imernQfiono1 Agency

General

Dppartment of Safeguards q F fi- STK 150

D e t a i l e d D e s c r i p t i o n of an
ôoAi" at t h e F a c i l i t y I.I'VIM for a

L o w K n r i r h e d U r a n i u m C o n v e r s i o n a n d Fuel F a b r i c a t i o n Facility

Kal ph J . JotiiT
Gyst'Tn Ctudier. S e c t i o n
D i v i s i o n uf n e v e 1opment a n d T e c h n i c a l S u p p o r t
Iinpart m<Mit of S a f e g u a r d s
I n t e r n a t i o n a ] A t o m i c fJnergy A g e n c y

E u g e n e V. W e ins took
W a l t e r K. K a ne
Technical Support Organization
Bruokhavon National Laboratory
Upt o n , N Y . , U S A

Gept o m b e r
 This safeguards technical report has been produced as a joint IAEA/US
research and development effort under the U.S. Support Programme in the
Systems Study Section of the Department of Safeguards. It is based on all
information and expertise available including that from consultants from a
number of Member States.

Neither the subject matter nor the content of the report reflects any
policy, expressed or implied, by the IAEA, any of the Member States, or the
U.S. Government.
 Foreword

During the course of the preparation of this report many have questioned
the purpose of the report and its role in international safeguards in relation
to existing documents such as INFCIRC/153(corrected), INFCIRC/66/Rev.2 and
IAEA/SG/INF/2. The first two documents provide requirements for IAEA
safeguards systems while the latter provides the basic structure for State
Systems of Accounting for and Control of Nuclear Material (SSAC). This report
is intended to provide the technical details of an effective SSAC which Member
States may use, if they wish, to establish and maintain their SSACs. It is
expected that systems designed along the lines described would be effective in
meeting the objectives of both national and international systems for nuclear
material accounting and control. There is no intention that this report would
add to, subtract from, or amend in any way the rights and obligations defined
for the IAEA and the States in documents INFCIRC/66/Rev. 2, INFCIRC/153
(corrected) or any specific safeguards agreement concluded with the IAEA
pursuant to these documents.
 TABLE OF CONTENTS

Part 1. Introduction 1
1 .1 Genera 1 1
1.2 Principal Elements of an SSAC 1
1.3 Purpose and Scope of this Document 2
1.4 The Mode 1 Plant 3
References 4

Part 2 . Nue lear Material Measurements 5

2 .1 Introduction 5
2.2 Material Balance Areas 5
2.3 Key Measurement Points 12
2.4 Materials and Material Types 13
2.5 Measurement Methods 15
2.5.1 Bulk Measurements 18
2.5.2 Sampling , 21
2.5.3 Analytical Measurements 23
2.5.3.1 Elemental Analysis 24
2.5.3.1.1 Feed, Product, and Intermediate Materials. 24
2.5.3.1.2 Scrap Materials 25
2.5.3.1.3 Waste Materials 25
2.5.3.2 Isotopic Analysis 26
2.5.3.3 Nondestructive Assay 27
2 .6 Documentation...... 30
2.6.1 Measurement Procedures Manual.... 30
2.6.2 Measurement Quality Procedures Manual 30
2.6.3 Measurement Logs 31
2.6.4 Quality-Control Logs 31
2.6.5 Material Records 31
2.7 Example 33
References 39

Part 3. Measurement Quality 42

3.1 Introduction 42
3.2 System Qualification 43
3.3 Standards and Calibrations 47
3.3.1 Bulk Measurement 48
3 .3.1.1 Weight 48
3.3.1.2 Volume 49
3 .3 .2 Analytical Measurements 50
3.3.3 Nondestructive Analysis (NDA) 53
3.4 Sampling 55
3.5 Control Programme 57
3.5.1 Measurements Monitoring 57
3 .5 .2 Control Data Evaluation 58
3.5.3 Documentation 60
3.6 Administration 60
3.7 Examples 61
3.7.1 Introduction 61
3 .7.2 Technical Requirements 61
 TABLE OF CONTENTS (CONT'D)

Page

3.7.2.1 Reference Standards 61

3.7.2.2 Standard Replication Program 62
3.7.2.3 Calibration Standards 62
3.7.3 Mass Measurements 62
3.7.3.1 Reference Standards 62
3.7.3.2 Measurement of Mass Standards 62
3.7.3.3 Mass Calibration...., 63
3.7.4 Analytical Measurements 63
3.7.4.1 Reference Standards 63
3.7.4.2 Measurement of Analytical Standards 63
3.7.5 NDA Measurements - 66
3.7.5.1 Calibration Standards 66
3.7.6 Samp ling 68
3.7.6.1 Homogeneous Materials 68
3.7.6.2 Heterogeneous Materials 68
3.7.7 Special Experiments 68
3.7.7.1 Heterogeneous Scrap 68
3.7.7.2 Solid Waste 69
3.7.8 Statistics.. 69
3.7.8.1 General 69
3.7.8.2 Statistical Applications 71
3.7.8.3 Error Estimation 71
3.7.8.3.1 Random Errors 71
3.7.8.3.2 Systematic Errors 71
3.7.8.3.3 Application to Overall Program... 72
3.7.9 Documentation 73
References 75

Part 4. Records and Reports 79

4.1 Introduction 79
4.2 Accounting Records 79
4.2.1 Ledgers 80
4.2.2 Inventory Change Journals 80
4.2.3 Supporting Documents SI
4.3 Operating Records 86
4.3.1 Nuclear Material Measurements 89
4.3.2 Measurement Quality 89
4.3.3 Physical Inventory Taking 89
4.3.4 Losses 90
4.4 Accounting Reports 90
4.4.1 Inventory Change Reports 90
4.4.2 Material Balance Reports 91
4.4.3 Inventory Report 97
4.5 Data Handling 97
4.5.1 MBA and Facility Records 97
4.5.2 Source Data and Operating Records 103
4.5.3 Data Flow 103
4.5.4 Shipper-Receiver Differences 107
 TABLE OF CONTENTS (CONT'D)

Page

4.5.5 Batch Fol low-Up 107

4.5.6 Internal Control 108
4.6 Examples of Records and Reports 108
4.6.1 Introduction 108
4 . 6 . 2 Inventory and Material Flow Assumptions 110
4 . 6 . 3 Example Transaction Recording 110
4 . 6 . 3 . 1 Off-Site Receipts Ill
4 . 6 . 3 . 2 Transfers from MBA-1 to MBA-2 112
4 . 6 . 3 . 3 Transfers from MBA-2 113
4.6.3.4 Shipments and Transfers from MBA-3 114
4 . 6 . 4 Example Inventory Change J o u r n a l s
and General Ledgers 115
4.6.5 Example Accounting Reports 115
4.6.6 Batch Follow-up 116
References 117

Part 5. Physical Inventory Taking 137

5 .1 Introduction 137
5 .2 Inventory Me a sûrement 137
5.2.1 Precision and Accuracy 137
5.2.2 Inventory Measurement Quality 138
5.2.3 Factors, Nominal Values, and Calculated Values 138
5.2.4 Validity of Prior Measurement 139
5.2.5 In-Process Measurement 139
5.2.5.1 Process Draindown and Cleanout 139
5.2.5.2 Dynamic Inventory 140
5.3 Inventory Organization and Planning 141
5.3.1 Assignment of Responsibilities 141
5.3.1.1 Facility Inventory Supervisor 141
5.3.1.2 Area Inventory Supervisor 141
5.3.1.3 Inventory Teams 142
5.3.2 Cutoff Procedures 142
5.3.2.1 Receipts, Shipments, and Other Removals 142
5.3.2.2 Internal Transfers 143
5.3.2.3 Process 143
5.3.2.4 Records 144
5.3.3 Facility Inventory Listing and Stratification 144
5.3.4 General Inventory Instructions 146
5.3.5 Area Inventory Instructions 147
5.3.6 Preliminary Inspection and Review 148
5.3.6.1 Process Conditions.... 148
5.3.6.2 Area Preparations 148
5.3.6.3 Records 148
5.3.6.4 Measurements 148
5.3.6.5 Inventory Teams 148
5.4 Conduct of Inventory 149
5.5 Post-Inventory Activities 149
5.5.1 List and Tag Accuracy Check 149
5.5.2 Cutoff Verification.'. 149
 TABLE OF CONTENTS (CONT'D)

Page

5.5.3 Inventory Summary and Reconciliation 150

5.6 Example of Physical Inventory Taking Procedures,
Instructions and Data Handling 151
5.6.1 Introduction 151
5.6.2 General Inventory Procedures 151
5.6.3 Responsibility and Authority 153
5.6.3.1 Facility Inventory Supervisor 153
5.6.3.2 Area Inventory Supervisor 154
5.6.3.3 Inventory Teams 155
5.6.4 Inventory Instructions 155
5.6.4.1 General Schedule and Cutoffs 155
5.6.4.1.1 Receipts, Shipment and
Other Removals 155
5.6.4.1.2 Internal Transfers... 155
5.6.4.1.3 Process 156
5.6.4.1.4 Records 156
5.6.4.2 MBA-1 Shipper Receiver Area 156
5.6.4.3 MBA-2 Process Area 156
5.6.4.4 MBA-3 Assembly Area 157
5.6.4.5 Physical Inventory List and Stratification.. 158
5.6.4.6 Inventory Team Instructions 158
5.6.5 Post-Inventory Activities 160
5.6.5.1 List and Tag Check 161
5.6.5.2 Cutoff Verification 161
5.6.5.3 Inventory Summary and Reconciliation 161
References 163

