 C, and μ is the drift mobility of the electrons in the silicon crystal lattice. Electron
mobility depends on the temperature, and it is in the range of 1220~1500 cm2/V·s. In normal
conditions at 300 K, it is usually 1350 cm2/V·s for silicon as shown in Table 1. Then we can
easily obtain the relationship between the dopant concentration and the resistivity:

KP
ρ= for phosphorus doping, (5.14)
[P]

21
 KB
ρ= for boron
b dopin
ng, (5.15)
[B]

where KP, KB are thhe proportio

onal constannts and [B] is n atomic conncentration in cm-3.
i the boron

Figure 6 shows thee relationsh hip betweenn dopant con ncentration and resistivvity. The reesistivity
is basicaally inverseely proportio
onal to the ddoping conccentration, but
b in the heheavily dopeed range,
this relaationship is broken as shown in F Figure 6. Th he differencce betweenn the resistiv
vity of p
type andd n type siliicon is due to
t the differrence in the mobility off the electroons and holees. Since
the minnimum resisstivity practically obtaiinable throu ugh NTD is regarded aas several oh hms due
to the reesidual radiioactivity off an irradiatted ingot, NTD-Si
N is in
n the regionn of proporttionality.
Thus, EEqs. (5.14), (5.15)
( can be
b applied too NTD irrad diation.

FIG. 6. Relation
nship betweeen doping concentratio
c on and resisstivity.

At the Japan Atom mic Energy

y Agency ((JAEA), preeviously Japan Atomicc Energy Research
R
Institutee (JAERI), the
t followin
ng constantss were used
d [56]:

5 ×1015 1.3 × 10016

ρ= f n-type, ρ =
for for p-ty
ype, (5.16)
[ P] [ B]

while att the Harweell reactors [28],

[ the folllowing con
nstant was used:

4.63 × 10155
ρ= . (5.17)
[ P]

In orderr to determiine the amouunt of impuurity neededd to attain a target resisstivity ρt, firrst of all,
the conccentration of
o the impuurities alreaddy existing in the crystal should bbe known, which w is
usually representedd by an in nitial resistiivity ρi. Th
hen we can n easily dedduce the ad dditional
dopant cconcentratioon needed to
t achieve thhe target ressistivity:

22
 1 1
[P]doping = K P  −  for initially n-type silicon, (5.18)
 ρt ρi 
K K
[ P]doping = P − B for initially p-type silicon. (5.19)
ρt ρi

5.3. Neutron flux and cross-section

From the above equations, we can determine the increment of phosphorus atom concentration
to be achieved by neutron irradiation. When silicon is irradiated in a neutron field, the neutron
capture reaction rate of 30Si in the silicon is given by [57]

∞
R =  φ( E ) σ
30 30 30 30
Si Si Si Si
(E)N dE = N σ φT , (5.20)
0

where φ(E ) and σ(E ) are the energy dependent flux and cross-section for the (n,γ) neutron
capture reaction, respectively, φT is the thermal flux intensity, σ is the average neutron capture
cross-section and N is the atomic density of target isotope.

The thermal flux is usually defined as an integral of the energy dependent flux over a thermal
energy range. For a well thermalized condition as in a reflector region of a research reactor,
the energy dependent flux is represented by a Maxwellian spectrum. The Maxwellian
spectrum dies off rapidly beyond 5kT, that is,

~ 5 kT ∞
φT =  φ( E ) dE ≅  φ( E ) dE . (5.21)
0 0

For a Maxwellian neutron spectrum,

1/ 2 1/ 2
2πn  2  ∞ 2  2kTn  2

− E / kTn
φT =   Ee dE = n  = nvT = 1.128nvT , (5.22)
(πkTn )3 / 2  m  0 π  m  π

where m, n is the neutron mass and density, Tn is the neutron temperature in Kelvin, k is the
Boltzmann constant and vT is the speed of the neutrons having the most probable energy.

Since the most probable energy of the above distribution is kTn, vT is given by

1 2
mvT = kTn . (5.23)
2

From the above equations, vT and the neutron flux depend on the neutron temperature even
though the neutron density is unchanged.

If the absorption cross-section of 30Si varies as 1/v, then

30
Si 30 E0 v 30
σ ( E ) = σ 0 Si = σ 0 Si 0 , (5.24)
E v

23
where E0 is an arbitrary relative energy, but for usual practice is taken to be 0.0253 eV, which
is the most probable energy in the Maxwellian spectrum at 20.46 °C and also corresponds to a
neutron speed of 2200 m/s. σ0 is the cross-section at E0.

Then, the reaction rate is given by

1/ 2
2πn  2 E0  ∞
 σ0 0 E e
30 30
Si Si − E / kT
R=N 3/ 2 
dE . n
(5.25)
(πkTn )  m 

For integration, we can use the parameter change x = E/kTn, then,

2n ∞
v0 σ 0 Si  x e − x dx .
30 30
Si
R=N (5.26)
π 0

Performing the integration, and if we insert a 2200 m/s flux,

φ0 = nv0 . (5.27)

Then,
30 30 30 30
Si
R=N nv0 σ 0 Si = N Si
σ 0 Si φ 0 . (5.28)

Thus, the (n,γ) capture reaction rate of the 30Si inside the neutron field can be simply deduced
by the 2200 m/s flux and the cross-section. From Eq. (5.22), we can obtain

π φT
n= . (5.29)
2 vT

30
Also, if we introduce an average cross-section at 0.0253 eV, σ 0 Si ,

30
Si π v0 Si 30 30
Si v0 Si
30
R=N σ 0 φT = N σ 0 φT . (5.30)
2 vT vT

If we define a useful quantity known as the average cross-section at a given neutron

temperature,
30 30
Si Si
R=N σ φT , (5.31)

where the average cross-section is defined as

1/2
30
Si π  T0  Si 30
σ =   σ0 . (5.32)
2 T 

Therefore, we can obtain that

24
 30 30 30 30
Si Si Si
R=N σ 0 Si φ 0 = N σ φT . (5.33)

Thus, thhe reactionn rate can be b deducedd by eitherr a 2200 m/s m formulaation or an integral
formulaation, with varying
v flux
x and crosss-section vaalues. The 2200
2 m/s foormulation doesd not
b howeverr, it has beeen often useed because tthe neutron flux can
represennt an actual situation, but
be easily determineed with 1/v neutron fluux monitors and commonly availabble 2200 m/s m cross-
section tables regarrdless of thee neutron teemperature and spectru um. In contrrast, reactorr physics
calculattions usuallyy give the neutron
n fluxx based on th
he integral formulation
f n. When thee neutron
spectrumm is a well thermalized Maxwelllian at 20.46°C, the integral theermal flux is 1.128
times thhat of a 2200 m/s flux x as shownn in Eq. (5..22). Thereffore, the thhermal neuttron flux
should bbe given wiith its definiition in ordeer to avoid confusion.
c

For som
me materialss with non-1 1/v cross-seections, we must
m correcct the deviattion from 1/v cross-
sectionss by introduucing a new factor,
30 30
Sii
R = g (T ) N σ 0 Si φ 0 , (5.34)

where gg(T) is the non-1/v

n facttor, and is aalso called the g-factor. It is tempperature dependent,
and for 30Si capturre at room temperature
t e, the g-facctor is 1.007
71. Figure 7 shows thee typical
30
capture cross-section of Si [58]. From the figure,, it is confirmed that tthe neutron n capture
cross-seection of 30Si
S has very good
g 1/v chharacteristics up to ~1 keV.
k

FIG. 7. Neutron ccapture cro o 30Si.

oss-section of

5.4. Esttimation off the K valu

ue

From thhe above eqquations, we w can obtaain the prod duction of phosphoruss atoms by neutron
irradiatiion of silicoon crystal fo
or an irradiaation time ti, namely

ti ∞
[ P ]doping =  
30
0 30 30 30
Si Si Si Si
φ( E , t ) σ (E)N ddEdt = N σ φT t i . (5.35)
0 0

25
Then, the neutron fluence that provides a phosphorous concentration corresponding to the
target resistivity can be deduced from Eq. (5.19) as follows:

 1 A
φT ti = K  −  , (5.36)
 ρt ρi 

where:

A = 1 for initially n type,

A = KB/KP for initially p type.

A new constant K is introduced to represent the product of fluence and resistivity for an
infinite case of ρi, and it is defined by

KP
K= 30 30
. (5.37)
Si Si
N σ

Since σ is dependent on the neutron spectrum including non-thermal neutrons, the K value
must be determined at each irradiation site. If the neutron spectrum is a well thermalized
Maxwellian, σ is a Maxwellian-averaged value at a specific neutron temperature. In a real
situation, the K value for an irradiation site should be determined for each customer and each
diameter based on the measured resistivity values’ feedback. An example of an estimation of
K is shown in Table 2. It is found that the K values show a rather wide range as listed in
Table 3. Since the neutron spectrum does not vary significantly across NTD irradiation sites,
it cannot be the major reason for this wide range. The value of resistivity for the silicon wafer
is an accepted standard in the industry. It seems that different definitions of the neutron flux
and differences in neutron flux measurement would be major causes of these variations.

TABLE 2: AN ESTIMATION OF K FOR A MAXWELLIAN SPECTRUM AT 20.46°C

Specifications Typical values
30
Si abundance 0.0310
Silicon nominal density 2.33 g/cm3
Silicon atomic weight 28.086
Avogadro number 6.023×1023
30
Si atomic density 1.549×1021 cm-3
30
Si capture cross-sections at 25.3 meV 0.1075 barns [10-24 cm2]
Maxwellian avg. capture cross-section 0.09599 barns [10-24 cm2]
Electronic charge 1.609×10-19 C
Electron mobility in silicon crystal at 300 K 1350 cm2/s
KP 4.630×1015
K for 2200 m/s flux 2.780×1019
K for Maxwellian integral flux 3.114×1019

26
 TABLE 3: K-VALUES USED IN SEVERAL SITES

Irradiation site K value

Topsil 0.8~1.0 times of 2.612×1019
Harwell 2.54×1019
SAPHIR 2.628×1019
JAERI 2.94×1019 (in 2200 m/s flux)
FRM-II [44] 4.86×1019 for 150 mm ingot,
4.26×1019 for 200 mm ingot (in 2200
m/s flux)
HANARO [59] 2.40×1019 (in 2200 m/s flux)

6. TECHNICAL REQUIREMENTS FOR NTD

The foremost advantage of NTD compared to other doping methods is the high uniformity of
the final resistivity after doping a whole silicon ingot, which entitles NTD-Si to the claim of
the best quality silicon for high power applications. Since the neutron flux in a reactor is not
uniform, the overall uniformity in an ingot depends on the ingot size and irradiation
technology. It is usually possible to control the neutron fluence very accurately, and therefore
the resistivity target can be specified and accomplished within a very tight tolerance [60]. The
accuracy of the target resistivity is a basic requirement for the doping process. Thus, the
accuracy of doping combined with the overall uniformity will govern the quality of the
product.

The physical volume of a batch of ingots is rather large, and the flux variation in such a
volume is much more vital than the required accuracy in the target resistivity. Irradiation
uniformity is expressed by radial and axial, or longitudinal, variations of the final resistivity
after irradiation. Thus, radially and axially uniform irradiation is the prime target in the design
and operation of an irradiation device.

The initial resistivity variation has a direct impact on the final resistivity variation after
irradiation. The initial variation can be only partially offset (smoothed) by irradiation.
Therefore large local variation in initial resistivity is not a major cause of final resistivity
variation in NTD wafers. In order to minimize the impact of initial resistivity variation, the
initial resistivity should be high enough compared to the target resistivity. Also, uncertainty
(error) in the resistivity measurement impacts on accuracy of the results received from the
customer. In general, uncertainty is larger when measuring high resistivity. Therefore, these
considerations need to be acknowledged during irradiation of high resistivity targets.

In practice typically several Si ingots are put together to form a batch of material to be
irradiated simultaneously. Obviously it is important to combine only ingots with similar
starting value of resistivity to obtain high accuracy with respect to the target resistivity to be
met by the irradiation

Neutrons used for NTD are mainly thermal neutrons, and a higher thermal neutron flux results
in a shorter irradiation time. If the irradiation time is too short, however, an accurate
irradiation would be difficult unless the ingots are inserted to and withdrawn from the neutron

27
field very rapidly. If the flux is too low, a very long irradiation time is needed, which limits
the practical application for NTD. The resistivity of NTD Si ranges from about 10 to about
1000 Ω•cm. For a neutron flux of 1013 cm-2s-1, about 17 hours are needed to obtain 50 Ωcm.
When one hour of irradiation is applied, about 800 Ω•cm would be a reasonable upper bound
at a flux of 1013 cm-2s-1. In the case with a flux five times higher, the upper bound would be
about 160 Ω•cm. Low flux reactors with a thermal neutron flux approximately1012 cm-2s-1
would be proper for the accommodation of high resistivity targets.

Since fast neutrons create extended charged lattice defects in a crystal, the fast neutron flux in
the irradiation position must be as low as possible. This requirement is in particular important
for high target resistivity NTD Si, because the screening radius of charged defects (Thomas
Fermi screening length) is longer in materials exhibiting a low charge carrier density.

Gamma rays are the major source of heat generation in the ingot, so the gamma ray flux
should also be as low as possible, and the ingot must be sufficiently cooled during the
irradiation. In addition, due to the short half-life of only 2.62 h also the β decay of 31Si itself
contributes considerably to the heat generation within the Si ingot under irradiation.

The diameter of an irradiation hole should be large enough to accommodate an irradiation rig
containing the ingots, coolant paths and online flux monitors. Because the NTD service
charge is usually based on the number of irradiations (cycles), the irradiation of large volume
ingots is profitable. However, loading and unloading of large ingots significantly distort the
neutron flux at nearby experimental facilities and neutron detectors for reactor operation. This
interference should be carefully examined.

Since a silicon ingot is very brittle, careful design and operation of the ingot handling tool is
required so as not to damage the ingot. Particular attention has to be paid to the use of gloves
during all handling procedures and to the cleanliness of all storage locations within and
outside the reactor pool. Si ingots must only be in contact with Si, Al or plastics. The contact
to steel and other metals must be strictly avoided.

Irradiated ingots must be cooled to reach the exemption level of residual activity, so a proper
space is needed to cool down the irradiated silicon. In addition, a cleaning facility, e.g. an
ultrasonic bath to be connected to the radioactive waste water tanks of the reactor, is needed
to remove potential contamination from the contact of the Si ingot with reactor pool water.
Equally, a surveying system for the residual radioactivity of the irradiated ingot is part of the
necessary equipment. In addition, it is crucial for the reactor operator to have the license to
issue an official radioactivity release document. Finally, for commercial irradiation services,
systematic quality assurance, traceability of the product and control are necessary. Many
customers of NTD service require the certification of the irradiation facility according to the
ISO 9001:2008 standard because they are urged by their customers to only cooperate with
ISO 9001 certified subcontractors.

The uniformity and accuracy criteria of NTD are determined in an individual contract
between an irradiation site and a wafer company. When irradiations do not meet the criteria,
the vendor of the spoiled ingots typically will require compensation.

6.1. Radial uniformity

The planar distribution of the resistivity of a NTD-Si wafer is represented by the radial
resistivity gradient (RRG) defined by

28
 ρ max − ρ min
RRG = 100 × (in %). (6.1)
ρ min

where ρmin and ρmax are the minimum and maximum values, respectively, among the
measured resistivities at several points on a wafer.

In general, there is a radial flux gradient in the reflector region of any research reactor, i.e. the
neutron flux becomes smaller away from the core centre. The effects of the radial non-
uniformity of the neutron flux are compensated by rotating the ingot during irradiation. Due
to the comparatively long irradiation times a rotational speed of only few rpm is sufficient.
Although silicon is rather transparent to neutrons, some attenuation of the neutrons inside the
ingots will occur. Therefore, the neutron fluence at the inner part of an ingot becomes lower
than the periphery. This difference becomes larger as ingot diameter increases. If the RRG is
only due to the neutron attenuation in the ingot, tight specifications from customers can be
met even with eight inch ingots. On the other hand, asymmetric irradiation of an ingot can
increase the RRG. Also, the initial resistivity and uncertainty in the resistivity measurement
can affect the RRG directly. Above effects may be reflected in the specifications. While the
specification of the RRG would depend on the customer, it is usually less than or equal to 4–5%
nowadays.

6.2. Axial irradiation uniformity

The longitudinal variation of the resistivity in a cylindrical silicon ingot is represented by an

axial resistivity variation (ARV) given by

plane plane
ρ max − ρ min
ARV = 100 × plane
(in %). (6.2)
ρ min

plane
where ρmin plane
and ρ max are the minimum and maximum values among the plane resistivities
in an ingot. The plane resistivity may be represented by an average, a medium or a centre
value in the plane, depending on the customers. When the plane resistivity is represented by
the value at centre, since the ARV is determined by only one measured value each for the
maximum and minimum resistivity planes as similar to RRG, it is more sensitive to the initial
resistivity variation and uncertainties in resistivity measurements than other variables.

The requirement for axial resistivity uniformity depends on the customers, but this
requirement has become stricter. In the early 1990s, a general variation of 10–20% was
allowed, but in the 2000s a variation within 5–8% has been required. If the uncertainty in the
resistivity measurement and the deviation from the target resistivity are combined into the
axial variation, the requirement is extremely precise. Therefore, it is recommended to design
the irradiation facility to minimize the axial variation of the neutron fluence as much as
possible.

6.3. Irradiation accuracy

The irradiation accuracy (IRRACC) may be represented by the deviation of the average
resistivity in an ingot from the target resistivity.

29
 ρaverage − ρ target
IRRACC = 100 × . (6.3)
ρ target

where ρtarget is the target resistivity, and ρaverage is the average or medium of all the measured
resistivity values for an ingot or plane. When many IRRACC measurements are accumulated,
their statistics may be close to a Gaussian distribution. The irradiation site should give an
effort to make its average zero and its standard deviation as small as possible. The standard
deviation is usually less than 3% at an irradiation site with sufficient experience in NTD. The
standard deviation becomes larger in cases of higher target resistivity due to the propagation
of the uncertainties.

Wafer companies want every measured resistivity to be within their criteria for maximum
yield. The maximum deviation of plane resistivity values from the target may be specified
rather than the ARV and the IRRACC, depending on customers.

6.4. Interfering radiation and NTD

Exposure to fast neutrons and gamma rays is not desirable for NTD. Fast neutrons are a major
source of permanent lattice defects, and gamma rays are a major source of heat generation in
ingots.

