Safeguards Techniques and Equipment

Safeguards Techniques
and Equipment
2003 Edition
International
Nuclear Verification
Series No. 1
(Revised)
SAFEGUARDS TECHNIQUES
AND EQUIPMENT
2003 Edition
The following States are Members of the International Atomic Energy Agency:
AFGHANISTAN GREECE PERU
ALBANIA GUATEMALA PHILIPPINES
ALGERIA HAITI POLAND
ANGOLA HOLY SEE PORTUGAL
ARGENTINA HONDURAS QATAR
ARMENIA HUNGARY
REPUBLIC OF MOLDOVA
AUSTRALIA ICELAND
AUSTRIA INDIA ROMANIA
AZERBAIJAN INDONESIA RUSSIAN FEDERATION
BANGLADESH IRAN, ISLAMIC REPUBLIC OF SAUDI ARABIA
BELARUS IRAQ SENEGAL
BELGIUM IRELAND SERBIA AND MONTENEGRO
BENIN ISRAEL SEYCHELLES
BOLIVIA ITALY SIERRA LEONE
BOSNIA AND JAMAICA SINGAPORE
HERZEGOVINA JAPAN SLOVAKIA
BOTSWANA JORDAN
SLOVENIA
BRAZIL KAZAKHSTAN
BULGARIA KENYA SOUTH AFRICA
BURKINA FASO KOREA, REPUBLIC OF SPAIN
CAMEROON KUWAIT SRI LANKA
CANADA LATVIA SUDAN
CENTRAL AFRICAN LEBANON SWEDEN
REPUBLIC LIBERIA SWITZERLAND
CHILE LIBYAN ARAB JAMAHIRIYA SYRIAN ARAB REPUBLIC
CHINA LIECHTENSTEIN TAJIKISTAN
COLOMBIA LITHUANIA THAILAND
COSTA RICA LUXEMBOURG
THE FORMER YUGOSLAV
CÔTE D’IVOIRE MADAGASCAR
REPUBLIC OF MACEDONIA
CROATIA MALAYSIA
CUBA MALI TUNISIA
CYPRUS MALTA TURKEY
CZECH REPUBLIC MARSHALL ISLANDS UGANDA
DEMOCRATIC REPUBLIC MAURITIUS UKRAINE
OF THE CONGO MEXICO UNITED ARAB EMIRATES
DENMARK MONACO UNITED KINGDOM OF
DOMINICAN REPUBLIC MONGOLIA GREAT BRITAIN AND
ECUADOR MOROCCO NORTHERN IRELAND
EGYPT MYANMAR UNITED REPUBLIC
EL SALVADOR NAMIBIA OF TANZANIA
ERITREA NETHERLANDS UNITED STATES OF AMERICA
ESTONIA NEW ZEALAND URUGUAY
ETHIOPIA NICARAGUA
UZBEKISTAN
FINLAND NIGER
FRANCE NIGERIA VENEZUELA
GABON NORWAY VIETNAM
GEORGIA PAKISTAN YEMEN
GERMANY PANAMA ZAMBIA
GHANA PARAGUAY 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, 2003
Permission to reproduce or translate the information contained in this publication may be
obtained by writing to the International Atomic Energy Agency, Wagramer Strasse 5, P.O. Box 100,
A-1400 Vienna, Austria.
Printed by the IAEA in Austria
August 2003
IAEA/NVS/1 (Revised)
INTERNATIONAL NUCLEAR VERIFICATION SERIES
No. 1 (Revised)
SAFEGUARDS TECHNIQUES
AND EQUIPMENT
2003 Edition
INTERNATIONAL ATOMIC ENERGY AGENCY

VIENNA, 2003
IAEA Library Cataloguing in Publication Data
Safeguards techniques and equipment. — Rev. ed. — Vienna :
International Atomic Energy Agency, 2003.
p. ; 24 cm. — (International nuclear verification series, ISSN,
1020-6205 ; no. 1 (Revised))
IAEA/NVS/1/(Revised)
ISBN 92–0–109403–5
Includes bibliographical references.
1. Nuclear industry—Materials management. 2. Nuclear reactors—
Materials—Analysis. 3. Nuclear reactors—Containment. 4. Environ-
mental sampling. I. International Atomic Energy Agency. II. Series.
IAEAL 03–00330
FOREWORD
The 1990s saw significant non-proliferation related developments in the

world, resulting in a new period of safeguards development. Over several years
an assessment was made of how to strengthen the effectiveness and improve
the efficiency of IAEA safeguards. In May 1997 this culminated in the adoption
by the IAEA Board of Governors of a Protocol Additional to Safeguards
Agreements which significantly broadens the role of IAEA safeguards. As a
consequence, the IAEA safeguards system entered a new era.
Together with the introduction of the strengthened safeguards systems, in
1997 the IAEA began to publish a new series of booklets on safeguards, called
the International Nuclear Verification Series (NVS). The objective of these
booklets was to help in explaining IAEA safeguards, especially the new
developments in safeguards, particularly for facility operators and government
officers involved with these topics.
The current booklet, which is a revision and update of IAEA/NVS/1, is
intended to give a full and balanced description of the techniques and
equipment used for both nuclear material accountancy and containment and
surveillance measures, and for the new safeguards measure of environmental
sampling. A completely new section on data security has been added to
describe the specific features that are included in installed equipment systems
in order to ensure the authenticity and confidentiality of information. As new
verification measures continue to be developed the material in this booklet will
be periodically reviewed and updated versions issued.
EDITORIAL NOTE
Although great care has been taken to maintain the accuracy of information con-
tained in this publication, neither the IAEA nor its Member States assume any responsi-
bility for consequences which may arise from its use.
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 territo-
ries, 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
2. NON-DESTRUCTIVE ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.1. Gamma ray spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2.1.1. Gamma emission and detection of
nuclear materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.1.2. Hand-held multipurpose gamma spectrometry . . . . . . . . 6
2.1.3. Multichannel analysers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1.4. IAEA high resolution gamma spectrometry
techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2. Neutron counting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.1. Neutron emission and detection for
non-irradiated fissile fuel . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.2. Gross neutron counting . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.3. Neutron coincidence counting . . . . . . . . . . . . . . . . . . . . . . 14
2.3. Spent fuel measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3.1. Neutron emission and detection . . . . . . . . . . . . . . . . . . . . . 18
2.3.2. Gross neutron and gamma ray detection . . . . . . . . . . . . . 20
2.3.3. Gamma ray energy spectral analysis . . . . . . . . . . . . . . . . . 20
2.3.4. Gamma ray intensity scanning . . . . . . . . . . . . . . . . . . . . . . 21
2.3.5. Cerenkov radiation detection . . . . . . . . . . . . . . . . . . . . . . . 22
2.4. Other NDA techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.4.1. Radiation measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.4.2. Physical property measurement . . . . . . . . . . . . . . . . . . . . . 24
3. DESTRUCTIVE ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.1. Elemental analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.1.1. Uranium by potentiometric titration . . . . . . . . . . . . . . . . . 27
3.1.2. Plutonium by potentiometric titration . . . . . . . . . . . . . . . . 29
3.1.3. Plutonium by controlled potential coulometry . . . . . . . . . 29
3.1.4. Uranium by ignition gravimetry . . . . . . . . . . . . . . . . . . . . . 29
3.1.5. Uranium, thorium or plutonium by K-edge X ray
densitometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.1.6. Plutonium by K X ray fluorescence analysis . . . . . . . . . . . 30
3.1.7. Plutonium and/or uranium by wavelength
dispersive X ray fluorescence spectrometry . . . . . . . . . . . 31
3.1.8. Uranium or plutonium by isotope dilution
mass spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.2. Isotopic analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.1. Uranium or plutonium isotopic composition by
thermal ionization mass spectrometry . . . . . . . . . . . . . . . . 33
3.2.2. Plutonium isotopic composition by high resolution
g ray spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.2.3. Uranium-235 in solution by g ray spectrometry . . . . . . . . 34
3.3. Other DA techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4. CONTAINMENT AND SURVEILLANCE . . . . . . . . . . . . . . . . . . . 36
4.1. Surveillance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.1.1. Installed single cameras for easy to access locations . . . . 43
4.1.2. Installed single camera for difficult to access
locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.1.3. Installed multi-camera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.1.4. Short term surveillance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.1.5. Underwater TV for attended applications . . . . . . . . . . . . 46
4.1.6. Surveillance review software . . . . . . . . . . . . . . . . . . . . . . . . 47
4.1.7. Miscellaneous surveillance systems and options . . . . . . . 48
4.2. Seals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.2.1. Single use seals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.2.2. In situ verifiable seals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5. UNATTENDED MONITORING . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
6. REMOTE MONITORING SYSTEMS . . . . . . . . . . . . . . . . . . . . . . . . 61
6.1. Remote monitoring equipment . . . . . . . . . . . . . . . . . . . . . . . . . . 61

6.2. Future development activities . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
6.2.1. Data reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
6.2.2. Alternative communication methods . . . . . . . . . . . . . . . . . 63
7. DATA SECURITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
7.1. Information protection requirements . . . . . . . . . . . . . . . . . . . . . 65

7.2. IAEA requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
7.2.1. Verification data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
7.2.2. Technical data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
7.2.3. Control data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
7.3. Member State requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
8. ENVIRONMENTAL SAMPLING . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.1. IAEA Clean Laboratory for Safeguards . . . . . . . . . . . . . . . . . . . 75

8.2. Screening of samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
8.2.1. Low level g ray spectrometry . . . . . . . . . . . . . . . . . . . . . . . 77
8.2.2. X ray fluorescence spectrometry . . . . . . . . . . . . . . . . . . . . 78
8.2.3. Alpha/beta counting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
8.3. Isotopic and elemental analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 78
8.3.1. Pulse counting thermal ionization mass spectrometry . . . 78
8.3.2. Scanning electron microscopy with
electron probe analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
8.3.3. Fission track method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
8.3.4. Secondary ion mass spectrometry . . . . . . . . . . . . . . . . . . . 82
1. INTRODUCTION
The IAEA has the task of providing continuing assurance to the

international community that States that have entered into safeguards
agreements with the IAEA are meeting their obligations. This requires that
assurance be given that any diversion of safeguarded nuclear material to a
proscribed military purpose would be detected and that all nuclear material in
the State has been declared. To this end, the IAEA must be able to verify the
correctness and completeness of the statement that it receives from States
concerning the nuclear materials included in safeguards agreements.
In addition, the IAEA has the right and obligation to verify States’
commitments under their safeguards agreements and, where applicable,
Additional Protocols. In particular, as part of the implementation of Additional
Protocol and Integrated Safeguards, inspectors must be able to confirm the
absence of undeclared nuclear materials and activities during the inspections
and complementary access visits.
The basic verification measure used by the IAEA is nuclear material
accountancy. In applying nuclear material accountancy, IAEA safeguards
inspectors make independent measurements to verify quantitatively the
amount of nuclear material presented in the State’s accounts. For this purpose,
inspectors count items (e.g. fuel assemblies, bundles or rods, or containers of
powdered compounds of uranium or plutonium) and measure attributes of
these items during their inspections using non-destructive analysis (NDA)
techniques, and compare their findings with the declared figures and the
operator’s records. The purpose of this activity is to detect missing items (gross
defects). The next level of verification has the aim of detecting whether a
fraction of a declared amount is missing (partial defect) and may involve the
weighing of items and measurements with NDA techniques such as neutron
counting or g ray spectrometry. These techniques are capable of measuring the
amount of nuclear material with an accuracy in the order of a few per cent. For
detecting bias defects, which would arise if small amounts of material were
diverted over a protracted length of time, it is necessary to sample some of the
items and to apply physical and chemical analysis techniques having the highest
possible accuracy, typically less than one per cent. In order to apply these
destructive analysis (DA) techniques, the IAEA requires access to laboratories
which use such accurate techniques on a routine basis.
Containment and surveillance (C/S) techniques, which are
complementary to nuclear material accountancy techniques, are applied in
order to maintain continuity of the knowledge gained through IAEA
verification, by giving assurance that nuclear material follows predetermined
1
SAFEGUARDS TECHNIQUES AND EQUIPMENT
routes, that the integrity of its containment remains unimpaired and that the
material is accounted for at the correct measurement points. They also lead to
savings in the safeguards inspection effort, e.g. by reducing the frequency of
accountancy verification. A variety of C/S techniques are used, primarily
optical surveillance and sealing. These measures serve to back up nuclear
material accountancy by providing means by which access to nuclear material
can be monitored and any undeclared movement of material detected.
Unattended and remote monitoring is a special mode of application of
NDA or C/S techniques, or a combination of these, that operates for extended
periods of time without inspector access. In remote monitoring, the unattended
equipment transmits the data off-site. For unattended and remote monitoring,
additional criteria must be met, including high reliability and authentication of
the data source. Data communication costs have dropped dramatically in
recent years. Consequently, expanded deployment of unattended and remote
monitoring systems has become an increasingly important element of IAEA
safeguards in efforts to maintain or increase effectiveness without increasing
inspector resources or overall costs.
Data security is an important feature of unattended and remote
monitoring systems. In fact, those types of safeguards systems, permanently
installed at facilities and periodically visited by inspectors, transmit data
between the components of different systems and between systems and IAEA
headquarters through insecure transmission paths. These data need to be
verified to guarantee their authenticity and may need to be encrypted to avoid
disclosure of specific information and/or to ensure confidentiality to the
Member States.
Environmental sampling, which allows detection of minute traces of
nuclear material, was added to the IAEA’s verification measures in the early
1990s as a powerful tool for detecting indications of undeclared nuclear
activities. The non-detection of minute traces of a specific nuclear material can
provide assurance that there were no activities utilizing the material in the area
where the environmental samples were taken.
The complexity and diversity of facilities containing safeguarded nuclear
material require a correspondingly diverse set of verification techniques and
equipment. Table I lists the main types of facility where inspections are
performed and the primary verification techniques that are implemented at
these facilities.
Development of equipment and techniques for safeguards is continuing
with the help of national support programmes that assist the IAEA in keeping
pace with the evolution of new technology. The IAEA defines the safeguards
needs, co-ordinates the support programmes, and tests and evaluates the
techniques and the resulting equipment being developed. All aspects of
2
TABLE I. MAIN TYPES OF FACILITY UNDER IAEA SAFEGUARDS
(data based on 2002 Safeguards Implementation Report)
Enrichment Fuel fabrication Power reactors and Spent fuel

plants plants separate storage facilities reprocessing plants
Number of facilities safeguarded in 2002
10 41 239 power reactor units 6