Part 6. Material Balance Closing 173

6.1 Introduction 173
6.2 The Material Balance Equation 173
6.3 Evaluation of MUF 174
6.4 Other Parameters Affecting MUF 176
6.5 Example of Material Balance Closing and MOF Evaluation 177
6.5.1 Introduction 177
6.5.2 Facility Material Balance 178
6.5.3 Facility MOF Evaluation 178
References 193

Appendix A - Reference Facility A-l

A.I General Characteristics A-l
A.2 Process Description A-l
A.2.1 Conversion Process A-l
A.2.2 Fabrication and Assembly Process A-l
A.2.3 Solid Waste Treatment Process A-2
A.2.4 Scrap Recycle Process A-2
A.3 Nuclear Material Flow and Key Measurement Points A-2
 List of Tables

Page

2.1 Typical Inventory of Model Plant 14

2.2 Effect of Measurement Errors for Different Categories of

Material on Overall Uncertainty in Material Balance 16

2.3 Capabilities .of Selected Weighing Devices 20

2.4 Example of Measurement Report 32

2.5 Example of a Measurement Uncertainty Report 34

2.6 Example of Item Label 34

2.7 Typical Operator's Measurement Methods and Accuracies at LEU

Conversion and Fuel Fabrication Plant 36-38
3.1 Target Values for Standard Deviation {% Relative) of
Analytical Measurements of Element and Concentration
in Fuel Materials in LEU Fabrication Facility 44

3.2 Target Values for Standard Deviation (X Relative) of

U-235 Determinations in LEU Fabrication Facility 44

3.3 Standard Reference Materials Available for the Determination

of Uranium or of Impurities in Uranium 52

3.4 Example of Mass Standards „. 64

3.5 Analytical Standards 65

3.6 Waste-Assay Calibration Standards 67

3.7 Scope of S t a t i s t i c a l Programs. 70

3.8 Records and Reports 74

5.1 Example Reference Facility Inventory 152

5 .2 Phys ica 1 Inventory Summary 162

6.1 Facility Record Data Summary 179

6.2 Measurement Error Da ta - MBA-2 183

6.3 MBA-2, Material Balance Interval Strata 184

6.4 S t a t i s t i c a l Notation from IAEA-TECDOC-261 186

6.5 Data Tabulation - MBA-2 , 187

 List of Tables (Cont'd)

Page

A-l Description of Flow Key Measurement Points Shown on Figure A.I.. A—4—A-6

A-2 Description of Inventory Key Measurement Points Shown on

Figure A. 1 A-7—A-8

A-3 Inventory Material Description A-9—A—10

 List of Figures

Page

2.1 Example of Operator's Material Control Areas 8

2.2 Schematic Representation of Conversion - Fabrication Plant with

tvo Parallel Lines 9

4.1 Inventory Change Journa1. , 82

4.2 Inventory Change Journal - Receipts 83

4.3 Inventory Change Journal - Removals 84

4.4 Inventory Change Document - Material Transfer 87

4.5 Inventory Change Document - Journal Entry Form 88

4.6 Inventory Change Report (ICE.) Form R.01/c 93

4.7 Inventory Change Report (ICR) Form R.01.2 95

4.8 Material Balance Report (MBR) Form R.03 99

4.9 Physical Inventory Listing (PIL) Form R.02/c 191

4.10 Data Flow and Document Relationships 104

4.11 General Ledger 106

4. i 2 Genera 1 Ledger 119

4.13 General Ledger 120

4.14 General Ledger 121

4.15 Inventory Change Document - Material Transfer 122

4.16 Inventory Change Journal - Receipts 123

1.17 Inventory Change Document - Material Transfer 124

4.18 Inventory Change Document - Material Transfer 125

4.19 Inventory Change Journal - Removals 126

4.20 Inventory Change Journal - Receipts 127

4.21 Inventory Change Journal - Removals 128

4.22 Inventory Change Document - Material Transfer 129
 List of Figures (Cont'd)

Page

4.23 Inventory Change Journal - Receipts 130

4.24 Inventory Change Document - Journal Entry Form 131

4.25 Inventory Change Document - Material Transfer 132

4.26 Inventory Change Journal - Removals 133

4.27 Inventory Change Document - Material Transfer 134

4.28 Inventory Change Report (ICR) Form R.01.1/c 135

5.1 Physical Inventory I tarn List , 165

5.2 Physical Inventory Item List 166

5.3 Physical Inventory Item List 167

5.4 Physical Inventory Item List 168

5.5 Physical Inventory Item List 169

5.6 Inventory Tag Reconciliation 170

5.7 Physical Inventory Listing (PIL) Form R.02/c 171

6.1 Material Balance Report (MBR) Form R.03 181

A.I Process Scheme of LEU Conversion and Fuel Fabrication A-3

 Part 1. INTRODUCTION

1.1 General

There are two primary objectives toward which the development and
maintenance of a state system of accounting for and control of nuclear
material (SSAC) can be directed. There is a national objective to provide for
detection of nuclear material losses or unauthorized use or removal of nuclear
material in the interest of public health and safety as well as for economic
interests. There is also an international objective to provide the basis for
the application of IAEA safeguards pursuant to the provisions of agreements
between the State and the IAEA. While systems designed specifically for one
or the other of these objectives may differ in some details there are many
technical elements which are the same. To assist Member States with regard to
the system directed toward the international objective the IAEA issued
IAEA/SG/INF/2, "Guidelines for States' Systems of Accounting For and Control
of Nuclear Materials". The guidelines describe the legal structure and
organizational and functional elements of an SSAC at both the State and the
facility level "for the organization and functions of the SSAC with respect to
obligations arising from Safeguards Agreements concluded by a State with the
IAEA". Although the guidelines provide a basic structure for an SSAC, some
States have expressed a need for more detailed guidance with regard to the
technical elements in the design and operation of SSACs for both the national
and the international objectives. To meet this need the present document has
been prepared, describing the technical elements of an SSAC in considerable
detail.

It is intended that this document will not only assist States in

establishing their national systems, but contribute to their understanding of
the IAEA's needs in carrying out its obligations, and thereby help make
international safeguards more effective, efficient, and less burdensome to the
State and operator. It is also hoped that the document will be a useful
training tool for those responsible for the operation of an SSAC at either the
state or the facility level.

It is emphasized that the descriptions in this document are of possible

SSAC's, and are not meant to be prescriptive in nature. There are many ways
to design an effective SSAC which will meet national objectives for nuclear
material accounting and control and at the same time will enhance the IAEA's
ability to achieve its objectives. In fact, the SSACs of a number of states
show considerable variation, while still being satisfactory from the IAEA's
point of view. Consequently, the material presented here should be regarded
as illustrative and representative of good practice. That is to say, on the
basis of its experience the IAEA is able to say that SSACs designed and
operated at the facility level along the lines described here are capable of
meeting national system objectives as well as fulfilling the undertakings of
the State with respect to international safeguards, and of enhancing the
ability of the IAEA to carry out its responsibilities.

1.2 Principal Elements of an SSAC

The guidelines presented in IAEA/SG/INF/2 provide a basic structure for an

SSAC but do not address the details of the specific elements of SSACs. The
technical and functional elements presented in the structure are inter-
 -* 2 —

related, in that the same subjects are addressed in different ways, depending
on the application. For example, material measurements are considered with
respect to establishing MBAs, with respect to material flows, and with respect
to physical inventory taking. The guidelines in IAEA/SG/INF/2 address seven
categories of information which can be defined and described so that they
include all of the elements of an SSAC. By considering thèse seven categories
of information as principal elements of an SSAC, a detailed description for
each of the seven can be prepared to cover all the subject matter and to show
the relationships of the SSAC functional elements and requirements identified
in IAEA/SG/INF/2. The seven principal elements are:
(1) Nuclear Material Measurements
(2) Measurement Quality
(3) Records and Reports
(4) Physical Inventory Taking
(5) Material Balance Closing
(6) Containment and Surveillance
(7) Organization and Management

Two levels of an SSAC may be defined: the State Authority level and the
facility level. While the functional separation between these two levels may
vary from State to State, in general the Authority level establishes the
requirements and criteria for the SSAC and the facility operator implements
the system in compliance with Authority requirements and criteria. The
guidelines presented in IAEA/SG/INF/2 addressed both levels of an SSAC. The
detailed descriptions of the principal elements presented in this document
address only the facility level of an SSAC with the Authority level being
addressed only to the extent necessary to show relationships between the two
levels.