Lattice defects are the displacement of silicon atoms from their normal lattice positions. Low
energy neutrons create lattice defects in an indirect way due to their absorption by 30Si and the
subsequent ß decay of 31Si, the recoil of which leads to the production of single interstitials
and vacancies. These defects, however, are usually recovered by appropriate heat treatment
after irradiation. The same applies to defects caused by energetic gamma rays. The most
important interference in the NTD process is the fast neutron knock-on displacement leading
to defect clusters which cannot be easily recovered. These defects introduce defect levels into
the band gap that cause free carrier removal, reducing carrier mobility and minority carrier
lifetime. It should be emphasized that an in-core fission neutron spectrum produces about 103
more displaced silicon atoms than a graphite moderated spectrum. Thus, a low fast neutron
flux is needed to avoid the irreversible lattice defects in a silicon crystal caused by fast
neutrons [61, 62]. After neutron irradiation, the ingot is annealed by the wafer company in
order to recover the crystal defect [63]. Wafer companies may judge the quality of neutron
fields by measuring the carrier lifetime.

In the IAEA consultants meeting, it was recommended that the thermal-to-fast flux ratio
should be at least of the order 7:1 [64]. Particular high values for the thermal/fast neutron flux
can be achieved in heavy water moderated reactors. E.g. in the case of the OPAL reactor, they
designed their thermal-to-fast neutron flux to be 900:1; at FRM II a ratio of thermal/fast flux
of > 1000 was measured This ratio for the irradiation facility of SIDONIE in (the light water
moderated) BR2 was reported to be about 10:1. SIDONIE also reported a cadmium ratio of
approximately 25 [30]. For the new Si doping facility POSEIDON at BR2 a cadmium ratio of
50 is obtained [32]. It is noteworthy that the efficiency of lattice defects as scattering centres
for the charge carriers is the more important the lower the charge carrier density is.
Consequently high target resistivity (>400 Ωcm) NTD Si is preferentially to be produced in a
well thermalized neutron spectrum (i.e. a heavy water moderated reactor).

The thermal-to-fast flux ratio varies depending on the low energy boundary of the fast neutron
flux. The Cd ratio, on the other hand, reveals information on the epithermal neutron flux,

30
which is not included in the thermal-to-fast flux ratio, but depends on neutron detectors. For
example, the classical Cd ratio calculated or measured by the reaction rate of gold is surely
different from that by silicon or cobalt. Thus, the thermal-to-fast flux ratio and the Cd ratio
should be explicitly specified in order to be meaningful. In brief, it is recommended that a Cd
ratio larger than 10 in the case of gold would be appropriate for NTD.

Gamma rays and ß decay of 31Si are the major source of heat generation, which may cause
excessive ingot temperature and formation of bubbles at the ingot surface. The IAEA
consultants meeting recommended the temperature at the irradiation position be below 180°C
[64] to avoid the diffusion of the minority carrier and to minimize lattice defects such as a
swirl. In BR2, the gamma ray heating was rather high, 3 W/g, and the target temperature of
the NTD facility process was set below than 250°C [30]. Although it is difficult to define a
concrete limit on the ingot temperature during the irradiation process, but it is certain that the
temperature should be as low as possible. Table 4 lists the neutron flux and gamma ray
heating data reported by various reactors.

TABLE 4: THE NEUTRON FLUX AND GAMMA RAY HEATING OF SEVERAL

IRRADIATION SITES

Thermal Irradiation Neutron flux Gamma ray heating or

Reactor
power rig [cm-2s-1] temperature
MARIA in 2.1×1013, (thermal),
30 MW 0.5 W/g
Poland Fast (>1 MeV)/thermal=0.02
56 MW
BR2 in 1.74×1014 (thermal),
(Nominal: < 200°C Si core temp.
Belgium 1.9×1013 (fast),
85 MW)
SAFARI-1 in 4 inch Measured temperature
20 MW 2.5×1013-8×1013 (thermal)
South Africa SILIRAD at Si: ~80°C
FRM II in 1.6×1013 (thermal) Max. temp.: 110°C
20 MW 8 inch
Germany Thermal/fast=1700 (Si core temp.) [44]
OPAL in 2.5×1012 –1.5×1013 (thermal)
20 MW 5, 6, 8 inch
Australia Thermal/fast=900
HANARO in
5,6 inch Thermal/fast=400
the Republic 30 MW 0.2–0.9 W/cm3
NTD2 Cd ratio for gold: 16~22
of Korea

6.5. Residual radioactivity

There are three kinds of regulations specified regarding the radiation level of irradiated silicon
[28]:

— Transport regulations — the radiation level at any point on the external surface of a
package shall not exceed certain criteria;
— Non-fixed contamination;
— The level of radioactivity of the material itself.

To ship the irradiated ingot back to the customer, its residual radioactivity must be lower than
the international criteria for exemption from regulations for radioactive material. Reactions of
interest in the NTD process according to the aspect of the residual radioactivity are as follows:

31
 30
Si(n,γ)31Si → (β-) 31P (T1/2 = 2.62 h);
31
Si(n,γ)32Si → (β-) 32P (T1/2 = ~172 y) → (β-) 32S (T1/2 = 14.3 d);
32
P(n,γ)32P → (β-) 32S (T1/2 = 14.3 d).

In addition to the above reactions, there are several threshold reactions involving fast neutrons
such as (n,p), (n,np)Al, (n,α) and (n,2p)Mg, but their short half-lives together with small
cross-sections make their contribution to the residual radioactivity negligible. Also, since the
impurity concentration inside the silicon ingot is extremely low, their activation is negligible.
31
Si has a relatively short half-life of 2.62 hours, and it will sufficiently decay away in a
proper cooling time of two or three days. The activity of 32Si is also very low because of its
very low production rate despite very long half-life. Therefore, the main source of residual
radioactivity is 32P (T1/2=14.3 d) produced by the 31P(n,γ) reaction. The activity of 32P is
slightly dependent on the neutron flux level, but it is essentially governed by the neutron
fluence, i.e. the inverse of resistivity.

The exemption criterion from the regulation for 32P may be different in each country. At the
Harwell reactors in 1985, the criterion of 7.4 Bq/g was applied [28] according to the IAEA
criterion that was published in 1977. They also used a surface contamination limit of
3.7 Bq/cm2. 74 Bq/g was used in Japan, which produces the largest amount of NTD wafers.
The level of 31Si or 32P for exemption from regulation is 1000 Bq/g in the current IAEA
Safety Standards.

As the resistivity is inversely proportional to the irradiation time, a lower resistivity causes a
higher residual radioactivity. The specific activity of 32P increases very rapidly as the
resistivity decreases below 10 Ω•cm. Since the half-life of 32P is 14.3 days, a very long
cooling time is needed when the resistivity is less than a few Ωcm. This is one of the reasons
why NTD is not practical for specimens with a low resistivity field. NTD is mainly applied
when resistivity surpasses about 20 Ω•cm. The actual demand for NTD Si ranges between 30
Ωcm and 1100 Ωcm with an increasing interest in high target resistivity. The average
resistivity of NTD Si was for a long time usually estimated as 40–50 Ωcm; hence an
irradiation time of a 5 Ω•cm sample is about 10 times of that for the average resistivity. The
practical lower limit is about 10–15 Ω•cm.

In practice, the activation of impurities and contaminants on the ingot surface sometimes
causes a problem in particular since the Si ingot is in most of the irradiation facilities in direct
contact with reactor pool water Careful cleaning of ingots not only after irradiation but also
before irradiation is important. At FRM II good experience was made by cleaning Si ingots by
isopropanol before the start of the irradiation and cleaning the irradiated Si ingot in an
ultrasonic bath at 65°C using a special decontamination non foaming, non-ionic detergence
which, is free of halides followed by keeping the ingot in de-ionized water for 24 h.

6.6. Interference with other facilities

An irradiation hole for NTD is relatively large considering the size of an average research
reactor. If the hole is filled with light water, the neutron flux around the hole changes
significantly when ingots are inserted into or withdrawn from the hole. If other experimental
facilities or neutron detectors for reactor operation exist near the hole, NTD irradiation could
perturb the experiments or reactor operation. Thus, the location of an NTD hole must be
carefully placed during the design stage and, if possible, a method minimizing the

32
perturbation should be considered. For example, a material having similar neutronic
characteristics to silicon (typically Al or a Si dummy) may replace the ingot when it is
withdrawn from the irradiation hole.

On the other hand also interference of neighbouring installations (like uranium target
irradiation positions for 99Mo production as a very drastic example) has to be taken into
account. Today, however, careful mapping of an entire reactor with all its installations is
possible by full scale Monte Carlo programs like MCNP(X) [65].

6.7. Handling of silicon ingots

A silicon ingot is a glassy cylindrical single crystal, so its handling tools should be designed
and utilized very carefully. Of course, a rig for efficient and safe irradiation should be
furnished. A vacuum sucking system is used in Japan and at FRM II to handle silicon ingots,
and an automated mechanical transfer system using canisters is used at OPAL. The silicon
ingot is also very brittle. The edge of an ingot is easily broken upon impact, especially during
insertion to the irradiation rig or to the irradiation canister. Generally, the clearance between
the ingot and the inner surface of the rig should be minimal to locate ingots at the centre of
irradiation rig. Thus, a specially designed handling tool is required for the easy and safe
manipulation of the ingot. As a general practice, if an ingot is broken, the vendor should be
reimbursed.

An ingot’s diameter has some tolerance from a typical diameter. The tolerance range may
depend on customers. The irradiation rig or the irradiation canister should be designed to
accommodate all ingots in the tolerance range. Since at FRM II for example only one
irradiation hole is available for NTD, the irradiation canisters may be equipped with ring-
shaped spacers made like the canister itself from an AlMg3 alloy to allow the irradiation of
ingots having a diameter of 5, 6 and 8 inches.

6.8. Neutron fluence monitoring

Generally, at an NTD irradiation site, online monitoring of the neutron flux is needed to
achieve accurate irradiation. It is usually accomplished by the self-powered neutron detector
(SPND). When the irradiation time is not long enough to neglect the response time of the
SPND, a dynamic compensation of SPND signals may be needed. As the SPND can be
located only outside of the ingot, its signal must be calibrated to the neutron flux in the ingot,
which can be measured by attaching neutron activation foils or wires in between ingots. But
the calibration factor may vary depending on many factors like geometrical core configuration,
control rod position, fuel burn up, SPND burn up, etc. Therefore, a careful investigation of the
calibration factor over a long term is required.

6.9. Washing, decontamination and radioactivity monitoring

Ingots are usually cleaned by chemicals before shipment to the wafer companies, but
additional cleaning before irradiation is recommended. While the liquid used for cleaning
before irradiation can be disposed as non-radioactive waste, decontamination after irradiation
produces radioactive waste. The pre-cleaning may reduce the burden incurred by
decontamination, so a washing procedure using an appropriate cleanser should be determined.
Since in most cases the Si ingots are already polished before being irradiated it is sufficient to
clean the un-irradiated ingots with a soft tissue and alcohol in order to remove traces from
fingerprints, etc.

33
It is, however, even more important to clearly define the washing procedure for the irradiated
Si, including levels of temperature of the cleaning bath, duration of the cleaning procedure,
and chemicals to be used. Before carrying the ingot out of the reactor, an inspection against
radioactive contamination is necessary. Therefore, all necessary equipment and procedure
must also be well established and available.

7. METHODS FOR UNIFORM IRRADIATION

The uniformity of neutron irradiation is usually expressed by radial (planar) and axial
(longitudinal) uniformities. The principle to achieve uniform irradiation is rather simple, and
the methods can be classified into a few types. However, even when the same method is used,
the design and operation of an NTD facility can vary greatly depending on the circumstances
of each irradiation site. Therefore, it is recommended that a reactor starting an NTD project
should search for the appropriate method by considering the characteristics and conditions of
their own reactor, rather than imitating an example.

7.1. Radially uniform irradiation

The radial uniformity of resistivity in a semiconductor ingot can be represented by the RRG
(radial resistivity gradient) given by Eq. (6.1). Wafer companies always require that the RRG
be as low as possible. Because there is a neutron flux gradient in the radial direction within a
reactor, the core side neutron flux in an irradiation hole is higher than that in a position around
the vessel. Therefore it is to be recommended to choose the location for a Si doping
irradiation hole in a position where the flux gradient is known to be low. In addition, the
silicon ingot must be rotated during irradiation in order to guarantee the complete radial
uniformity. The rotational speed should be chosen high enough as to guarantee a sufficiently
high number of revolutions, typically at least 100 during the irradiation time. The sufficiently
high number of revolutions during the irradiation time makes sure that the rotation will not be
the reason for a major RRG. It also should be noted that for higher neutron fluxes, with a
larger gradient and shorter irradiation time, a faster rotation might be needed.

Nevertheless, the RRG cannot be zero because the neutron flux at the inner part of a silicon
ingot is slightly lower than that at the periphery due to neutron self-attenuation by silicon.
Also, the non-uniformity of the initial resistivity distribution contributes to the RRG. Figure 8
shows the calculated radial neutron flux distribution inside a silicon ingot [66]. In this case,
RRGs by neutron attenuation are expected to be less than 1.5% and 2% for 5 and 6 inch
silicon ingots, respectively. If the neutron cross-section of non-crystal silicon is used for the
calculation, the RRG becomes larger by a factor between two and three because the scattering
cross-section of a single crystal at a thermal neutron energy range is smaller than that of
powder. In this calculation, neutron cross-sections at 300 K were used. In the actual
irradiation of the ingot, since the ingot temperature is usually higher than 300 K, a slightly
higher RRG is expected.

The misalignment and asymmetry in relation to the rotation centre of ingots and irradiation
rigs also increase the RRG. Furthermore, it is even more sensitive for an irradiation facility
cooled by light water since these neutrons attenuate rapidly in light water.

The rotation concept is simple but its actual realization is not so easy. In a vertical irradiation
hole, a metallic chain connected to a stepping motor can be used for rotation. A water bearing
concept has been adopted in the OPAL reactor.

34
 FIIG. 8. Exampple of calcu
ulated radiaal neutron fllux distributtion inside a silicon ing
got.

7.2. Axiially uniforrm irradiattion

Actuallyy, a uniform
m axial irraadiation is tthe most chhallenging in NTD from m the view wpoint of
irradiatiion technollogy, speciffically durinng the design stage ofo an irradiiation devicce when
determining the beest irradiation method at each irrradiation site. Basicallly, three irrradiation
processees have beeen found to achieve an axial uniforrmity. The choice
c of ann irradiation
n process
is primaarily dependdent on the characteristtics of each
h irradiation
n hole, and tthe selectedd process
must alsso be adaptted for various equipm ment and devices and additionally,
a , the preferrences of
designers.

7.2.1. R
Reciprocatinng motion

If an ingot moves inside an irrradiation hhole from on ne end to th

he other as shown in Figure
F 9,
the ingoot experiencces the entiire axial vaariation of the
t neutron flux. Conssequently, when
w the
moving speed is coontrolled app propriately,, a uniform irradiation is possible.

FIIG. 9. The uniform irradiation metthod to use the

t reciproccation motioon of an ing
got.

This meethod is useeful for an irradiation hole that has enough space at booth ends of the hole
such as a penetratinng beam tu ube. The irraadiation uniiformity in this methodd is less afffected by
changess in reactor core condittions such aas the controol rod positiion than in oother methoods since
every loongitudinal position in an ingot paasses through the same neutron fl flux field. However,
H
the movvement of thhe silicon in ngot itself m
may changee the neutron flux distriribution. Alsso, since
a siliconn ingot unddergoes thee average n eutron flux x of the trav velling zonne, the efficciency in
neutronn utilization is relatively
y low. SIDO ONIE at BRR2 [30], GÉK-4 at the T Tomsk com mplex [47]
and SILLIRAD at SA AFARI-1 [4 42] use this method.

35
7.2.2. Inversion

If the neutron flux distribution in a certain part of the irradiation hole is linear in the
longitudinal direction, an ingot can be irradiated through this linear region for half of the total
irradiation time, and then the ingot is inversed for the remaining half as shown in Figure 10.

FIG. 10. The uniform irradiation method using the inversion of an ingot.

This method can be applied at an irradiation hole having sufficiently linear neutron flux
distribution. Its axial irradiation uniformity is relatively less sensitive to core conditions as
compared to the flux screen method. However, the improvement of the uniformity is limited if
the linearity of the flux distribution is not sufficient. While it utilizes the neutron flux in the
irradiation zone fully, the highest flux region may not be utilized as shown in the figure. Also,
this method needs more manpower for the two irradiations of a batch, and the time loss
between the two irradiations decreases neutron utilization. JRR-3M and JRR-4 in Japan use
this method [56] as well as the POSEIDON facility at BR2 in Belgium [64].

7.2.3. Flux screen

If a proper flux screen is installed to flatten a neutron flux as shown in Figure 11, a uniform
neutron irradiation can be achieved. The flux screen is composed of relatively strong neutron
absorbers in a high neutron flux region and weak absorbers in a low flux region. This may be
accomplished by using different materials with different neutron absorption cross-sections or
by appropriate thickness variation of the neutron absorbers. The choice of neutron absorbents
depends on the available thickness for the neutron screen. If the thickness is a few millimetres,
stainless steel or nickel can be used. As the neutron flux in a high flux region decreases, the
flux at both ends of the screen decreases as well. In order to increase neutron efficiency, an
effort to increase the flux at the both ends of screen should be given.

The irradiation procedure of this method is relatively simple compared to other methods.
However, since the axial neutron flux distribution varies especially according to the control
rod position, significant planning should be considered before undertaking this method. An
appropriate methodology in the development of a suitable flux screen is to irradiate a batch of
Si ingots which had been equipped with flux monitors in small holes, drilled into the ingots.
The analysis of such experiment would typically follow by complementary modelling by
means of MCNP calculations [45]. If the location of the flux screen can be adjusted to
compensate for a change of control rod position, rather consistent axial uniformity can be
achieved regardless of rod height. This method has already in the 1990s been used at the
ASTRA reactor in Austria and has more recently been adopted at OPAL, HANARO [67] and
FRM II [44, 45)].

36
 FIG. 11. The uniform irradiation method using a neutron screen.

7.3. Effect of initial resistivity

The relationship between the initial and final resistivity is expressed by Eq. (5.36). There can
be an irradiation request for an ingot with a rather low initial resistivity, as in the case of a
repeated irradiation or an ingot doped by another method. If there is no local distribution of
the initial resistivity value, an irradiation site can easily meet the final resistivity as long as the
initial resistivity value is known. In case of a high RRG or ARV the required irradiation time
has to be calculated from the average initial resistivity and the target resistivity. In a second
step, the effect of this irradiation time on the parts of the ingot exhibiting the minimum and
maximum initial resistivity has to be evaluated in order to decide whether or not an acceptable
accuracy and homogeneity in the irradiated ingot may be achieved. So, it is important to
acquire information on the initial resistivity and its distribution within the ingot from a wafer
company, and to convince the company of the necessity to irradiate only ingots showing high
homogeneity of the initial resistivity. The same applies of course also when several ingots are
put together to form a batch of Si to be irradiated simultaneously.