80 separate storage facilities
INTRODUCTION
Main techniques deployed
Materials: UF6 Materials: U and Pu oxides, MOX Materials: Spent fuel Materials: U and Pu nitrates
Gamma ray spectrometry Gamma ray spectrometry Cerenkov glow detection Destructive analysis
Weighing Neutron counting Gross g ray and neutron Neutron counting
Destructive analysis detection
Isotopic determination
IAEA summary statistics for 2002 (approximate numbers)
2 430 inspections at 603 facilities and 706 NDA systems used in 1 639 inspections at 553 facilities
766 DA samples analysed 1581 analytical results were reported
489 video cameras deployed for optical surveillance
24 572 seals detached and verified
232 environmental samples taken in 11 enrichment installations and 36 other installations,
including facilities with hot cells
3
equipment performance are evaluated, including compliance with

specifications, reliability and transportability and, most importantly, suitability
for use by IAEA inspectors in nuclear facilities. The IAEA has an established
quality assurance procedure to authorize equipment and software for routine
inspection use.
The equipment and techniques highlighted in this booklet are those in
frequent use for inspection purposes or in the late stages of development. The
overall objective of this publication is to provide a comprehensive overview of
the techniques and equipment underlying the implementation of IAEA
safeguards.
4
2. NON-DESTRUCTIVE ANALYSIS
2.1. GAMMA RAY SPECTROMETRY
2.1.1. Gamma emission and detection of nuclear materials
Most nuclear materials of concern in IAEA safeguards emit g rays that

can be used for NDA of the materials. Gamma rays have well defined energies
that are characteristic of the isotopes emitting them. Determination of the g ray
energies serves to identify the isotopic composition of the materials. When
combined with a measurement of intensities, the g ray energies can provide
quantitative information on the amount of material that is present:
(a) Enriched uranium fuel, for example, has a strong 186 keV g ray associated
with the alpha decay of 235U, and the 235U enrichment can be verified by
measuring this g ray.
(b) Plutonium samples generally contain the isotopes 238Pu, 239Pu, 240Pu and
241Pu as well as decay products, which give rise to a highly complex mix
of characteristic g ray energies. Plutonium spectra can be analysed to

determine the isotopic composition.
(c) The date of discharge of irradiated fuel from a reactor can be verified by
measuring the relative intensities of g rays associated with fission and
activation products. The 662 keV g ray from 137Cs is particularly
important for this type of determination.
To detect g rays, the radiation must interact with a detector to give up all
or part of the photon energy. The basis of all g ray detector systems is the
collection of this liberated electrical charge to produce a voltage pulse whose
amplitude is proportional to the g ray energy. In a g ray spectrometer, these
pulses are sorted according to amplitude (energy) and counted using
appropriate electronics, such as a single or multichannel analyser. With a
multichannel analyser, the g rays from different energies can be displayed or
plotted to produce a g ray energy spectrum which provides detailed
information on the measured material.
The g ray detectors most commonly used are either scintillators (usually
activated sodium iodide (NaI) crystals) or solid state semiconductors (usually
high purity germanium (HpGe) and cadmium–zinc–telluride (CdZnTe)
crystals).
5
(1) The NaI detectors can be made with large volumes and generally have
higher g ray detection efficiencies than Ge detectors. Their safeguards
applications include, for example, the verification of 235U enrichment in
fresh fuel as well as the presence of spent fuel through detection of fission
product g radiation. Their ability to distinguish between g rays of different
energies, however, is relatively poor and of the three types of detector
they have the worst energy resolution.
(2) Germanium detectors have far superior energy resolution to NaI
detectors and are better suited to the task of resolving complex g ray
spectra and providing information about the isotopic content of materials.
The Ge detectors used by the IAEA range in size from small planar types
to large (80–90 cm3) coaxial detectors. A disadvantage of these detectors
is that they must be operated at a very low temperature, which is usually
achieved by cooling with liquid nitrogen.
(3) Standard CdZnTe detectors (and CdTe detectors) do not need cooling
and of the three detectors they have the highest intrinsic detection
efficiency. Recent progress in fabrication techniques has substantially
improved CdZnTe resolution. Until 1997 the standard volumes available
were relatively small (20 and 60 mm3), but now relatively large 500 mm3
and 1500 mm3 CdZnTe detectors are available. The portability and small
size of CdZnTe and CdTe detectors have made them especially suitable
for use in a wide range of applications, including use in confined spaces
such as in situ verification of fresh fuel assemblies whose design permits
insertion of only a small detector probe into the assembly interior and of
spent fuel bundles stored underwater in closely packed stacks.
Figure 1 illustrates the capabilities of various types of detector with low,

medium and high resolution. Several g ray spectrometers (multichannel
analysers and detectors) that differ mainly in their resolution and analytical
capability are being used for safeguards purposes. These are summarized in
Table II and described below.
2.1.2. Hand-held multipurpose gamma spectrometry
HM-5 (fieldSPEC). This device is the successor to the Hand-held Assay

Probe (HM-4). Due to technological progress, the HM-5 (Fig. 2) is far superior
in design and performance to its predecessor. The HM-5 is a modern, hand-
held, digital gamma spectrometer combining various functions such as dose
rate measurement, source search, isotope identification, active length
determination for fuel rods and assemblies, and Pu/U attribute verification.
6
NON-DESTRUCTIVE ANALYSIS
NAID U235 Spectrum

16000
14000
12000
Counts
10000
8000
6000
4000
2000
0
0 100 200 300 400 500 600
Channel
CDZT U235 Spectrum

3500
3000
2500
Counts
2000
1500
1000
500
0
0 100 200 300 400 500 600
Channel
GDET U235 Spectrum

45000
40000
35000
30000
Counts
25000
20000
15000
10000
5000
0
0 1000 2000 3000 4000 5000
Channel
FIG. 1. Comparison of g ray spectrometric performance of various types of detector (low,

medium and high resolution).
7
TABLE II. GAMMA RAY SPECTROMETERS
Code Equipment name Primary applications

HM-5 Hand-held Assay Probe Qualitative determination of the
presence of U, Pu and
other isotopes
IMCN, I-2000 Multichannel Analyser Verification of U enrichment, spent

IMCC, (IMCA) used with NaI or CdZnTe fuel and Pu isotopic composition
IMCG or Germanium Detector
MMCN, Miniature Multichannel Analyser Verification of U enrichment and

MMCC, (MMCA) used with NaI or CdTe or spent fuel
MMCG Germanium Detector
The basic HM-5 modular design includes an NaI detector. For special
applications the NaI detector can be replaced with a more stable, higher
resolution CdZnTe detector. Up to 50 gamma spectra, each with 1024 channels,
can be stored in the non-volatile memory of the HM-5 and later be transferred
to a computer for further processing or plotting.
With such versatility, the HM-5 is used for traditional safeguards
inspections and for investigations under the conditions of the Additional
Protocol. The HM-5 is also particularly useful for law enforcement services to
detect and identify nuclear and radioactive materials being smuggled across
borders.
FIG. 2. The HM-5 fieldSPEC.
8
2.1.3. Multichannel analysers
IMCA (InSpector2000). The new InSpector 2000 (I-2000) multichannel

analyser (IMCA), based on digital signal processing (DSP) technology, can be
combined with the various types of detector that are now used for inspection
purposes, namely HpGe, CdZnTe and NaI, allowing high, medium and low
resolution spectrometry. The device provides unsurpassed count rate and
resolution performance coupled with environmental stability in a very small
and compact package. The IMCA (Fig. 3) follows the ongoing evolution in the
MCA market, including the latest developments in this domain. Its high
performance is derived from the application of DSP technology, which digitizes
preamplifier signals at the very beginning of the signal processing chain. The
use of analog circuitry in the instrument is reduced, resulting in a compact
instrument that has increased stability, accuracy and data reproducibility while
improving the overall signal acquisition performance.
Because the I-2000 is a commercially available instrument that is being
used worldwide, the IAEA can use I-2000 based operator equipment for
inspection purposes. Use of such operator equipment is contingent on the
IAEA being able to authenticate the data and on applicability of common
software. The transition to this new equipment has enabled the IAEA to phase
out obsolete equipment such as the old InSpector MCA and the PMCA and
continue the trend toward standardization of equipment types used for
inspection purposes. The I-2000 has been authorized for inspection use since
2001.
MMCA. The Miniature Multichannel Analyser is a developed
miniaturized spectrometry system that supports all detectors used by the
IAEA including NaI (called MMCN for this application), CdZnTe (MMCC)
FIG. 3. IMCA shown with an NaI detector and portable computer.
9
FIG. 4. MMCC: Miniature Multichannel Analyser (MMCA) with CdZnTe detector.
and HpGe (MMCG). The MMCA (Fig. 4) is significantly smaller and lighter
than the previous IAEA portable unit, the PMCA, and during operation the
battery lifetime is three times longer (at least 12 hours for CdZnTe and NaI
detectors). The MMCA has the footprint of a palmtop computer and weighs
680 g, including the lithium ion battery. Combined with a palmtop computer
and a CdZnTe detector it makes a powerful yet versatile system that fits into
half a briefcase, making it very convenient for many inspection activities.
2.1.4. IAEA high resolution gamma spectrometry techniques
When coupled to a Ge detector, the MMCA or IMCA becomes a high

resolution g ray spectrometer (MMCG or IMCG, Fig. 5). This type of
spectrometer is often used to determine the 235U enrichment of uranium
hexafluoride (UF6) in shipping cylinders. When selecting the UF6 measurement
procedure from the options menu in the applications firmware, an inspector is led
through a series of predetermined steps to measure and calculate the
FIG. 5. IMCG: Inspector Multichannel Analyser with HpGe detector.
10
enrichment. The cylinder wall thickness must also be determined so that

corrections for g ray attenuation in the container wall can be made. The thickness
is measured with an ultrasonic thickness gauge.
Inspectors can also use the MGAU software which simplifies the
measurement and analysis of high resolution uranium spectra. MGAU can
provide results with an accuracy of 1–2%, provided that the wall thickness of
the steel container is less than 10 mm and that the activity of the thorium
daughter is in equilibrium with the parent 235U and 238U activities. The MGAU
analysis procedure eliminates the need to measure the container wall thickness
or to provide an enrichment calibration for the measurement system.
Another important application of high resolution gamma spectrometry is
the determination of the isotopic composition of plutonium. Plutonium emits a
complex spectrum of X and g rays which are interpreted using dedicated software
such as MGA and FRAM. These codes take advantage of the high energy
resolution of the spectra from a HpGe detector to separate and evaluate the
contributions of the different plutonium isotopes. Isotopic determination of
plutonium is used to verify the nature of the material and as an input parameter
for the interpretation of the neutron measurements. The recently developed
TARGA software provides a user friendly environment with the MGA code to
determine the isotopic composition of plutonium samples. The combination of
the I-2000 system with TARGA software replaces the previously used Medium
Count Rate System combined with PUIS software.
2.2. NEUTRON COUNTING
The IAEA uses a number of different types of neutron counting

equipment (Table III). This section gives information on the source of the
neutrons, the importance of neutron coincidence counting to obtain the mass of
fissile fuel, and a few examples of passive and active detector systems.
2.2.1. Neutron emission and detection for non-irradiated fissile fuel
Neutrons are primarily emitted from non-irradiated nuclear fuel in three

ways:
(1) Spontaneous fission of uranium and plutonium, particularly involving the

even isotopes of plutonium;
(2) Induced fission from fissile isotopes of uranium and plutonium, typically
by means of a low energy neutron source;
(3) Alpha particle induced reactions, (a, n) involving light elements such as
oxygen and fluorine.
11
TABLE III. COINCIDENT NEUTRON DETECTOR SYSTEMS FOR

NON-IRRADIATED FISSILE FUEL
Passive Neutron Coincidence Counters

BCNC Bird Cage Counter Verification of Pu mass in special
storage configurations
CNCM Compact Neutron Coincidence Counter Verification (qualitative) of MOX
fuel assemblies in shipping crates
DRNC Drawer Counter Verification of Pu mass in facility
specific containers
FAAS Fuel Assembly/Capsule Assay System Verification of Pu mass in MOX
fuel assemblies
FPAS Fuel Pin/Pallet Assay System Verification of Pu mass in MOX
fuel pins in facility specific storage
trays
GBAS Glovebox Assay System Semiquantitative determination of
Pu hold-up in gloveboxes
HBAS Hold-up Blender Assay System Semiquantitative determination of
Pu hold-up in facility blenders
HLNC High Level Neutron Coincidence Counter Verification of Pu in 20–2000 g
canned samples (pellets, powders,
scrap)
INVS Inventory Sample Counter Verification of Pu in 0.1–300 g
samples. Modified version can be
attached to gloveboxes
LNMC Large Neutron Multiplicity Counter Verification of Pu in
contaminated/impure items
MAGB Glovebox Counter Verification of Pu mass in facility
gloveboxes
PCAS Canister Counter Verification of Pu mass in MOX
canisters
PNCL Plutonium Neutron Coincidence Collar Verification of Pu mass in MOX
fuel assemblies
PSMC Plutonium Scrap Multiplicity Counter Verification of Pu in 1–300 g
canned samples of scrap
PWCC Passive Well Coincidence Counter Verification of Pu mass in
CANDU MOX fuel bundles
UFBC Universal Fast Breeder Counter Verification of Pu (up to 16 kg)
in FBR fuel
UWCC Underwater Coincidence Counter Underwater verification of Pu in
fresh MOX fuel assemblies
12
TABLE III. (cont.)
Active Neutron Coincidence Counters

AWCC Active Well Coincidence Counter Verification of 235U in high
enriched U samples
UNCL Uranium Neutron Coincidence Collar Verification of 235U in low
enriched U fuel assemblies;
a variety of collar configurations
are available.
WCAS Waste Crate Assay System Verification of waste materials
WDAS Waste Drum Assay System Interrogation of low level waste
drums for Pu mass
Fission neutrons in the first two categories are emitted in groups of two or
more per fission event. The multiple neutron signature is detected as a neutron
coincidence. Nearly all the isotopes of U, Pu and other transuranic elements emit
alpha particles. These interact with light elements present in compounds (e.g.
oxides and fluorides) or as impurities (e.g. B, Be and Li) to form an undesirable
neutron background. Neutron coincidence counting discriminates against this
background by processing neutron pulses to select time correlated (coincident)
signatures of multiple neutrons emitted during fission and eliminating the (a, n)
neutron pulses that are emitted singly and are thus uncorrelated.
Passive coincidence detector systems determine the mass of Pu based on
spontaneous fission primarily in the even numbered isotopes (238Pu, 240Pu and
242Pu, with 240Pu being the dominant contributor). The major fissile isotope,
239Pu, has a typical abundance in fuel of 60% or higher, yet it makes an
insignificant contribution to the spontaneous fission neutron signal. Isotopic