1.3 Purpose and Scope of this Document

The purpose of this document is to provide a detailed description of a

system for the accounting for and control of nuclear material in a model low
enriched uranium conversion and fuel fabrication facility which can be used by
a facility operator to establish his own system in a way which will provide
the necessary information for compliance with a national system for nuclear
material accounting and control and for the IAEA to carry out its safeguards
responsibilities. A discussion of the IAEA safeguards approach for low
enriched uranium conversion and fuel fabrication facilities is presented in
IAEA STR-157.

The scope of this document is limited to descriptions of the first five of

the seven principal SSAC elements identified in 1.2 above. The last two
elements, Containment and Surveillance and Organization and Management have
not been included because it was thought more useful to concentrate on the
nuclear material accounting aspects of the SSACs. The other two elements may
be treated at a later time. The descriptions will be at the facility level,
with specific examples for each of the five principal elements for a facility
like the model. The effect of the size of the facility, when it differs from
that of the model, will be considered as appropriate. Examples will be
presented for each of the principal elements of an SSAC so that, by the
conclusion of the document, a detailed description of an SSAC for the model
plant will have been provided. Some of these examples will be based on
discussions in the published safeguards literature cited in the list of
references at the end of each Part. These discussions will not be repeated in
detail here. To understand the erimples more fully the reader is urged to
 - 3 -

consult the references, especially the more comprehensive ones such as the
IAEA Safeguards Technical Manual. Part E, Methods and Techniques (IAEA-174),
and Part F, Statistical Concepts and Techniques, Vol. 3 (IAEA-TECDOC-261),
Statistical Methods in Nuclear Material Control (TID-26298), by J.L. Jaech,
and Selected Measurement Methods for Plutonium and Uranium in the Nuclear Fuel
Cycle (TID-7029), by C.J. Rodden.

Some discussion is also devoted to the IAEA's philosophy and practices in

the belief that a clearer understanding of its needs will foster that spirit
of cooperation so essential to the application of effective and efficient
safeguards with a minimum of intrusion.

The descriptions in this document are not intended to add to, subtract
from, or amend in any way the rights and obligations defined for the IAEA and
the States in documents INFCIRC/66/Rev.2, INFCIRC/153 (Corrected), or any
specific Safeguards Agreements concluded with the IAEA, but only to provide
information which States may find useful in the exercise of their rights and
the discharge of their obligations under their national requirements and under
their agreements with the IAEA.

1.4 The Reference Facility

It is useful to base the example in this guide upon a reference facility.

The one chosen for that purpose is described in detail in Appendix A.
Briefly, the facility has an annual throughput of up to 300 tonnes of UO2
with a nominal enrichment of 3.0 weight percent of U-235. Physical
inventories, which include clean-out of process equipment, are ',aken twice per
year. The feed materials are low enriched UFg, uranyl nitrate solution
(UNH) received from off site, and UO2 powder received from off site and from
scrap recovery. Products are light water reactor (LWR) assemblies of both the
PWR and BWR type.
References for Part 1;

IAEA/SG/INF/1: IAEA Safeguards Glossary, IAEA, Vienna, 1980.

IAEA/SG/INF/2: Guidelines for States Systems of Accounting for and Control of

Nuclear Material, IAEA, Vienna, 1980.

INFCIRC/66/Rev. 2: The Agency's Safeguards System, IAEA, Vienna, 1968.

INFCIRC/153 (Corrected): The Structure And Content of Agreements Between The

Agency And States Required In Connection With The Treaty On The Non-
Proliferation Of Nuclear Weapons, IAEA, Vienna, 1971.

IAEA-TECDOC-261: IAEA Safeguards Technical Manual, Part F: Statistical Con-

cepts and Techniques, Vol. 3, IAEA, Vienna, 1982.

IAEA-174: IAEA Safeguards Technical Manual, Part E: Methods and Techniques,

IAEA, Vienna, 1974.

Jaech 1973: J.L. Jaech, Statistical Methods in Nuclear Material Control, TID-
26298, Superinterident of Documents, U.S. Government Printing Office,
Washington, D.C. 20402, 1973.

Rodden 1972: C.J. Rodden, Selected Measurement Methods for Plutonium and Ura-
nium in the Nuclear Fuel Cycle, TID-7029 (2nd Ed.), 1972, National Techni-
cal Information Service, U.S. Department of Commerce, Springfield, VA.
22151.

STR-157: Safeguarding Low Enriched Uranium Conversion and Fuel Fabrication

Facilities, System Studies Section, Division of Development and Technical
Support, IAEA Department of Safeguards, STR-157, 1984.
 Part 2. NUCLEAR MATERIAL MEASUREMENTS

2.1 Introduction

A basic requirement for any nuclear material accounting system is the estab-
lishment and maintenance of a measurement system for the determination of the
quantities of nuclear material received, produced, shipped, or otherwise removed
from inventory, and the quantities on inventory. Measurement systems for nu-
clear materials in bulk processing plants are also required by the domestic
authorities in many states for the protection of the public security, health,
and safety. The operator will also be impelled to measure nuclear materials
involved in his process in order to maintain the quality of his product, to
enhance operational safety and -efficiency, and for purposes of financial
accountability.

According to IAEA/SG/INF/1 the nuclear material measurement system should

provide for:

Identification of the key measurement points in the process and the

characteristics of the nuclear material to be measured

Specification of measurement performance desired

Specification of the measurement technique(s) to be employed

Specification for measurement equipment

Procedures for maintenance of equipment

Operator's qualifications and provision for training

Calibration standards and procedures

Routine measurement and data analysis procedures

Procedures to control the quality of measurements and to maintain

performance at the desired level.

Key measurement points are defined and discussed in Sec. 2.3. They
are most easily understood in the context of the material balance area (MBA),
discussed below.

2.2 Material Balance Areas

IAEA material balance verification is made on the basis of material balance

areas (MBAs), which are defined in IAEA/SG/INF/1 as "an area in or outside of a
facility such that:

(a) The quantity of nuclear matrial in each transfer into or out of each
'material balance area' can be determined; and
 (b) The physical inventory of nuclear material in each 'material balance
area' can be determined when necessary, in accordance with specified
procedures,

in order that the material balance for IAEA safeguards purposes can be
established". MBAs are chosen through negotiation between the IAEA and the
State, and are defined in the Safeguards Agreement Subsidiary Arrangements. It
may be helpful to consider here some of the factors that go into the choice.

The IAEA uses several criteria to establish MBAs, including accuracy

considerations, the ability to apply containment and surveillance measures so as
to help ensure the completeness of flow measurements, the desire of the State to
protect certain commercially sensitive information, and the nature of the
activities and form of the nuclear materials within the MBA.

(î'or facilities coming under INFCIRC/66/Rev. 2 agreements, instead of MBAs

there are accounting areas with accounting sub-areas; also, only materials on in-
ventory (i.e., not flows) are measured].

Recognizing that for domestic control purposes the facility operator may
choose to establish MBAs different from IAEA MBAs, this section will also ad-
dress those factors that should be considered by the facility operator so that
the facility MBAs can at least be compatible with IAEA MBAs, if not the same.
Having the same or compatible MBAs for both facility and IAEA purposes should fa-
cilitate both facility material control efforts and IAEA safeguards efforts. As
used here, compatible means that the MBAs are whole combinations of each other.
That is, an MBA of one type should not cut across an MBA of the other type.

Since it must be possible to measure all transfers into and out of an MBA,
and material unaccounted for (MUF) is reported on an MBA basis, it would be dis-
advantageous to choose MBAs in such a way that inputs and outputs are in a form
difficult to measure. Thus, an MBA should not cut across a process if the mate-
rials within that process are changing form, inaccessible for measurement, or
for other reasons not amenable to accurate measurement. Usually it is not diffi-
cult to choose boundaries that avoid these problems. In most cases the
boundaries of MBAs can be chosen so as to coincide with physical boundaries in
the facility, which may then act as a containment for the process, thus
facilitating the application of surveillance devices, or with locations in the
process at which the feed and the product are uniform and the latter is in a
final or near-final form.

In some plants more material may be held in feed and product stores than in
the process itself, and feed and product may even exceed the annual throughput.
In that case, a large part of the feed and product may be static, i.e.,
unchanging, over .a material balance period, and should not contribute to the
operator's material unaccounted for. It may be convenient, then, to establish
those areas as separate MBAs. It should be noted that unless the static stores
are also sealed by the IAEA, the IAEA would have to verify them at the beginning
and end of each material balance period, and therefore, even though static, they
will contribute to the uncertainty in the IAEA's verification of the material
balance.
 7 •

Also, if some of the material is in the form of discrete, identifiable

items, for which the MUF is zero, it may be desirable to make separate MBAs for
such items. Examples are feed in the form of UFg cylinders, or product in the
form of rods or assemblies. This also has the advantage of making it possible
to treat the feed storage area, which includes receipts, as a shipper-receiver
MBA, in accordance with IAEA practice, and therefore to report shipper-receiver
differences on a batch basis. Sometimes, the receiving area is set up as a sepa-
rate shipper-receiver MBA.