8. NEUTRON FLUENCE MONITORING

Neutron flux measurements are indispensable in the NTD process to ensure the irradiation
quality. First, in the commissioning stage of an irradiation facility, the neutron flux
distribution over the ingots, Cd ratio, etc., should be measured to confirm the design and
performance of the installed NTD equipment. This can be accomplished by the activation
method. During routine irradiation of the silicon ingot, the neutron flux or fluence should be
monitored in order to guarantee accurate irradiation. The fluence may be monitored by real
time flux monitors, neutron activation monitors or simply by tracking the irradiation time. If
the neutron flux in the ingot including its trend during a reactor operation cycle is fully
understood and reliable, accurately tracking irradiation time alone may be sufficient. Against
potential influence on the neutron flux at the NTD hole from changes of reactor conditions,
however, real time neutron monitoring calibrated to the neutron flux in the ingot is
recommended. Activation monitors calibrated to the resistivity can give the basic data needed
for the determination of the irradiation time and for the calibration of online flux monitors.
Routine use of activation monitors is reliable until the NTD process is fully stabilized, and
afterwards may be occasionally utilized for quality control purposes.

37
8.1. Self-powered neutron detector

In the NTD process, real time neutron flux monitoring is usually accomplished by SPNDs.
The SPND operates by directly measuring the beta decay current following the capture of
neutrons, so it is always used in the current mode. The SPND is widely used for
measurements of the flux variation in nuclear reactors and is available with a small size and
diameter. Since it needs no external bias voltage applied to the detector, it requires simple
electronics. A good current meter and proper wiring are enough to measure the neutron flux.
However, its response is slow, and therefore dynamic compensation may be needed when
irradiation time is not sufficiently long as compared to the response time. Since the SPND
cannot be installed inside an ingot, the monitored flux is different from the flux in the ingot.
This relation between the SPND signal and neutron flux in the ingot should be well
understood.

Figure 12 shows a typical structure of a rhodium SPND [16]. The rhodium wire emits beta
particles after absorbing neutrons, and these beta particles are transferred to the current meter
through a lead wire. MgO or Al2O3 is used for insulation.

FIG. 12. Typical structure of a rhodium based SPND.

An SPND’s emitter should be a conductor so that the electric current may flow through it. Its
neutron absorption cross-section should be large enough to produce measurable current, but
not be too large for a sufficient life time. The most common emitter materials are rhodium and
vanadium, and their basic characteristics are shown in Table 5 [16].

TABLE 5: THE BASIC CHARACTERISTICS OF THE RHODIUM AND VANADIUM

Half-life of
Activation cross- Beta end- Typical neutron
Emitter Nuclide of induced
section at thermal point energy sensitivity*
material interest beta activity
energy [barns] [MeV] [A·cm2s]
[s]
51
V
Vanadium 4.9 225 2.47 5×10-23
(99.750 %)
103
Rh 139 44
Rhodium 2.44 1×10-21
(100 %) 11 265
*Sensitivity quoted for emitter of 1 cm length and typical diameter

The vanadium SPND has a simple neutron absorption reaction scheme producing 52V, which
has a half-life via beta decay of 225 s. Therefore, the current from the SPND after an abrupt
reactor shutdown decreases in a simple decay form. Thus the compensation for reactor power
variation is simple. Compared to rhodium, vanadium has a smaller absorption cross-section
and the half-life of its activation product is longer. Thus the vanadium SPND has a slower

38
response and a smaller burn-up rate than the rhodium SPND. In a typical reactor flux, the
vanadium SPND can be used for several years.

Rhodium has two neutron absorption reactions:

103
Rh + n → 104Rh → (β-) 104Pd (T1/2 = 44 sec),
103
Rh + n → 104mRh → (γ) 104Rh (T1/2 = 265 sec)→ (β-) 104Pd (T1/2 = 44 s)..
Thus, the decay scheme is somewhat more complex but can be easily evaluated numerically.
The high neutron sensitivity of the rhodium SPND provides a great advantage for low neutron
flux applications.

8.2. Flux monitoring using activation

Neutron activation wires and foils for NTD applications are sufficiently small to attach at the
surface of ingots. Thereby, neutron fluxes at the surface, including those between ingots, can
be measured. There are several neutron flux monitors that can be utilized, and the
radioactivity of the irradiated monitor can be measured by several detectors such as a high
purity germanium (HPGe) detector, Geiger-Müller counter, proportional counter and
coincidence method, etc. The choice of an activation detector may depend on the purpose of
the measurement and reactor power. The fluence measurements would be compared with the
measured resistivity values by the customer, and both are used for fitting or calibration of the
SPND signals.

8.2.1. Activation and radiation counting

Neutron monitoring using activation is an indirect neutron measurement that uses the
radioactivity induced in materials by neutron interactions. The activation interaction rate R of
neutrons within a material is given by:

R = N t σφ , (8.1)

where Nt is the number of target nuclei in the sample, σ is the activation cross-section
averaged over the neutron spectrum and φ is the neutron flux averaged over the sample.
Thus, the rate of activation per unit mass is a direct indicator of the neutron flux. As the
sample is irradiated, the radioactive nuclear species that are formed also undergo radioactive
decay. Thus, the activity of the sample is given by

A(t ) = R(1 − e − λt ) , (8.2)

where λ is the decay constant. The induced activity builds up over time and approaches the
saturated activity, A∞, after an infinitely long irradiation time, as given by

A∞ = R = N t σφ . (8.3)

If the irradiation has proceeded for a time t0, an activity A0 at the time t0 is given by

A0 = A∞ (1 − e − λt ) .
0
(8.4)

39
If a radiation counting of the material is carried out over an interval between t0+a and t0+b,
the number of counts will be

b
C = ε  A0 e − λt dt + B , (8.5)
a

where ε is the overall counting efficiency, and B is the number of background counts [16].
Then, the saturated activity is

λ(C − B)
A∞ = N t σφ = − λt0
. (8.6)
ε(1 − e )(e − λa − e − λb )

Finally, the neutron flux can be obtained by

A∞ λ(C − B)
φ= = . (8.7)
N t σ N t σε(1 − e − λt )(e − λa − e − λb )
0

In addition to the above procedure, the emission probability of radiation from the nuclide of
interest and the dead time of the counting system must also be considered and accounted for.

8.2.2. Activation detectors

General criteria for choosing an activation detector are as follows [16]:

— Shape of the activation cross-section;

— Magnitude of the cross-section;
— Decay constant of the induced activity;
— Purity and interfering activities;
— Nature of the induced activity;
— Physical properties.

The shape of the activation cross-section describes the variation of the neutron capture cross-
section for a given neutron energy. It usually varies inversely with neutron velocity, v, in the
thermal neutron energy range, but above the thermal region, it is more dependent upon the
material. Since the reaction of interest in NTD is the 30Si(n,γ)31Si reaction, an activation
detector having a similar activation cross-section shape to 30Si is preferable. However, the
usual 1/v detector can be used as well because a well thermalized neutron spectrum is usually
employed in NTD.

If the magnitude of the cross-section is high, the mean free path of neutrons in the detector is
small. If the thickness or diameter of the detector is not sufficiently small compared to the
mean free path, the neutron flux inside the detector material is significantly lower than on the
surface, which is called the neutron self-shielding effect. The insulation of the neutron
detector also depresses the neutron flux. This may cause a wrong deduction of the neutron
flux. Therefore, the common geometric form of the material is a thin foil or a small diameter
wire. Most valuable information can be obtained by the irradiation of flux monitors which are
placed within a Si ingot, being equipped with tiny drilled holes for this purpose. Because of
their negligible self-shielding suitable flux monitor are e.g. an Al:Au wires containing 0.1%
or 0.2% of gold [44]

40
The deccay constannt of the indduced activiity must be small to haave a sufficiiently long half-life
as comppared to thee irradiation and countinng times duuring the flu
uence measuurement. Th he purity
should be high annd interferin ng activitiess should bee as small as
a possible.. The naturre of the
inducedd activity must
m have characteristi cs for easy
y and accurrate detectioon, such ass gamma
rays witth appropriate energy levels and high yield. The physiccal propertiies must be suitable
for handdling and coorrosion resistant.

Cobalt aand gold arre useful in NTD appliccations as the activatio on detector for thermall neutron
detectioon [68]. Theey are stable elementss, so there is i neither innterfering rradioactivity
y nor an
uncertaiinty in the abundance.
a Gold has a strong ressonance cap pture at 4.9 eV, which must be
accountted for throough the caadmium diffference meethod in ord der to meaasure accuraately the
thermal neutron fluux. As its highh capturre cross-secction may cause
c flux ddepression near the
monitorr, the monittor must bee very smalll or thin. Gold
G per for low power experiments
is prop
like a performancee test of an irradiation facility sincce it has advantages suuch as a hig gh cross-
section, 100% abbundance, high h detecttion efficieency for in nduced gam mma rays and an
approprriately longg half-life. For
F routinee fluence monitoring,
m however, ththe cross-seection of
gold is too high annd the half--life of 98Auu is not lon ng enough. Therefore, cobalt is popularly
p
used forr fluence monitoring
m th
hough its crross-section n is also very high. If tthe detector is pure
cobalt, tthe consequuently tiny detector
d maakes handlin ng difficult. A cobalt ddetector dilu
uted with
aluminium may bee convenien nt for fluennce monito oring. The cross-sectio
c on shape off 94Zr is
rather similar to 300Si, and a zirconium fooil with eno ough size foor easily haandling can be used
for flueence monitooring due to t its low ccross-sectio on. Figure 13 shows a typical acctivation
cross-seection for reelated materrials [58].

FIG. 133. Neutron capture

c cross-sections of Cd, 30Si and typica
al activationn detectors for
f NTD
applicattion (see figgure caption
n).

8.2.3. C
Cadmium diffference meethod

The obsserved activvity of an acctivation deetector correesponds to a mixture oof activation

n caused
by thermmal, epitherrmal and fast neutrons.. In a well thermalized
t d region, sinnce the conttribution
of fast neutrons iss almost neegligible, a majority of o the non-thermal neeutron reacttions are
resonannce reactionns with epiithermal neeutrons. To o confirm the total coontribution of non-
thermal neutrons too the total reeaction rate , the cadmiu nce method can be appllied.
um differen

41
The neutron absorption cross-section of cadmium is very large for neutron energies up to
about 0.4 eV, above which it drops abruptly and remains low for high neutron energies as
shown in Figure 13. Cadmium with a thickness of only 0.5 mm acts as a selective neutron
filter, blocking low energy neutrons below 0.4 eV but allowing higher energy neutrons to pass
through with little attenuation. The cadmium ratio is defined as follows,
Activity of the bare detector
Cd ratio = , (8.8)
Activity of the cadmium covered detector
Thermal and resonance neutron contribution
= .
Only the resonance contribution

The Cd ratio can show the degree of thermalization for a given neutron field. However, this
value depends on the material used in neutron detection. For example, cobalt and gold have
different cross-section shapes as shown in Figure 13, so the Cd ratios for cobalt and gold are
different for an identical neutron spectrum.

8.2.4. Procedure of neutron flux determination

A procedure of neutron flux determination using gamma ray detection from an activated
monitor is as follows:
1) Set up gamma ray detection electronics;
2) Measure the gamma ray spectrum of standard sources for calibration;
3) Calculate the absolute photo-peak efficiency;
4) Draw the calibration curve using experimental condition;
5) Choose the proper activation monitor;
6) Place the activation monitor in the irradiation facility and activate it for a certain time;
7) After proper cooling time, transfer it to a counting position and count until sufficient
statistics are attained; one should note that the geometry of the flux monitor must be as
similar as possible to the one of the calibration source.
8) Read the data from the multi-channel analyzer and determine the area under the
photopeak;
9) Remove the sample and measure the background spectrum;
10) Substitute the data values into the proper equation;
11) Determine the neutron flux of the irradiation facility.

9. PROCEDURES OF NTD FOR SILICON

9.1. Overall procedure of NTD-Si wafer production

9.1.1. Production of ultra pure polysilicon rods

The production procedure of ultra pure polysilicon is as follows:

Silica (SiO2) → (Refining/Reduction) → Metallurgical Grade Silicon → (Purification)

→ Ultra Pure Trichlorosilane → Ultra Pure Polysilicon Rod.

This process is common for CZ-Si and FZ-Si, but the polysilicon rod of FZ-Si has a higher
price because it is purer than one of CZ-Si. The starting material, silica, is very abundant in
earth. After refining, metallurgical grade silicon with about 98% purity is obtained through

42
the reduction process using graphite in a kiln. Ultra pure trichlorosilane is made from a
reaction with hydrochloric acid followed by distillation. Finally, it is reduced by hydrogen to
obtain the ultra pure polysilicon rod. Its purity is about the ration of eleven-to-nine. The
reactions corresponding to the respective steps are as follows:

SiO2 + 2C → Si + 2CO2 (obtaining metallurgical grade silicon);

Si + 3HCl → HSiCl3 + H2 (obtaining trichlorosilane);
HSiCl3 + H2 → Si + 3HCl (obtaining polysilicon rod).

9.1.2. FZ single crystal growth

This process is common for both gas doped FZ-Si and NTD Si:

(Polycrystal Grinding) → (Polycrystal Etching) → (FZ Single Crystal Growth) →

(Block Cutting) → (Ingot Inspection) → (Ingot Surface Grinding).

While the CZ-Si crystal is made from broken polysilicon pieces, the FZ-Si crystal is made
from a polysilicon rod whose surface is ground and etched. The FZ-Si crystal is then cut into
a cylindrical ingot. An orientation of the crystal is measured by X-ray and its surface is
polished. For the case of gas-doped FZ-Si, doping is accomplished during crystal growth.
MCZ-Si is also sometimes treated with NTD. The NTD process for MCZ-Si after single
crystal growth is the same as FZ-Si.

9.1.3. NTD

An overall NTD process is as follows:

(Ingot Etching) → (Initial Resistivity Measurement) → (Ship to a Reactor) → (Storage

under suitable conditions e.g. no contact with stainless steel from storage racks) →
(Neutron Irradiation and Decay of Induced Radioactivity) → (Cleaning and Residual
Radioactivity Measurements) → (Ship Back) → (Heat Treatment) → (Resistivity
Measurement) → (Feedback of Measured Resistivity to the Reactor).

A silicon ingot is etched to remove impurities on the surface and the initial resistivity on the
top and bottom are measured before sending the ingot to a reactor site. The measured initial
resistivity is forwarded to the irradiation site to determine the irradiation fluence. The
procedure of NTD at the reactor site is shown in Figure 14.

FIG. 14. NTD procedure at the reactor site.

43
Ingots may arrive at the reactor site earlier or later than their data, including target resistivity.
The ingots are inspected and washed before irradiation. An irradiation plan may be set for a
reactor operation cycle or for a certain time period. After irradiation and sufficient cooling,
irradiated ingots are decontaminated. In parallel, the neutron fluence may be measured by
activation detectors irradiated along with the ingots. The resistivity of each irradiated ingot
may be predicted based on the fluence determined either by the activation detectors or
irradiation records. The residual radioactivity is inspected for each decontaminated ingot. If
the radioactivity is low enough, the ingot is shipped back to the customer. Preferably the
packaging which had been provided by the supplier for the shipment to the reactor is reused.
Final resistivity data returned by the customer are compared with predictions and utilized for
the improvement of future neutron irradiations.

9.1.4. Post NTD process

The post NTD process in the wafer company is as follows:

(Ingot Inspection) → (Annealing) → (Surface Grinding) → (Slicing) → (Lapping) →

(Annealing) → (Etching) → (Polishing) → (Resistivity Measurement). After inspection,
the ingot is heat treated for annealing. A wafer is produced from each end of the ingot. Then
the resistivity of each wafer is measured.

The method for measuring the resistivity is a four probe method. The uncertainty in the
measurement is about 2%. The measuring positions are usually seven points — at centre, four
points at a half radius away from the centre, and two points at five mm inside the edge.
During the test irradiation step that begins the irradiation procedure, several characteristics
such as resistivity variations, the annealing temperature, the carrier life time, impurities and
the crystal deficiency may be investigated. The main impurities are usually oxygen and
carbon.

After their return from the reactor, the ingots go through a special thermal process to anneal
irradiation induced lattice damage. Thermal neutrons merely generate interstitial-vacancy
pairs, whereas fast neutrons produce a large disordered region around the recoil track. In the
case of excessive crystal damage, extensive trapping results in a behaviour similar to an
intrinsic semiconductor and the irradiated silicon’s resistivity is immeasurably high [53]. At
any rate, it has been confirmed that the composition of the fast neutron spectrum affects the
performance of NTD-Si. Thus, it seems that NTD customers prefer to irradiate their silicon
ingots in heavy water reactors.

9.2. NTD procedures at the reactor site

9.2.1. Ingot undertaking

Some terminologies used by the wafer companies for the NTD irradiation are as follows:

— Batch: A quantity of silicon ingots which will be irradiated at the same time at an
irradiation hole.
— Carton: A packing box for holding ingots at an irradiation site. It could contain multiple
ingots.

44
— Packing list: A document containing a list of items related to the packing of the ingots.
Generally, it contains the quantity of ingots, identification numbers, dimensions and
weights, information on n or p type conduction in the starting material and initial and
target resistivities.
— Invoice: A customs invoice. It may be required for the import and export processes of
ingots.
— Instruction on neutron irradiation: A document containing the items related to neutron
irradiation. It may be enclosed in the packing carton or transferred by fax or e-mail.
— Order: In addition to the instructions for neutron irradiation, a document which requests
irradiations for specific ingots.

When a carton containing the ingots arrives at an irradiation site, first of all, a packing list or
invoice should be confirmed. Then items inside the carton should be checked with the packing
list. If the packaging was broken or raises concerns over ingot damage, it has to be recorded
with a photograph and the ingot supplier should be notified immediately. If multiple
companies have ordered the ingot irradiation, the documentation has to be classified by each
company.

9.2.2. Ingot inspection and washing

After arrival of an ingot shipment, the ingots should be examined for defects, whether they
originated before arriving at the irradiation site or not. The inspection should be done very
carefully because a very small crack on the ingot may cause a large break during processing.
Thus, if a small crack is found on the ingot, then the ingot supplier should be notified
immediately. Ingots should be treated very carefully and only using gloves to avoid any
mechanical damage and contamination. In some cases, each silicon ingot is covered by
aluminium foil to protect its surface or to protect it against gross contamination [63, 68]. Also,
it may be protected by a specially fabricated aluminium container [69].

The dimensions, including diameter, length and any eccentricity in shape, may be measured.
During the measurements, the following matters should be confirmed. If the measured values
are larger than the criteria below, a normal irradiation is impossible:

— Longitudinal length of the silicon ingot: Lengths at four points 90° apart are usually
measured to confirm the eccentricity of the ingot;
— Deviation in the flank: Upper, middle and lower points are measured. Large deviations
can cause an abnormal rotation of the irradiation rig or its precession;
— Diameter: The diameter should be within the allowed tolerance.