abundance must be known or verified — typically by means of a high
resolution g ray measurement. Using the isotopic abundance, the 240Pueff mass
determined from coincident neutron count rates can be converted into the total
Pu mass for the sample. For uncontaminated, well characterized samples,
measurement accuracy can be of the order of 1% or less.
The fissile isotope 235U does not undergo sufficient spontaneous fission
for practical passive detection. In this case, an active system incorporating
AmLi neutron sources is used to ‘interrogate’ the 235U content through
neutron induced fission. For low energy incident neutrons, induced fissions in
the 238U of a sample contributes insignificantly to the measured coincident
13
neutron count rate, even though 235U may be enriched to only a few per cent
(e.g. low enrichment fuels).
Neutron detectors employ various neutron capture reactions to generate
pulses. The reactions produce energetic particles which in turn ionize a gas and
produce a charge pulse in response to a neutron interaction. The choice of
detector (i.e. the capture material) is based mainly on the neutron detection
sensitivity required and on the insensitivity to other radiation, e.g. g rays. Nearly
all detectors are most sensitive to low energy neutrons. Consequently in many
neutron measurement systems, detectors are surrounded with a moderator
material such as polyethylene to slow energetic neutrons down to thermal
energies.
2.2.2. Gross neutron counting
Gross neutron counting refers to the sum of all neutrons detected. Here
the neutron source cannot be characterized since coincidence requirements are
not applied. The presence of significant numbers of neutrons is often a
sufficient indication that fissile nuclear material is present. All the neutron
coincidence detection systems (discussed below) determine total neutron count
rates as well as coincidence count rates.
Other detector systems, such as the Fork Detector and the Unattended
Fuel Flow Monitor, employ gross neutron counting as their primary signature.
These systems mainly measure spent fuel materials as described in another
section of this booklet.
2.2.3. Neutron coincidence counting
Neutron coincidence counting has evolved into a very stable, reliable and
accurate technique for determining Pu and 235U content. Modern, well
designed neutron coincidence systems are capable of reliably processing pulses
over a very large range of input count rates (i.e. over more than six orders of
magnitude). The stability is achieved by judicious selection and placement of
amplifier electronics to minimize noise interference. The electronics boards,
when located at the detector head, amplify and shape the pulses, apply lower
level discrimination to remove g pulses or noise, and feed out very narrow
(50 ns wide) logic pulses to an external pulse processor (the electronics
controller).
Reliable coincidence counting is also due to a sophisticated pulse
processing circuit (shift register electronics) in the external electronics
controller. Pulses occurring within a specified time period (normally set at
64 µs) of one another may be termed correlated (i.e. ‘coincident’) neutron
14
pulses. The correlation time is associated with the slowing down of neutrons in
the moderator of the detector head. The shift register electronics circuitry
keeps track of coincidences between pulses separated by about 1000 ms (called
accidentals) and coincidences in the first 64 µs (called real coincidences plus
accidentals). Analysis software provided with a coincidence counter system
subtracts the accidentals data from the reals+accidentals data to determine real
coincidences. In analysing the information, various small corrections are also
automatically applied.
Passive detector systems have two basic geometrical configurations: well
detectors that completely enclose the sample or collar detectors that encircle
the sample (e.g. a fuel assembly). Well detectors have the preferred geometry
since they have the capability to detect all the neutrons emanating from the
sample. Collar detectors are an alternative detector design that is appropriate
when the sample is too large for placement inside a well detector. Whereas
calibrated passive well detectors measure the total mass of Pu in a sample,
collar detectors measure Pu mass per unit length of a fuel assembly. The linear
density must be multiplied by an effective length to determine the total Pu
mass in the assembly.
About twenty versions of passive detector systems are currently used for
nuclear safeguards, with design features optimized for specific sample sizes,
shapes or Pu mass ranges. The passive detector systems are listed in Table III
along with their primary applications. Two representative systems are described
below.
HLNC. The High Level Neutron Coincidence Counter is shown in Fig. 6.
This detector system is typical of IAEA well detector coincidence counting
systems used for measuring non-irradiated Pu materials. The HLNC includes a
head which houses the neutron detectors (3He gas proportional counters)
connected to special amplifiers. The electronics controller, JSR-12, provides
power to the amplifiers and 3He tubes and processes the train of pulses to
determine coincidence events. A portable computer connected to the JSR-12
automates the collection of data and analyses and archives the data. A printer,
which presents the results in a concise report format, completes the detector
package. This 60 kg detector features a large sample cavity and 18% neutron
detection efficiency. By removal of the top end cap, a can containing Pu (in
pellet, powder or scrap form) can be centred in the large cavity. The sample is
given an identification number in the computer, an appropriate calibration
curve is selected and a run time (typically 100 s) is designated. Upon initiation
of the measurement, the IAEA neutron coincidence counting (INCC) code
automatically runs through a sequence of measurements (typically three), each
of which must pass all built-in quality control criteria for acceptable results.
When the measurements are completed, the Pu mass is calculated and
15
FIG. 6. HLNC: High Level Neutron Coincidence Counter.
compared with the declared value to provide a quantitative verification that for
typical high purity Pu inventories is accurate to 1%.
INVS. For small samples (bagged Pu pellets, powders and solutions in
vials) with much lower total Pu content than those typically measured with an
HLNC, the Inventory Sample Counter (INVS) is the detector system of choice.
The INVS has nearly double the neutron detection efficiency of an HLNC.
Fig. 7 shows one of four versions of this portable detector system. In another
FIG. 7. INVS: Inventory Sample Counter.
16
version, the INVS has an inverted geometry and is permanently attached to the
floor of a glovebox so that samples can be assayed for Pu content without the
inconvenience and inefficiency of removing the samples from the glovebox.
Although the cavity of an INVS is typically only about 6 cm in diameter by
16 cm high, it is well suited for samples available at facilities such as fuel
fabrication plants or on-site laboratories. The INVS provides highly reliable
Pu content verification with an accuracy of up to 1% on individual
measurements. Measurement procedures are automated with the INCC
program and are essentially the same as for the HLNC.
Active detector systems use neutron sources (typically AmLi) to
interrogate (induce fission in) the 235U in a sample. A well geometry is again
preferred but a collar geometry is needed when the sample is a fuel assembly.
The active neutron detectors in use by IAEA safeguards are listed in Table III.
Details of an active well detector and an active collar detector are presented
below. The full detector system includes a detector head, which detects the
neutrons and houses a neutron interrogation source; the electronics controller,
which powers the detector and determines the neutron coincidence rates; a
portable computer for controlling the measurements and for analysing data to
determine 235U content; and the printer for generating reports.
AWCC. The Active Well Coincidence Counter has a large (150 kg)
detector head permanently attached to a wheeled cart for transportability
(Fig. 8). The AWCC has 42 3He counters embedded in thick polyethylene
cylinders, resulting in a relatively high (nearly 30%) neutron detection
FIG. 8. AWCC: Active Well Coincidence Counter.
17
efficiency. The 235U content in a sample is interrogated with two AmLi neutron
sources placed in the top and bottom end caps to provide a uniform
distribution of interrogation neutrons over the sample volume. The cavity size
is adjustable up to about 20 cm in diameter and 23 cm in height by removal of
inserts and reflectors accommodating samples such as metal disks, canned
metal oxide powders and fuel pebbles in carousels. The INCC program is used
to automate the measurement procedure and data analysis to enable assaying
the 235U content to high accuracy.
UNCL. The Uranium Neutron Coincidence Collar for determining the
linear mass density of uranium in fresh fuel assemblies is mounted on a cart.
When it is being used at a fuel fabrication or reactor facility, the side containing
the neutron source is removed or swung open and the collar is wheeled into
position surrounding a fuel assembly. Once the door with the AmLi source is
closed, the measurement cycle is initiated. After a specified number of
measurements have passed the quality assurance acceptance criteria in the
INCC program, the 235U mass per unit length is determined. The linear density
data are combined with results of a measurement of the effective length to
determine the 235U content of the entire fuel assembly.
2.3. SPENT FUEL MEASUREMENT
2.3.1. Neutron emission and detection
Spontaneous fissions in the 242Cm and 244Cm isotopes are the major
source of neutrons emanating from spent fuel. These isotopes are produced
through multiple neutron capture events when a fuel assembly is exposed to
high neutron fluxes in a nuclear reactor. Fission products in the irradiated fuel
produce an extremely high radiation background in which the neutrons must
be detected. The high radiation environment influences the type of techniques
that can be deployed for spent fuel verification. One approach is to choose a
detector which is basically insensitive to g rays. Another approach is to shield
against the g rays while allowing neutrons to pass through the shield into the
neutron detector. Spent fuel verification methods include not only neutron
detection but also g ray and ultraviolet light (Cerenkov radiation) detection.
Table IV lists the spent fuel measurement systems in use by the IAEA. The
Fork Detector (FDET) incorporates both neutron and g ray detectors for gross
defect verification of fuel assembly characteristics such as irradiation history, initial
fuel content and number of reactor cycles of exposure. Detector systems are avail-
able to measure the g ray energy spectra from irradiated fuel (SFAT and IRAT),
and g ray intensity as a function of fuel bundle storage position (CBVB and
18
TABLE IV. SPENT FUEL MEASUREMENT SYSTEMS
Code Equipment name Description and applications
FDET Fork Detector Irradiated Fuel Detector system that straddles

Measuring System LWR fuel assemblies with pairs of
neutron and g ray detectors. Gross
g ray and neutron intensities and
ratios of intensities can give specific
information on the fuel assembly.
SFAT Spent Fuel Attribute Tester Gross defect device used for
verifying the presence of fission
product or activation product at
the top of the irradiated fuel
assembly.
IRAT Irradiated Fuel Attribute Tester Gross defect device used for
verifying fission product presence
in an irradiated fuel assembly.
ICVD Cerenkov Viewing Device Hand-held light intensifying device
optimized to view Cerenkov light
(near ultraviolet) in a spent fuel
storage pond. System can be used
in a lighted area. Primarily used to
identify irradiated LWR fuel
assemblies.
DCVD Digital Cerenkov Viewing Device Highly sensitive digital device for
viewing Cerenkov light from long
cooled, low burnup fuel.
CBVB CANDU Bundle Verifier for Baskets Attended radiation monitoring
systems that scan storage baskets
CBVS CANDU Bundle Verifier for Stacks or stacks of CANDU fuel bundles
and record gamma intensity as a
function of detector position.
GBUV Gamma Burnup Verifier Facility specific system used to
make high resolution g ray
measurements of spent fuel
assemblies. Collimator in front of
the Ge detector is built into the
facility.
19
FIG. 9. FDET: Fork Detector Irradiated Fuel Measuring System (detector head, GRAND
electronics unit and portable computer).
CBVS). Cerenkov glow viewing devices (ICVD and eventually a digital device,
DCVD) examine the ultraviolet light that appears in the water surrounding spent
fuel. The various measurement systems are described in more detail below.
2.3.2. Gross neutron and gamma ray detection
FDET. The Fork Detector Irradiated Fuel Measuring System shown in

Fig. 9 includes the detector head, a several metre long extension pipe (not
shown), a Gamma Ray and Neutron Detector electronics unit (currently the
GRAND3 but eventually to be replaced by the MiniGRAND) and a portable
computer. The detector head incorporates g ray insensitive neutron detectors
(four gas filled fission chamber proportional counters) and g ray detectors
suitable for measuring extremely high g ray intensities (two gas filled ionization
chambers). The neutron and g ray signatures measured by the detectors are
used to verify the highly radioactive spent fuel assemblies stored underwater in
spent fuel ponds. The FDET is positioned about 1 m above the tops of
neighbouring assemblies. The irradiated fuel assembly being measured is lifted
so that the tines of the detector straddle the fuel portion of the assembly in
order to collect the neutron and gross gamma data.
The ratio of the neutron to g ray data, when combined with other,
complementary information, is used to characterize a particular type of fuel
assembly, giving information related to its neutron exposure in the reactor, its
initial fissile fuel content and its irradiation history (e.g. the number of cycles
for which the assembly was in the reactor).
2.3.3. Gamma ray energy spectral analysis
SFAT. The Spent Fuel Attribute Tester (Fig. 10), consisting of a

multichannel analyser electronics unit and a NaI or CdZnTe detector, is used for
20
FIG. 10. SFAT: Spent Fuel Attribute Tester.
taking measurements from the top of a fuel assembly as it sits in the storage rack.
The SFAT provides a qualitative verification of the presence of spent fuel
through detection of particular fission product g rays — either from 137Cs
(662 keV) for fuel that has cooled for longer than four years or from short lived
fission products such as 144Pr (2182 keV) for fuel with short cooling times.
Activation products such as 60Co are also identifiable. The SFAT is particularly
helpful in situations where Cerenkov viewing cannot provide verification, e.g.
when Cerenkov radiation is weak because the spent fuel has low burnup and/or
a long cooling time, or when water in the storage pond is insufficiently clear. The
SFAT and its lead shielding are housed in a stainless steel watertight container
which is submerged in a storage pond and positioned over the item to be exam-
ined. A watertight collimator pipe is attached below the detector housing to
permit only radiation from the principal assembly rather than from adjacent
assemblies to reach the detector. A multichannel analyser provides for acquisi-
tion, recording and analysis of data, as well as supplying power to the detector.
The intensity of the selected g rays from a specific fuel assembly is compared with
the spectrum from the gap separating the assembly from its neighbour to confirm
the presence of fission or activation products in the measured assembly.
IRAT. The Irradiated Fuel Attribute Tester (Fig. 11) is a small, lightweight
CdZnTe based detector that can be suspended from a spent fuel pond bridge
and used to measure a fission product spectrum from a spent fuel assembly
partially raised from a storage rack. The detector is housed in a stainless steel
cylinder that includes shielding and a collimator. A multichannel analyser
collects and analyses spectral information from a spent fuel assembly. The
presence of fission product isotopes such as 137Cs, 134Cs, 144Pr, 154Eu and others
is used to confirm the irradiated fuel characteristics.
2.3.4. Gamma ray intensity scanning
CBVB, CBVS. The CANDU Bundle Verifier, suspended on an automatic

winch whose speed can be set for scanning either storage baskets or storage
21
FIG. 11. IRAT: Irradiated Fuel Attribute Tester.
stacks, includes a highly collimated and shielded CdTe detector. The verifier is
attached to an amplifier and a portable computer. The computer can be used
either with an external analyser for high count rate conditions or with an
internal multichannel analyser card for moderate count rate applications. The
662 keV g ray line from 137Cs generally dominates a spectrum for spent fuel
that has cooled longer than two years and provides a useful signature for
verifying the spent fuel. For shorter cooling times the 757 keV line from
95Nb/95Zr is used to verify the presence of spent fuel. The particular g ray line
to be used is selected in the SCANDU program. The detector head is moved at

a selected speed vertically across the face of the stacked fuel and a scan
sequence is initiated in the computer. The g ray intensity is measured as a
function of the vertical position. The high intensity peaks, indicating irradiated
fuel bundles, are counted and compared with the declared information on the
number of stored fuel bundles.
2.3.5. Cerenkov radiation detection
ICVD, DCVD. The Cerenkov Viewing Device (ICVD) and Digital

Cerenkov Viewing Device (DCVD) are image intensifier viewing devices
sensitive to the ultraviolet radiation in the water surrounding spent fuel
assemblies. The hand-held ICVD is shown in Fig. 12. The viewing device is
capable of operating with facility lights turned on in the spent fuel pond area.
The ICVD is optimized for ultraviolet radiation by filtering away most of the
visible light and by having an image intensifier tube primarily sensitive to the
ultraviolet light frequencies. Cerenkov radiation is derived from the intense
g radiation emanating from spent fuel, which when absorbed in the water
produces high energy recoil electrons. In many cases these electrons exceed the
speed of light and therefore must lose energy by emitting radiation (Cerenkov
22
FIG. 12. ICVD: Cerenkov Viewing Device.
radiation). Spent fuel also emits b particles (which are also energetic electrons),
adding to the Cerenkov radiation. Spent fuel assemblies are characterized by
Cerenkov glow patterns that are bright in the regions immediately adjacent to
the fuel rods. The variation in light intensity is apparent when viewed from a
position aligned directly above the fuel rods. With careful alignment and
appropriate assessment of the object being viewed, an irradiated fuel assembly
can be distinguished from a non-fuel item that may look the same to the naked
eye. Typically, a row of fuel assemblies is viewed vertically from the bridge
while the facility operator slowly runs the bridge down the row. One inspector
views the items in the row through the ICVD and verbally declares each item
as spent fuel, as a void or as some other object, while a second inspector
compares the observed results with the facility declarations. The DCVD is
currently being developed for use in verifying assembles with long cooling
times and/or low burnups which have weak Cerenkov signals that cannot be
seen with a standard ICVD.
2.4. OTHER NDA TECHNIQUES
2.4.1. Radiation measurement
KEDG. The K-edge Densitometer is facility owned equipment used by

the IAEA to determine the Pu concentration in solutions. The system consists
of a high resolution Ge detector, a multichannel analyser and a portable
computer. A 57Se/57Co source of low energy g rays is positioned to permit the g
radiation to pass through a small sample vial of the solution. Determination of
the amount of absorption of this radiation provides a highly sensitive measure
of the concentration of Pu in the sample.
23
FIG. 13. LCBS: Load Cell Based Weighing System.
2.4.2. Physical property measurement
The IAEA also uses equipment to measure such quantities as the weight
of an object (LCBS), the wall thickness of a container (ULTG) and the liquid
level in a tank (PTMS).
LCBS. The Load Cell Based Weighing System, shown in Fig. 13, operates
in two load ranges up to 5000 and 20 000 kg and provides a convenient and
rapid means of determining the gross weight of bulky, massive objects such as
UF6 shipping cylinders. The load cell construction includes two shackles
separated by a load supporting element that is bonded to a strain gauge. When
a load is lifted with the hoist, the strain gauge deforms, changing its electrical
resistance. The resistance change is converted into a weight displayed on a
digital readout unit that is attached through a cable to the load cell. Typically,
gross weights are determined with this system to an accuracy of better than 1%.
FIG. 14. ULTG: Ultrasonic Thickness Gauge.
24
ULTG. The Ultrasonic Thickness Gauge (Fig. 14) is a small hand-held

device with a digital readout that measures the wall thickness of an object
based on the round trip flight time of ultrasonic waves that are reflected from
the inner wall. The thickness information is sometimes needed to adjust for
radiation attenuation in walls of containers such as UF6 shipping containers,
UO2 hoppers and UO2 cans. These corrections are particularly important when
container wall thicknesses vary. Using the ULTG standard probe, the typical
measurement range for steel extends from 1.2 to 200 mm. In the standard mode
of operation, the speed of the ultrasonic waves in a particular medium is stored
in the memory of the ULTG, so that the flight time can be internally converted
directly into a wall thickness and displayed on the readout.
25
3. DESTRUCTIVE ANALYSIS
Destructive analysis (DA) measurements for element assay and

determination of isotopic composition can be made on all types of solid and
liquid materials encountered in bulk handling nuclear plants. DA is used in the
following ways:
(a) To verify that protracted diversion of safeguarded nuclear materials has

not occurred,
(b) To certify working standards used for the calibration of NDA and
installed verification instruments,
(c) To provide assurance of the quality and independence of on-site
measurements (e.g. validation of facility specific procedures),
(d) To carry out periodic verifications of the operator’s measurement system.
DA verification measurements involve the following steps:
(1) The taking of independent samples;

(2) Their conditioning at the facility to ensure that they are in a chemical
form adequate for maintaining their integrity during transport;
(3) Their packaging, sealing and shipment to the IAEA Safeguards
Analytical Laboratory (SAL) in Seibersdorf, near Vienna;
(4) Their analysis by SAL or the Network of Analytical Laboratories
(NWAL), consisting of laboratories in different States that have been
certified to analyse safeguards relevant samples;
(5) The statistical evaluation of the results of their analysis.
To obtain meaningful and sufficiently accurate results, it is necessary to

apply optimized and validated procedures for each of these steps.
Bulk measurement is generally considered to be part of sampling. The
sample related bulk data collected on-site by the inspector concomitantly with
the sampling include the weights or volumes of the sampled items or batches as
declared by the operator and verified by the inspector. In addition to the bulk
data, the operator’s declarations for the elemental and isotopic compositions of
the material sampled are recorded in a working paper. This working paper
provides instructions for the sample amounts to be taken and the most
appropriate sample bottle to be used. Specific types of sample bottle have been
selected and tested by the IAEA for taking and shipping samples of various
types of materials (Fig. 15).
The main analytical techniques applied in DA measurements are
summarized in Table V. The measurement precisions and accuracies reflected in
26
DESTRUCTIVE ANALYSIS
FIG. 15. Sample bottles used for IAEA verification samples. Top left: vials for Pu, MOX
or high enriched U powder. Top right: hard polyethylene bottle for hard or solid materi-
als. Bottom left: glass bottle for depleted, natural or low enriched U powder. Bottom right:
UF6 container.
the table by the random and systematic uncertainties, respectively, are values
achieved in the analysis of materials of nuclear grade or similar chemical purity.
They include the contributions of all uncertainties occurring after sampling.
The effects of sampling, impurities and foreign components will vary with the
type of material, to the extent that sampling uncertainties can become the
dominant factor in the total measurement error.
3.1. ELEMENTAL ANALYSIS
3.1.1. Uranium by potentiometric titration
The New Brunswick Laboratory Davies and Gray titration is the basic
method for the determination of U content in gram size samples of all types of
non-irradiated materials. An automated titration system (Fig. 16), developed at
27
TABLE V. MAIN ANALYTICAL TECHNIQUES USED BY THE SAL

AND THE NWAL
Analysed Type of Uncertainty (% rel.)