An alternative to setting up separate MBAs in the cases discussed above is

to select key measurement points in those areas. Key measurement points are
discussed below.

Facility requirements may cause the operator to divide the facility into a
relatively large number of MBAs. For example, if there are parallel production
lines, say one for each enrichment or one for PWR and one for BWR fuel, making
a separate MBA out of each line may improve the capability of detection of
anomalies in each of them. It may also make it easier to analyze such anoma-
lies. However, from the point of view of international safeguards, it is the
timely detection of the diversion of significant quantities of nuclear material
from the entire plant that matters, and this overall detection capability usu-
ally is not improved by dividing the plant in such a manner, for the same number
and quality of measurements. Therefore, although the operator may wish to use
smaller MBAs to satisfy domestic requirements, he might find it more convenient
to combine them into larger MBAs for international safeguards.

Another important consideration is that reporting to the IAEA is based on

MBAs. Minimizing the number of MBAs therefore minimizes the reporting
requirements. Facilities with computerized accounting systems have the advan-
tage of easily being able to extract from the accounting data for the operator's
MBAs the information necessary to generate the reports for the IAEA's MBAs.

Examples of MBAs

A possible choice of MBAs for the model plant described in Appendix A is

shown in Fig. A.I. MBA-1 is a shipper-receiver MBA. Cylinders of UFg and con-
tainers of UNH and TJO2 are received from off site. Their contents are assayed
and the results compared with the shipper's numbers. The materials are then
transferred to the process MBA (MBA-2) as feed. MBA-2 consists of the entire
process (conversion plus fabrication) including the rod loading procedure.
MBA-3 consists of the rest of the process: cleaning and welding of rods, assem-
bly of rods, storage of assemblies, dismantling of defective rods, etc. The ad-
vantages of this arrangement are that (1) receiving, which does not generate
MUF, is treated as a separate area, for comparison of shipper's and receiver's
values, (2) the process MBA has mostly clean, accurately measurable inputs and
outputs (the waste streams tend to be small), and (3) the inventory in MBA-3 is
in discrete, identifiable form (rods and assemblies) and should not contribute
to MUF. Therefore, only MBA-2 generates MOF.

An example which illustrates how different the MBA structure may be for do-
mestic purposes and for international safeguards is shown in Fig. 2,1. This ex-
ample is taken from a description of a plant used as a model for a recent train-
ing course in SSACs (LA-8901-C). Shown in the figure are the material balance
 SOLAR EVAPORATION FONDS (Liquid Waste Discard)
HUA-1

ICA-1 1CA-3K
Shipping-Receiving Uaatc
Darrel
Sloruge
ICA-3B-II
Storage
ICA
1IBA-1 -3A Warehouse & Trailers
Powder
Conversion Storage
Scrap
recovery 1IBA-2 HOA-3
Pellet Preparation Rod Loading

IIUA-6
Analytical Roa Storage and
Laboratory Bundle Assembly
ICA-2

UO 2 PLANT

ICA-1
Shipping-Receiving

WAREHOUSE

Figure 2,1, Example of Operator's Material

Control Areas
areas chosen by the operator. The basis on which the choices were made is
summarized as follows in the reference:

"The material control areas for the Model plant were selected on the
basis of process, administrative, accounting, and physical considera-
tions. The operation of each MBA is the responsibility of one of the
Plant supervisors. Each MBA has defined physical boundaries and each
MBA represents a natural grouping of like processing and handling
operations. Lastly, the natural flow of material between MBAs in-
volves the transfer of discrete measureable items so that flows into
and out of the MBA can be accurately accounted for."

In the figure, areas in which material is in discrete, identifiable forms are

called Item Control Areas (ICAs); in the other areas (MBAs) the material appears
in bulk form. Note that in the process building there are four MBAs. Although
the total number of material control areas for domestic purposes is large, for
international safeguards the entire plant was made into a single MBA.

The effect of the choice of MBA on the accuracy of the material balance is
illustrated by the following highly simplified example, in which it is assumed
that all measurement errors are random and uncorrelated - that is, systematic
errors and covariances, which would occur in any real case, are ignored. De-
spite this simplification, the conclusion derived below is quite general. Con-
sider a combined conversion and fabrication plant with two parallel lines of
equal throughput. This is shown in Fig. 2.2, in which Cj and C2 and F^ and F 2
refer to the conversion and fabrication portions of lines 1 and 2, respectively.

C2

Figure 2.2. Schematic representation of conversion-fabrication

plant with two parallel lines.
 10

Two cases will be considered:

Case I, a single MBA for the entire plant.

Case II, each line, C^ + F^, a separate MBA - i.e., two parallel MBAs.

The characteristics of the plant are assumed to be as follows:

Throughput 300,000 kg U/year

Batch size * 1000 kg 0

Inventory in process area: 80,000 kg U

In-process inventory 15,000 kg 0

Intermediace store 65,000 kg U

DO2 inventory 15,000 kg 0

Fuel inventory 115,000 kg U

Total inventory 210,000 kg U

a
6-month material balance period, throughput 150,000 kg U

Case I. Single MBA

Measurement uncertainties (la):

Throughput, ±0.252 per batch » ±2.5 kg U/batch

a
In-proces<? inventory ±0.52 ±75 kg 0

Intermediate store, ±0.25% * ±162.5 kg 0

UO 2 store, ±0.252 - ±37.5 kg D

Total bulk inventory - 95,000 ± / ( 7 5 ) 2 + (162.5) 2 + (37.5) 2

- 95,000 ± 182.9 kg (± 0.19%)

Throughput - 150,000 ± /l50 x (2.5) 2 * 150,000 ± 30.6 kg 0

Assume receipts * shipments (R • S) and beginning inventory • ending inven-

tory (BI*EI), over one material balance period. Then the uncertainty in MUI" »
BI + R - EI-S is

/2(30.6) 2 + 2 x (182.9) 2 - 262.3 kg U

 11

Case II. Parallel MBAs

The same batch size (1000 kg) will be assumed, but each line will handle
only half as many batches in a material balance period. The relative measure-
ment error per batch will be assumed to be the same as for Case I. The inven-
tory will be divided into two equal parts, with the- relative measurement error
for each part being /2 times the relative error for the entire inventory.

Line 1 :

Input - 75 batches @ 1000 kg/batch ± 0.25% - 75,000 ± 21.65 kg D

Process: 40,000 kg D

In-process inventory « 7500 ± 0.712 = 7500 ± 53.0 kg U

Intermediate store - 32,500 ± 0.35% • 32,500 ± 114.9 kg tJ

UO 2 store « 7500 ± 0.35% - 7500 ± 26.5 kg U

Total bulk inventory - 47,500 ± A 5 3 . 0 ) 2 (114.9) 2 + (26.5) 2

« 47,500 ± 129.3 kg

0MOF! - /Zx(21.65) 2 + 2x(129.3) 2 = 185.4 kg

185
Line 2: a^j^ " -4 k
S

As can be seen, the uncertainty in the material unaccounted for is lower in

each of the two parallel MBAs than in the single MBA case. The sensitivity of
detection for a diversion of a given size is therefore improved by dividing th%
plant into two MBAs, provided that all the diverted materai is taken from one of
the two MBAs; however, if half the amount is taken from each MBA (for the same
total diversion from the entire plant) the overall detection sensitivity in Case
II is the same as in Case I — i.e.,

a
MUF ftotal plant) 262.2 kg.

Since for planning purposes the IAEA may not assume that diversion would be lim-
ited to one MBA, its detection sensitivity would not be increased by dividing
the plant into multiple MBAs in the manner shown, although such a scheme might
be advantageous from the viewpoint of physical protection or domestic
safeguards.*

*In fact, it can be shown that for the same overall false alarm probability,
dividing the plant into multiple MBAs would actually decrease the detection
sensitivity.
 12

2.3 Key Measurement Points

An IAEA MBA is defined as an area such that the quantity of nuclear mate-
rial transferred into or out of it and the physical inventory of nuclear mate-
rial in it can be determined to establish a material balance. This determina-
tion is made through the use of strategic points and key measurement points
(KMPs), which are defined as follows:

A strategic point is "a location selected during examination of

design information where, under normal conditions and when combined
with the information from all "strategic points" taken together, the
information necessary and sufficient for the implementation of
safeguards measures is obtained and verified; a "strategic point"
may include any location where key measurements related to material
balance accountancy are made and where containment and surveillance
measures are executed." (lAEA/SG/INF/1, para. 112).

A key measurement point is "a location where nuclear material appears

in such a form that it may be measured to determine material flow or
inventory. 'Key measurement points' thus include, but are not limited
to, the inputs and outputs (including measured discards) and storages
in material balance areas. (IAEA/SG/INF/1, para. 113).

Thus, while all key measurement points are strategic points, not all strate-
gic points are key measurement points. Some strategic points may be assigned
solely for the deployment of containment and surveillance measures.