Careful washing of the ingot before irradiation is recommended to reduce the production of
radioactive waste during decontamination. An ingot may be washed by demineralized water
in an ultrasonic cleaner, and the moisture on the surface of the ingot is then dried. Further
cleaning by wiping it with a soft tissue containing alcohol to remove organic contamination
like fingerprints etc. is strongly recommended. Finally, any fine powder remaining on the
surface is removed by a dry air gun.

Due to the delicacy of single crystalline Si ingots the efficacy of the cleaning effort is difficult
to determine quantitatively. In any case careful optical inspection should ensure that the
surface is homogeneously shiny, depending on the finish of its polish which may vary
according to the supplier, and without any streaks or reams. In addition, once contamination
of an irradiated ingot has been observed, the contaminating isotopes need to be identified, i.e.

45
by gamma spectrometry analysis of the solvent used, in order to optimize the cleaning
procedure before further irradiations.

9.2.3. Irradiation planning

Irradiation planning includes:

— Determination of the neutron fluence for each ingot based on the initial and target
resistivity and an estimation of the irradiation time using the neutron flux predicted
from previous irradiation results and other available data;
— Preparation of an irradiation schedule;
— Determination of the loading scheme of the ingots, taking into account the initial
resistivity and length of each ingot;
— Determination of limiting radiation levels during irradiation and unloading of Si ingots
and provision of storage positions sufficiently deep below the surface of the reactor pool
to accept ingots with an unacceptably high dose rate due to, i.e. unexpected, highly
activated trace elements within the ingot;
— Providing a data sheet for each ingot that contains relevant information on the ingot and
will include the irradiation record, feedback resistivity, etc.;
— Working schedule and manpower assignment;
— Information of the wafer company on the irradiation and return schedule.

During irradiation planning, the following matters should be taken into account:

— The constant K may be slightly different depending on the wafer company if various
irradiation channels or ingot canisters are used. A proper K value should be used for
each company;
— The type of silicon ingot (initially n or p type) should be checked. Initially n type ingots
are the usual case, but sometimes p type ingots are also found;
— The deviation of ingot diameter from the nominal diameter should be checked. This
may influence the neutron flux at the ingot;
— The number of ingots per batch and their total length should be checked. When the total
length of an irradiation batch is significantly short for a uniform irradiation, an addition
of dummy silicon ingots with an appropriate length is recommended;
— The utilization plan for nearby irradiation facilities around the NTD hole should be
checked. The neutron flux at the NTD hole can be affected by changes at nearby
experimental facilities, especially at irradiation holes between the NTD hole and the
core centre;
— If the irradiation time is not long enough compared to the time for ingot loading and
unloading as well as to the ingot rotation speed, its impact upon a uniform irradiation
has to be checked;
— Initial resistivity values for both ends of the ingot are usually provided from the wafer
company. They are usually different, and proper compensation may be possible by
careful arrangement of ingots in the irradiation rig.

9.2.4. Ingot irradiation

After finishing the preparations for an irradiation, ingots are moved into the reactor site.
Subsequently, neutron activation monitors are attached if necessary, and then ingots are

46
loaded iinto the irraadiation rig
g. The rig iss inserted in
nto the irraadiation holee for irradiation, as
shown iin Figure 155. The irrad diation time and measu urements of the neutronn fluence by y on line
neutronn flux monittor are marrked. Whenn the irradiaation time or o the neutrron fluence reaches
pre-set targets, thee rig is un nloaded. Thhe process may be co omputerizedd for an au utomatic
unloadinng of the rig.
r When the flux sccreen metho od is applieed, the rig may function as a
neutronn screen as well, and its axial pposition maay be adjusted for an optimum location. l
Rotationn of the rig reduces thee radial nonuuniformity of the irradiiation.

In the case that the relation between the ssignal of thee on line flu ux monitor aand the neuttron flux
in the inngot is not well
w known n, a partitionned irradiatiion would beb preferablle. An accurate flux
in the inngot and thee axial non--uniformity of the irraddiation can beb obtainedd from the acctivation
fluence monitors at a the first irrradiation. T hould be apppropriately short to
The first irrradiation sh
achieve an accuratee irradiation n during thee second irraadiation.

In autom
matic or seemiautomatiic irradiatioon facilities an automaatic stop off the irradiaation and
unloadinng of the irrradiation caanisters cann be foreseeen in case of unstable rrotation of the
t ingot
or unexppected channges in reactor power.

matic facilityy at FRM II. The irrradiation

FIG. 155. Start off Si doping using the semiautom
channell is marked by a yellow
w circle.

47
9.2.5. Ingot cooling

After irradiation, the ingots are unloaded from the rig, and they are moved to a proper place
for cooling. When a water pool is used for cooling, baskets may be utilized to store the ingots
in the pool. If the baskets are not irradiated, they can be used for the withdrawal of the ingots
from the pool as well. It is important not to use steel baskets but only baskets made from Al or
organic materials. The cooling time is dependent on the fluence, but generally, two days of
cooling is enough for the handling of ingots without any shielding. If the number of silicon
ingots to be handled is large, sufficient space for cooling should be available.

9.2.6. Decontamination and radiation inspection

After enough cooling to enable simple handling of the ingot, it should be decontaminated.
Various cleansers such as pure water, soapy water, alcohol or acetone may be used. The
highly contaminated areas are washed with an acid solution [69]. It is highly recommended to
agree upon the chemicals to be used for cleaning the ingots with the Si supplier and to provide
strict written rules for the cleaning procedures to the staff in charge. Finally, the ingot is
sufficiently rinsed with pure water and dried. The use of a chemical for decontamination
complicates waste management. The careful washing of the ingot before irradiation reduces
the burden for decontamination and waste management.

A primary contamination inspection is done by a survey for surface contamination. The

smearing test is a typical way to check the surface contamination. A scanning device using a
Geiger-Müller detector for measuring β rays coming from the ingot is also used [70]. If the
residual radiation level is higher than the criterion for carrying an ingot out of the irradiation
site, it should be decontaminated again. It is desirable that this inspection should be done by
members of a health physics team who are independently responsible for radiological safety at
the irradiation site.

For the final radiation inspection an automatic device for release measurements, commercially
available from various suppliers, is recommended. The advantage of such a device is the fixed
geometry and calibration, and the issue of a reliable standard release document which will be
accepted not only by reactor operators and customers but also by supervision authorities. Of
course these devices need to be recalibrated periodically, and the calibration records need to
be kept available for future control purposes.
Finally the acceptable radiation level for release needs to be determined. Typically this value
is defined in national legislation for the producing reactor’s country, but also additional or
more stringent limits may be in place in the customer’s home country. At FRM II in
Germany, for example, the surface contamination must be <0.5 Bq/cm² and the specific
activity <0.09 Bq/g in order to allow the unrestricted release of the ingot.
9.2.7. Neutron fluence measurement and quality control

The neutron fluence measurement after irradiation can be done by using neutron activation
monitors irradiated with the ingot simultaneously. In most cases the readings of the SPND
monitors and/or irradiation time are sufficient. When the ingot comes out of cooling, the
fluence monitors are detached from the ingot. If the monitors were attached on the ingot
surface by adhesive tapes, these tapes and residual bonding agent should be completely
removed. These materials can be extra sources of residual radioactivity.

48
Various neutron activation monitors can be utilized, and the radioactivity of the irradiated
monitor can also be measured by various methods. The measured neutron fluence is compared
with the signals of real time neutron monitors during the irradiation so that the signals can be
calibrated. If the neutron fluence is within the irradiation criterion, it can be concluded that the
irradiation was successfully completed.

9.2.8. Ingot return shipment

If the radiation level is low enough, then the ingot can be taken out of the irradiation site and
shipped back. Documents required for shipping are usually an invoice, a traveller’s sheet, a
certification sheet from the radiation inspection and if applicable a customs clearance. On the
traveller’s sheet, the number of packages, ingot identification number, weight, target
resistivity and the surface radioactivity contamination level may be included. In the return
shipment, ingots are usually packed in the same cartons as in the initial undertaking.

9.2.9. Irradiation improvement

Achieving an accurate neutron fluence corresponding to the target resistivity as well as a

uniform irradiation are the prime targets of neutron irradiation with NTD in order to maintain
a high irradiation quality. Though the proportional constant between the resistivity and the
neutron fluence can be determined theoretically, in practice this constant is determined
experimentally for each company. Neutron fluence data measured at the irradiation site are
compared with the measured resistivity data from NTD customers so the proportional
constant can be precisely determined. As the amount of data increases, a more reliable
proportional constant can be generated, and the quality of NTD can be continuously improved.

9.3. Warnings and irradiation control

9.3.1. Warnings during irradiation and treatment

Warnings for abnormal situations during the irradiation and ingot treatments should be
conveyed by sound or lamps on a display [71]. A failure in ingot rotation during irradiation is
one of the most frequent cases. If the time without rotation is not short in comparison with the
total irradiation time, it could severely affect the final radial resistivity distribution and result
in resistivity gradient. The irradiation system may be designed to unload the irradiation rig
automatically when rotation is stopped. These warnings can be connected to the
instrumentation panel, the reactor control room and the NTD workers. Warnings may be
needed for the following cases:

— In the unlikely event of a failure in ingot rotation;

— In the unlikely event of a non uniform reactor power;
— Abnormal signal from the SPND;
— Abnormal temperature rise due to a failure in ingot cooling;
— Abnormal high or low level of an ingot radioactivity in the storage rack or during the
transportation of an ingot;
— When an ingot is strongly shocked;
— Occurrence of a transport error;
— End of an irradiation.

49
9.3.2. A
Abnormal reesults

The abnnormal results that can occur after an irradiatiion process of silicon inngots are ass follows:

— D
Damage to ann ingot duriing the irraddiation proccess;
— Prreposterous over- or unnder-irradiattion;
— D
Deviation froom the irraddiation criterria for accu
uracy and un
niformity;
— Siignificantly different data
d betweenn the final resistivity
r an
nd the valuue received from
f the
inngot supplieer;
— U
Unexpected highh dly removable contaminnation;
residuaal radioactivvity or hard
— A
Abnormally highh RRG.

For the case of connsiderable under-irradi

u iation, it can
n be correccted by addiitional irrad
diation if
the reacctor operatioon time is available.
a Foor other abnnormal casees, they musst be reporteed to the
ingot suupplier as soon as posssible. Closee communiccation with the ingot ssuppliers is the best
way to ggain trust annd resolve such
s abnormmal events ono mutual an nd transpareent basis.

9.3.3. C
Computer baased irradia
ation controol

Computter based coontrol of irrradiation, suuch as typified by Figure 16, is nnot only connvenient
but alsoo can guaraantee more accurate irrradiation. It may include autom matic driving of the
irradiatiion rig based on the pre-set
p irraadiation tim ment of rig position
me or fluencce, adjustm
dependiing on thee control rod r positioons, monito oring of th
he rotationn status, au utomatic
withdraawal of the rig in casee of no rotaation, monittoring of reeal time siggnals, generration of
warningg signals, traansmission of data to rremote placees, etc.

FIG.116. Computed assisted pprocess con

ntrol of Si doping
d at FR
RM II.

50
 10. DOCUMENTATION
To ensure the efficient control of silicon throughout all the stages of the irradiation process, a
good documentation system should be maintained. Generally, documentation for the NTD
process can be divided into two groups, such as documentation for business purposes and for
quality assurance. At the Harwell reactors, each piece of silicon was documented by the
customer’s name, consignment number (i.e. a collection of crystals delivered at the same
time), arrival date, irradiation number, crystal identifications, diameter, length, neutron dose
required, irradiation facility used, date and times of irradiation, details of rotation, date of
health physics check, financial details, invoice numbers and date of dispatch [28]. Also, at the
MURR reactor, the database was divided into three groups, such as the lot file, the schedule
file and the position table [72].

The following items may be included in the documentation for NTD quality:

— Customer name;
— Ingot identification number;
— Rod number ;
— Weight;
— Length;
— Diameter ;
— Initial resistivity of upper side;
— Initial resistivity of lower side;
— Average initial resistivity;
— Target resistivity;
— Requested thermal neutron;
— Delivery date;
— Name of irradiation facility;
— Target resistivity;
— K value for each company or each period;
— Control rod position;
— Irradiation position and direction in the rig;
— Expected thermal neutron flux;
— Expected irradiation time;
— Number of irradiations for each ingot;
— Signals of SPND before, during, after irradiation;
— Started timing of irradiation;
— Ended timing of irradiation;
— Measured irradiation fluence;
— Ratio of irradiated fluence to targeted fluence;
— Estimated resistivity after irradiation;
— Result of the residual radioactivity check;
— Abnormal events;
— Feedback resistivity from ingot supplier.

Since more and more wafer companies are obliged by their customers to cooperate only with
sub-contractors being certified according to the regulations of the EN ISO 9001 quality
control system, it is recommended to research reactors working in the field of NTD to get
licensed according to this system as it has been done long in the past by DR3 in Risø
(Denmark) and more recently by BR2, FRM II and other major NTD players.

51
 11. POSSIBLE EMERGENCIES
In general the physical properties of Si make it a favourable and easy going target for being
irradiated in a nuclear reactor. In this respect the most important parameters are:

• Its moderate neutron capture cross section;

• Only one natural isotope of silicon is radioactive;
• The short half-life of Si-31 (2.62 h);
• The absence of activation induced gamma radiation.

Nonetheless there are few concerns which must be kept in mind during the design of a Si
doping facility. First, single crystalline Si ingots are extremely brittle. Consequently the
reactor facility should provide a loading/unloading area guaranteeing that Si flakes, which
will unavoidably occur during handling of large amounts of Si, are collected and not
conveyed into more sensitive regions of the reactor pool. In addition the loading/unloading
area has to be easily accessible for cleaning, for example, by means of a vacuum assisted
system.

Also, Si ingots are considerably heavy. At FRM II, for example, Si batches up to 35 kg are
irradiated simultaneously using a single irradiation canister. Consequently devices for lifting
and transporting Si batches within the reactor pool must be laid out with a sufficient safety
margin against undesired drops that might cause serious damage to important reactor
installations. The same applies to the mechanical design of the interlocking device between
the lifter and the canister containing the Si ingots.

Finally, Si is usually provided by suppliers in batches possessing excellent purity.

Nonetheless a strategy dealing with irradiated Si ingots containing an unacceptable amount of
activated impurities should be available before the start of the irradiation service. If an
unexpected high level of radioactivity has been detected upon measurement of an irradiated Si
ingot, it must be verified at once whether the unexpected radioactivity is due to contamination
or activation. In case of contamination the entire process of Si handling at the reactor must be
checked in order to determine at which stage the ingot was polluted. In case of activation, on
the other hand, the supplier’s fabrication process most likely introduced the foreign atoms,
and the supplier must be informed immediately to allow him to inspect his manufacturing
process. In that case the Si ingot will most probably need to be disposed by the reactor
according to national regulations regarding radioactive waste classification and treatment,
since Si suppliers usually do not have a handling license for radioactive materials. Developing
a protocol in response to such scenarios is a recommended course of action during
negotiations between the reactor facility and Si supplier and should be included in the
subsequent contract.

52
 REFERENCES
[1] MEESE, J.M., ed., “Neutron Transmutation Doping in Semiconductors”, Neutron
Transmutation Doping in Semiconductors (Proc. 2nd Int. Conf. Columbia, Missouri,
1978), Plenum Press, New York (1979).
[2] GULDBERG, J., ed., “Neutron-Transmutation-Doped Silicon”, Neutron Transmutation
Doping of Silicon (Proc. 3rd Int. Conf. Copenhagen, 1980), Plenum Press, New York
(1981).
[3] LARRABEE, R.D., ed., “Neutron Transmutation Doping of Semiconductor Materials”,
Neutron Transmutation Doping (Proc. 4th Conf., Gaithersburg, Maryland, 1982) Plenum
Press, New York (1984).
[4] INTERNATIONAL ATOMIC ENERGY AGENCY, Silicon Transmutation Doping
Techniques and Practices (Proc. IAEA Consultants Mtg, Otwock-Świerk, Poland, 1985),
IAEA-TECDOC-456, IAEA, Vienna (1988).
[5] HEIJMINK LIESERT, B.J., GODLEWSKI, M.,, GREGORKIWICZ, T.,
AMMERLAAN, C.A.J., Neutron transmutation doping of GaP: Optical studies, Applied
Surface Science 50 (1991) 245.
[6] BOUDART, B., MARÍ, B., PRÉVOT, B., SCHWAB, C., Efficiency of neutron
transmutation doping of InP investigated by optical and electrical methods, Nuclear
Instruments and Methods B 63 (1992) 101.
[7] MARÍ, B., SEGURA, A. CHEVY, A., Electrical properties of neutron-transmutation-
doped InSe, Applied Surface Science 50 (1991) 415.
[8] ZHAO, W., LUKIC-ZRNIC, R., GORMAN, B.P., COTTIER, R.L., GOLDING, T.D.,
LITTLER, C.L., DINAN, J.H., ALMEIDA, L.A., DURA, J.A., LINDSTROM, R.M.,
SCHAAKE, H.F., LIAO, P., Magnetoconductivity tensor analysis of anomalous
transport effects in neutron irradiated HgCdTe epilayers, Physica E 20 (2004) 246.
[9] HALLER, E.E., PALAIO, N.P., RODDER, M., HANSEN, W.L., KREYSA, E., “NTD
germanium: A novel material for low temperature bolometers”, from Reference [3], p.21.
[10] SHLIMAK, I.S., Neutron transmutation doping in semiconductors: Science and
applications, Physics of The Solid State 41 (1999) 716.
[11] PASCA, E., BARUCCI, M., GUILIANI, A., OLIVIERI, E., RISEGARI, L., VENTURA,
G., Bolometers in magnetic field: Use of NTD Ge sensors, Nuclear Instruments and
Methods A 575 (2007) 433.
[12] BEEMAN, J., SILVER, E., BANDLER, S., SCHNOPPER, H., MURRAY, S.,
MADDEN, N., LANDIS, D., HALLER, E.E., BARBERA, M., The Constellation-X
Focal Plane Microcalorimeter Array an NTD-Germanium Solution, NASA Technical
Reports Server (http://ntrs.nasa.gov/search.jsp, Retrieved on 2008-01-29), 2001.
[13] YOUNG, M.H., HUNTER, A.T., BARON, R., MARSH, O.J., WINSTON, H.V.,
“Neutron transmutation doping of p type Czochralski-grown gallium arsenide”, from
Reference [3], 1.
[14] AFSAR, M.N., “Reliable identification of residual donors in high purity epitaxial
gallium arsenide by transmutation doping”, from Reference [3], 37.
[15] ALEXIEV, D., BUTCHER, K.S.A., EDMONDSON, M., TANSLEY, T.L., Neutron
transmutation doping of liquid phase epitaxial gallium arsenide, Nuclear Instruments
and Methods B 86 (1994) 288.
[16] KNOLL, G.F., Radiation Detection and Measurement, 3rd edn, John Wiley & Sons, New
York, (2000) 816 pp.
[17] SZE, S.M., IRVIN, J.C., Resistivity, mobility, and impurity levels in GaAs, Ge, and Si at
300 K, Solid State Electronics 11 (1968) 599.
[18] Retrieved from “http://en.wikipedia.org/wiki/Czochralski_process” on 2008-01-29.
[19] Provided by Siltronic AG, Munich, Germany.