Analytical technique
for material Random Systematic
Elemental analysis
NBL Davies and Gray U U, U–Pu, U–Tha 0.05 0.05
titration
MacDonald and Savage Pu Pu materialsa 0.1 0.1
titration
Controlled potential Pu Pure Pu 0.1 0.1
coulometry materials
Ignition gravimetry U, Pu U oxides 0.05 0.05
K-edge C ray densitometry U, Th, Pu U, Pu, U–Pu, 0.2 0.2
U–Tha
K C ray fluorescence Pu Pu materialsa 0.2 0.2
analysis
Wavelength dispersive C ray Pu, U Pure U and Pu 0.3 0.3
fluorescence spectrometry oxides, and MOXa
Isotopic dilution mass U, Pu Spent fuel input 0.1 0.1
spectrometry solutions, Pu and
U–Pu materials,
HALW
Plutonium (VI) Pu Pu, U–Pua 0.2 0.2
spectrophotometry
Alpha spectrometry Np, AM, HAWL, Spent 5.0 5.0
CM fuel input
Isotopic analysis
Thermal ionization mass U and Pu All Pu and U 0.05b 0.05b
spectrometry isotopes materials, and
spent fuel input
solutions
High resolution g ray Pu Pure U and Pu 0.5–2.0 0.5–2.0
spectrometry (Ge detector) isotopes, materials
Am, Np
Gamma ray spectrometry 235U Low enriched 0.2–0.5 0.2–0.5
(NaI detector) U materials
Alpha spectrometry 238Pu Pu materials 0.2 0.3
a Except spent fuel.

b For ratios of major isotopes.
28
FIG. 16. Automatic U titrator.
SAL, achieves measurement precisions and accuracies of 0.05%rel. or better in

routine operation. This method is applicable to samples of any U materials
containing at least 50 mg U so that at least four replicate aliquots of the sample,
containing at least 10 mg U each, can be titrated.
3.1.2. Plutonium by potentiometric titration
The MacDonald and Savage titration is used for the determination of Pu

content in gram size samples of non-irradiated nuclear materials. It provides
precisions and accuracies of 0.1%rel. or better. This method has been designed
to determine 2–4 mg Pu in nitric acid solutions. It is suitable for the direct
determination of Pu in nuclear materials ranging from Pu from a reprocessing
plant to fresh reactor fuel materials with a U:Pu ratio of up to 30.
3.1.3. Plutonium by controlled potential coulometry
Controlled potential coulometry (Fig. 17) is used to determine 4–10 mg

Pu. Coulometry can also be used to determine Pu in samples of industrial
materials, provided that chemical separation is first carried out to remove
potential interfering elements. The technique applies to gram size Pu samples,
such as Pu product solutions, Pu metal and Pu oxide powders or pellets, and to
mixed U and Pu oxides after dissolution of the solid sample.
3.1.4. Uranium by ignition gravimetry
Ignition gravimetry is applied for determining U concentrations in

nuclear grade U and Pu oxides. An accurately weighed sample is converted to
stoichiometric U3O8 by ignition in air to a constant mass at 900°C ± 10°C for U.
29
FIG. 17. Coulometry cell for determination of Pu concentration.
The amount of U or Pu in the sample is calculated using a gravimetric

conversion factor for U3O8 to U, depending on the isotopic composition of the
sample. The precision and accuracy for nuclear grade oxides containing less
than 200 ppm of impurities are of the order of 0.05%rel. or better.
The presence of non-volatile impurities (the more frequent and abundant
being Fe, Si, Al and Ca) requires a correction, based on the impurity content
determined by C ray fluorescence spectrometry and/or inductively coupled
plasma–mass spectrometry.
3.1.5. Uranium, thorium or plutonium by K-edge C ray densitometry
K-edge C ray densitometry (or hybrid K-edge densitometry, HKED) is

applicable to all U, Th and Pu materials and to mixed U–Th or U–Pu samples
containing a sufficient amount of the analyte: a precision and accuracy of about
0.2%rel. may be achieved when the concentration of the analyte ranges from
50 to 150 g/L. The method is very selective.
3.1.6. Plutonium by K C ray fluorescence analysis
K C ray fluorescence analysis is applied to samples of PuO2 and Pu nitrate

solutions containing at least 3–4 mg Pu with the addition of known amounts of
30
FIG. 18. Uranium/plutonium by K-edge X ray densitometry (HKED).
U as an internal standard. It is also used for U assay in samples of MOX with a

determination of the Pu content by titration. A precision and accuracy of
0.2%rel. are achievable.
3.1.7. Plutonium and/or uranium by wavelength dispersive C ray fluorescence

spectrometry
C ray fluorescence spectrometry is used in conjunction with a commercial

high frequency furnace for fast analysis of Pu and mixed U–Pu oxides (Fig. 19).
About 0.3 g of the sample is melted in a lithium borate flux and the melt cast
into a platinum dish, producing a borate disk of very homogeneous
composition. The concentrations of Th, U, Np, Pu and Am can be determined
simultaneously by measuring the fluorescence of the La lines. The calibration
FIG. 19. X ray fluorescence spectrometer.
31
curves are linear over at least a tenfold change in concentration. A complete

analysis can be performed in about 15 min, with a reproducibility of about 0.3%
for the concentrations of the major heavy elements.
C ray fluorescence spectrometry is also used to determine, semi-
quantitatively in various types of samples, the concentrations of major, minor
and trace elements with atomic masses from 9 (fluorine) through to 89–103 (the
actinide elements).
3.1.8. Uranium or plutonium by isotope dilution mass spectrometry
Isotope dilution mass spectrometry is applied for U or Pu determinations

in all samples of spent fuel input solutions, but also for samples of low content,
such as milligram size U–Pu samples and wastes.
For U and/or Pu determinations in high burnup spent fuel input solutions,
an aliquot of the sample solution is spiked with a known amount of a certified
tracer containing enriched 235U and 239Pu. For pure U materials, a spike of 233U
is used; for pure Pu materials or low burnup spent fuel, a spike of 242Pu or 244Pu
is used. Spiked solutions of Pu bearing materials are chemically treated to
attain an isotopic equilibrium of Pu. Two spiked aliquots and an unspiked
aliquot are separately purified by chromatography in order to provide pure
fractions for thermal ionization mass spectrometry (see Section 3.2.1.). The
chemical treatment of spent fuel samples is performed with a fully automatic,
robotized system (Fig. 20). The resulting U and Pu fractions are then
FIG. 20. Robotized system for separation of spent fuel input solution samples.
32
evaporated to dryness and redissolved in nitric acid to yield solutions

containing about 1 mg U and 50 ng Pu per microlitre. The isotopic ratios of both
the spiked and unspiked aliquots are measured by thermal ionization mass
spectrometry and the U and Pu contents are calculated accordingly. When the
original sample can be spiked directly and total evaporation mass spectrometry
measurements are done, the elemental assays have a precision and accuracy of
0.1%rel. or better.
3.2. ISOTOPIC ANALYSIS
3.2.1. Uranium or plutonium isotopic composition by thermal ionization mass

spectrometry
Thermal ionization mass spectrometry, employing three multidetector

mass spectrometers, each equipped with nine Faraday cups, is used to measure
the U or Pu isotopic composition of all samples of nuclear materials submitted
to SAL (Fig. 21). Comprehensive new software, developed in co-operation with
the Institute of Reference Materials and Measurements in Geel, Belgium, the
Los Alamos National Laboratory in the United States of America and SAL,
includes routines for basic calibration steps, such as the cup linearity test,
relative cup efficiency measurements and a system calibration of mass
fractionation effects.
Isotope ratios of 0.05–20 can be measured with a precision and accuracy
of 0.05%rel. using a data collection procedure involving total evaporation of
the sample loaded on the filament. This procedure greatly reduces the mass
fractionation effects.
FIG. 21. Thermal ionization mass spectrometer.
33
3.2.2. Plutonium isotopic composition by high resolution g ray spectrometry
High resolution g ray spectrometry is used to screen all Pu samples as

they are received at SAL. The Pu content of samples containing Np is analysed
by isotope dilution mass spectrometry rather than by potentiometric titration.
SAL has thereby acquired considerable experience in isotopic analyses using a
multipurpose g ray spectrometry analysis program called MGA (Multi-Group
Analysis).
Samples containing Pu are placed, in their original packaging, on a planar
Ge detector and a spectrum is acquired in the energy range 0–614 keV. The
spectrum is then analysed using MGA, which calculates the abundances of
238Pu, 239Pu, 240Pu and 241Pu. The isotope 242Pu is estimated from isotopic
correlation. The abundances of 235U and 237Np, if present in the Pu sample, as

well as that of 241Am, are determined simultaneously. Typical precisions and
accuracies range between 0.5 and 2%rel. for all isotope abundances except
242Pu.
3.2.3. Uranium-235 in solution by g ray spectrometry
Gamma ray spectrometry with a NaI detector is used as a backup

procedure for mass spectrometry in the determination of 235U enrichment of U
samples, which are dissolved and analysed for U concentration by
potentiometric titration.
In this procedure, 5 mL aliquots of a U solution containing about 80 g/L
U in 1M nitric acid are weighed into identically shaped glass tubes and the
tubes closed with a stopper. A set of five standard solutions, containing known
concentrations of NBS standards (U-010, U-015, U-020A, U-030A and U-050)
is used for calibration. The number of counts at the 185.7 keV energy peak of
235U is calculated and related, in weight per cent, to the total U content of the
sample which has been assayed by titration. In the absence of radioisotope

interferences, the results have a precision and accuracy ranging between
0.5%rel. for natural and 0.2%rel. for enriched U.
3.3. OTHER DA TECHNIQUES
Alpha spectrometry is applied for the measurement of 238Pu abundance

with Si (Li) or ion implanted detectors. This method is used in parallel with
isotope dilution mass spectrometry for the determination of 238Pu abundance
or for the measurement of Pu in spent fuel samples. Neptunium-237 (241Am)
and 244Cm are also measured by alpha spectrometry in combination with
34
chemical separations. 237Np (241Am) and 244Cm are also measured by alpha
spectrometry in combination with chemical separations.
Plutonium(VI) spectrophotometry is applicable to the determination of
milligram amounts of Pu in small samples of products, with accuracies similar
to those of titration.
Inductively Coupled Mass Spectrometry (ICP/MS) can determine most
elements at the parts per billion (ppb) level in solution. It is used for the
quantitative and qualitative determination of impurities in various matrices
(including many inspection sample types).
35
4. CONTAINMENT AND SURVEILLANCE
Containment and surveillance techniques are extensively used by the

IAEA because they are flexible and cost effective. The two main C/S categories
are optical surveillance and sealing systems.
Optical surveillance is most effective in storage areas (such as spent fuel
storage ponds) with relatively few activities that could be interpreted as the
removal of nuclear material. A typical application would consist of two or more
cameras positioned to completely cover the storage area. The field of view of
the cameras is such that any movement of items that could be the removal of
nuclear material is easily identified. This means that items have to be
sufficiently large within the field of view to be identified and that one or more
images have to be recorded during the movement of material. The image
recording may be set at a periodic frequency (significantly shorter than the
fastest possible removal time) or the motion (i.e. scene change) may trigger the
recording. Optical surveillance is intrinsically an unattended operation that
may be enhanced by the remote transmission of image data or system
operation data (i.e. the status of the surveillance system). Unattended and
remote monitoring systems are discussed in Section 5.
Seals are typically applied to individual items containing nuclear material.
A seal can help to indicate that material was neither introduced into or
removed from a container. At the same time, sealing provides a unique identity
for the sealed container. Unattended IAEA monitoring equipment is also
sealed. Most IAEA seals are applied for extended periods of time, typically
several months to years. Seals may be either single use seals that are replaced
when checked or seals that are verifiable in situ, i.e. they can be checked for
integrity and identity in the field. If the seals are verifiable in situ then the
verification activity must be efficient (to limit radiation exposure to the
inspector) and extremely reliable. The in situ verification activity must consist
of checking the item containment, the seal integrity and the method of the
seal’s attachment to the item.
4.1. SURVEILLANCE
Surveillance includes both human and instrument observation. Because it

is prohibitively expensive to provide round-the-clock human surveillance, the
IAEA has developed a range of optical surveillance systems that can provide
effective, ongoing surveillance when an inspector is not physically present on-
site. Unattended optical surveillance techniques are used widely by the IAEA
to support and complement nuclear material accountancy and to provide
36
CONTAINMENT AND SURVEILLANCE
continuity of knowledge about nuclear materials and other items of safeguards

significance between on-site inspection visits.
Effective surveillance is achieved when a camera’s field of view covers the
entire area of safeguards interest to capture the movement of safeguarded
items. Additionally, the picture taking interval is set to record at least two
images, should the item be moved, so that its direction of movement can be
determined. The image recording frequency may be set at a fixed time interval,
which is significantly shorter that the fastest removal time, or may be triggered
by scene change detection or other external triggers.
Optical surveillance is intrinsically an unattended technique that can be
used to record images only, or it may be integrated with other unattended
monitoring equipment to provide nuclear measurement, containment history
and other data. The IAEA’s surveillance systems can also automatically
transfer data to IAEA Headquarters or to an IAEA regional office.
Surveillance equipment is designed to meet several basic application
requirements. Chiefly, those requirements are as follows:
(a) Single camera — for easy to access locations,

(b) Single camera — for difficult to access locations,
(c) Multi-camera — for larger and more complex facilities,
(d) Short term surveillance — for activities that include open core
monitoring,
(e) Surveillance — for remote monitoring,
(f) Underwater closed circuit TV — for attended applications in fuel storage
ponds.
IAEA surveillance equipment has evolved from film cameras, through

systems based on videotape technology, to today’s digital image surveillance
(DIS) systems. The evolution of IAEA surveillance equipment has been
mandated mostly by strong commercial trends that dictate the availability of
applicable technologies on the market. With a significant reduction in the
number of moving parts, DIS is inherently more reliable than previous film and
videotape technologies. Other benefits include enhanced digital data
evaluation, assisted review capabilities, improved authentication and
encryption and its facilitation of remote monitoring.
In 1995, the IAEA embarked upon a replacement programme to phase
out old and obsolete surveillance equipment. In 1998, the Department of
Safeguards decided that surveillance systems based on the DCM14 digital
camera module (Fig. 22) most closely met the essential user requirements for
the IAEA surveillance systems and that they were the most suitable equipment
for the replacement of the existing film and videotape based systems. While
37
FIG. 22. DCM14 with video CCD camera (CCD: Charge Coupled Device).
very compact, the DCM14 performs many tasks required for a safeguards
surveillance system, including:
(1) Digitization of a standard video camera image;