The IAEA distinguishes betweo.n two types of key measurement points: flow
key measurement points and inventory key measurement points. KMPs of each type
are established in order to facilitate the verification of both the flow and in-
ventory components of the material balance. Flow KMPs are usually located at
points at which receipts, shipments, and transfers to or from MBAs occur.
Inventory KMPs are located at points where materials are stored or where they
are expected to appear during physical inventories. It is a convention that
flow KMPs are designated by Arabic numerals and inventory KMPs by upper case
letters.

As mentioned in Sec. 2.2, the judicious use of KMPs can sometimes eliminate
the need for a separate MBA. For example, a shipper-receiver difference KMP can
be set up in the receiving area of a facility in place of a separate MBA; the ad-
vantage would be the reduction in reporting requirements. The use of inventory
key measurement points for material in stores also makes it possible to distin
guish th'.s material from in-process material, for accounting purposes. Again,
the same objective can be attained by defining stores as separate MBAs, but, at
the expense of increased reporting requirements.

There is a tendency to think of key measurement points as actual

geographical locations within the plant at which measurements are made. Al-
though this was the original meaning of the term, over the years the definition
has been extended to include specific categories or strata of nuclear material.
Thus, a KMP may not be simply "receipts", but receipts of a specific kind, such
 13

as UFg cylinders or containers of UO2. A typical inventory 7CM? could be all fin-
ished pellets in intermediate storage, or all waste of a given kind being held
for eventual discard^ or all fuel rods awaiting assembly. A KM? could even con-
sist of several locations within a facility at which material of the sane kind
is measured. KMPs are therefore a convenient way to group materials of the same
kind for measurement purposes. They are then closely related to the material
strata (using this word in the statistical sampling sense) in a facility, and
their proper selection can simplify the calculation of uncertain-ties in the ma-
terial balance. That is, associated with each KM? would be a set of measurement
errors applying to all material passing through that point.

A possible choice of KMPs for the model plant is shown on the diagram of
the process, Fig. A.I in Appendix A. A description of the flow and inventory
KMPs is given in tabular form in Tables A.I and A.2 respectively, in Appendix
A. The tables describe the form of the material, the containers in which it
appears, the sampling method, and the measurement method. As an illustration
of the principles discussed above in connection with the selection of KMPs,
note that instead of a single KMP in the receiving MBA, there are three KMPs,
(1A, IB and 1C) one for each category of receipts (UFg, UNH, and U0 2 >, and
each characterized by its own types of measurements and therefore by its own
set of measurement errors.

2.4 Materials and Material Types

Low enriched uranium fabrication facilities process and handle many differ-
ent material types in loose form, e.g., liquids, gases, powders, pellets, etc.,
in addition to discrete, identifiable items such as fuel rods and assemblies.
The bulk material types will include not only pure or "clean" materials such as
D?g, DO2 powder, UO2 pellets, etc., but also impure or "dirty" materials such as
scrap and waste which will have to be processed to restore them to usable form
or discarded if such processing is uneconomical. These dirty materials may in-
clude off specification powder and pellets, grinder sludge, solid and liquid
waste, and incinerator ash.

The accuracy with which these materials can be measured depends on their
type. Relative uncertainties are usually smaller for the clean materials, such
as feed, product, and well characterized intermediate fores, than for the dirty
materials, mainly because of the inhomogeneity of the latter. The inhomo-
geneities may introduce large sampling errors if chemical analysis is used.
Nondestructive assay (NDA) is often used to measure inhotnogeneous materials, in
order to avoid sampling errors, but NDA methods tend to be inherently less accu-
rate than chemical methods. Their accuracy is also dependent to a certain ex-
tent on material homogeneity. Therefore, avoidance of large inventories of
scrap and waste will reduce the uncertainties in the material balance. This can
be accomplished in the case of scrap by processing it to an accurately
measurable fora before the physical inventory taking, so that large accumula-
tions of difficult to measure material do not appear on the inventory.

The kinds of materials and the amounts of each that might appear in a typi-
cal inventory of a plant of the type under discussion here is shown in Table
2.1. In Column 2 of the table is listed the number of items in each category on
inventory and the type of container. The total amount of material on inventory
is 135,225 kg U, but of this 90,500 kg is in discrete, identifiable iteas (rods
 14

Table 2.1. Typical Inventory of Model Plant

Category No. of Items :"otal Uraniu

28 cylinders 28,000

OO2 Powder 280 buckets 7,000

Sintered Pellets 1000 trays 5,000

Dirty Powder 90 drums 2,250

ADU Scrap 15 drums 375

Green Scrap 20 drums 500

Sintered Scrap 20 drums 500

Grinder Sludge 100 buckets 500

Fuel Rods 50 bins (10,000 rods) 50,000

Reject Rods 100 rods 500

Liquid Wastes 100 drums 200

Solid Wastes 80 barrels 400

Fuel Assemblies 80 assemblies 40,000

TOTAL 135,225
 15

and assemblies), so that only ^5,000 kg need be measured during the inventory
taking. Of course, the material in the rods and assemblies was measured as
pellets, before being inserted into the rods.

The effect of different categories of materials, either in the flows or the

inventory, on the overall measurement uncertainties can be illustrated by a sim-
ple calculation. Assume the receipts, shipments, and beginning and ending inven-
tory are as shown in Table 2.2, for a one year material balance period. For
purposes of illustration the receipts are assumed to equal shipments, and the
beginning inventory to equal the ending inventory (in practice that will usually
not be the case). Note that waste discards are classed as shipments, since they
are assumed to be removed from the MBA, which comprises the entire processing
.area of the facility. The estimated relative measurement error (in percent) for
each category of material is shown in Column 4 of the table, the absolute error
(in kg U) in Column 5, and the variance in Column 6. The total variance and
Cjfljp are shown at the bottom of the table, as is the overall error relative to
throughput.

As can be seen from the example, even though waste discards amount to less
than 2" of the total throughput, because of their large measurement error they
contribute almost 85Z of the total variance in the material balance. Doubling
the amount of waste would increase the Ojfljp by ^36!?. In this particular exam-
ple, the scrap quantities on inventory are small, presumably because during the
material balance period a considerable amount of scrap was processed and
recycled. Consequently, the effect of scrap measurement uncertainties on the
overall uncertainty is quite small (only </>4» of the overall variance). Since or-
dinarily the scrap generation rate may amount to 10-1531 of the throughput, if
the scrap accumulation had not been kept low by processing it would have had a
much larger effect on the material balance uncertainties. (States under
INFCIRC/66/Rev.2 are required to recover scrap and waste within a reasonable
time, if practicable, or arrange jointly with the IAEA for its accounting and
disposal).

2.5 Measurement Methods

For efficient operation, to satisfy health, safety and domestic safeguards

requirements, and to fulfill contractual obligations, a facility operator will
want to establish an effective nuclear material accounting system. Such a sys-
tem is also necessary if the IAEA is to carry out its safeguards function at the
facility. A basic principle of the IAEA's safeguards system is that of indepen-
dent verification. Under this principle, the IAEA does not repeat all the
operator's measurements, but' verifies a random sample of them. By selecting
this random sample from all strsta entering into the material balance, the IAEA
can estimate any long-range systematic error or bias on the part of the opera-
tor, and verify his material balance. A great advantage of this approach is
that it minimizes both the intrusiveness of, and the resources required for,
inspections. It follows that for effective application these advantages can be
fully realized only if the operator measures all receipts, shipments, transfers
between material balance areas, discards, and materials on inventory; if measure-
ment of certain materials (e.g., certain kinds of hold-up) is impracticable or
impossible and they are not included in the inventorv, they will appear in the
MUF.
 16

Table 2.2. Effect of Measurement Errors for Different Categories

of Material on Overall Uncertainty in Material Balance

U Mass Measurement
Stratum (kg) Error, 7. a(kg U) a 2 (kg2)

Receipts UF6 300,000 0.08 240 57600

Shipments
or Discards Pellets 295,290 0.02 59.06 3488.08
Solid Waste 2,500 25.0 625 390625
Liquid Waste 1,460 3.0 43.8 1918.44
Incinerator Ash 750 25.0 187.5 35156.25

Subtotal 488787.77

Beginning
and Ending
Inventory UO 2 Scrap 4,800 1.9 91.2 8317.44
ADU Scrap 500 2.98 14.9 222.01
Pellet Scrap 500 1.9 9.5 90.25
Sludge Scrap 750 2.24 16.8 282.24
UO2 Powder 5,000 0.04 2.0 4.0

Subtotal 17831.88

a2 = 488787.77 + 2 (17831.88) = 506619.65 kg 2

MUF
(?,„„, - 711.77 kg U

* Throughput - 0.00237 (or >/"0.24Z)

MUF
 17

Since the IAEA's inspection goals are stated for uranium in terms of
quantities of fissile isotopes (see below), the operator will need to measure
and make available to the IAEA data on not only the amounts of the element
(uranium in the present case) but the isotopic composition, or at least the quan-
tity of D-235. He would ordinarily measure these anyway, for one or more of the
reasons listed above.