53
[20] BALKANSKI, M., WALLIS, R.F., Semiconductor Physics and Applications, Oxford
University Press, New York, (2000), 521 pp.
[21] HÄRKÖNEN, J., et al., Particle detectors made of high-resistivity Czochralski silicon,
Nucl. Instr. and Meth. A 541 (2005) 202.
[22] Retrieved from “http://www.tf.unikiel.de/matwis/amat/elmat_en/kap_6/advanced/
t6_1_3.html” on 2008-01-29.
[23] LARK-HOROVITZ, K., "Nucleon-bombarded semiconductors”, Semiconductor
Materials (Proc. Conf. Reading, 1951), Butterworth, London (1951) 47-69.
[24] TANENBAUM, M., MILLS, A.D., Preparation of uniform resistivity n-type silicon by
nuclear transmutation, J. Electrochemical Society 108-2 (1961) 171-176.
[25] SCHNOLLER, M.S., IEEE Transactions on Electron Devices 21 (1974) 313.
[26] HEYDORN, K., ANDRESEN, K., “Neutron transmutation doping of silicon at Risø
National Laboratory”, from Reference [4], 17.
[27] NIELSEN, K.H., “New horizontal facility for neutron transmutation doping of silicon”,
Proc. 6th Mtg International Group on Research Reactors (IGORR), Taejon, Republic of
Korea, 1998, Korea Atomic Energy Research Insitute (KAERI), Daejeon (1998) 151.
[28] CRICK, N.W., “Silicon irradiations at the Harwell Laboratory of the United Kingdom
Atomic Energy Authority”, from Reference [4], 65.
[29] CRICK, N.W., “Silicon irradiations in the Harwell reactors”, Multipurpose Research
Reactors (Proc. Int. Conf., Grenoble, 1987), IAEA-SM-300/71P, IAEA Vienna (1988)
379.
[30] CUNDY, D.R., et. al., “SIDONIE - A new silicon irradiation facility in BR2”, Proc. Int.
Conf. on Irradiation Technology, Saclay, France, 1992.
[31] ALBERMAN, A., BLOWFIELD, H., “A review of silicon transmutation doping and its
practice at French and Belgian research reactors”, IAEA Technical Report Series 455,
IAEA Vienna (2007) 152.
[32] PONSARD, B., BLOWFIELD, H., “Production of radioisotopes and NTD silicon in the
BR2 reactor”, Products and Services of Research Reactors (Proc. IAEA Technical Mtg,
Vienna, 2010), IAEA, Vienna (in press).
[33] CHAWLA, R., CHRISTEN, R., HAMMER, J., LEHMANN, E., “Irradiation
Techniques employed at the 10 MWth SAPHIR reactor”, from Ref [30].
[34] GUNN, S.L., MEESE, J.M., ALGER, D.M., “High precision irradiation techniques for
NTD silicon at the University of Missouri Reactor, from Reference [1], 197.
[35] KOLIN, N.G., Neutron-Transmutation doping and radiation modification of
semiconductors: Current status and outlook, Russian Physics Journal 46 (2003) 543.
[36] PEAT, G., “The HIFAR 10VGR NTD Silicon Production Rigs and Facilities”, from Ref
[30].
[37] KOONEN, E., “BR2: The first year of operation after refurbishment”, Proc. 6th Mtg
International Group on Research Reactors (IGORR), Taejon, Republic of Korea, 1998,
Korea Atomic Energy Research Insitute (KAERI), Daejeon (1998) 25.
[38] KUZMINOV, V., BLOWFIELD, H., “Optimisation of the poolside facility for neutron
doping of silicon in high flux materials testing reactor BR2”, Transactions of the Joint
Conference on Research Reactor Fuel Management (RRFM) and 11th International
Group on Research Reactors (IGORR), Session IV, Lyon, 2007, European Nuclear
Society, Brussels (2007) 20.
[39] KUSUNOKI, T., MAGOME, H., TAKEUCHI, M., KOMEDA, M., KOBAYASHI, S.,
YAMASHITA, K., “Present status of production of neutron-transmutation-doped silicon
in JRR-3 and JRR-4”, Proc. 6th Int. Conf. on Isotopes, Seoul, 2008.
[40] KIM, M.S., KIM, H., KIM, Y.C., PARK, S.J., “Optimization of neutron transmutation
doping for silicon at HANARO”, Proc. 6th Int. Conf. on Isotopes, Seoul, 2008.

54
[41] AMOS, P.E., KIM, S., “New silicon irradiation rig design for OPAL reactor”,
Transactions of the Joint Conference on Research Reactor Fuel Management (RRFM)
and 11th International Group on Research Reactors (IGORR), Session IV, Lyon, 2007,
European Nuclear Society, Brussels (2007) 20.
[42] LOUW, P.A., STRYDOM, W.J., “Neutron transmutation doping of silicon at the
SAFARI-1 research reactor”, Proc. 6th Int. Conf. on Isotopes, Seoul, 2008.
[43] STYRDOM, W.J., LOUW, P.A., “Neutron transmutation doping of silicon at the
SAFARI-1 research reactor”, IAEA Technical Report Series 455, IAEA Vienna (2007)
179.
[44] LI, X., GERSTENBERG, H., NEUHAUS, I., “Silicon doping system at the research
reactor FRM II”, Proc. 6th Int. Conf. on Isotopes, Seoul, 2008.
[45] GERSTENBERG, H., LI, X., NEUHAUS, I., “Silicon doping at FRM II”, Transactions
13th Topical Mtg on Research Reactor Fuel Management (RRFM), Session V, Vienna,
2009, European Nuclear Society, Brussels (2009).
[46] Retrieved from “http://web.mit.edu/nrl/www/” on 2008-01-29.
[47] VARLACHEV, V.A., KUZIN, A.N., LYKHIN, S.V., SOLODOVNIKOV, E.S., USOV,
Y.P., FOTIN, A.V., TSIBUL’NIKOV, Y. A., Tomsk complex for neutron-transmutation
doping of silicon, Atomic Energy 79 (1995) 447.
[48] INTERNATIONAL ATOMIC ENERGY AGENCY, Research Reactor Database
(RRDB), available at http://nucleus.iaea.org/RRDB/ (November 2011).
[49] KHALIL, M.Y., SULTAN, M.A., HASSAN, M.H., ABDEL-FATTAH, A.Y.,
Parametric analysis of the neutron transmutation doping (NTD) facility at the Egyptian
Second Research Reactor (ETRR-2), Annals of Nuclear Energy 32 (2005) 355.
[50] SHEIBANI, S., MOATTAR, F., GHANNADI MARAGHEH, M., KHALAFI, H.,
Investigation of a simple and efficient method for silicon neutron transmutation doping
process in Tehran Research Reactor, Annals of Nuclear Energy 29 (2002) 1195.
[51] CARBONARI, A.W., PENDL, W., SEBASTIÃO, J.R., SAXENA, R.N., DIAS, M.S.,
An irradiation rig for neutron transmutation doping of silicon in the IEA-R1 research
reactor, Nuclear Instruments and Methods B 83 (1993) 157.
[52] SAXENA, R.N., “The IEA-R1 research reactor: 50 years of operating experience and
utilisation for research, teaching, and radioisotopes production”, Proc. Int. Conf. on
Research Reactors: Safe Management and Effective Utilization, Sydney, 2007, IAEA
Vienna (2008).
[53] from Reference [3], vii.
[54] SMITH, T.G.G., “Future reactor capacity for the irradiation of silicon”, from Reference
[3], 83.
[55] SZE, S.M., NG, K.K., Physics of Semiconductor Devices, John Wiley & Sons, Hoboken,
(2007) 815 pp.
[56] TOKAI RESEARCH AND DEVELOPMENT CENTER, The Handbook of Research
Reactor Utilization, Japan Atomic Energy Research Institute (formerly Japan Atomic
Energy Agency), 1994, 213 pp. (in Japanese)
[57] LAMARSH, J.R., Introduction to Nuclear Reactor Theory, Addison-Wesley, New York
(1966) 598 pp.
[58] CHADWICK, M.B., YOUNG, P.G., HETRICK, D., Los Alamos National Laboratory,
Oak Ridge National Laboratory EVAL-JUN97, Retrieved from “http://atom.kaeri.re.kr/”
on 2008-01-29.
[59] KIM, K.S., KIM, H., KIM, Y.C., PARK, S.J., “Optimization of neutron transmutation
doping for silicon at HANARO”, Proc. 6th Int. Conf. on Isotopes, Seoul, 2008.
[60] VON AMMON, W., Neutron transmutation doped silicon - Technological and economic
aspects, Nuclear Instruments and Methods B 63 (1992) 95.

55
[61] LI, H., WANG, R., ZHANG, H., ZHAO, J., LIU, G., CHEN, Y., DUAN, S., Annealing
behavior comparison of NTD FZ (H) Si irradiated in light-water-reactor and heavy-
water-reactor, Material Science and Engineering B 107 (2004) 119.
[62] KULIKOV, D.V., et al., Computer simulation of neutron transmutation doping of
isotopically engineered heterostructures, Nuclear Instruments and Methods B 228 (2005)
230.
[63] MEESE, J.M., Processing of semiconductor materials and devices by neutron irradiation,
Journal of Nuclear Materials 108 (1982) 715.
[64] General recommendations and summary, from Reference [4], 7.
[65] MCNPX user's manual, version 2.4.0. LA-CP-02-408, Los Alamos National Laboratory
(2002).
[66] KIM, M.S., LEE, C.S., OH, S.Y., HWANG, S.Y., JUN, B.J., Radial uniformity of the
neutron irradiation in the silicon ingots for the neutron transmutation doping at
HANARO, Nuclear Engineering and Technology 38 (2006) 93.
[67] KIM, H.S., OH, S.Y., JUN, B.J., KIM, M.S., SEO, C.G., KIM, H.I., Design of a neutron
screen for 6-inch neutron transmutation doping in HANARO, Nuclear Engineering and
Technology 38 (2006) 675.
[68] MEESE, J.M., GUNN, S.L., “Factors affecting phosphorous production rate in NTD
silicon”, from Reference [2], 223.
[69] WINKLER, H., “Silicon irradiation at Saphir”, from Reference [4], 29.
[70] DIAS, M.S., SEBASTIÃO, J.R., KOSKINAS, M.F., Methodology for Monitoring the
Residual Activity in Silicon Rods Irradiated with Thermal Neutrons, Applied Radiation
and Isotopes 57 (2002) 801.
[71] HANSEN, K., STENDAL, K., ANDRESEN, K., HEYDORN, K., “An automatic
controlled, heavy water cooled facility for irradiation of silicon crystals in the DR 3
reactor at RISØ National Laboratory, Denmark”, from Reference [3], 91.
[72] BERLINER, R., WOOD, S., “A computer controlled irradiation system for the
University of Missouri Research Reactor”, from Reference [1], 215.

56
 ANNEX I.
Database of Research Reactors Involved in NTD
This document has been developed as a supporting database of research reactors to describe
their functional characteristics and NTD capabilities. The intention is to promote, share and
provide an overall picture of the usefulness and effectiveness of irradiation facilities available
among the Member States as well as to increase cooperation between different research
reactor centres.

The information was collected through questionnaires sent to research reactor owners. All
data pertaining to the general description of reactor parameters; irradiation holes or facilities;
available expertise; and experience and availability for NTD of Si are stored in the database.
It is expected that this database will be useful to nuclear facility operators, nuclear system
designers, professional engineers and scientists among the Member States. This database
should be in particular useful for potential customers of NTD of Si.

It has to be recognized that in addition to 18 research reactors included in this database,

several other research reactors offer the service of Si doping on commercial basis and
contribute considerably to the worldwide production of NTD-Si, i.e. in the Russian Federation
and the US. Unfortunately more detailed information on these facilities was not made
available due to confidentiality restrictions.

The IAEA is grateful to all research reactor managers who have responded to the
questionnaires and made this information available for public use for other Member States.

I-1. Australia: OPAL (20 MW)

I-1.1. General description of the reactor parameters

Reactor name OPAL

Reactor type Pool
First criticality August 2006
Power level 20 MW
Maximum flux [cm-2s-1] (thermal) 2×1014 (fast) 2×1014
Material/Density U3Si2-Al 4.8 g/cm3
Enrichment 19.7% wt.
Fuel description
Type (element/bundle) MTR, flat plate
Clad Al
Absorber Hf
Coolant Light water
Moderator D2O
Reflector D2O
Core cooling Forced light water at 2000 m3/h

I-1.2. Irradiation hole or facility

Name Position Utilization field Flux [cm-2s-1] Status

(thermal/fast)
HF (2) Reflector Isotope 2×1014 In operation

57
 prroduction
MF (3) Reflector IIsotope ×1014
1.3× In operatio
on
prroduction
LF
F (12) Reflector IIsotope 8×1013 In operatio
on
prroduction
LRT
T (55) Reflector IIsotope 2×1012 – 1×1014 In operatio
on
prroduction
SR
RT (2) Reflector IIsotope 6 ×1012 – 2×1013 In operatio
on
prroduction
NT
TD (6) Reflector IIsotope 1.7 ×1012 – 1.6×1013 In operatio
on
prroduction

FIG
G. I-1.1. Coore view of OPAL reactor.

I-1.3. A
Available exxpertise

Field Computeer code Equ

uipment Capability
C Applicatio
on
or facility inccluding man or
power experience
Neutronnic design MCN NP
Thermoohydraulic design CFX
X
Mechannical design CAD
D
Safety analysis PARET/R
RELAP
Instrum
mentation andd control
Radiatiion dosimetryy
Radiatiion protectionn or residuall MCN NP/
radioacctivity monitooring Microsshield
Deconttamination
Theorettical backgroound for
materiaal irradiation

58
I-1.4. Experience and available facility for silicon doping

Objective Si doping
Irradiation material Silicon
Position Channel in Reflector
Name of hole or facility NTD1 - NTD6 (6 facilities in total)
Thermal neutron flux (with or without target) 1.7 ×1012 – 1.6×1013
Fast neutron flux 1 ×108 – 2×1010
Gamma ray flux
Longitudinal length of hole or facility 600 mm (usable)
Diameter or size 237 mm
Diameter of the accessible irradiated material 4”, 5”, 6” and 8”
Way of uniform irradiation SS and Al flux screen, Rotation at ≈ 15 min-1
Way of reduction for background radiation Light water
Irradiation time (max., min. and average) 3 h – 2 d; average 15 h
Way of rotation Hydraulic turbine system
Way of cooling for irradiation object Forced light water
Way of monitoring the irradiation fluence SPND + irradiation time
Way of decontamination Wash in demin water and cloth dry
Way of residual radioactivity survey Gamma ray counter
Exemption criteria for residual radioactivity 0.09 Bq/g, 0.5 Bq/cm²
Cadmium ratio for gold or other material > 30
Capacity per year ≈ 25 t (current)
Production during last 5 years ≈ 43 t (last 2 years)
Present status In operation

I-2. Belgium: BR2 (100 MW)

I-2.1. General description of the reactor parameters

Reactor name BR2

Reactor type Tank in Pool MTR
First criticality June 1961, major refurbishment in 1995 - 1997
Power level up to 100 MW
Maximum flux [cm-2s-1] (thermal) 1×1015 (fast) 6×1014
Material/Density Cermet UAlx; 1.3 g Utot/cm³
Enrichment 92,5 % wt
Fuel description
Type (element/bundle) Typically 32 assemblies
Clad Ag3
Absorber Hf
Coolant Light water
Moderator H2O + Be
Reflector Be
Core cooling forced light water stream at 1.8 m³/s

59
I-2.2. Irradiation hole or facility

Name Position Utilization Flux [cm-2s-1] Status

field (thermal/fast)
6 PRF Reflector Mo-99 2.5×1014 / In operation
production.
7 H1 Central Be plug Isotope Up to 1×1015 / In operation
production
In core Within fuel Isotope Up to 6×1014 (fast In operation
assembly production by neutrons)
fast neutrons
BRIGITTE pool Gamma --/ commission
irradiation

I-2.3. Available expertise

Information not available.

I-2.4. Experience and available facility for silicon doping, gemstone colouration and
membrane fabrication (Separately described for each facility)

ITEM-A DESCRIPTION
Objective NTD-Silicon
Irradiation material Silicon
Name of hole or facility SIDONIE
Position Within Be channel
Thermal neutron flux (with or without target) 5.5×1013 cm-2s-1
Fast neutron flux 0.2×1013 cm-2s-1
Gamma ray flux 2 W/g Al
Longitudinal length of hole or facility Batch length up to 800 mm
Diameter or size 200 mm
Diameter of the accessible irradiated material 125 mm
Way of uniform irradiation Translation and rotation
Way of reduction for background radiation ALARA principle
Irradiation time (max., min. and average) 15 min – 25 h; average 3.5 h/35 Ωcm
Way of rotation Electrical drive
Way of cooling for irradiation object Forced convection by H2O
Way of monitoring the irradiation fluence SPND + irradiation time
Capacity per year 15 t
Cd ratio 25

ITEM-B DESCRIPTION
Objective NTD-Silicon
Irradiation material Silicon
Position 6 channels in pool outside reactor vessel
Name of hole or facility POSEIDON
Thermal neutron flux (with or without target) 5.3×1012 cm-2s-1
Fast neutron flux 0.1×1013 cm-2s-1
Gamma-ray flux 0.8 W/cm³

60
 Longituudinal lengthh of hole or facility
f Hole len ngth 1275 mm, batch lenngth: 500 mm m
Diametter or size 200 mm m
Diametter of the acccessible irrad
diated materiaal 150 mm m, 200 mm
Way off uniform irraadiation Inversioon and rotatioon
Way off reduction foor backgroun nd radiation Irradiatiion in graphiite moderatorr
Irradiattion time (maax., min. and
d average) averagee 27 h/48 Ωcm m
Way off rotation Electriccal drive
Way off cooling for irradiation object
o Natural convection by b H2O
Way off monitoring the irradiatio on fluence SPND + irradiation time
Way off decontaminnation Ultrasou und in a dem
mineralized wwater medium m
Way off residual raddioactivity su
urvey By indeependent auth hority
Exempttion criteria for residual radioactivity
r y IAEA reegulations fo or 'exempt' raadioactive
materiall
Capacitty per year 60 t (SIDDONIE + PO OSEIDON)
Production during last
l 5 years 10 t (In core) and ~22 t (Out core))
Presentt status In operaation

FIG.
F I-2.1 Co
Core view off BR2 reacto
or.