(2) Image and data authentication, ensuring genuineness;
(3) Image and data encryption, ensuring confidentiality;
(4) Image compression to reduce image and data storage requirements;
(5) Local storage to ensure redundancy when data are transmitted out of the
camera housing;
(6) Detection of changes in the camera’s field of view (scene change
detection);
(7) Power management to ensure maximum possible operation should the
local facility’s power fail;
(8) Secure remote surveillance when connected to a communications server.
Safeguards surveillance systems are relatively unique in that the

equipment must operate unattended for extended periods in harsh conditions
and with a high degree of security. Despite repeated attempts over the years to
identify commercial off-the-shelf equivalents, no immediately applicable
systems were found. Systems that nearly meet the requirements invariably
require some degree of modification, entailing additional costs.
Because of its inherent flexibility, the introduction of the DCM14 also
provided a means to consolidate and standardize future surveillance systems.
Using the DCM14 in different configurations it became possible to assemble
single and multiple camera systems for easy and difficult to access locations
from a standard array of basic building blocks. Since 1998, the DCM14 has been
used to construct 5 basic digital surveillance systems, meeting the full range of
safeguards applications, often in difficult environments. The replacement
strategy is summarized in Table VI.
38
TABLE VI. REPLACEMENT AND CONSOLIDATION PLAN FOR

SURVEILLANCE SYSTEMS
Film, videotape and early

DIS and other systems in
digital systems currently in
Application current and future use
use, but to be phased out
(2003–)
(1995–2002)
Installed Single-Camera CSMS Compact ALIS All in one

-for easy to access Surveillance and surveillance, mains
locations Monitoring System operated
(COSMOS)
PHSR Photo Surveillance ALIP All in one
Unit (Minolta) surveillance, portable
battery operated
Installed Single-Camera GDTV Gemini Digital DSOS Digital single-

-for difficult to access Video System camera optical
locations surveillance
MIVS Modular Integrated
Video System
Installed Multi-Camera MXTV Multiplex TV SDIS Server digital image

Surveillance System surveillance
Up to 6 cameras
MOSS Multi-Camera DMOS Digital Multi-
Optical Surveillance Camera Optical
System Surveillance
Between 6 and 16
cameras
UEMS Upgraded Euratom FAST FAST company
Multi-Camera surveillance system
Optical Surveillance Developed by
System (EMOSS) Euratom for joint
inspection use
VSEU Video System
Multiplex
(DigiQuad)
Short Term Surveillance STVS Short Term TV ALIP All in one

System surveillance portable
Surveillance for Remote SDIS Server digital image

Monitoring surveillance
39
TABLE VI. (cont.)
Film, videotape and early

DIS and other systems in
digital systems currently in
Application current and future use
use, but to be phased out
(2003–)
(1995–2002)
DMOS Digital multi-

camera optical
surveillance
Underwater TV UWTV Underwater TV UWTV Underwater TV

-for attended UWVD Underwater Viewing UWVD Underwater viewing
applications Device device
Surveillance Review GARS General Advanced GARS General Advanced

- hardware and 6.3 Review Station 6.4 Review Station
software Version 6.3 Version 6.4
MARS MIVS Advanced
Review Station
MORE Multi-system Optical
Review Station
Surveillance continues to play an important role in safeguards. Over the

past few years there has been a steady increase in the number of camera units
deployed in safeguarded facilities.
Currently, the IAEA maintains about 800 cameras connected to
400 surveillance systems in 170 safeguarded sites worldwide. While
replacement of the older systems is well under way, the programme is not
expected to be fully completed before 2005. Until that time, old and new
systems will continue to coexist. Table VII provides an overview of the IAEA’s
main systems for the coming years.
Equipment has also been developed to provide an increasingly
sophisticated review capability for surveillance. Following the same technology
trends, review stations have evolved from film review tables, through videotape
systems (some with advanced features such as scene change detection) to the
IAEA’s most recent GARS review software that can be run on a personal
computer equipped with the appropriate digital media peripherals. Further
details of the IAEA’s new and most widely used digital surveillance systems
follow.
40
TABLE VII. OPTICAL SURVEILLANCE SYSTEMS
Photographic system
PHSR Photo Surveillance Unit (Minolta) Battery powered, twin and triple
Minolta film cameras used for general
and short term surveillance.
Phased out.
Videotape: single camera surveillance systems

CSMS Compact Surveillance and Battery or mains powered, self-
Monitoring System (COSMOS) contained single camera surveillance
system for easy to access locations.
Phased out.
MIVS Modular Integrated Video System Single camera surveillance systems
for difficult to access locations.
Phased out.
SIDS Sample Identification System Facility specific surveillance system
integrated with an HLNC and
triggered by neutrons above a pre-set
threshold, allowing MOX sample
identification in a fuel fabrication facility.
STVS Short Term TV System Single camera and recorder, developed
from MXTV equipment, for short term
surveillance applications.
To be replaced by ALIS and ALIP.
UWTV Underwater TV Commercial underwater closed circuit
TV system (CCTV) for inspector
attended fuel ID verification in storage
ponds.
Digital: single camera surveillance systems

ALIP All In One Surveillance Portable Battery powered, single camera for easy
to access locations or for portable
surveillance applications.
ALIS All In One Surveillance Mains powered, single camera for
installation in easy to access locations
DSOS Digital Single-Camera Optical Single camera for installation in difficult
Surveillance to access locations.
GDTV Gemini Digital Video System Early single camera, digital surveillance
system for difficult to access locations.
A replacement for CSMS and MIVS.
To have been phased out by early 2003.
41
TABLE VII. (cont.)
Videotape: Multi-camera surveillance systems

FTPV Fuel Transfer Video Facility specific CCTV system used at
fuel transfer ponds.
MOSS Multi-Camera Optical Videotape based, multiple camera
Surveillance System surveillance system for up to 16 cameras.
To be phased out by 2005/2006.
MXTV Multiplex TV Surveillance System Videotape based, multiple camera
surveillance system for up 16 cameras.
To be phased out by 2004/2005.
VSEU Video System Multiplex Dual use surveillance systems developed
(DigiQuad) by Euratom.
VSPC Video system Facility specific CCTV system for up to
4 cameras on a split display screen.
Digital: Multi-camera surveillance systems

DMOS Digital Multi-Camera Optical Multiple camera surveillance system for
Surveillance up to 16 cameras with remote monitoring
capability.
SDIS Server Digital Image Surveillance Multiple camera surveillance system for
up to 6 cameras with remote monitoring
capability.
UEMS Upgraded Euratom Multi-Camera Upgraded EMOSS system for up to
Optical Surveillance System 4 cameras.
To be phased out by 2004.
Surveillance review systems

GARS General Advanced Review Station For the review of ALIS, ALIP, DMOS,
Software DSOS, GDTV, SDIS surveillance.
MARS MIVS Advanced Review Station For MIVS surveillance only
Phased out.
MORE Multi-system Optical Review For COSMOS, MIVS, MXTV, MOSS,
Station VSEU.
To be phased out by end 2005.
42
FIG. 23. ALIS: All In One Surveillance Unit.
4.1.1. Installed single camera for easy to access locations
ALIS. The All In One Surveillance Unit (Fig. 23) is a mains operated,
fully self-contained digital surveillance system based on the DCM14 digital
camera module. All the components fit within a standard IAEA camera
enclosure with all the functionality of the DCM14 plus an intergraded inspector
interface terminal. Images and associated log files are stored on PCMCIA
flashcards. With a 660 Mbyte flashcard installed, ALIS can record between
40 000 and 50 000 images, depending on the compression used.
4.1.2. Installed single camera for difficult to access locations
DSOS. The Digital Single Camera Optical Surveillance System (Fig. 24)
is based on DCM14 technology and is designed for applications where the
camera must be placed in a difficult to access location. DSOS consists of a
DCM14 based digital camera connected to a recording unit by a special
FIG. 24. DSOS: Digital Single Camera Optical Surveillance System.
43
composite cable. The recording unit, which is also based on DCM14 technology,
allows an inspector to service the system at a more convenient and safe location
using procedures similar to those used when servicing an ALIS. DSOS may also
be used as a direct replacement for MIVS.
4.1.3. Installed multi-camera
SDIS. The Server based Digital Surveillance System (Fig. 25) was initially
developed for remote monitoring applications. Its primary function is the
collection of images and data from up to 6 DCM14 surveillance cameras. It may
also be used for the direct interrogation of VACOSS seals. The SDIS server
sorts and classifies image and other data and can securely transfer images and
data to IAEA offices. An uninterrupted power supply unit is an integral part of
SDIS and has been designed to keep the system in full operation for about
48 hours without an external mains power supply. Figure 25 shows the internal
parts of SDIS.
Two modes of operation are available:
(1) Unattended: The data are stored on a removable Jaz disk and are
physically carried to the GARS equipped review station.
(2) Remote monitoring: The data are transferred to an IAEA office by
telephone line (PSTN), ISDN, ADSL, frame relay or satellite link and
subsequently reviewed on a GARS equipped review station.
DMOS. The Digital Multi-Camera Optical Surveillance (Fig. 27) is

designed for unattended and remote monitoring applications. DMOS is used
for applications requiring between 6 and 16 cameras connected to a central
recording and communications console. DMOS is based on DCM14 technology
and as with the SDIS each camera is interrogated by a server computer. Images
FIG. 25. SDIS: Server based Digital Surveillance System.
44
FIG. 26. SDIS server (‘blue box’ — lid open).
and data from each camera are initially stored on a large RAID array prior to
final storage on a removable digital linear tape (DLT).
4.1.4. Short term surveillance
ALIP. The All In One Surveillance Portable Battery Unit (Fig. 28) is a
battery operated, fully self-contained digital surveillance system based on the
FIG. 27. DMOS: Digital Multi-Camera Optical Surveillance.
45
FIG. 28. ALIP: All In One Surveillance Portable Battery Unit.
DCM14 digital camera module. It consists of a camera, a video terminal, the

DCM14 digital camera module, a mains operated power supply and a set
of batteries, all of which are enclosed in a camera housing that has the same
footprint as the standard IAEA camera housing but has been extended
vertically to accommodate the batteries. With fully charged batteries, the
system can perform surveillance duties for up to 100 days with no external
power. Images and associated log files are stored on PC cards. With a
660 Mbyte flashcard installed, ALIP can record between 40 000 and
50 000 images, depending on the compression used.
4.1.5. Underwater TV for attended applications
The portable UWTV system (Fig. 29) is mainly used for verifying bundles
in spent fuel ponds of CANDU type reactors. It can also be used for all other
FIG. 29. UWTV system.
46
kinds of underwater inspections. A complete system consists of a radiation

hardened camera, a camera control unit (CCU) and various accessories such as
a motorized 90 degree rotating head and a light system. Light accessories are
available for long and short distance verification activities. For bundle ID
verifications, the camera must be capable of reading small letters under limited
light conditions and withstand a very high level of radiation, still remaining
watertight down to a depth of 15 metres in water. The CCU has a built-in
monochrome monitor for on-site review. The video can also be recorded on an
external videocassette recorder.
4.1.6. Surveillance review software
MORE. The Multi-System Optical Review Station (Fig. 30) was designed
to assist inspector review of CSMS, MIVS, MXTV and MOSS videotapes. With
the phasing out of CSMS and MIVS in 2002, MORE will continue to be used
for the review of MXTV and MOSS until those surveillance systems in the field
can be replaced by digital multi-camera equivalents (e.g. DMOS).
Each MORE system comprises an IBM compatible computer running
MORE software (with a built-in DAT drive to archive digitized images), a
display unit for the computer, a monochrome video monitor with automatic
CCIR/EIA-170 video standard detection, three videotape recorders to replay
surveillance tapes and a printer for reports. To utilize the scene change
detection option it is first necessary to create set-up files. Regions of interest
are defined within the recorded image captured by the camera in the field.
Regions of interest are defined in the field of view as areas of safeguards
significance (e.g. possible paths for the removal of safeguarded material).
FIG. 30. MORE: Multi-System Optical Review Station.
47
FIG. 31. GARS software.
GARS. The General Advanced Review Station software (Fig. 31) was
developed to run on a personal computer with the appropriate media drives to
review the recorded images from ALIP, ALIS, DSOS, DMOS, GDTV and SDIS.
At its simplest, GARS provides a flexible and user friendly inspector
interface (similar to popular commercial media players) for the review of
images and data from flashcards, Jaz disks, removable hard drives, CD-ROMS
and DLTs. GARS also has advanced features that can be used to reduce an
inspector’s review effort. Those features include image and data authentication
verification, image and data decryption, scene change detection of recorded
images, digital image enhancement and multiple camera display options.
4.1.7. Miscellaneous surveillance systems and options
In addition to the systems described above, new surveillance systems and

equipment to enhance the capabilities of existing surveillance equipment are
both under development and under evaluation. Table VIII summarizes those
systems.
4.2. SEALS
Seals, sometimes referred to as tamper indicating devices, are used to

secure materials, documents or any other important items in a tamper-proof
48
TABLE VIII. OPTICAL SURVEILLANCE SYSTEMS
FAST FAST company surveillance Multiple camera digital surveillance

system system, developed by Euratom for
joint use applications.
Under evaluation.
LRFO Laser Range Finder Option Option for the attachment of DCM14
based cameras to counter in-front-of-lens
tampering.
Under development.
VMOS VACOSS-S/MOSS System Option that allows the integration of the

MOSS multi-camera surveillance system
with a remotely verifiable VACOSS seal.
To be phased out with MOSS.
WCSS Wall Containment Sensor System Wall penetration detection for triggering
surveillance images.
Under evaluation.
containment. The purpose of the seals is to provide evidence of any

unauthorized attempt to gain access to the secured material. The seals also
provide a means of uniquely identifying the secured containers. Depending on
the type of application, several seals are in use by Operations Divisions as
shown in Table IX. It must, however, be pointed out that the seals do not
provide any kind of physical protection, nor were they designed to provide such
protection.
4.2.1. Single use seals
CAPS. The Metallic Seal is extensively used for sealing material

containers, material cabinets and IAEA safeguards equipment. Typically, 20
000 of these metal cap seals are verified each year. The seal is detached in the
field and brought to IAEA Headquarters for identification. The primary
advantages of CAPS are that it is simple, inexpensive and easily attached or
detached by the inspector. Attachment and detachment efficiency is important
to limit the radiation exposure of the inspector. Unique identification of each
seal is obtained by imaging random scratches on the inside surface of the metal
cap and by comparing the installation and removal images (Fig. 32).
49
FIG. 32. Comparison of metal cap seal images for seal validation.
VOID. The Improved Adhesive Seal is made of special material which

cannot be detached without being destroyed. The seal breaks along special slots
cut into the material in a way that does not enable the seal to be reattached. As
for all adhesive seals, the seal is intended only for temporary applications
(24 hours or less).
4.2.2. In situ verifiable seals
In situ verifiable seals are a kind of seal that is uniquely identifiable and
verifiable in the field. They fall into the three main categories of fibre optic,
ultrasonic and electronic seals.
FBOS. In the Fibre Optic General Purpose Seal the seal wire as used in
CAPS is replaced by a multi-strand plastic fibre optic loop with its ends
enclosed in a seal in such a way that a unique random pattern of fibres is
formed. This can be verified by shining a light into the ends of the loop and
observing the magnified pattern of the fibre ends either photographically or by
means of a digital recording of the image pattern. The COBRA Seal System II
50
(FBOS) employs this technology. Immediately after it is installed, the seal is

inserted into a verification assembly that records a reference image of the seal
signature pattern. The verifier consists of a verifier head, a still video camera
and a liquid crystal display monitor. The verifier head holds the body of a
COBRA seal while an image of the seal signature is recorded by the video
camera. The image can then be printed and compared with the reference image
of the same seal.
TABLE IX. SEALING SYSTEMS
Single use seals