The measurements can be accomplished in many ways, depending on many fac-

tors such as the material and material types themselves, the accuracy required,
and the resources available. The measurements will fall into three categories:
(1) bulk, i.e. volume and/or weight, (2) sampling and analytical and (3)
nondestructive. In the discussion that follows these categories will be
addressed in terms of their applicability to material and material types in the
general categories identified in the previous section. They also will be
addressed in terms of accuracy and limitations such as error sources,
interferences, and availability of standards and calibrations.

An important determinant in the choice of a measurement method for a partic-

ular category of material is the contribution that the uncertainty in that
method will contribute to the overall uncertainty in the material balance over
some material balance period. For low enriched uranium fuel fabrication and con-
version plants the recommended frequency of physical inventory taking is once or
twice a year (Hough 1979). In the words of the reference, "To facilitate effec-
tive safeguards, the minimum accuracy of material balances which (is) to be
achieved by facility operators should correspond to international standards."
For plants of the type under consideration here, the maximum expected uncer-
tainty (la) is 0.32 of throughput (IAEA/SG/INF/1, para. 100).

In choosing a particular measurement method the operator should therefore

consider the uncertainties resulting from the use of that method, taken together
with the uncertainties resulting from all the other measurements on the flows
and inventories, in relation to the expected measurement uncertainties
corresponding to international practice. This does not mean that all categories
of material will be measured with the same accuracy, since, as noted in Sec.
2.4, the contribution that a particular category makes to the overall uncer-
tainty in the material balance depends not only on the accuracy with which it
can be measured but on the amount of material in that category relative to the
other terms in the material balance equation. The operator may therefore have
considerable latitude in the choice of measurement method for various categories
of material, and the final selection will depend on other factors as well, such
as cost, convenience, availability of the required equipment or skills, safety,
and timeliness of results with respect to production schedules. Indusi and
Marcuse have described a method of optimizing the allocation of measurement re-
sources so as to minimize the costs for a given uncertainty in the material unac-
counted for (Indusi 1974).

An important characteristic of any measurement system used for interna-

tional safeguards is that of traceability, which is defined as "the ability to
relate individual measurement results to internationally recognized standards or
nationally accepted measurement systems through an unbroken chain of comparison"
(IAEA/SG/INF/1, para. 156). The standards referred to may consist of standard
materials, standard artifacts (e.g., weights or radioactive counting standards),
or standard procedures. Traceability ensures that all safeguards related
 18

measurements in facilities under safeguards are performed to common or at least

to comparable standards. It also facilitates the resolution of anomalies. It
is a desirable characteristic for bulk, analytical, and nondestructive measure-
ment systems.

In bulk measurements, traceability is provided through calibration of the

measurement equipment against working standards which, in turn, can be related
to higher standards: secondary standards, primary standards, and, ultimately,
the accepted national or international standard. Thus, for weighings there is
a hierarchy of standard weights, with working standards at the lowest level and
the international kilogram at the highest. In between, in descending order of
accuracy, are the national standards and at the facility, primary standards, sec-
ondary standards,..., and working standards.

Traceability for analytical (chemical and isotopic) measurements is pro-

vided through the distribution and use of standard reference materials (SRMs).
For chemical analysis the hierarchy, in descending order, is reference or atomic
weight standards, ultimate standards, primary standards, and working standards
(Rodden 1972). Primary standards, in this case, are commercially available sub-
stances with a purity of 100*0.022.

The use of standard reference materials is not, of itself, sufficient to

ensure traceability. The measurement procedures themselves should also conform
to national or international standards, if such exist (e.g., those promulgated
by national standards agencies or nationally accepted standards institutes).
Thus, the American National Standards Institute (ANSI) publishes standards for
analytical.procedures for the accountability of UFg (ANSI N15.7-1972), for mass
calibration techniques (ANSI N15.18-1975), for calibrating NDA systems (ANSI
N15.20-1975), and for volume calibration techniques (ANSI N15.19-1975). Confor-
mity can also be demonstrated by participation in interlaboratory comparisons,
round-robins, etc. Correct procedures are especially important for NDA
measurements, since the main source of error here is not the lack of
sufficiently pure and well-characterized standard reference standards but matrix
and geometrical effects which may not be properly reflected in the calibration
standards.

Traceability implies the existence of a system for documenting the

measurements, calibrations, and standardizations performed by the operator.
This aspect of the records system is discussed in Sec. 2.7. The subjects of
'traceability, standards, and their role in determining measurement bias and
accuracy are treated more fully in the detailed description of the Measurement
Quality Element.

2.5.1. Bulk Measurements

Bulk measurements are made to determine the total quantity of material,

expressed either as a mass (or weight) or as a volume. The amount of nuclear ma-
terial can then be obtained by multiplying the total mass or volume by the
analytically determined concentration of the element of interest, in this case
uranium, in mass per unit mass or per unit volume of material. Since volumes
are affected by densities, which in turn depend on temperatures, it is usually
desirable to convert volume to mass by using either a measured density or a
known density at a measured temperature.
 19

Solids are usually weighed. Either weight or volume of a liquid may be

measured. The weight, as explained above, is usually preferable, but when the
liquid is in fixed tanks weighing may not be practicable and, instead, liquid
levels and densities may be measured and converted to volumes or masses by a
suitable calibration, using known masses or volumes of some standard liquid such
as water*

Weighing may be performed with scales, balances, or load cells; the last
may be of either the hydraulic or the mechanical type, in which case a trans-
ducer (e.g., a strain gauge) may convert the mechanical force into an electrical
signal for more accurate and convenient readout. Typical precisions of selected
weighing devices are shown in Table 2.3.

The maintenance of optimum accuracies in weighing depends most importantly

on the frequent use of standard weights for calibration or comparison, or, in
certain cases where buoyancy effects or the distribution of weight affects the
response of the weighing device, of replica mass standards which physically re-
semble the production object to be weighed (e.g., OT5 cylinders).

For optimum tccuracy in weighing, the tare weight (the weight of the empty
container) must be accurately known, and any residual material (e.g., UFg heels)
left in the container after transfer of its contents must also be measured and
taken into account.

Volume measurements of liquids are usually made by measuring the liquid

levels in a tank or by measuring flows. Since the bulk measurement of liquids
plays a relatively minor role in the accountability of nuclear materials in a
conversion and fuel fabrication plant, measurement methods will be described
only briefly here. The reader should consult the references for details.

The traditional technique for measuring volume in a tank accurately and re-
motely has been that of dip-tube manometry (Rodden 1972). In this method the
pressure differences are measured between three tubes through which air is
slowly flowing. One tube is located in the liquid at the bottom of the tank, a
second also in the liquid at a known height above it, and the third in the vapor
space above the liquid. The pressure difference between the first and second
tubes and the known difference in heights give the density of the liquid. This
in turn is used to convert the pressure difference between the first and third
tubes to a liquid height. Typically, the differential pressure measurements are
carried out with the use of manometers, differential pressure transmitters, or
electromanometers. Typical attainable accuracies for these instruments are
0.02, 0.2, and 0.0032 of full scale, respectively.

À simpler and more direct method of measuring liquid levels is the sight
glass. In a recent survey of measurements at actual facilities using less than
ideal tank geometries random errors of 1.5Z and a systematic error of 0.5Z were
found (Rogers 1982).
 20

Table 2.3. Capabilities of Selected Weighing Devices*

Precision,
Type of Device Capacity Precision 7. at Full Scale

1. Equal-Arm Balance 500 8 ±0.3 g ±0.06

2. Equal-Arm Scale 2 kg ±0.4 g ±0.02

3. Top-Loading Scale 120 g ±0.0005 g ±0.0004

1000 S ±0.005 g ±0.0005

4. Platform Scales 800 lbs. ±0.25 lb. ±0.03

Platform Scales,
Warehouse Type 20,000 lbs. ±6 lbs. ±0.03

5. Load Cells

a. Hydraulic 200 kg ±0.1 kg. ±0.05

b. Strain gauge 2 tons ±0.8 lb. ±0.02

* From LAEA-174, with modifications.

 21

Liquid volumes can also be measured by integrating the fluid flows over a
period of time. Flows may be measured with turbine type flow meters or with dif-
ferential pressure orifice flow meters (Strohmeier 1974). The former have been
found to be capable of precisions of ±0.0252 and accuracies of 0.252 over a wide
range of flow rates, and the latter of accuracies of 1-3% typically.

Further details of volume measurement techniques in liquids and attainable

accuracies may be found in IAEA-174, Ellis 1981, Ehinger 1979, and Keisch 1980.

Because of the form of the materials in an LEU fuel fabrication plant, most
of the bulk measurements would consist of weighings. Of the categories of mate-
rial listed in previous sections, the ones that would be weighed are the UFg cyl-
inders (and heels), UNH and UOj receipts and UO2 product, sintered pellets, pel-
let, stacks (just prior to rod loading), rejected sintered pellets, ammonium
diuranate, green and sintered scrap, grinder sludge, and dirty powder.