I-3. Brazzil: IEA-R1

1 (5 MW)

I-3.1. G
General description off the reactoor parameteers

Reactorr name IEA-R1

Reactorr type MTR Open swimming pool
p reactor
First crriticality September 1957
Power llevel 5 MW
Maxim m-2s-1]
mum flux [cm (thermal) 2..7x1014 (fast)) 1.1x1014
Material/Deensity U3Si2-A m3
Al , 3.0 Ut/cm
Enrichmentt 19.75%
%
Fuel deescription
Type (elemeent/bundle) 31 MTRR assembliess/ core
Clad Aluminnium
Absorbber Cd+In+Ag
Coolannt Light waterr
Moderaator Light waterr
Reflecttor Graphite, B
Be
Core coooling Forced lightt water

61
I-3.2. Irradiation hole or facility

Name Position Utilization Flux [cm-2s-1] Status

field (thermal/fast)
Beam tubes (9) diffractometry Operational
radiography
BNCT
Photonuclear
reactions
In core positions Isotope Thermal:4.6×1013 Operational
production Fast: 1.3×1014
Material testing
Pneumatic Reflector border NAA Operational

I-3.3. Available expertise

Information not available.

I-3.4. Experience and available facility for silicon doping

ITEM DESCRIPTION
Objective Si doping
Irradiation material Silicon ingots
Position within graphite reflector
Name of hole or facility Si irradiation rig
Thermal neutron flux (with or without target) Without up to 7 x 1012 cm-2s-1
Fast neutron flux
Gamma ray flux
Longitudinal length of hole or facility Ingot length: 500 mm
Diameter or size 175 cm
Diameter of the accessible irradiated material 125 mm
Way of uniform irradiation Inversion of 2 ingots and rotation
Way of reduction for background radiation Position in graphite reflector
Irradiation time (max., min. and average)
Way of rotation Electrical drive
Way of cooling for irradiation object Natural convection
Way of monitoring the irradiation fluence SPND, time
Way of decontamination Distilled water
Way of residual radioactivity survey Beta counter
Exemption criteria for residual radioactivity 7.4 Bq/g,
Cadmium ratio for gold or other material ≈ 29 for Cobalt
Capacity per year ≈ 1.2 t – 30 Ωcm
Production during last 5 years
Present status Operational

62
 I-4. China: CARR (60MW)
I-4.1. General Description of the Reactor Parameters
Reactor name CARR
Reactor type Pool
First criticality 2010
Power level 60 MW
Maximum flux [cm-2s-1] (thermal) 8×1014 (fast) 1×1015
Material/Density U3Si2-Al/4.3 g/cm3
Enrichment 19.75 % wt.
Fuel description
Type (element/bundle) MTR
Clad Al
Absorber Hf
Coolant Light water
Moderator D2O
Reflector D2O
Core cooling Force

I-4.2. Irradiation hole or facility

Name Position Utilization Flux [cm-2s-1] Status
field (thermal/fast)
CI Reflector Cooling needed 8×1014 / prepared
IRS
I-125 Reflector I-125 4×1014 / prepared
MT Reflector Material /1×1014 prepared
irradiation
monitoring
AT Reflector NAA 6×1014/ prepared
NI Reflector RIs production 6×1014 / prepared

I-4.3. Available expertise

Field Computer code Equipment Capability Application
or facility including man or
power experience
Neutronic design WIMS/CITAT
ION/MCNP
Thermohydraulic design CFX/CFD
Mechanical design CAD/TDM
Safety analysis Relap
Instrumentation and control
Radiation dosimetry
Radiation protection or residual
radioactivity monitoring
Decontamination
Theoretical background for
material irradiation

63
I-4.4. Experience and available facility for silicon doping, gemstone colouration and
membrane fabrication

ITEM DESCRIPTION
Objective Silicon doping
Irradiation material Silicon
Position D2O tank
Name of hole or facility NTD Channel
Thermal neutron flux (with or without target) 1×1014 cm-2s-1 (without target)
Fast neutron flux <1×1012 cm-2s-1
Gamma ray flux
Longitudinal length of hole or facility 600 mm for NTD
Diameter or size
Diameter of the accessible irradiated material 3 in (1), 4 in (1), 5 in (5)
Way of uniform irradiation upside down and rotate
Way of reduction for background radiation
Irradiation time (max., min. and average)
Way of rotation Hydraulic
Way of cooling for irradiation object nature
Way of monitoring the irradiation fluence SPND
Way of decontamination rinsed by hydrochloric or hydrofluoric acid
Way of residual radioactivity survey
Exemption criteria for residual radioactivity surface < 10-4 μCi/cm2
Cadmium ratio for gold or other material
Capacity per year 50 t/yr (estimate)
Production during last 5 years
Present status Under construction and installation

I-5. China: HFETR (125 MW)

I-5.1. General description of the reactor parameters

Reactor name High Flux Engineering Test Reactor (HFETR)

Reactor type Pressure vessel
First criticality Dec. 27, 1979
Power level 125 MW
Maximum flux [cm-2s-1] (thermal) 6.2×1014 (fast) 1.7×1015
235
Material/Density U-Aluminium
Enrichment High: 90%, LFU: 20%
Fuel description
Type (element/bundle) Multi-tubular fuel element
Clad 305-Aluminium
Absorber Cd-In-Ag
Coolant Water
Moderator Water
Reflector Beryllium
Core cooling Force water cooling

64
I-5.2. Irradiation hole or facility

Name Position Utilization field Flux [cm-2s-1] Status

(thermal/fast)
Φ 230 mm Reflector Fuel, NTD 1.2×1014/3.4×1014 In use
Φ 150 mm Core Fuel, material 4.2×1014/1.6×1015 Core center
Φ 150 mm Core NTD 2.9×1014/1.0×1015 In use
Φ 120 mm Reflector Material, NTD 1.2×1014/2.5×1014 In use
Φ 63 mm Core Fuel 3.1×1014/1.2×1015 Other

I-5.3. Available expertise: Division and dedicated persons for each field

Information not available.

I-5.4. Experience and available facilities for silicon doping, gemstone colouration and
membrane fabrication

ITEM DESCRIPTION
Objective Silicon doping
Irradiation material Silicon
Position HFETR’s hole
Name of hole or facility Φ230, Φ150, Φ120 mm
Thermal neutron flux (with or without target) 1.2×1014 cm-2s-1 ~ 4.2×1014 cm-2s-1
Fast neutron flux 0.25×1015 ~ 1.6×1015 cm-2s-1
Gamma ray heating 0.9 ~ 11.0 W/g
Longitudinal length of hole or facility 1000 mm (reactivity), 500 mm for NTD
Diameter or size Φ120 ~ Φ230 mm
Diameter of the accessible irradiated material six inches
Way of uniform irradiation upside down and rotate
Way of reduction for background radiation
Irradiation time (max., min. and average)
Way of rotation automatic rotation (hydraulic)
Way of cooling for irradiation object water-naturally cooling
Way of monitoring the irradiation fluence
Way of decontamination supersonic wave + hydrochloric
Way of residual radioactivity survey clean, cleanse
Exemption criteria for residual radioactivity survey β< criteria(GB)=10-4 μCi/cm2
Cadmium ratio for gold or other material
Capacity per year 10 t
Production during last 5 years
Present status in use

65
 I-6. China: MJTR (5 MW)

I-6.1. General description of the reactor parameters

Reactor name Ming Jiang Test Reactor(MJTR)

Reactor type Pool
First criticality 1992
Power level 5 MW
Maximum flux [cm-2s-1] (thermal) 8.03×1013 (fast) 1.43×1014
235
Material/Density U-Aluminium
Enrichment HFU: 90% , LFU: 20%
Fuel description
Type (element/bundle) Multi-tubular fuel element
Clad 305-Aluminium
Absorber Cd-In-Ag
Coolant Water
Moderator Water
Reflector Beryllium
Core cooling Fore water cooling

I-6.2. Irradiation hole or facility

Name Position Utilization Flux [cm-2s-1] Status

field (thermal/fast)
Φ220 mm Reflector NTD 4.23×1013/0.96×1014 In use
Φ180 mm Reflector NTD 4.52×1013/0.98×1014 In use
Φ120 mm Reflector NTD 5.82×1013/1.09×1014 In use
Φ63 mm Core Material, NTD 6.03×1013/1.21×1014 Other

I-6.3. Available expertise

Information not available.

66
I-6.4. Experience and available facility for silicon doping, gemstone colouration and
membrane fabrication

ITEM DESCRIPTION
Objective Silicon doping
Irradiation material Silicon
Position reflector, core
Name of hole or facility Φ 220 ~ Φ63 mm
Thermal neutron flux (with or without target)
Fast neutron flux
Gamma ray flux 0.4 W/g
Longitudinal length of hole or facility 1000 mm (reactivity), 500 mm for NTD
Diameter or size Φ220 ~ Φ63 mm
Diameter of the accessible irradiated material 2 ~ 6 in
Way of uniform irradiation upside down and rotate
Way of reduction for background radiation
Irradiation time (max., min. and average)
Way of rotation automatic rotation (hydraulic)
Way of cooling for irradiation object water-naturally cooling, nature
Way of monitoring the irradiation fluence
Way of decontamination supersonic wave + Hydrochloric
Way of residual radioactivity survey cleanse
Exemption criteria for residual radioactivity survey β< criteria (China GB)=10-4 μCi/cm2
Cadmium ratio for gold or other material
Capacity per year 20 t
Production during last 5 years
Present status in use

I-7. Czech Republic: LVR-15 (10 MW)

I-7.1. General description of the reactor parameters

Reactor name LVR-15

Reactor type Tank reactor
First criticality 1989
Power level 10 MW
Maximum flux [cm-2s-1] (thermal) 1.8×1014 (fast) 2×1014
Material/Density UO2+Al
Enrichment 36%/19,7%
Fuel description IRT-2M/IRT-4M, 4/8 quarter
Type (element/bundle)
tube in one fuel assembly
Clad Al, SAV 1
Absorber Boron carbide
Coolant Light water
Moderator Light water
Reflector Be blocks/light water
Core cooling Forced light water flow 1750 m3/h

67
I-7.2. Irradiation hole or facility

Name Position Utilization Flux [cm-2s-1] Status

field (thermal/fast)
DONA C, D 9, 10 Si 2.7×1013/2.2×1012 In operation
Behind reflector
Vertical channel H5, H6, H8 Radioisotope H5, H8 1.3×1014 In operation
Reflector Production H8 8×1013
Central channel D.E,4,5 Ir or U for Mo99 1.7×1014 In operation
Irradiation
Rabbit system H1 reflector NAA 5×1013 In operation

I-7.3. Available expertise

Field Computer code Equipment Capability Application

or facility including man or
power experience
Neutronic design MCPN,
NODER
Thermohydraulic design RELAP 5,
LVR-15 STA
Mechanical design CAD
Safety analysis 2010

I-7.4. Experience and available facility for silicon doping

ITEM DESCRIPTION
Objective Si doping
Irradiation material Silicon
Position C, D, 9, 10 channel behind reflector
Name of hole or facility DONA 5
Thermal neutron flux (with or without target) 2.7×1013 cm-2s-1
Fast neutron flux 2.2×1012 cm-2s-1
Gamma ray flux
Longitudinal length of hole or facility 300 mm
Diameter or size 4”
Diameter of the accessible irradiated material 2 – 4”
Way of uniform irradiation Stainless steel shielding, rotation of ingot 1.3
times per min
Way of reduction for background radiation -
Irradiation time (max., min. and average) 20 min – 20 h; average 2 h
Way of rotation Electrical driver
Way of cooling for irradiation object Forced reactor cooling water
Way of monitoring the irradiation fluence Neutron coax and irradiation time
Way of decontamination Bath and decontamination
Way of residual radioactivity survey Gamma ray counter
Exemption criteria for residual radioactivity 4 kBq/m2 for P32
Cadmium ratio for gold or other material 5.3

68
 Capacitty per year 500 kg/y depending on orders
Production during last
l 5 years 1500 kgg
Presentt status In operaation

FIG
G. I-7.1. Corre view of LVR-15
L reacctor.

I-8. Francee: ORPHÉE (14 MW))

I-8.1. G
General description off the reactoor parameteers

Reactorr name ORPHEE

Reactorr type pool
First crriticality 1980
Power llevel 14 MW
Maxim m-2s-1]
mum flux [cm ×1014 (fast) 3×10
(thermal) 3× 3 12

Material/Deensity UAl
Enrichmentt 93%
Fuel deescription
Type (elemeent/bundle) plates
Clad Al
Absorbber Hf
Coolannt H2O
Moderaator D2O
Reflecttor H2O + Beryyllium
Core coooling forced lightt water stream
m

I-8.2. Irrradiation hole or facility

N
Name Position
n Utilizzation field Flux [ccm-2s-1] Status
(thermaal/fast)
I1 D2O tank
k IIsotope (thermal) 6×1013 In operation
n
prroduction
R11, R2 D2O tank
k Siliicon NTD (thermal) 1×1013 In operation
n
for Ø 125 mm
E33, E4 D2O tank
k Siliicon NTD (thermal) 5×1011 In operation
n
for Ø 151 mm

69
I-8.3. A
Available exxpertise

Informaation not available.

I-8.4. E
Experience and availab
ble facility for silicon doping

ITEM DESCR
RIPTION
Objectiive Si dopinng
Irradiattion materiall Silicon
Positionn 4 channnels in reflecttor (D2O tankk)
Name oof hole or faccility -
Thermaal neutron fluux (with or without
w targeet) 1.0×1013
1
cm-2s-1
Fast neeutron flux 1.0×10 cm-2s-1
10
1

Gammaa ray flux

Longituudinal lengthh of hole or facility
f 600 mm m
Diametter or size 160 mm m
Diametter of the acccessible irrad
diated materiaal 152 mm m
Way off uniform irraadiation flux screen, Rotationn of ingot at 2 rpm
Way off reduction foor backgroun nd radiation -
Irradiattion time (maax., min. and
d average) 2 h – 200 h; average 8 h
Way off rotation Electriccal drive
Way off cooling for irradiation object
o Natural convection byb H2O
Way off monitoring the irradiatio on fluence SPND + irradiation time
Way off decontaminnation Ultrason nic bath +
Way off residual raddioactivity su
urvey Gammaa ray counter
Exempttion criteria for residual radioactivity
r y IAEA
Cadmiuum ratio for gold or otherr material 500
Capacitty per year ≈ 10 t (ddepending onn Si diameterr)
Production during last
l 5 years ≈ 30 t
Presentt status In operaation

FIG
G. I-8.1. Corre view of Orphée
O reacctor.

70
 I-9. France: OSIRIS (70 MW)

I-9.1. General description of the reactor parameters

Reactor name OSIRIS

Reactor type pool
First criticality 1966
Power level 70 MW
Maximum flux [cm-2s-1] (thermal) 3×1014 (fast) 2.3×1014
Material/Density U3Si2Al
Enrichment 19.75%
Fuel description
Type (element/bundle) plates
Clad Al
Absorber Hf
Coolant H2O
Moderator H2O
Reflector H2O + Beryllium
Core cooling forced light water stream at 5300 m3/h

I-9.2. Irradiation hole or facility

Name Position Utilization field Flux [cm-2s-1] Status

(thermal/fast)
12 channels Core Material testing Fast 1–2.3×1014 In operation
15 channels reflector Material or fuel Thermal 1–3×1014 In operation
testing
Mo-99 & In operation
isotope
production

I-9.3. Available expertise

Information not available.

71
I-9.4. E
Experience and availab
ble facility for silicon doping

ITEM DESCR
RIPTION
Objectiive Si dopinng
Irradiattion materiall Silicon
Positionn 2 channnels in reflecttor (pool)
Name oof hole or faccility DIODO ON P1 and P2 2
Thermaal neutron fluux (with or without
w targeet) 4.0×1013
1
cm-2s-1
Fast neeutron flux 5.0×1012
1
cm-2s-1
Gammaa ray flux
Longituudinal lengthh of hole or facility
f 500 mm m
Diametter or size 130 mm m
Diametter of the acccessible irrad
diated materiaal 125 mm m
Way off uniform irraadiation Ni flux screen, Rotaation of ingott at 2 min-1
Way off reduction foor backgroun nd radiation -
Irradiattion time (maax., min. and
d average) 1 h – 100 h; average 5 h
Way off rotation Electriccal drive
Way off cooling for irradiation object
o Natural convection byb H2O
Way off monitoring the irradiatio on fluence SPND + irradiation time
Way off decontaminnation Ultrason nic bath +
Way off residual raddioactivity su
urvey Gammaa ray counter
Exempttion criteria for residual radioactivity
r y IAEA
Cadmiuum ratio for gold or otherr material 10
Capacitty per year ≈ 12 t (ddepending onn Si diameterr)
Production during last
l 5 years ≈ 20 t
Presentt status In operaation

FIG
G. I-9.1. Coore view of Osiris reacttor.

72
 I-10. Germany: FRM II (20 MW)

I-10.1. General description of the reactor parameters

Reactor name FRM II

Reactor type Pool
First criticality 2004
Power level 20 MW
Maximum flux [cm-2s-1] (thermal) 8×1014 (fast) 2×1014
Material/Density U3Si2-Al 3 g/cm3 , 1.5 g/cm³
Enrichment 92.5% wt.
Fuel description
Type (element/bundle) Compact Core
Clad AlFeNi
Absorber Hf
Coolant Light water
Moderator D2O
Reflector D2O
Core cooling forced light water stream at 300 kg/s

I-10.2. Irradiation hole or facility

Name Position Utilization field Flux [cm-2s-1] Status

(thermal/fast)
KBA Reflector Cooling needed 1.3×1014 / In operation
Isoptope prod.
RPA Reflector NAA Up to 7×1013 / In operation
JBE70 Reflector NAA, Fission 2×1014 In operation
track
GAB pont Gamma --/ commission
irradiation

I-10.3. Available expertise

Field Computer code Equipment Capability Application

or facility including man or
power experience
Neutronic design MCNP
Thermohydraulic design CFX/CFD
Mechanical design CAD
Safety analysis
Instrumentation and control
Radiation dosimetry
Radiation protection or residual
radioactivity monitoring
Decontamination
Theoretical background for
material irradiation

73
I-10.4. E
Experiencee and availa
able facilityy for silicon
n doping, gemstone
g coolouration and
membrrane fabricaation

ITEM DESCR
RIPTION
Objectiive Si dopinng
Irradiattion materiall Silicon
Positionn Channel in moderato
or tank
Name oof hole or faccility SDA
Thermaal neutron fluux (with or without
w targeet) 1.7×1013
1
cm-2s-1
Fast neeutron flux 1.0×1010
1
cm-2s-1
Gammaa ray flux
Longituudinal lengthh of hole or facility
f Batch leength 500 mm m
Diametter or size 237 mmm
Diametter of the acccessible irrad
diated materiaal 125, 150, 200 mm
Way off uniform irraadiation Ni flux screen, Rotaation of ingott at 5 min-1
Way off reduction foor backgroun nd radiation
Irradiattion time (maax., min. and
d average) 20 min – 2 d; averag ge 1-2 h
Way off rotation Electriccal drive
Way off cooling for irradiation object
o Natural convection byb H2O
Way off monitoring the irradiatio on fluence SPND + irradiation time
Way off decontaminnation Ultrason nic bath + Decopur FS5000
Way off residual raddioactivity su
urvey Gammaa ray counter
Exempttion criteria for residual radioactivity
r y 0.09 Bq q/g, 0.5 Bq/cm
m²
Cadmiuum ratio for gold or otherr material > 50
Capacitty per year ≈ 15 t (ddepending onn Si diameterr)
Production during last
l 5 years ≈ 40 t
Presentt status In operaation

FIG
G. I-10.1. Coore view of FRM
F II reactor.