CAPS Metallic Seal Cap seal applied to a wide range of
containments for continuity of knowledge
of the contents. Verified at IAEA
Headquarters after removal.
Improved Adhesive Seal Commercial sealing tape that cannot be

removed without destroying the seal.
In situ verifiable seals

ACIV Automatic COBRA Automatic verifier for COBRA seals.
Image Verifier
FBOS Fibre Optic General In situ verifiable fibre optic seal.
Purpose Seal (COBRA)
ULCS Ultrasonic Seal (ARC) Seal used for underwater stack sealing of
fuel bundles. It uses a random coil which
gives the seal a unique signature. An
automated reader compares the signature
with a stored value of the seal in situ.
USSB Ultrasonic Sealing Bolt General purpose bolt seal primarily used
underwater to seal the lids of spent fuel
assembly containers.
VCOS VACOSS-S Electronic Reusable seal consisting of a fibre optic
Seal (Variable Coding loop and electronic seal. Light pulses
Seal System) monitor the loop and every opening and
closing of the seal is stored in the seal. A
palmtop computer reads the seal.
VMOS VACOSS-S/MOSS Unattended system that records the
System closing (or opening) of VACOSS electronic
seals by means of a specially adapted MOSS.
51
A much more logistically convenient COBRA seal verifier has recently

been developed which stores digital images and is able to compare the patterns.
This procedure enables the inspector to automatically verify the seal identity
and integrity in situ and to conveniently store the pattern in a computer.
Figure. 33 illustrates the Automatic COBRA Image Verifier (ACIV) with a
COBRA seal.
ULCS, USSB. The Ultrasonic Seal (ULCS) and Ultrasonic Sealing Bolt
(USSB) are constructed to contain a unique random pattern of, for example,
inclusions of metal pieces in a lighter substrate or a randomly oriented coil of
wire. They are applied in a variety of ways, with special designs for different
applications. Verification is accomplished by transmitting ultrasonic pulses
through the seal with a suitable transducer and observing the unique pattern of
reflections. Verification consists of comparing the pattern obtained when
installed with that obtained during subsequent in situ checks. These types of
seals have proven particularly effective for underwater applications such as for
stacks of CANDU fuel bundles (ULCS) or for bolts closing shipment and
storage containers of LWR spent fuel assemblies (USSB).
VCOS. Electronic seals will be used with increasing frequency in IAEA
applications as remote monitoring becomes more universally applied and as
seal manufacture becomes less expensive. The first IAEA electronic seal,
originally designed in the late 1970s, was the Variable Coding Seal System
(VACOSS-S), shown in Fig. 34. This seal uses electronic encoding methods in
conjunction with fibre optic loops. The VACOSS-S Electronic Seal is intended
for high reliability, long duration surveillance in applications that require
periodic access. The time, date and duration of openings and closings of the
loop are recorded internally for later retrieval. The fibre optic loop is
interrogated with a light pulse every 250 ms for continuity of the light path.
There is no known method of splicing the fibre in the interval between
FIG. 33. ACIV: Automatic COBRA Image Verifier with COBRA seal.
52
FIG. 34. VACOSS-S Electronic Seal with fibre optic loop, interface box and palmtop
computer.
interrogations. The internal batteries have a two year operational lifetime. For
installations with multiple seals in proximity, the seals may be connected in
series. All seals connected in this fashion can be read in sequence without
changing the connection. The electronics are potted in an X ray resistant
compound of epoxy and ceramic particles to frustrate any possible attempt at
reverse engineering. A tamper switch detects any opening of the seal housing.
The seal housing is opened only to replace the internal batteries and openings
are recorded as tamper events.
53
5. UNATTENDED MONITORING
Unattended monitoring systems (UMS) run 24 hours a day, 365 days a

year, continuously monitoring activities in a variety of nuclear facilities. Each
system has redundant sensors and data storage capability and a backup power
supply for short term power outages. The systems are designed to maintain
continuity of knowledge in a cost effective manner while reducing the costly
inspection burden on the IAEA through reduced inspector days in the field. As
the number of nuclear facilities continues to increase worldwide, the use of
unattended systems to reduce inspection effort in the field becomes more and
more critical. Current unattended monitoring systems primarily employ
radiation detection sensors to detect the flow of nuclear material past key
points in the facility process area. However, the suite of sensors also includes
ones capable of measuring temperature, flow, vibration and electromagnetic
fields. For complex nuclear facilities where the plant is automated (remotely
operated), unattended assay and monitoring techniques are an integral part of
the safeguards implementation approach.
Unattended use necessitates that special considerations be included in
the instrument design if the system is to be reliable and cost effective in
providing credible, independent data. This means that the system must operate
without the loss of safeguards relevant data over extended periods, including
times when the power supply to a facility is interrupted. The unit should
automatically record its status periodically. If data are to be sent over
unsecured transmission pathways then the data must be authenticated and
encrypted. If data are to be shipped off-site (remote monitoring) then they
must be encrypted to meet the requirements of the facility and the State for
confidentiality of information and the IAEA requirements for data security.
Because of stringent design considerations, unattended and remote monitoring
equipment typically has to be flexible, modular and highly reliable.
As part of the preparation for field installation, all systems are carefully
tested at the IAEA Safeguards Equipment Support Facility in Vienna using
simulated signals. A testing period of at least 90 days, representative of the
current unattended period between visits by inspectors, is used. With such
testing, early component failure, configuration errors and manufacturing
defects can be eliminated while the system and its components are easily
accessible during the testing phase in Vienna. Once the system has successfully
operated for a full inspection period or approximately 90 days without failure
it is ready for field installation.
Unattended safeguards instruments are often deployed in facility areas
with limited personnel access, such as areas with a high radiation level.
Depending on the facility and the process being safeguarded, the optimum
54
UNATTENDED MONITORING
placement of appropriate instruments, even those requiring custom designs,

more than justifies the initial high cost when the long term overall economics
are considered. Nevertheless, the IAEA is focusing on standardizing all
equipment and systems where possible to allow the maximum efficiency in
utilizing its limited resources. Custom designs and duplication of capability
using different equipment are avoided wherever possible.
The future of UMSs is moving toward fully integrated systems which
incorporate direct integration of surveillance systems with local area networks
to provide a server-type central collect computer. This is the next logical step in
providing the most cost effective approach to the collection of relevant
safeguards data. Such a system would use UMS-type sensors to trigger cameras.
In this way, image data are significantly reduced to relevant events and the
review capability of inspectors is enhanced. Such a system is already in
operation at the BN-350 complex in Kazakhstan. A newer integrated version is
currently being tested at the Safeguards Equipment Support Facility and
scheduled for installation in the Chernobyl Spent Fuel Conditioning Facility in
FIG. 35. The Chernobyl conditioning facility monitoring system being tested in the
Safeguards Equipment Support Facility.
55
the near future. The elements of these systems include an intelligent local
operating network (iLON) that interconnects all UMS devices, digital cameras
and the collect computer. The iLON controls triggering, time synchronization,
local authentication and startup functions. Multi-instrument collect (MIC)
software is used to poll all UMS devices to collect the data on a single collect
computer using the iLON. After an inspector extracts the data, Radiation
Review (RAD Review) is used to examine the data. RAD Review has the
capability of viewing the radiation signals received from the system. Its normal
suite of options includes peak counting algorithms based on various threshold
settings against various expected backgrounds and peak searches. The version
at the BN-350 includes a new integration tool in which the digital camera
images are also part of the database. For each sensor peak, the inspector can
‘click’ on the peak of interest and the associated camera image will be
displayed. This provides a powerful review tool that integrates two previously
separate review functions, allowing for the most cost effective review approach
for IAEA inspectors to draw the required safeguards conclusion.
In summary, the primary advantages of unattended verification
techniques are:
(a) More effective safeguards through continuous monitoring,

(b) Reduced inspection efforts,
(c) Reduced radiation exposure of inspectors and facility staff,
(d) Reduced level of intrusiveness in the operation of nuclear facilities.
Unattended monitoring systems in use are listed in Table X.
ATPM. Advanced Thermo-hydraulic Power Monitoring is used to

monitor the power output of a research reactor and can verify that the output
is consistent with the operator declared power level. This system monitors the
temperature and water flow in the reactor’s primary cooling loop. Because
research reactors can modify their core layout and in turn the associated
radiation level, this system provides a verification method that is independent
of radiation signature. To ensure the necessary fault tolerance, all sensors are
installed as duplicate independent sensors (both temperature and flow), whose
signals are then independently collected in the main cabinet where they are
compared.
This makes it more difficult to falsify the signal and results in a robust
system that continues to provide the required inspection data with multiple
single point component failures. Aside from the use of tamper enclosures and
tamper indicating conduits for the signal cabling, the high sampling rate from
the sensors makes it more difficult to send a false signal to the collecting unit.
56
TABLE X. UNATTENDED AND REMOTE MONITORING SYSTEMS
Unattended Monitoring Systems

ATPM Advanced Thermohydraulic Monitoring system that calculates the
Power Monitor power output of research reactors by
measuring the flow and temperature
differential on the primary coolant loop.
CONS Input Flow Verification Radiation monitoring system that tracks
System the movement of irradiated fuel at a
large scale reprocessing plant
(CONSULHA).
ENGM Entrance Gate Monitor Radiation monitoring system that
monitors non-irradiated fuel assemblies
containing Pu that are brought into the
reactor facility.
REPM Reactor Power Monitor Neutron monitoring system placed
outside the reactor biological shield to
monitor the power level of the reactor.
UFFM Unattended Fuel Flow Monitor Generic radiation monitoring system
that monitors the flow of fresh and
irradiated fuel in a reactor facility.
Placement and details of radiation
detectors are facility dependent.
VIFB CANDU Spent Fuel Radiation monitoring system that counts
Bundle Counter irradiated fuel bundles as they are
discharged to the spent fuel storage bay
of an on-load refuelled power reactor.
VIFC CANDU Core Discharge Radiation monitoring system that
Monitor monitors the discharge of spent fuel
bundles from an on-load refuelled
power reactor core (reactor may be on-
power or shut down).
UFFM. The Unattended Fuel Flow Monitor consists of separate neutron

or g ray detector assemblies permanently installed in a reactor facility. These
detectors are installed in pairs allowing for the necessary fault tolerance so that
no critical safeguards data are lost. The detectors monitor the movement of
fresh fuel assemblies to the reactor (for those reactors utilizing MOX — mixed
oxide fuel that contains plutonium — such as breeder reactors and LWRs on a
MOX cycle), spent fuel assemblies from the reactor to the fuel storage pond,
57
FIG.. 36. Advanced Thermo-hydraulic Power Monitoring.
and spent fuel assemblies out of the storage pond. Neutron detection in the
UFFM is contingent on shielding the neutron detectors from the intense
gamma radiation from the spent fuel. The sealed detector systems may be
attached to massive transport vehicles or to the storage pond wall near the
underwater entrance. Using both neutron and gamma detectors also makes
signal substitution by an adversary extremely difficult.
A typical transfer sequence involving several UFFM units could include
a fresh fuel assembly being brought to the reactor core and a spent fuel
assembly being retrieved and brought to the storage pond. The combination of
neutron and g ray signatures at the successive units characterizes the
transferred material as fresh fuel, spent fuel or another material (e.g. neutron
irradiated blanket material at a breeder reactor facility). Although individual
systems are facility specific, the neutron detectors on the transport vehicles are
typically 3He proportional counters or fission chambers and the g ray detectors
are typically NaI scintillators or ionization chambers. Underwater in the
storage pond the neutron detector is usually a 10B lined gas filled proportional
58
counter and the g ray detector is a gas filled ionization chamber. UFFMs are
designed to monitor neutron and g ray counts continuously but store only data
that are significantly above background levels. The IAEA is in the process of
upgrading all data generators that collect this data with removable flash
memory capable of storing at least 100 days of data. This is equivalent to the
current inspection period and ensures the necessary fault tolerance in case of
computer failure. Surveillance cameras normally complement a UFFM over
the fuel transfer route.
ENGM. The Entrance Gate Monitor is included at plutonium fuelled
reactor facilities which incorporate the UFFM. This is a permanently installed
passive neutron coincidence collar detector (PNCL). Fresh fuel assemblies
entering the reactor facility must pass through the ENGM so that their Pu
content can be verified. Therefore the ENGM is the system which verifies the
amount of fresh fissile fuel in an assembly and serves as the first detector in a
sequence of detector systems which follow the movement of fuel assemblies
within the reactor facility.
VIFB. The CANDU Spent Fuel Bundle Counter (Fig. 37) is an
unattended system that monitors a strategic location in the spent fuel bundle
pathway of an on-load refuelled power reactor. Collimated g ray detectors
detect the fuel bundle as it passes. The proper placement of detectors and the
use of the appropriate algorithm for the facility enable the device to count the
bundles as they pass and record the direction in which they are moving, even
when two bundles are moving together, which is normally the case.
High operational reliability, great dynamic detection sensitivity (to
include all operational possibilities) and insensitivity to power outages are
some of the important features of the bundle counter. Sufficient redundancy
was built in to accommodate individual component failures without
compromising operation.
VIFC. The CANDU Core Discharge Monitor is a typical unattended
monitoring system operating in an inaccessible area. The VIFC detects
irradiated fuel upon discharge from the core face of a CANDU reactor.
Both neutron (normal on-power discharge signal) and g ray intensities are
continuously monitored. The inspector, upon reviewing the data, is able to
identify in a straightforward, unambiguous manner the abrupt but
characteristic change in count rate associated with fuel bundle discharge. The
review technique is valid for irradiated fuel discharge both when the reactor is
on-power and when it is shut down. Because of the linear increase in
background signal, the system can also track the operating power level of the
reactor.
The VIFC was designed to be fail-safe. Sufficient redundancy was built in
to accommodate individual component failures without compromising
59
FIG. 37. VIFB: CANDU Spent Fuel Bundle Counter.
operation (failure of the VIFC would be exceedingly difficult to recover from

in safeguards terms). The detection modules are designed to last the lifetime of
the reactor, since their location inside the containment area limits possibilities
for maintenance and repair. Automatic performance and failure monitoring
have been incorporated in the VIFC.
60
6. REMOTE MONITORING SYSTEMS
Remote monitoring in the safeguards context is generally considered to

mean that data are transmitted off-site to IAEA Headquarters or to an IAEA
field office. Cost effectiveness is a prime justification for adding this feature to
surveillance, electronic seals and other unattended systems. If data can be sent
to the inspector then the frequency of inspection visits to the facility can be
reduced, thus saving both time and expense. In principle, a remote monitoring
system with ‘state of health’ reporting can also function significantly more
reliably than an unattended system that is serviced at a set frequency. Some
events that would ultimately lead to a failure of the system can be remotely
evaluated and reported in time for appropriate action to be taken. Limited
IAEA human resources, an ever growing stockpile of nuclear material and
economics are likely to accelerate the implementation of remote monitoring in
the near future.
Utilizing remote monitoring (RM) as an additional safeguards tool has
been under consideration by the IAEA for over two decades. In October 1996,
the IAEA Department of Safeguards established its Remote Monitoring
Project (RMP). Its chief objectives were the definition and development of a
remote monitoring infrastructure and its implementation using available
technologies. More specifically, the RMP devised a safeguards policy for
remote monitoring, defined approaches and procedures for LWRs, storage sites
and on-load reactors, specified basic RM equipment, conducted field tests,
undertook cost–benefit analysis of two countries and developed a
Departmental plan for the implementation of RM. Upon its completion in
December 1998, the plan was approved by the Department, allowing RM to
move to the implementation phase with responsibilities divided among the
Operations and Support Divisions.
As of June 2002, the IAEA had 54 digital surveillance systems with
RM capability installed worldwide. Of those, 33 systems with 71 cameras are
operating in the RM mode. In addition to the deployed systems, further work
is currently under way to provide additional monitoring features so that RM
can be extended to other equipment types.
6.1. REMOTE MONITORING EQUIPMENT
The IAEA’s RM equipment is based on systems which can be integrated

with a server running the Microsoft Windows NT operating system. At present,
only C/S components have reached a sufficient level of maturity to be
61
authorized for inspection use. However, other unattended monitoring systems