Liquid flows such as DNH would be measured by flow meters of the orifice or
turbine type, whose signal would be integrated to give the total volume, or by
sight glass readings of the liquid level in the collecting tanks. Liquid wastes
could be measured in the same way or by weighing drums of the material.

2.5.2. Sampling

Analytical measurements are performed on samples of material removed from

the production streams or inventories. The analytical results will therefore
apply .to the material from which the sample is taken only to the extent that
that sample is representative of the material. Since often the materials are
inhomogeneous (especially scrap and waste materials), care must be taken to as-
sure representativeness. In an ideal sampling process all portions of the mate-
rial should have an equal probability, per unit mass or volume, of being taken
as a sample: for example, all parts of a drum or can of powder, or all regions
of a tank containing a solution. This may require taking multiple samples from
a container. In extreme cases of heterogeneity it may be necessary to resort to
nondestructive assay, which can measure the entire contents of a small container
or a major fraction of a large container, thereby averaging out local
inhomogeneities. Consideration should be given to the stability of the mate-
rial; properties important to the analysis must be preserved through proper
packaging, handling, and environment. In certain instances the long term reten-
tion of "archive" samples for later comparison with production material may be
required.

If the material to be sampled can be divided into different strata with

required.

If the material to be sampled can be divided into different strata with

non-identical properties, each stratum should be sampled independently. Various
sampling patterns or strategies may be adopted: A completely random sampling
routine can be carried out with aid of random numbers, either from a table or
generated by a computer or hand calculator. Alternatively, for certain purposes
a systematic pattern in which, for example, every tenth or one-hundreth item is
taken for analysis, may be used.
 22

In a conversion fabrication plant of the type under consideration here, the

materials that have to be sampled are usually liquids and solids, the latter
typically in the form of a powder or pellet (UFg as received in cylinders is a
solid, but for sampling purposes is liquified by heating).

Liquids usually must be mixed before being sampled. Various mixing methods
are used, such as stirring with a motor driven propeller, pump recirculation of
the contents, and sparging (bubbling air through the liquid at a rapid rate).
Sparging may be ineffective if the solution contains suspended or settled
solids. Samples may be removed by dip cups, by air or vacuum lift devices, by
drip taps in process lines, or by drain valves at the bottom of process tanks.
An especially effective method for the sampling of heterogeneous liquids is the
continuous or semi—continuous stream sampler, which, as the name implies, in-
volves the continuous or periodic sampling of the liquid stream as it passes by
(Lovett 1974). Detailed descriptions of procedures for sampling UFg in small
and large cylinders, uranium solutions in shipping containers, and uranium solu-
tions in process tanks are available in the literature (Rodden 1972).

It is usually more difficult to obtain representative samples of powders

than of liquids, because of a tendency towards segregation caused by particle-
size effects and the greater difficulty of achieving homogeneity through mixing.
Gross differences in particle size can be reduced by milling to a uniform powder
(e.g., with a ball mill) or by sieving and sampling each mesh size. Powders
with smaller particle 3ize differences can be rendered more homogeneous by using
a V-blender or similar device to blend the material.

A number of devices or procedures are available for ensuring the representa-

tiveness of powder samples, The grab sampler is the simplest: a scoop is used
to remove samples from various locations within the container being sampled. A
more sophisticated device is the core sampler, which may consist of a pipe
driven down through the material or a rotating auger inside a metal pipe (Lovett
1974). Either of these is capable of withdrawing an entire core or vertical
profile of the material. A vacuum thief, a pipe connected to a vacuum system,
can also be used for this purpose. If azimuthal heterogeneities are suspected,
a sampling template containing holes at some angular interval (e.g., 120°) can
be used in connection with these devices (Sodden 1972). The cores withdrawn in
this manner can then be further homogeneized, to eliminate vertical segregation
or layering.

A more satisfactory device is the riff1er or sample-splitter. Commonly,

this device consists of one or more V-shaped troughs (although other devices,
for example the "spinning" riffler, are also available), each of which divides
a flowing stream of powder into two equal streams. Samples of the desired size
can be obtained by using a riffler with the appropriate number of stages.

Whatever method of sampling is used, it is important, in calculating the un-

certainty in the overall material balance, to include the contribution from sam-
pling error. A carefully designed measurement quality program will therefore in-
clude procedures for both minimizing sampling error and for estimating its magni-
tude. One method by which this may be done is to analyze duplicate samples,
estimating the standard deviation by the method of paired differences (Brouns
1978; Jaech 1973). This will reflect both the sampling variance and the
 23

analytical variance; the latter may be known from separate experiments, or the
two effects may be separated by analysis of variance techniques.

In a recent survey of measurement practices and accuracies in uranium

facilities, chemical sampling errors for the more homogeneous materials (feed,
product, and intermediates) were found to be on the order of 0.032. Sampling
errors for clean scrap covered a wide range (0.03 to 19%), and for dirty scrap
an even wider range (0.6 to nearly 402)• Isotopic sampling errors (resulting
from the mixing or blending of materials of different isotopic composition) were
also larger for scrap than for feed, product and intermediate materials, but the
variation was much smaller than for the chemical composition. (Rogers 1982).

In the LEU conversion fabrication plant, OO2 powders would ordinarily be

thief sampled, possibly after blending (especially dirty powders), or riffle-
sampled. Pellets would be selected at random from trays, according to a statis-
tical sampling schedule. UNH and liquid wastes would be sampled by any of the
techniques described above for liquids, a recirculating sampler usually being
used for the latter. ADD is usually grab sampled from random locations in the
container or, if dry enough, auger-sampled. Containers of grinder sludge are
stirred and grab sampled.

Detailed procedures for sampling DFg cylinders have been described in the
literature (Rodden 1972; KelleT 1967). Briefly, they consist of heating the cyl-
inder at temperatures between 200 and 250°F to liquify its contents, attaching
a sample pipet and sample container to the cylinder value, purging and
evacuating both, and withdrawing the sample, under gravity, into the sample con-
tainer .

2.5.3. Analytical Measurements

The choice of analytical method for the measurement of uranium in the vari-
ous material categories will depend on the nature of the material, its purity,
the desired accuracy, possible interferences, and on practical considerations
such as cost, analysis time, and availability of equipment and skills. For
fuel, product, or intermediate materials, which are relatively pure, the methods
of choice are usually the gravimetric or ignition method, potentiometric titra-
tion methods such as the modified Davies-Gray method, and other electrometric
methods. These methods are treated in detail by Rodden (Rodden 1972). Special
methods applicable to the analysis of UFg are treated in the American National
Standard "American National Standard Analytical Procedures for Accountability of
Uranium Hexafluoride" (ANSI N15.7). Analytical procedures for certain materials
in the nuclear fuel cycle are specified in the standards published by the
American Society for Testing and Materials (ASTM 83). For uranium present in
lower concentrations, for example, in various waste materials, in addition to
the Davies-Gray method other preferred methods are fluorometry, spectrophoto-
metry, and x-ray fluorescence. Surveys of the analytical methodology and
detailed references are given in IAEA 174 and in Bakkila 1981. A brief summary
of some of the more important or commonly used methods for the materials
appearing in a typical LEU conversion fabrication plant follows. Properly used,
these methods have been demonstrated to be capable of achieving accuracies
corresponding to international standards. The descriptions are mostly based on
those in IAEA-174 and Rodden 1972; other references are also given in the text.
 • 24

2.5.3.1. Elemental Analysis

Very pure materials such as feed, product, and intermediate materials

can be analyzed with a minimum of sample preparation. Less pure materials such
as scrap and waste will usually require a more elaborate pre-treatment t;o remove
impurities.

2.5.3.1.1. Feed, Product, and Intermediate Materials

A widely used method for the determination of uranium in high pu-

rity compounds like UFg, DNH, OO2, and O3O8 is the ignition or gravimetric
method (Rodden 1972). Accuracy is acceptable for materials containing less than
0.05% impurities. The main pieces of equipment required are a pyrohydrolysls
furnace and an emission spectrograph for the determination of impurities. The
method consists of converting the sample to U3O8 by ignition to constant weight
at a carefully controlled temperature in the range of 850-900°C. The uranium
concentration is then obtained from a combination of stoichiometric and
gravimetric data, with corrections for nonvolatile impurities, which are
analyzed spectrographically. Uranium nitrate samples are evaporated to dryness
before ignition and UFg samples are hydrolized in water and then evaporated to
dryness. For the types of materials under discussion typical accuracies and
precisions, expressed as a coefficient of variation, are on the order of ±0.05%.
In a recent survey of performance in a number of plants, the random and system-
atic errors in the determination of uranium in feed, product, and intermediates
were 0.04 and 0.02%, respectively, and in clean scrap 1.6 and 0.02% (Rogers
1982).