74
 I-11. Indonesia: RSG-GAS (30 MW)

I-11.1. General description of the reactor parameters

Reactor name RSG-GAS (Multipurpose Reactor)

Reactor type Open pool
First criticality Jul. 29, 1987
Power level 30 MW
Maximum flux [cm-2s-1] (thermal) 2.43×1014 (fast) 2.8×1014
Material/Density 2.96 g/cm3
Enrichment 19.75%
Fuel description
Type (element/bundle) MTR
Clad AlMg
Absorber Ag-In-Cd
Coolant H2O
Moderator H2O
Reflector Be
Core cooling H2 O

I-11.2. Irradiation hole or facility

Name Position Utilization field Flux [cm-2s-1] Status

(thermal/fast)
IP B6 Topaz 1.67×1014 Operation
IP D9 Isotope 2.14×1014 Operation
IP G7 Isotope 1.50×1014 Operation
IP E4 Topaz 2.04×1014 Operation
CIP D6 Isotope 2.23×1014 Operation
CIP D7 Isotope 2.43×1014 Operation
CIP E6 Isotope 2.39×1014 Operation
CIP E7 Isotope 2.36×1014 Operation

I-11.3. Available expertise

Field Computer code Equipment Capability Application

or facility including man or experience
power
Neutronic design WIMSD, SRAC PC Excellent, 4 Core design,
2006, Batan-3DIFF, In-core fuel
Batan-EQUIL-2D. management,
MCNP, MCNPX Irradiation
service
Thermohydraulic COOLOD-N, PC Excellent, 3 T/H core
design PARET, EUREKA design
Mechanical design

75
 Safety analysis RELAP PC Excellent, 3 T/H accident
analysis
Instrumentation and
control
Radiation dosimetry GammaVision-32 HPGe Excellent, 2 NTD, NAA,
system Radionuclide
analysis
Radiation protection
or residual
radioactivity
monitoring
Decontamination
Theoretical
background for
material Irradiation

I-11.4 Experience and available facilities for silicon doping, gemstone colouration and
membrane fabrication

ITEM DESCRIPTION
Objective Gemstone colouration
Irradiation material Topaz
Position B6, E4, D9 (In core) and TIF (Out core)
Name of hole or facility IP (In core) and TIF (Out core)
Thermal neutron flux (with or without target) 1.50×1014 up to 2.43×1014 cm-2s-1
Fast neutron flux
Gamma-ray flux
Longitudinal length of hole or facility 60 cm
Diameter or size 7.3 cm
Diameter of the accessible irradiated material 7.0 cm
Way of uniform irradiation rotation at TIF
Way of reduction for background radiation N/A
Irradiation time (max., min. and average) 2 up to 10 hours (In core) and 8 up to 12
days (Out core)
Way of rotation manually (by hand)
Way of cooling for irradiation object Primary Water (In core) and Air cooling
(Out core)
Way of monitoring the irradiation fluence Surveymeter
Way of decontamination washing by ultrasonic
Way of residual radioactivity survey Surveymeter
Exemption criteria for residual radioactivity N/A
Cadmium ratio for gold or other material N/A
Capacity per year 1.8 t (In core) and ~400 kg (Out core)
Production during last 5 years 10 t (In core) and ~2 t (Out core)
Present status Operation

76
 FIG. I-11.1. Corre view of RSG-GAS
R reactor.

I-12. Republic
R ANARO (30 MW)
off Korea: HA

I-12.1. G
General deescription of
o the reacttor parameeters

Reactorr name HANARO

H
Reactorr type Open
O tank inn pool
First crriticality Feb.
F 1995
Power level 30
3 MWth
Maxim m-2s-1]
mum flux [cm (thermal)
( 4.00×1014 (fast) 2.1×1014
Material/Den
M nsity U3Si/
Enrichment
E 19.75 w/o
Fuel deescription element : rod
r
Type
T (elemennt/bundle) bundle: heexagonal withh 36 rods
circular with
w 18 rods
Clad
C Aluminium m
Absorbber Hafnium
H
Coolannt H2O
Moderaator D2O
Reflecttor D2O
Core coooling Forced
F up floow

77
 FIG. I-12.1. Corre view of HANARO
H reactor.

I-12.2. IIrradiation
n hole or fa
acility

N
Name Position
n Utilizzation field m-2s-1]
Flux [cm Status
(thermal//fast)
C
CT Inner corre Maaterial test, 014
4.4×10 In service
RI pproduction
IIR1 Inner corre FTL 014
3.9×10 Coommissioning
g
14
IIR2 Inner corre Maaterial test, 3.9×10
0 In service
RI pproduction
O
OR Outer corre Maaterial test, 014
3.4×10 In service
RI pproduction
C
CN Reflector Colld neutron 9×1013 D
Developing
source
NT
TD-2 Reflector Siliccon doping 5×1013 In service
NT
TD-1 Reflector Siliccon doping 5×1014 Testing
H
HTS Reflector RI pproduction 1×1011 In service
IP-44, 5, 15 Reflector RI 013
~ ×10 In service
NAA
A-1, 2, 3 Reflector NAA 6×109~ 1.5
5×1014 In service

I-12.3. A
Available expertise
e

F
Field Computer code Eqquipment or Capabilitty Appplication or
facility includin
ng exxperience
man powwer
Neutronnic design HANAFM MS, Workstation,
W Excellentt, 4 Core ddesign, In-corre
HELIOS, MCNP,
M Cluster, PC.
PC C fuel m
management,
etc. Irradiiation servicee

78
Thermohydraulic MATRA-h, CFX, Workstation, Excellent, 4 T/H core design,
design etc. PC Cluster, PC. CFD analysis
Mechanical ANSYS PC Excellent, 4 HANARO
design construction,
New facility
installation
Safety analysis MARS(RELAP5) PC Excellent, 2 T/H accident
analysis
Instrumentation - DCS, PLC Excellent, 4 Reactor power
and control automatic control
Radiation GammaVision-32 HPGe System Excellent, 2 NTD, NAA,
dosimetry Radionuclide
analysis
Radiation - RMS Excellent, 5 Radiation
protection or monitoring
residual
radioactivity
monitoring

I-12.4. Experience and available facility for silicon doping, gemstone colouration and
membrane fabrication

ITEM DESCRIPTION
Objective Neutron Transmutation Doping for Si ingots
Irradiation material Single crystal ingot
Position Vertical irradiation hole at D2O reflector
Name of hole or facility NTD1 & NTD2
Thermal neutron flux (with or without target) ~5.2×1013 cm-2s-1
Fast neutron flux
Gamma ray flux
Longitudinal length of hole or facility 120 cm
Diameter or size 220 cm (NTD1), 180 cm (NTD1)
Diameter of the accessible irradiated material 6” & 8”(NTD1), 5 & 6” (NTD2)
Way of uniform irradiation Rotating and Neutron screen
Way of reduction for background radiation N/A
Irradiation time (max., min. and average) 0.2 ~ 9.5 hours
Way of rotation Stepping motor
Way of cooling for irradiation object Natural convection and semi-forced pumping
by lower floater
Way of monitoring the irradiation fluence SPND (for real time monitoring)
Zr foils (for actual fluence)
Way of decontamination Washing with water
Way of residual radioactivity survey Smearing test
Exemption criteria for residual radioactivity B/G/ level, Beta < 4 kBq/m2
Cadmium ratio for gold or other material ~ 20
Capacity per year 20 t using NTD2 only
Production during last 5 years 45 t (2003~2007)
Present status In service at NTD2
Test irradiation at NTD1

79
 I-13. The Netherlands: HFR (45 MW)

I-13.1. General description of the reactor parameters

Reactor name High Flux Reactor (HFR)

Reactor type Tank in pool
First criticality 1961
Power level 45 MW
Maximum flux [cm-2s-1] Depending on experiment details
Material/Density U3Si2
Enrichment About 19.75%
Fuel description
Type (element/bundle) Plate
Clad Aluminium-alloy
Absorber Cd
Coolant Water
Moderator Water
Reflector Beryllium
Core cooling Force water cooling

I-13.2 Irradiation hole or facility

Information not available.

I-13.3. Available expertise

Information not available.

I-13.4. Experience and available facilities for silicon doping

ITEM DESCRIPTION
Objective Silicon doping
Irradiation material Silicon
Position PSF1-4
Name of hole or facility SIFA
Thermal neutron flux (with or without target) 2.5×1013 cm-2s-1
Fast neutron flux
Gamma ray heating Approximately 0.5 W/g
Longitudinal length of hole or facility 500 mm for NTD
Diameter or size ~ Φ160 mm
Diameter of the accessible irradiated material Six inches
Way of uniform irradiation Neutron absorption shield
Way of reduction for background radiation
Irradiation time (max., min. and average) A few hours
Way of rotation Automatic rotation (electric)
Way of cooling for irradiation object Water cooling
Way of monitoring the irradiation fluence SPND
Way of decontamination Rinse
Way of residual radioactivity survey Radiation monitoring

80
 Exemption criteria for residual radioactivity IAEA regulations
Cadmium ratio for gold or other material
Capacity per year
Production during last 5 years
Present status In use

I-14. Norway: JEEP II (2 MW)

I-14.1. General description of the reactor parameters

Reactor name JEEP II

Reactor type Tank
First criticality 1966
Power level 2 MW
Maximum flux [cm-2s-1] (thermal) 3×1013 (fast) 3.5×1013
Material/Density UO2/10.3 g/cm3
Enrichment 3.50%
Fuel description
Type (element/bundle) Bundle
Clad Aluminium
Absorber Cadmium
Coolant Heavy water
Moderator Heavy water
Reflector Heavy water
Core cooling Forced

I-14.2. Irradiation hole or facility

Name Position Utilization Field Flux [cm-2s-1] Status

(thermal/fast)
I1 - I5 Reflector NTD 9×1012/7×1010 Operating
I6 Reflector Isotope prod. 1.3×1013/NA Operating
E3,E4 Reflector NTD 4×1012/1010 Operating
E2 Reflector Isotope prod. 8×1012/NA Operating

I-14.3. Available expertise

Information not available.

81
I-14.4. Experience and available facility for silicon doping

ITEM DESCRIPTION
Objective Si-doping
Irradiation material Silicon
Position Reflector
Name of hole or facility I1 – I5, E3, E4
Thermal neutron flux (with or without target) 4×1012 – 9×1012 cm-2s-1 (Meas. Co-activat.)
Fast neutron flux

Gamma ray flux 1010 – 7×010 cm-2s-1 (Calc. E > 0.82 MeV)
Longitudinal length of hole or facility 90 cm
Diameter or size 12.7 cm
Diameter of the accessible irradiated material 12.7 cm
Way of uniform irradiation Rotation
Way of reduction for background radiation
Irradiation time (max., min. and average) Max 80 h, Min 2 h, Avg. 14 h
Way of rotation Electrical drive
Way of cooling for irradiation object Heavy water, forced
Way of monitoring the irradiation fluence Irradiation time
Way of decontamination
Way of residual radioactivity survey
Exemption criteria for residual radioactivity
Cadmium ratio for gold or other material
Capacity per year
Production during last 5 years 20 t
Present status Operating at full capacity

I-15. Poland: MARIA (30 MW)

I-15.1. General description of the reactor parameters

Reactor name MARIA

Reactor type Pool
First criticality December 17, 1974
Power level 30 MW
Maximum flux [cm-2s-1] (thermal) 2.5×1014 (fast) 5.0×1013
Material/Density UO2 – Al/2.8 g/cm3
Enrichment 36%
Fuel description
Type (element/bundle)
Clad Al Alloy
Absorber B4C
Coolant Light water
Moderator Be/H2O
Reflector Graphite, H2O
Core cooling Forced light water flow (6-9 m/s)

82
I-15.2. Irradiation hole or facility

Name Position Utilization field Flux [cm-2s-1] Status

(thermal/fast)
Irradiation graphite Radioisotopes
Φth = 2.5×1014 in operation
channels reflector production
Irradiation
beryllium Radioisotopes
channels with Φth = 1.3×1014 in operation
reflector production
rabbit system
Facility for graphite
Φf = 2.2×1012
minerals reflector/special Topaz colouring in operation
(E>0.5 MeV)
irradiation boral shielding

I-15.3. Experience and available facility for silicon doping

ITEM DESCRIPTION
Objective NTD of Silicon (Neutron Transmutation
Doping of Single Crystals)
Irradiation material Single silicon crystals
Position Channel located in graphite reflector zone
Name of hole or facility NTD-1
Thermal neutron flux (with or without target) 6.0×1012 n/cm2 s
Fast neutron flux 4.5×1010 n/cm2 s
Gamma heat generation < 0.1 W/g
Longitudinal length of hole or facility 500 mm
Diameter or size 157 mm
Diameter of the accessible irradiated material 5” or 6”
Way of uniform irradiation Neutron flux flattering: rotation plus
flipping with flux profile linearization
Way of reduction for background radiation Well moderated neutron spectrum in
graphite reflector
Irradiation time (max., min. and average) Dependent on target resistivity (7.5 Ω.cm
– 100 Ω.cm)
Way of rotation Electric drive
Way of cooling for irradiation object Forced convection by pool cooling flow
Way of monitoring the irradiation fluence SPND – detector system
Way of decontamination Ultrasonic bath
Way of residual radioactivity survey Gamma & beta ray counter
32
Exemption criteria for residual radioactivity P: < 2.5 Bq/g
Cadmium ratio for gold or other material Epithermal index r= 0.02
Capacity per year 6200 kg/y (for ingots diameter 5 in)
9500 kg/y (for ingots diameter 6 in)
Production during last 5 years ca. 300 kg (tests)
Present status Ready for operation; calibration tests for
new customer

83
 I-16. South Africa: SAFARI-1 (20 MW)

I-16.1. General description of the reactor parameters

Reactor name SAFARI-1
Reactor type Tank in pool, Oak Ridge design type
First criticality 1965
Power level 20 MW
Maximum flux [cm-2s-1] (thermal) 2.7×1014 (fast) 1.1×1014
Material/Density U3Si2-Al , 4.88Ut/cm3
Enrichment 19.75%
Fuel description
Type (element/bundle) 19 fuel plates/element
Clad Aluminium
Absorber Cd control rod section
Coolant Light water
Moderator Light water
Reflector Be
Core cooling Forced light water

I-16.2. Irradiation hole or facility

Name Position Utilization field Flux [cm-2s-1] Status
(thermal/fast)
Beam tubes (6) diffraction ≈2×107 thermal Note 1
radiography Operational
SANS
In core positions Isotope Thermal:2.7×1014 Operational
production Fast: 1.1×1014

Hydraulic Reflector border Activation Thermal: 2×1014 Operational

Isotope Fast: 1.1×1014
production
Ringas Reflector border Activation Thermal: 1×1014 Operational
analysis Fast: 1.1×1014
Note 1: Under
reconstruction

I-16.3. Available expertise

Field Computer code Equipment Capability Application
or facility including man or
power experience
Neutronic design MCNP
Thermohydraulic design RELAP,
ANSYS
Mechanical design Solid Edge
Safety analysis
Instrumentation and control
Radiation dosimetry

84
 Radiattion protectioon or residuaal
raddioactivity monitoring
m
Decontaminnation
Theooretical backkground for
m
material irraddiation

I-16.4. E
Experiencee and availa
able facilityy for silicon
n doping

ITEM DESCR
RIPTION
Objectiive ng
Si dopin
Irradiattion materiall Silicon ingots
Positionn Pool sid
de
Name oof hole or faccility SILIRAAD
Thermaal neutron fluux (with or without
w targeet) Withoutt 1.5 x 1014 cm
c -2s-1
Fast neeutron flux Withoutt 2.6 x 1013 cm
c -2s-1
Gammaa ray flux
Longituudinal lengthh of hole or facility
f Ingot leength ≤ 600 mm
m
Diametter or size
Diametter of the acccessible irrad
diated materiaal Diameteer ≤ 152 mm m
Way off uniform irraadiation Translattion and rotaation
Way off reduction foor backgroun nd radiation
Irradiattion time (maax., min. and
d average) ≥ 1 houur
Way off rotation Electriccal drive
Way off cooling for irradiation object
o Natural convection
Way off monitoring the irradiatio on fluence SPND, time
Way off decontaminnation Ultrason nic wash
Way off residual raddioactivity su
urvey Beta cou unter
Exempttion criteria for residual radioactivity
r y 1000 Bq q/g, 1MBq to
otal activity
Cadmiuum ratio for gold or otherr material ≈ 13 forr cobalt
Capacitty per year ≈ 20 t depending
d on
n resistivity aand diameter
Production during last
l 5 years ≈ 90t
Presentt status Operatio onal

FIG. I-16.1. Corre view of SA

AFARI-1 reeactor.

85
 I-17 Thaailand: TRR
R (2MW)

I-17.1. G
General deescription of
o the reacttor parameeters

Reactorr name Thai Researrch Reactor

Reactorr type Swimming pool type TR
RIGA Mark III Reactor
First crriticality Nov. 7, 19777
Power llevel 2 MW (maxximum)
Maxim m-2s-1]
mum flux [cm ×1013 (fast) 1×10
(thermal) 3× 1 13

Material/Deensity UZZrH
Enrichmentt 20%%
Fuel deescription
Type (elemeent/bundle) element
Clad SS--304
Absorbber B4C
Coolannt H2O
Moderaator H2O
Reflecttor Graphite
Core coooling Natural connvection

FIIG. I-17.1. C
Core view off TRR reacttor.