are currently under development and evaluation to complement and
supplement the basic C/S devices, allowing the remote monitoring of more
complex facility types.
The development of the DCM14 digital camera module was an important
milestone in making RM possible. The DCM14 provided the IAEA with a
single device that could digitize the output from a standard video camera,
convert the analog video into digital images that could be further compressed,
authenticated and encrypted if required. The device also provided the
capability to store images and status data on internal removable media and to
transmit those images and data to an external data collector.
The IAEA’s central data collection and communications controller for field
deployment is the Server Digital Image Surveillance (SDIS) system. Aside from
allowing the connection of digital cameras to VACOSS seals, the SDIS server
also provides the capability to encrypt data and communicate with the IAEA’s
offices over a variety of communication links, including PSTN, ISDN and
satellite. It is also planned that the SDIS server will be connected, via an Ethernet
link, with other UMSs to transmit encrypted data. In addition to the SDIS, the
IAEA is currently testing the Digital Multi-Camera Optical Surveillance
System for remote monitoring where comparatively large numbers of cameras
are required.
6.2. FUTURE DEVELOPMENT ACTIVITIES
From the analyses to date, it is clear that RM implementation costs

depend on facility type, monitoring system used and the country’s
communications costs and its nuclear fuel cycle. In addition, Integrated
Safeguards may not justify the incorporation of RM at LWRs in countries in
which the Additional Protocol is already in force or expected to be in force in
the near future. It has therefore been decided that all proposals for the
incorporation of RM into inspection activities will be subject to a cost–benefit
analysis prior to approval to proceed. To date, Operations divisions have
proposed installations in a further 10 countries. Those proposals will be
subjected to a rigorous cost–benefit analysis.
The challenge for safeguards now is to further reduce RM
implementation costs. Areas have already been identified where it has been
suggested that significant cost savings may be possible. The following are
examples.
62
REMOTE MONITORING SYSTEMS
FIG. 38. Remote monitoring schematic.
6.2.1. Data reduction
Currently, all images and data are collected on-site and transmitted to
IAEA offices. While the inspectorate has a high degree of confidence that
activities of safeguards significance will be fully captured if the picture taking
interval is set properly, excessive amounts of data are transmitted due to the
large number of images. This results in high communication costs even though
a large proportion of those transmitted images contain scenes of no safeguards
significance or are redundant.
By employing techniques to detect only images where a change has
occurred or where an associated event triggers surveillance, it would be
possible to reduce the amount of redundant and non-safeguards relevant data
transmitted. Based on previous IAEA studies, the use of scene change
detection on fixed time interval recordings (SCD) can reduce the number of
redundant scenes and scenes of no safeguards significance by up to 90%. All
DCM14 based systems are SCD capable. The IAEA is currently building up its
field experience over the wide range of facility types so that it may be in a
position to fully accept such techniques with a high degree of confidence.
6.2.2. Alternative communication methods
Currently, the IAEA uses PSTN, ISDN, frame relay and INMARSAT to
link remote sites to communication hubs or directly to the IAEA offices. High
costs are associated with the use of these conventional dial-up, network and
63
FIG. 39. Remote monitoring over a VPN.
satellite services. Where available, the Internet provides comparatively low cost
data communications. By establishing a virtual private network (VPN) using an
Internet service provider as the carrier, the IAEA may be able to take
advantage of the economies of scale associated with these recent, alternative
communication methods. Virtual private networkss can provide possible
solutions to security concerns. Some typical commercial estimates predict
savings of 20–40% over a leased line and from 60–80% over a dial-up service.
64
7. DATA SECURITY
Data security is an important feature of unattended and remote

monitoring systems. In fact those types of safeguards systems permanently
installed at facilities and periodically visited by inspectors transmit data
between different system components and between systems and IAEA
Headquarters through unsecured transmission paths. Those data need to be
cryptographically authenticated to guarantee that they are genuine and may
need to be encrypted to prevent disclosure to the host or to ensure
confidentiality to the Member States.
Security requirements are defined to meet a security target, and in terms
of the types of security services required and the required strength of those
services. Table XI provides a list of security services.
In the list, the ‘Authentication Security Service’ is identified as a separate
service although it is often used as a supporting security service for others. For
example, confidentiality cannot be guaranteed, even if the data are encrypted,
if the receiver’s identity is in doubt (i.e. the sender is unsure of who can decrypt
the data). Similarly, integrity protection is of little value if the originator’s
identity is in doubt.
7.1. INFORMATION PROTECTION REQUIREMENTS
To simplify the application of the security architecture and specifically to

allow the presentation of a limited set of ‘robustness levels’ for cryptographic
approaches, each information type is assessed in terms of the length of time its
security needs to remain uncompromised. This, along with the threat level, will
determine the strength (robustness) of the approach and the algorithms
selected.
Safeguards information used in unattended and remote monitoring
systems can be described by the general information model shown in Fig. 40
and summarized in Table XII. This model is derived from the IAEA Policy
Paper on Remote Monitoring1 and extended to add ‘control data’. The model
segregates the data on the basis of its intended use as follows:
1INTERNATIONAL ATOMIC ENERGY AGENCY, Remote Monitoring for

Safeguarding Nuclear Facilities, Safeguards Policy Series No. 16, IAEA, Vienna (1998).
65
TABLE XI. SECURITY SERVICES*
Security service Ensures that How it is accomplished (‘mechanism’)
Confidentiality The information is Information is encrypted. Data are

(including traffic kept private padded to prevent traffic flow analysis.
flow analysis and Barriers (e.g. firewalls) prevent
data separation) improper data flow (e.g. covert
channel). Confidentiality requires
authentication of receiver(s).
Integrity (including The information has ‘Signature’ over the data allows detection
replay/data not been altered and of alterations. Inclusion of counter/time
substitution is not a copy of value prevents replays. Integrity requires
protection) previous data authentication of origin.
Access control Only authentic user Users prove their identity and authority
with sufficient using trusted means
authority can access
the resource
Non-repudiation Users cannot later User includes information (signature)
deny their action only they could create. Non-repudiation
(‘assertion’) implies authentication of user. May
require time stamp.
Authentication The user is who they Users prove their identity using trusted
say they are authentication means (password,
possession of unique object, biometrics).
Availability The systems are Systems are immunized against attack
(including intrusion operating, unauthorized (e.g. firewalls are deployed), mechanisms
detection, resistance activities do not report attacks to the response agent.
to attacks) compromise operation Prevention of denial-of-service attacks.
Audit Accountability is Generate reports (audit messages) when
maintained for all significant events occur. Integrity and
significant events guaranteed delivery of all reports is
required.
Assurance All required functions The design and its implementation have
are present in been thoroughly reviewed.
deployed equipment
* Derived from the Open Systems Interconnection (OSI) Security Framework, the
Information Assurance Technical Framework (IATF) and the Common Criteria (CC).
66
DATA SECURITY
TABLE XII. SUMMARY OF UNATTENDED AND REMOTE

MONITORING DATA TYPES
Data type Sample data Data use Frequency of use
Verification Camera image, Used for safeguard Period between safeguard

data seal status, review. Only reviews depends upon
operator declaration ‘verification data’ arrangements with facility
may be used to draw but will be in the range of
conclusions about one to six months.
safeguards
implementation at a
facility.
Technical data, Number of triggered Used for technical For remote monitoring
including recordings, review for: systems technical reviews
summary and battery charge level, (1) Planning of are daily but may be
state of tamper indication, inspection activities delayed by up to three
health data equipment (2) Equipment working days. For other
temperature, maintenance and systems technical
failure indication, repair or other reviews are during on-
audit log data follow-up activities. site visit or soon
thereafter.
Control data Set time of day, Used for control by: Real time on-line
set sampling (1) Safeguards operations
interval, run application programs, (as implemented).
diagnostic, key (2) IAEA inspectors
management data and technical staff
(a) Verification data are used during safeguards reviews carried out by
approved IAEA inspectors,
(b) Technical data are used during technical reviews carried out by approved
IAEA technical staff,
(c) Control data are used for the ‘real time’ control of the equipment either
by automated safeguards application programs or manually during
inspection or maintenance activities.
Depending on the surveillance design, some technical data may also be

included in ‘verification data’ if deemed important for the safeguards review
(shown by dotted lines in Fig. 40). Although some general conclusions may be
67
FIG. 40. Unattended and remote monitoring information model.
drawn regarding the data types in the model, there will be exceptions,
depending upon the specific design of the unattended monitoring.
7.2. IAEA REQUIREMENTS
This section will address IAEA requirements for securing data. Member
State requirements will be addressed in Section 7.3..
7.2.1. Verification data
Unattended and remote monitoring systems data must be analysed using

trustworthy data to permit valid conclusions to be drawn. That is, the
information used for analysis must be known to have originated from the
intended source, must be known not to have been changed in transit, and must
be known not to be a repeat of previous data.
The IAEA Policy Paper on Remote Monitoring states the following (see
footnote 2):
“Measures shall be taken to ensure the authenticity of the transmitted

and stored data.”
In the above statement the term ‘authenticity’ refers to the combination

of integrity (including replay/data substitution) and authentication (of origin)
protection, as defined above.
68
DATA SECURITY
The IAEA requirements for confidentiality are defined as follows:
“As a general rule, the detailed safeguards information from the Agency
equipment should not be made available to the States. However,
arrangements could be made for sharing of certain data as part of the
cooperation arrangements with the State authorities.”
Although integrity and authentication are of primary importance, some

safeguards data may require confidentiality protection since knowledge of
actually measured values may allow States to exploit instrument characteristics
or inaccuracies. For example, if the actual measured value and the instrument
accuracy are known, declarations could be manipulated to ensure that each
declaration is within the bounds of measurement error while allowing a
protracted diversion of small amounts of material.
Non-repudiation (of origin) protection is required for operator
declaration data. Depending upon the security mechanism,2 basic non-
repudiation (of origin) may be inherent in the method used for integrity and
authentication. Typically, operator declarations will require the time of the
declaration to be included if there may be a significant delay in the delivery of
the declaration to the IAEA. If a declaration is delivered to the IAEA quickly,
it can securely archive the information upon receipt and be in a position to
confirm that it has not been replaced with a later declaration. On the other
hand, if there may be a delay in delivery, a time stamp would be required.
Non-repudiation, integrity and authentication protection, and confiden-
tiality protection when needed, must be strong enough to withstand external
attacks. As indicated above, not all of the information may require confiden-
tiality protection, and even when required this protection may only be needed
for a limited time period. For example, measurement data may only be sensitive
until the corresponding Member State declaration has been made.3
2 For example, if public key based signatures are used basic non-repudiation of
origin is provided. Stronger non-repudiation protection requires additional mechanisms

to bind the time of signing to the signature. For example, a copy (or digest, complete
with the user’s signature) of the message could be stored in a trusted storage facility so
that evidence is available that the signature was applied at a particular point in time.
Alternatively, a copy (or digest and signature) might be sent to a trusted time stamping
service that would add a time value and sign the overall combination. Trusted time
stamping services and trusted storage facilities may also be required to support other
security services such as trusted auditing and tracking.
3 This may not be true in all cases. Long term analysis of instrument data inaccu-
racy may allow the prediction of future inaccuracies. In such cases instrument data may
need to be protected until the instrument is recalibrated.
69
Thus, although non-repudiation, integrity and authentication protection

will be required throughout the life of the verification data (i.e. many years),
confidentiality protection may4 only be required until the safeguards review
has been completed (e.g. a few months).
7.2.2. Technical data
The IAEA Safeguards Glossary5 points out that:
“Receiving state of health data with a sufficient frequency makes it

possible to detect failures of equipment or tampering early enough for
remedial actions to be implemented to satisfy the timeliness
requirements.”
The IAEA Policy Paper on Remote Monitoring (see footnote 2) states that:
“Data remotely transmitted to Headquarters or Regional Offices are

normally subject to a daily technical review. If this is not possible, there
should be no more than three consecutive working days without review.
The technical review includes monitoring of state-of-health messages and
tamper reports.”
The objective is to transfer these data as quickly as possible to permit

remedial action to be initiated. In practice however, and especially for
unattended monitoring systems not connected to a data network, the receipt of
these data may be delayed until an inspector arrives at the site.6
All technical data require integrity and authentication protection to
ensure that failures, logs and indications of tampering are reliably reported to
IAEA staff. Additionally, for data which may indicate increased equipment
vulnerability, confidentiality protection will be required until remedial action
can be taken. For example, state-of-health data indicating a failure of internal
backup battery power may indicate an increased vulnerability to the removal
4 More sophisticated attacks might use historical data to predict future instru-
ment errors. In such cases confidentiality protection would be required until the equip-
ment is recalibrated.
5 INTERNATIONAL ATOMIC ENERGY AGENCY, IAEA Safeguards
Glossary, 2001 Edition, IAEA/NVS/3, IAEA, Vienna (2002).

6 A networking failure at a site may delay state-of-health data delivery even for
networked systems.
70
DATA SECURITY
of primary power. Such an indication might allow a threat agent to optimize

their attack on the equipment.
The integrity and authentication protection, and confidentiality
protection when needed, must be strong enough to withstand external attacks.
It would appear that such protection may only be needed for a limited period
of time. For example, state of health data may only be sensitive until the
equipment can be maintained. Thus protection may only be required until the
technical review is complete and remedial action has been taken.
It may be appropriate to provide integrity and authentication protection
to some audit and other information for longer periods if its relevance cannot
be determined during the initial technical review. Historical trends may have to
be analysed to arrive at conclusions. In such situations this information may
also be included in the verification data.
7.2.3. Control data
At present, equipment control data are not commonly used in unattended

and remote monitoring systems. However, control data will be used
increasingly to permit sensors to be adjusted and controlled from the collect
computer, and eventually to permit remote monitoring systems to be
controlled from IAEA premises. Control data are transmitted in ‘real time’ and
might include:
(a) The distribution of common time values to sensors by the collect computer,
(b) The control of sensor operation (e.g. camera focus, pan/tilt, sampling rate),
(c) The activation of test and calibration routines in sensors.
In the future, and provided that adequate security measures can be put
into place, the overall remote monitoring system might be controlled from
IAEA offices. In such cases it will be necessary to ensure that remote log-in and
access controls are adequate to ensure the secure control of the equipment.
Typical control information might include:
(1) Secure log-in to the collect computer,

(2) The initiation of file and data transfers,
(3) Operational control of the remote monitoring system,
(4) The activation of test routines in the collect computer,
(5) The secure entry of commands into the collect computer which may in
turn issue secure controls to attached sensors,
(6) The updating of software.
71
Integrity and authentication protection are required for all such data to
prevent their modification. Confidentiality is required for some control data
since observation may reveal sensitive information such as the criteria used to
adjust surveillance triggers.7 Access control and availability protection must be
provided to prevent threat sources from masquerading as authentic sources of
control data and to prevent denial of service attacks at the communications ports.
Control commands can also be susceptible to traffic analysis, whereby an
observer gathers knowledge from an analysis of traffic patterns. For example,
even if a command is encrypted, if the command results in a higher rate of data
transmission an observer can likely draw valid conclusions about the content of
the command.
All required protection measures must be strong enough to withstand
external attacks. These measures are required throughout the lifetime of the
data.
7.3. MEMBER STATE REQUIREMENTS
Member State concerns mainly relate to ensuring that their information

is kept confidential and ensuring that the safeguards programme reliably
reflects their conformance to the agreements.
The Model Safeguards Protocol8 states in part that:
“Article 14:
b. Communication and transmission of information … shall take due

account of the need to protect proprietary or commercially sensitive
information or design information which the Member State regards
as being of particular sensitivity.
Article 15:
a The Agency shall maintain a stringent regime to ensure effective

protection against disclosure of commercial, technological and
7 Confidentiality protection of time of day update control data may not be

needed.
8 INTERNATIONAL ATOMIC ENERGY AGENCY, Model Protocol
Additional to the Agreement(s) between State(s) and the International Atomic Energy
Agency for the Application of Safeguards, INFCIRC/540, IAEA, Vienna (1997).
72
DATA SECURITY
industrial secrets and other confidential information coming to its

knowledge, including such information coming to the Agency’s
knowledge in the implementation of this Protocol.”
The IAEA Policy Paper on Remote Monitoring (see footnote 2) states

that:
“Encryption should be applied during data transmission, as agreed with

the State.
Transmitted data shall be treated and stored as ‘Safeguards Confidential’
information.
States have the right to know the kind of information being transmitted,
and to the protection of the data through appropriate encryption. As a
general rule, the detailed safeguards information from the Agency
equipment should not be made available to the States. However,
arrangements could be made for sharing of certain data as part of the
cooperation arrangements with the State authorities. The arrangements
for sharing such data shall be approved by the Deputy Director General
— Safeguards on a case by case basis.”
It is noted that these Member State requirements are specifically directed

towards protecting the confidentiality of ‘commercial, technological and
industrial secrets’. This Member State requirement for confidentiality will not
normally require additional security mechanisms while the information is
contained within the physically secure facility provided by the Member State.
This Member State requirement will normally only apply to information while
it is within communications networks provided by the IAEA, on transportable
media, or within IAEA offices or inspection equipment. Exceptions to this may
occur if sensors and collect computers are not within a single contiguous facility
(and thus require a data link through unprotected areas) or if the facility
communications links are otherwise exposed (e.g. due to the use of wireless
links or the use of facility networks shared with other users).9
9 Such deployments are not expected. The use of wireless networks and shared
networks is not recommended but may be required due to the constraints on specific
installations.
73
For UMS and RMS, and for the purposes of this security architecture,
Member State confidentiality requirements impact two areas:
(1) Ensuring that adequate protection is provided for data links between
sensors and collect computers if not otherwise protected by physical
boundaries provided by Member States,
(2) Ensuring adequate protection of the data stored on transportable media.
In addition, if the security of laptop computers used by inspection and

technical staff is not addressed elsewhere, it will be important to ensure that
any Member State information stored or viewed on these devices is adequately
protected.
The confidentiality protection must be strong enough to withstand
external attacks. Since the owner of the information being protected is in the
best position to judge the value of their data, provision will need to be made to
satisfy the protection strength specified by Member States. In the absence of
specification by a Member State, the IAEA’s security architecture specifies a
minimum strength level suitable for general commercial practice and
equivalent to the ‘Safeguards Confidential’ specified by the IAEA.
74
8. ENVIRONMENTAL SAMPLING
Environmental sampling was introduced in 1996 as one of the new IAEA

safeguards measures which contribute to the confirmation of the absence of
undeclared nuclear material or nuclear activities. Collection of environmental
samples at or near a nuclear site combined with ultrasensitive analytical
techniques such as mass spectrometry, particle analysis and low level
radiometric techniques, can reveal signatures of past and current activities in
locations where nuclear material is handled. Initial implementation of
environmental sampling for safeguards is focused on the collection of swipe
samples inside enrichment plants and installations with hot cells.
Implementation is being extended to other types of nuclear facilities and
samples collected in connection with complementary access inspection
activities under the Additional Protocol.
Samples are analysed in either bulk or particle mode, depending on the
sampling objectives and the activity levels of the swipes. Bulk analysis involves
the analysis of an entire sample, usually by g ray spectrometry or isotope
dilution thermal ionization mass spectrometry; the analytical measurements
represent average results for the material contained. Particle analysis relies on
the detection and analysis of individual particles in the micrometre size range
and provides as results the U and Pu content as well as isotope ratios of U
and/or Pu in these particles.
8.1. IAEA CLEAN LABORATORY FOR SAFEGUARDS
The IAEA Clean Laboratory for Safeguards (Fig. 41) was inaugurated in
December 1995 with the goal of providing a Class 100 clean-room capability for
the provision and certification of sampling kits and for the receipt, screening
and distribution of environmental samples from safeguards inspections. This
facility significantly reduces the risk of cross-contamination that might lead to
incorrect safeguards conclusions. The Clean Laboratory consists of over 200 m2
of laboratory space, with approximately 50 m2 at the ISO Class 5 cleanliness
level (Fig. 42). The laboratory is equipped with a suite of analytical techniques,
including a, b, g and C ray fluorescence spectrometry, scanning electron
microscopy with electron probe analysis and high sensitivity thermal ionization
mass spectrometry.
Environmental swipe samples received at the Clean Laboratory are given
a code number to maintain confidentiality about their origin. The samples are
then measured by low background g ray spectrometry to detect the presence of
actinide elements (primarily U and Pu) and fission or activation products (such
75
FIG. 41. The IAEA Clean Laboratory for Safeguards.
as 60Co, 137Cs and 106Ru); the samples are then measured by C ray fluorescence
spectrometry to detect the presence of U, Pu or other important elements.
Alpha/beta counting is then applied to radioactive samples to detect actinides
or b emitting isotopes such as 3H, 90Sr or 99Tcm.
Following the screening measurements, subsamples are distributed to

laboratories of the NWAL for more detailed analysis. Selected samples are
chosen for measurement in the Clean Laboratory by isotope dilution thermal
ionization mass spectrometry, using a highly sensitive instrument equipped
FIG. 42. Analyst working in the clean module of the IAEA Clean Laboratory.
76
ENVIRONMENTAL SAMPLING
with pulse counting detection. The ultimate sensitivity of this method is in the
10-15 g range for U and Pu.
One of the main activities of the Clean Laboratory is the preparation of
clean sampling kits for collecting environmental samples. A kit for the
collection of swipe samples is shown in Fig. 43. This consists of all the supplies
needed by an IAEA inspector in the field: clean swipe cloths, plastic minigrip
bags, clean-room gloves, a sample data form, a pen and labels. A roll of
aluminium foil is provided to establish a clean working surface. A different type
of swipe sampling kit is required for sampling inside hot cells, where the
subsamples must be taken with remote manipulators and shipped back to the
IAEA in a special lead lined container because of their higher radiation level.
8.2. SCREENING OF SAMPLES
8.2.1. Low level g ray spectrometry
Immediately after receipt, environmental samples are measured with a

low background g ray spectrometer system. The spectrometer is based on a
90% efficient coaxial Ge detector enclosed in a high purity lead shield of 10 cm
thickness. The samples, in special beakers, are placed in a 15 position sample
changer and counted for 1 h each to provide a g ray spectrum in the energy
range from 5 keV to 3 MeV. The total g activity, corrected for background, is
obtained by this method and if sufficient activity is detected, an evaluation of
the spectral peaks can be performed to estimate the activity in the sample of
FIG. 43. Cotton swipe kit for environmental sampling.
77
individual g emitting isotopes such as 60Co, 95Zr, 106Ru, 134Cs, 137Cs and 241Am.
Depending on the number of counts collected, the precision and accuracy of
these measurements are in the range 2–5%rel. The absolute activity of
individual radioisotopes is not as important as the relative activity compared
with a selected isotope such as 137Cs.
8.2.2. C ray fluorescence spectrometry
X ray fluorescence spectrometry is used to detect microgram amounts of

U, Pu or other elements of interest on the surface of swipe samples. The sample
is held by a robot arm and irradiated by X rays from an X ray tube, resulting in
the emission of fluorescent X rays from elements present on the swipe. The
fluorescent X rays are detected using a 100 mm2 Si(Li) detector placed near the
sample. Counting is performed for 4–5 hours and the spectra are then evaluated
to determine the amount of the element present as well as its spatial
distribution. The screening method is completely non-invasive because the
subsample can be measured inside its plastic bagging.
8.2.3. Alpha/beta counting
A gridded ionization chamber counting system is used to screen

radioactive swipe samples for the presence of a or b emitting isotopes. The
swipes are subsampled with an adhesive carbon disc, which is placed in the
counting chamber and measured for one hour.
This system has high collection efficiency and has a sensitivity in the
millibecquerel range. Alpha emitting nuclides such as 210Po and b emitters such
as 3H, 90Sr and 99Tcm can be measured much more sensitively in this way than
by g or X ray methods.
8.3. ISOTOPIC AND ELEMENTAL ANALYSIS
8.3.1. Pulse counting thermal ionization mass spectrometry
Screening measurements are used to decide which samples should be sent

for more detailed analysis. Thermal ionization mass spectrometry is used to
measure U and Pu concentrations and isotopic compositions in environmental
samples. The basic technique was described in Section 3. However, for
measurements of environmental samples a much higher sensitivity is needed,
extending into the l0-9 and 10-12 g ranges. This is achieved by the use of special
78
sample treatment procedures, drop deposition of the sample elements onto the
mass spectrometer filament and use of a pulse counting detection system with
high detection efficiency. The mass spectrometer is shown in Fig. 44.
Isotopic spikes (233U, 242Pu or 244Pu) are added to the samples during
chemical processing to allow U or Pu concentrations to be determined using
the isotopic dilution method. Isotope ratios are measured for all isotopes of U
or Pu relative to the spike isotope and the isotopic composition of the sample
is estimated by subtraction of the known isotopic composition of the spike. The
accuracy and precision of this technique are about 1–10% for a U or Pu
concentration in the 10-9 g range and for the ratios of the major isotopes in the
sample.
8.3.2. Scanning electron microscopy with electron probe analysis
The Clean Laboratory is equipped with a scanning electron microscope

with wavelength and energy dispersive X ray fluorescence detectors (Fig. 45).
Particles of interest are removed from the sample using adhesive carbon discs,
which are introduced into the electron microscope. Under high magnification
(500—5000×) the particles are examined and the backscattered electron signal
is used to search for particles containing heavy elements. These particles can
then be measured by energy dispersive X ray fluorescence spectrometry to give
a semiquantitative elemental analysis. Particles containing U or Pu can be
identified in this way; their size and morphology, as well as other elements
present, will give information about the process that created them. This type of
FIG. 44. Mass spectrometer for environmental isotopic analysis.
79
FIG. 45. Scanning electron microscope for particle analysis.
analysis forms a part of the classical ‘particle analysis’ approach which is

applied in certain NWAL laboratories in conjunction with thermal ionization
mass spectrometry.
8.3.3. Fission track method
Traditional particle analysis first involves a step in which particles of

interest containing fissile isotopes such as 235U or 239Pu are selected by the
fission track method. The particles are then mounted onto the filament of a
thermal ionization mass spectrometer for measurement of the isotopic
composition of the U and Pu present.
The fission track method involves removal of particles from the
environmental sample by ashing (for vegetation or swipe samples) or physical
removal by ultrasoneration in an inert solvent. The particles are then spread
onto a plastic track etch film (e.g. Lexan) in a layer of collodion
(nitrocellulose). The film is then irradiated in a reactor with thermal neutrons
at a total dose of 1014 neutrons. Particles containing fissile isotopes leave
damage tracks in the film, which can be etched to make them visible under a
light microscope (Fig. 46). An experienced analyst can compare the size and
appearance of the particles with the number of fission tracks to decide which
particles should be measured further. The analyst can then pick up each particle
of interest and mount it directly onto a filament for thermal ionization mass
spectrometry.
For thermal ionization mass spectrometry, the particle is held in a

rhenium metal filament and heated in the ion source of the mass spectrometer
80
Particle Size (mm) Tracks %U-235 Compound
A 1.2 100 3.0 UO2

B 1.0 19 0.5 UO2
C 1.5 40 0.5 UO2
D 0.7 8 0.5 UO2
E 1.5 600 91.8% 239Pu PuO2
FIG. 46. Lexan film showing fission tracks.
at 1500–1800°C to produce ions of U or Pu, which are counted by a pulse

counting detection system. The mass spectrometer steps between the isotopes
of U or Pu to accumulate a mass spectrum. The abundance of the various
isotopes can be estimated from the collected ion counts with a precision and
accuracy of better than l%rel. for isotopes of 1–90% abundance in particles
with a diameter of 1–5 mm. Particles with diameters down to 0.1 mm can be
measured, but with less precision and accuracy.
81
8.3.4. Secondary ion mass spectrometry
Another technique used in SAL and certain network laboratories for

measuring the isotopic composition of micrometre size environmental particles
is secondary ion mass spectrometry (Fig. 47). The particles are deposited on a
conducting substrate and placed in the vacuum system of the instrument where
they are bombarded with energetic ions of oxygen. The ion bombardment
results in sputtering of the sample and the ejection of secondary ions which are
representative of the particle under examination. The secondary ions are
accelerated and mass analysed by the spectrometer and counted with either an
imaging or pulse counting ion detector. In the ion microscope mode of
operation, an image is generated using secondary ions of a given mass
(e.g. 235U+). Another image can then be taken using a different secondary ion
signal (such as 238U+) and the two images merged to obtain the 235U/238U ratio
for each particle in the field of view (typically 150 mm in diameter). By scanning
100–200 fields in one session, it is possible to interrogate several thousand
particles by this method, thus giving a distribution of the 235U enrichments
found in the particles from a sample.
Once an interesting particle has been identified in the ion microscope mode, it
can be measured to completion by focusing the primary ion beam on it and
stepping between the isotopes of interest. This will yield the complete isotopic
composition of the particle, including the minor isotopes such as 234U and 236U.
Depending on the size of the particle, the precision and accuracy of this
approach can be l%rel. for isotopes at the 1–90% abundance level and up to
10%rel. for minor isotopes.
FIG. 47. Secondary ion mass spectrometry (SIMS).
82

Safeguards Techniques and Equipment

Uploaded by

Copyright:

Available Formats

Safeguards Techniques and Equipment

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Safeguards Techniques and Equipment

Uploaded by

Copyright:

Available Formats

Safeguards Techniques

INTERNATIONAL ATOMIC ENERGY AGENCY

The 1990s saw significant non-proliferation related developments in the

2.1. Gamma ray spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3.1. Elemental analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

4. CONTAINMENT AND SURVEILLANCE . . . . . . . . . . . . . . . . . . . 36

6. REMOTE MONITORING SYSTEMS . . . . . . . . . . . . . . . . . . . . . . . . 61

6.1. Remote monitoring equipment . . . . . . . . . . . . . . . . . . . . . . . . . . 61

7.1. Information protection requirements . . . . . . . . . . . . . . . . . . . . . 65

8.1. IAEA Clean Laboratory for Safeguards . . . . . . . . . . . . . . . . . . . 75

The IAEA has the task of providing continuing assurance to the

Enrichment Fuel fabrication Power reactors and Spent fuel

Number of facilities safeguarded in 2002

10 41 239 power reactor units 6

IAEA summary statistics for 2002 (approximate numbers)

equipment performance are evaluated, including compliance with

2.1. GAMMA RAY SPECTROMETRY

2.1.1. Gamma emission and detection of nuclear materials

Most nuclear materials of concern in IAEA safeguards emit g rays that

of characteristic g ray energies. Plutonium spectra can be analysed to

Figure 1 illustrates the capabilities of various types of detector with low,

2.1.2. Hand-held multipurpose gamma spectrometry

HM-5 (fieldSPEC). This device is the successor to the Hand-held Assay

NAID U235 Spectrum

CDZT U235 Spectrum

GDET U235 Spectrum

FIG. 1. Comparison of g ray spectrometric performance of various types of detector (low,

TABLE II. GAMMA RAY SPECTROMETERS

Code Equipment name Primary applications

IMCN, I-2000 Multichannel Analyser Verification of U enrichment, spent

MMCN, Miniature Multichannel Analyser Verification of U enrichment and

FIG. 2. The HM-5 fieldSPEC.

2.1.3. Multichannel analysers

IMCA (InSpector2000). The new InSpector 2000 (I-2000) multichannel

FIG. 3. IMCA shown with an NaI detector and portable computer.

FIG. 4. MMCC: Miniature Multichannel Analyser (MMCA) with CdZnTe detector.

2.1.4. IAEA high resolution gamma spectrometry techniques

When coupled to a Ge detector, the MMCA or IMCA becomes a high

FIG. 5. IMCG: Inspector Multichannel Analyser with HpGe detector.

enrichment. The cylinder wall thickness must also be determined so that

2.2. NEUTRON COUNTING

The IAEA uses a number of different types of neutron counting

2.2.1. Neutron emission and detection for non-irradiated fissile fuel

Neutrons are primarily emitted from non-irradiated nuclear fuel in three

(1) Spontaneous fission of uranium and plutonium, particularly involving the

TABLE III. COINCIDENT NEUTRON DETECTOR SYSTEMS FOR

Code Equipment name Primary applications

Passive Neutron Coincidence Counters

TABLE III. (cont.)

Code Equipment name Primary applications

Active Neutron Coincidence Counters

insignificant contribution to the spontaneous fission neutron signal. Isotopic

2.2.2. Gross neutron counting

2.2.3. Neutron coincidence counting

FIG. 6. HLNC: High Level Neutron Coincidence Counter.

FIG. 7. INVS: Inventory Sample Counter.

FIG. 8. AWCC: Active Well Coincidence Counter.

2.3. SPENT FUEL MEASUREMENT

2.3.1. Neutron emission and detection