I-17.2. IIrradiation
n hole or fa
acility

Name Position Utilizaation field m-2s-1]

Flux [cm us
Statu
(thermal//fast)
In-coree CT,C8, C12,
C F3, F29
9, G5, G33, Isotope production,
p 1–3×10013 /
In usse
facilityy G22 NAA
N
Out- Rotary speecimen rack (lazy susan)), NAA
N 4×1011/1.22×1011 In usse
core A1, A22, A4, CA2, CA3, TC NAA
N
facilityy Wett tube and vooid tank Gemstonee colouration
n
Neutronn 8” south Neutron radiography 7×1055/ In usse
beam 8” North Neeutron
port 6” North diffracctrometry
PG
GNAA

86
I-17.3. Available expertise

Field Computer Equipment or Capability inclu. Application or

code facility man power experience
Neutronic design SRAC Personal computers, Design and Core design and
MVP Computer cluster Calculation analysis,
(4 persons) Core management
Thermohydraulic COOLODN Personal computers Design and Fuel Temperature
design Calculation distribution analysis
(2 persons)
Mechanical AUTOCAD Personal computers, Design and A number of irradiation
design Fabrication facility fabrication facility designs and
(4 persons) fabrications
Safety analysis EUREKA2/ Personal computers Analysis and Transient analysis of
RR Calculation TRR-1/M1
(2 persons)
Instrumentation - Electronics centre Design and Design and production
and control production of survey meter, Design
(3 persons) and production of MCA
Radiation - TLD program for Calibration and Providing TLD service
dosimetry neutron and gamma measurement within the country
dosimetry 2 personnel
Radiation - A wide range of Radiation Safety Health physics operation
protection or radiation protection program (5 for TRR-1/M1, Practice
residual equipment available Radiation Safety of Radiation Safety
radioactivity from Radiation officers) program
monitoring Safety Section
Decontamination - Waste management Decontaminatio Waste management and
centre n decontamination in
(5 persons) support of TRR-1/M1
Theoretical - XRF Analysis, R&D R&D in gemstone
background for XRD (6 persons) irradiation, R&D in
material NAA membrane fabrication,
irradiation UV R&D in polymerization
spectrophotometer by radiation processing

87
I-17.4. Experience and available facility for silicon doping, gemstone colouration and
membrane fabrication

ITEM DESCRIPTION
Objective Gemstone Colouration
Irradiation material Topaz
Position Wet tube, Void tank
Name of hole or facility W1 - 5
Thermal neutron flux (with or without target) 4×1011 cm-2s-1
Fast neutron flux 1.2×1011 cm-2s-1
Gamma ray flux
Longitudinal length of hole or facility 60 cm
Diameter or size 12.5 cm
Diameter of the accessible irradiated material -
Way of uniform irradiation rotation
Way of reduction for background radiation Cd Filter
Irradiation time (max., min. and average) 400 h
Way of rotation Manual
Way of cooling for irradiation object -
Way of monitoring the irradiation fluence flux and time control
Way of decontamination ultrasonic cleaner
Way of residual radioactivity survey NaI(Tl) system, HPGe system
Exemption criteria for residual radioactivity 2 nCi/g
Cadmium ratio for gold or other material 3.85
Capacity per year 100 kg
Production during last 5 years ~ 0.5 t
Present status Fully utilized

I-18. Vietnam: DNRR (500 kW)

I-18.1. General description of the reactor parameters

Reactor name Dalat Nuclear Research Reactor (DNRR)

Reactor type Pool type
First criticality Nov. 01, 1983
Power level 500 kW
Maximum flux [cm-2s-1] (thermal) 2.19×1013 (fast) 9.66×1012
U-235 /1.4 g/cm3 (36.0%
enrichment)
Material/Density
U-235 /2.5 g/cm3 (19.7%
enrichment)
Fuel description Enrichment 36% and 19.7%
Type
Fuel element of tube type
(element/bundle)
Clad Al alloy

88
 Absorbber B4C
Coolannt H2O
Moderaator H2O
Reflecttor Be, graphitee
Core coooling Natural connvection

FIG. I-18.1. Core vview of Dala

at Research Reactor.

I-18.2. IIrradiation
n hole or fa
acility

Naame Position
n U
Utilization fieeld Flux x [cm-2s-1] Status
(theermal/fast)
Neutron trap Centree RI 2.1×1013/2.7×1012 In use
12 11
Rotary sspecimen Graphite refflector RI, Gem 3.5×10 /1.0×10 In use
raack
Channnel 1.4 Cell 1.4
4 RI, Gem 9.6×1012/7.1×1012 In use
12 12
Channnel 7.1 Cell 7.1 NAA 6.2×10 /8.6×10 In use
Channnel 13.2 Cell 13..2 NAA 6.6×1012/7.9×1012 Upgradin
ng
10 7
Thermaal column Next to refllector NAA 9.2×10 /5.3×10 In use
12 110
Beam pport No.1 Raadially orien
nted tube 1.30×10 /1.35×10 Not
availablee
Beam pport No.2 Raadially orien
nted tube 5.85×1012/2.44×1099 Not
availablee
Beam pport No.3 Tangential tube (nn,2g) reactio
on 9.60×1011/9.35×1099 In use
study
Beam pport No.4 Raadially orien GNAA, neutrron 3.32×1012/8.75×10110
nted tube PG In use
ffiltered beam
ms

89
I-18.3. Available expertise

Field Computer Equipment or Capability Application or

code facility including man experience
power
Neutronic design WIMSD, Workstation 5 For Dalat
MCNP4C2, research reactor
SRAC2006 calculation
Thermohydraulic COOLODN, Workstation 4
design RELAP5,
EURECKA
Mechanical
design
Safety analysis RELAP5, Workstation 5
SRAC2006
Instrumentation ASUR-14R 1
and control
Radiation Gamma, 5 Dose control
dosimetry neutron
dosimeters
Radiation Radiation 2 Radiation
protection or monitoring monitoring
residual system around reactor
radioactivity building
monitoring
Decontamination
Theoretical
background for
material
irradiation

90
I-18.4. Experience and available facility for silicon doping, gemstone colouration and
membrane fabrication

ITEM DESCRIPTION
Objective Gem irradiation
Irradiation material Topaz
Position Cell 1-4, graphite reflector
Name of hole or facility Channel 1-4, Rotary specimen rack
Thermal neutron flux (with or without target) Without target: 9.6×1012 cm-2s-1 (Channel 1-
4), 3.5×1012 cm-2s-1 (Rotary specimen rack)
Fast neutron flux 7.1×1012 cm-2s-1 (Channel 1-4),
1.0×1011 cm-2s-1 (Rotary specimen rack)
Gamma ray flux
Longitudinal length of hole or facility 200 mm
Diameter or size 28 mm
Diameter of the accessible irradiated material
Way of uniform irradiation
Way of reduction for background radiation
Irradiation time (max., min. and average) 100 h, 30 h, 60 h
Way of rotation
Way of cooling for irradiation object
Way of monitoring the irradiation fluence
Way of decontamination H2 O
Way of residual radioactivity survey
Exemption criteria for residual radioactivity
Cadmium ratio for gold or other material 4.53 (Au) for Rotary specimen rack
Capacity per year 30 – 50 kg expecting
Production during last 5 years
Present status Studying

91
 GLOSSARY AND ABBREVIATIONS
ARV Axial Resistivity Variation

CCD Charged Coupled Device

CZ Czochralski Process

FZ Floating Zone Method

HPGe High Powered Germanium Detector

ICI International Conference on Isotopes

IGORR International Group on Research Reactors

IRRACC Irradiation Accuracy

JAEA Japan Atomic Energy Agency

JAERI Japan Atomic Energy Research Institute

KAERI Korea Atomic Energy Research Institute

NTD Neutron Transmutation Doping

RCA Regional Cooperative Agreement

RR Research Reactor

RRFM Research Reactor Fuel Management Conference

RRG Radial Resistivity Gradient

SCK/CEN Center for Nuclear Energy Research

SIDONIE Silicon Doping by Neutron Irradiation Experiment

SPND Self-Powered Neutron Detector

UKAEA United Kingdom Atomic Energy Authority

93
 CONTRIBUTORS TO DRAFTING AND REVIEW
P. Bode Technical University Delft, Delft, Netherlands

H. Gerstenberg Technical University of Munich, Munich, Germany

B.J. Jun Korean Atomic Energy Research Institute (KAERI), Daejeon,

Republic of Korea

T. Karma Australian Nuclear Science and Technology Organisation (ANSTO),

Lucas Heights, Australia

M.S. Kim Korean Atomic Energy Research Institute (KAERI), Daejeon,

Republic of Korea

A. Kornduangkaeo Thailand Institute of Nuclear Technology (TINT), Bangkok, Thailand

N. Peld International Atomic Energy Agency

D. Ridikas International Atomic Energy Agency

A. Tarigan National Nuclear Energy Agency (BATAN), Serpong-Tangarang,

Indonesia

This TECDOC was suggested and planned during two Technical Meetings:

— IAEA Technical Meeting on Irradiation Technology, held at IAEA Headquarters,

Vienna, Austria, from 13–17 May, 2007, and;
— IAEA Technical Meeting on Irradiation Technology, held in Pattaya City, Thailand,
from 20–22 February, 2008.

Both meetings were conducted under the regional IAEA TC project RAS4026 ‘Adding Value
to Materials through Irradiation with Neutrons,’ and each of its participants is to be thanked
for his or her participation and effort towards the publication of this document.

95
@
1R

:KHUHWRRUGHU,$($SXEOLFDWLRQV
,QWKHIROORZLQJFRXQWULHV,$($SXEOLFDWLRQVPD\EHSXUFKDVHGIURPWKHVRXUFHVOLVWHGEHORZ
RUIURPPDMRUORFDOERRNVHOOHUV3D\PHQWPD\EHPDGHLQORFDOFXUUHQF\RUZLWK81(6&2FRXSRQV

$8675$/,$
'$,QIRUPDWLRQ6HUYLFHV:KLWHKRUVH5RDG0,7&+$0
7HOHSKRQH‡)D[
(PDLOVHUYLFH#GDGLUHFWFRPDX‡:HEVLWHKWWSZZZGDGLUHFWFRPDX

%(/*,80
-HDQGH/DQQR\DYHQXHGX5RL%%UXVVHOV
7HOHSKRQH‡)D[
(PDLOMHDQGHODQQR\#LQIRERDUGEH‡:HEVLWHKWWSZZZMHDQGHODQQR\EH

&$1$'$
%HUQDQ$VVRFLDWHV)RUEHV%OYG6XLWH/DQKDP0'86$
7HOHSKRQH‡)D[
(PDLOFXVWRPHUFDUH#EHUQDQFRP‡:HEVLWHKWWSZZZEHUQDQFRP
5HQRXI3XEOLVKLQJ&RPSDQ\/WG&DQRWHN5G2WWDZD2QWDULR.--
7HOHSKRQH‡)D[
(PDLORUGHUGHSW#UHQRXIERRNVFRP‡:HEVLWHKWWSZZZUHQRXIERRNVFRP

&+,1$
,$($3XEOLFDWLRQVLQ&KLQHVH&KLQD1XFOHDU(QHUJ\,QGXVWU\&RUSRUDWLRQ7UDQVODWLRQ6HFWLRQ32%R[%HLMLQJ

&=(&+5(38%/,&
6XZHFR&=652.OHFDNRYD3UDKD
7HOHSKRQH‡)D[
(PDLOQDNXS#VXZHFRF]‡:HEVLWHKWWSZZZVXZHFRF]

),1/$1'
$NDWHHPLQHQ.LUMDNDXSSD32%2; .HVNXVNDWX ),1+HOVLQNL
7HOHSKRQH‡)D[
(PDLODNDWLODXV#DNDWHHPLQHQFRP‡:HEVLWHKWWSZZZDNDWHHPLQHQFRP

)5$1&(
)RUP(GLWUXH-DQVVHQ32%R[)3DULV&HGH[
7HOHSKRQH‡)D[
(PDLOIRUPHGLW#IRUPHGLWIU‡:HEVLWHKWWSZZZIRUPHGLWIU
/DYRLVLHU6$6UXHGH3URYLJQ\&DFKDQ&HGH[
7HOHSKRQH‡)D[
(PDLOURPXDOGYHUULHU#ODYRLVLHUIU‡:HEVLWHKWWSZZZODYRLVLHUIU

*(50$1<
8129HUODJ9HUWULHEVXQG9HUODJV*PE+$P+RIJDUWHQ'%RQQ
7HOHSKRQH‡)D[RU
(PDLOEHVWHOOXQJ#XQRYHUODJGH‡:HEVLWHKWWSZZZXQRYHUODJGH

+81*$5<
/LEURWUDGH/WG%RRN,PSRUW32%R[+%XGDSHVW
7HOHSKRQH‡)D[‡(PDLOERRNV#OLEURWUDGHKX

,1',$
$OOLHG3XEOLVKHUV*URXSVW)ORRU'XEDVK+RXVH-1+HUHGLD0DUJ%DOODUG(VWDWH0XPEDL
7HOHSKRQH‡)D[
(PDLODOOLHGSO#YVQOFRP‡:HEVLWHKWWSZZZDOOLHGSXEOLVKHUVFRP
%RRNZHOO1LUDQNDUL&RORQ\'HOKL
7HOHSKRQH‡)D[
(PDLOERRNZHOO#YVQOQHW

,7$/<
/LEUHULD6FLHQWLILFD'RWW/XFLRGL%LDVLR³$(,28´9LD&RURQHOOL,0LODQ
7HOHSKRQHRU‡)D[
(PDLOLQIR#OLEUHULDDHLRXHX‡:HEVLWHZZZOLEUHULDDHLRXHX
-$3$1
0DUX]HQ&RPSDQ\/WG1LKRQEDVKLFKRPH&KXRNX7RN\R
7HOHSKRQH‡)D[
(PDLOMRXUQDO#PDUX]HQFRMS‡:HEVLWHKWWSZZZPDUX]HQFRMS

5(38%/,&2).25($
.,16,QF,QIRUPDWLRQ%XVLQHVV'HSW6DPKR%OGJQG)ORRU<DQJ-DHGRQJ6HR&KR*6HRXO
7HOHSKRQH‡)D[‡:HEVLWHKWWSZZZNLQVUHNU

1(7+(5/$1'6
'H/LQGHERRP,QWHUQDWLRQDOH3XEOLFDWLHV%90$GH5X\WHUVWUDDW$1/%=+DDNVEHUJHQ
7HOHSKRQH  ‡)D[  
(PDLOERRNV#GHOLQGHERRPFRP‡:HEVLWHKWWSZZZGHOLQGHERRPFRP
0DUWLQXV1LMKRII,QWHUQDWLRQDO.RUDDOURRG32%R[&==RHWHUPHHU
7HOHSKRQH‡)D[
(PDLOLQIR#QLMKRIIQO‡:HEVLWHKWWSZZZQLMKRIIQO
6ZHWVDQG=HLWOLQJHUEY32%R[6=/LVVH
7HOHSKRQH‡)D[
(PDLOLQIRKR#VZHWVQO‡:HEVLWHKWWSZZZVZHWVQO

1(:=($/$1'
'$,QIRUPDWLRQ6HUYLFHV:KLWHKRUVH5RDG0,7&+$0$XVWUDOLD
7HOHSKRQH‡)D[
(PDLOVHUYLFH#GDGLUHFWFRPDX‡:HEVLWHKWWSZZZGDGLUHFWFRPDX

6/29(1,$
&DQNDUMHYD=DOR]EDGG.RSLWDUMHYD6,/MXEOMDQD
7HOHSKRQH‡)D[
(PDLOLPSRUWERRNV#FDQNDUMHYD]VL‡:HEVLWHKWWSZZZFDQNDUMHYD]VLXYR]

63$,1
'tD]GH6DQWRV6$F-XDQ%UDYR$(0DGULG
7HOHSKRQH‡)D[
(PDLOFRPSUDV#GLD]GHVDQWRVHVFDUPHOD#GLD]GHVDQWRVHVEDUFHORQD#GLD]GHVDQWRVHVMXOLR#GLD]GHVDQWRVHV
:HEVLWHKWWSZZZGLD]GHVDQWRVHV

81,7('.,1*'20
7KH6WDWLRQHU\2IILFH/WG,QWHUQDWLRQDO6DOHV$JHQF\32%R[1RUZLFK15*1
7HOHSKRQH RUGHUV ‡ HQTXLULHV ‡)D[
(PDLO RUGHUV ERRNRUGHUV#WVRFRXN‡ HQTXLULHV ERRNHQTXLULHV#WVRFRXN‡:HEVLWHKWWSZZZWVRFRXN
2QOLQHRUGHUV
'(/7$,QW%RRN:KROHVDOHUV/WG$OH[DQGUD5RDG$GGOHVWRQH6XUUH\.734
(PDLOLQIR#SURIERRNVFRP‡:HEVLWHKWWSZZZSURIERRNVFRP
%RRNVRQWKH(QYLURQPHQW
(DUWKSULQW/WG32%R[6WHYHQDJH6*73
7HOHSKRQH‡)D[
(PDLORUGHUV#HDUWKSULQWFRP‡:HEVLWHKWWSZZZHDUWKSULQWFRP

81,7('1$7,216
'HSW,5RRP'&)LUVW$YHQXHDWWK6WUHHW1HZ<RUN1<86$
81 7HOHSKRQHRU‡)D[
(PDLOSXEOLFDWLRQV#XQRUJ‡:HEVLWHKWWSZZZXQRUJ

81,7('67$7(62)$0(5,&$
%HUQDQ$VVRFLDWHV)RUEHV%OYG6XLWH/DQKDP0'
7HOHSKRQH‡)D[
(PDLOFXVWRPHUFDUH#EHUQDQFRPÂ:HEVLWHKWWSZZZEHUQDQFRP
5HQRXI3XEOLVKLQJ&RPSDQ\/WG3URFWRU$YH2JGHQVEXUJ1<
7HOHSKRQH WROOIUHH ‡)D[ WROOIUHH 
(PDLORUGHUGHSW#UHQRXIERRNVFRP‡:HEVLWHKWWSZZZUHQRXIERRNVFRP

2UGHUVDQGUHTXHVWVIRULQIRUPDWLRQPD\DOVREHDGGUHVVHGGLUHFWO\WR
0DUNHWLQJDQG6DOHV8QLW,QWHUQDWLRQDO$WRPLF(QHUJ\$JHQF\
9LHQQD,QWHUQDWLRQDO&HQWUH32%R[9LHQQD$XVWULD
7HOHSKRQH RU ‡)D[
(PDLOVDOHVSXEOLFDWLRQV#LDHDRUJ‡:HEVLWHKWWSZZZLDHDRUJERRNV
12-17861