Various potentiometric titration methods can be used when impu-

rity levels are too high for accurate results with the gravimetric method. One
of the most widely used is the modified Davies-Gray method (Eberle 1970) . In
this method, samples are converted to the sulfate form by treating with sulfuric
acid. This is then treated with an excess of ferrous sulfate to reduce U(VI) to
U(IV) in a concentrated phosphoric acid solution containing sulfamic acid.
After several more steps involving oxidation by nitric acid in the presence of
Mo(VI) and the addition of dilute sulfuric acid containing V(IV) as a catalyst
the uranium concentration is determined by potentiometric titration with stan-
dard potassium dichromate solution. Interfering elements (silver, mercury,
vanadium, and manganese) can be removed either by sample treatment or by TBP ex-
traction (Pietri 1978; Bingham 1975). Precisions are as low as 0.08% (RSD), al-
though a recent survey of actual performance in plants found an average random
error of 0.14% and a systematic error of 0.04% on feed, product, and intermedi-
ate materials; random errors on clean scrap were found to be 3.5%, however
(Rogers 1982).

A coulometric method in which U(VI) is reduced electrically to

U(IV) at a mercury cathode at a controlled potential is also widely used for the
determination of uranium. The uranium content is determined by comparison with
a standard known solution of uranium on which a similar procedure is performed.
In product materials interferences are few; in less pure materials, a chemical
separation of the sample may have to be performed first (Rodden 1972). Random
and systematic errors in product material determination have been found to be
comparable, in practice, with those for the Davies-Gray method (Rogers 1982).
 25

2.5.3.1.2. Scrap Materials

Scrap materials such as dirty UO2, reject pellets, and dissolver

solutions can be analyzed in a number of ways. The solid materials must first
be dissolved. Any insoluble residues in solution samples must be redissolved
and added to the solution. One of the most satisfactory and widely applicable
methods is the modified Davies-Gray method, discussed above. Precision and
accuracies in synthetic samples were each found to be on the order of a few hun-
dredths of a percent, relative (Rodden 1972).

The modified Davies-Grey and gravimetric methods can be used for

virtually all the dense (i.e., high concentration) materials (feed, product,
intermediates, scrap, in-process residues, etc.) found in LEU conversion-
fabrication plants.

2.5.3.1.3. Waste Materials

Waste materials are those which are relatively low in uranium con-
tent and have large admixtures of extraneous material. They may be liquids,
solids, or mixtures such as sludges or slurries, and may consist of process
residues, floor sweepings, contaminated lubricants or reagents, etc. As such,
they are often very inhomogeneous and difficult to sample representatively.
They may be measured chemically or nondestructively. For volume reduction
wastes may be incinerated, concentrated by evaporation, leached, etc. The wide
variety of forms requires a variety of pre-treatment methods for the preparation
of samples suitable for chemical analysis. These are discussed at length by
Rodden (Rodden 1972).

The most common chemical methods for assaying waste samples of

uranium are the modified Davies-Gray method, fluorometry, spectrophotometry, and
x-ray fluorescence. The modified Davies-Grsy method has already been discussed.

Fluorometry may be used to analyze samples in the 1-100 nanogram

range, although it is also used at concentrations of up to 12. The method is
based on the selective excitation of fluorescence in uranium by ultraviolet
light. To enhance the fluorescence the sample is fused with a carbonate or fluo-
ride flux. Concentration is calculated from a calibration curve prepared with
known samples of uranium. Although few substances interfere in the sense of
fluorescing in the same spectral region, small quantities of certain elements
may absorb the fluorescent radiation, leading to an underestimate of the amount
of uranium in the sample; this process is called "quenching." Prior treatment by
chemical separation can eliminate this effect, but an alternative is to measure
the fluorescence from the sample before and after adding a known amount of ura-
nium as a standard. Precisions are usually in the 1-102 range, but sometimes
worse. In a recent survey of facilities using this method to analyze liquid
wastes the average accuracy was found to be «T5" and the average precision 185
(Rodden 1972; Hakkila 1981; Wing 1981; Rogers 1982). An improved method using
a pulsed laser as the exciting source has been described by Zook et al. (Zook
1981). It can be used to assay nanogram quantities of uranium in solution
without requiring a fusion step to enhance the fluorescence. Precisions of
2-3Z and accuracies of better than 1Z have been obtained for sample concentra-
tions in the range of 0.01-4 yg U/g.
 26

Spectrophotometry is based on the principle of selective absorp-

tion of light by a compound in solution containing the element in question. At
low concentrations the response (absorption) is proportional to concentration.
The range of sample sizes to which the method is applicable is 1-40 mg. To
avoid interferences, prior chemical extractions may be required. Precisions
cited in the literature range from a few tenths of a percent to 202. An
automated version using a complexing agent and an extractant routinely achieves
a precision of 0.5-1.5Z, depending on the sample size. (IAEA-174; Hakkila 1981;
Rogers 1982; Jackson 1976).

The x-ray fluorescence technique is usually applied to solutions

having uranium concentrations in the Ug/ml to mg/ml range. High energy X-rays
from an X-ray generator or gamma rays from an istopic source are used to excite
the uranium K, L, or M X rays. X-ray generators provide greater intensity and
better signal to background ratios. For thin samples the intensity of the
excited radiation is proportional to concentration, which may be calculated from
a calibration curve obtained with known standards. An advantage of the tech-
nique is that measurement times are short (half an hour or less). Precisions
and accuracies on the order of 1-32 or less are reported (IAEA-174; Hakkila
1981).

2.5.3.2. Isotopic Analysis

Isotopic measurements are required by the facility operator to

meet specifications for the product, to satisfy domestic regulatory
requirements, and to assure operational safety. They are also required to sat-
isfy accountability requirements for international safeguards, since the inspec-
tion goals for uranium are stated in terms of significant quantities of U-235.
The IAEA's verification of the material balance will therefore involve a verifi-
cation of both the amount of uranium and the amount of U-235 — i.e., the IAEA
will normally verify both the chemical and the isotopic measurements of the oper-
ator.

If the facility fabricates fuel with various enrichments, as is

usual in LEU fuel-fabrication plants, isotopic measurements will be especially
important, to prevent or detect mixups. Isotopic measurements are also impor-
tant to determine the composition of blended, recycled material.. In addition,
the isotopic composition is required to compute the effective atomic mass, used
to determine the elemental composition in a chemical analysis. Measurements may
be made on UF5 receipts, to verify the shipper's values, and on product, interme-
diate, scrap, and waste materials.

The most accurate method of measuring isotopic composition is

mass spectrometry. Nondestructive methods, based on gamma-ray spectrometry, are
less accurate but much less time-consuming, and require relatively simple equip-
ment. They are especially suitable for "go no-go" measurements in which it is
desired merely to distinguish one enrichment from another, or for measurements
of waste, for which high accuracy is not required. They are discussed in more
detail in Sec. 2.5.3.3.
 27

The mass spectrometers employed fall into two classes, according

to the means by which the sample to be measured is introduced into the ion
source: 1« Gas phase, in which a sample of UFg is introduced through a
calibrated leak, and 2. Thermal ionization in which a small amount of uranyl
nitrate is dried on a filament in the ion source. Isotopic ratio measurements
are ordinarily carried out by comparing the observed peak ratios for the iso-
topes of interest from the sample measured with those from a known standard, or
preferably, from two standards whose isotopic ratios bracket that of the sample.
Replicate measurements of the sample and standards(s) are interleaved for a num-
ber of times sufficient to yield the desired1 precision.

Each type of mass spectrometer is subject to a systematic error

characteristic of the apparatus. The gas phase instrument is susceptible to
contamination by samples introduced at an earlier time, so that it may
"remember" an isotopic ratio other than that of the sample under investigation.
Procedures to determine the magnitude of the bias introduced by this effect have
been developed; evidently it may be minimized through the employment of stan-
dards as near as possible in isotopic composition to the sample under investiga-
tion. In the thermal ionization spectrometer a systematic error in the apparent
isotopic ratio arises from a preferential evaporation of the lighter isotope
from the heated filament. This error may be minimized and corrected for through
the control of the filament temperature and through the reproducible timing of
the sequence of measurements. Thermal ionization mass spectrometers are the
most widely used type in LEU fabrication plants. The procedures for carrying
out mass spectometric measurements with high accuracy and precision are
discussed in detail in Rodden 1972 and in NBS 260-27.

The precision and accuracy of the gas phase spectrometer for sam-
ples enriched to 1 to 5Z in U-235 is typically in the range of 0.05Z; for a
narrower range of enrichments it is possible to achieve uncertainties in the
range of 0.01 to 0.02Z. Typical uncertainties for the thermal ionization mass
spectrometer are JO.1Z.

Recently, a transportable quadrupole mass spectrometer has been

developed at ISPRA (Guzzi 1982) for the isotopic analysis of OF g samples at indi-
vidual facilities. In initial tests the instrument demonstrated a precison of
J"0.1Z and accuracy of -A).5". Improvements leading to smaller uncertainties ap-
pear feasible.

Methods of optical spectroscopy may also be employed (Rodden

1972) to determine the ratio of U-235 to U-238 in uranium samples. The
uncertainties in this method are of the order of IX.

2.5.3.3. Nondestructive Assay

In low enriched uranium conversion and fabrication plants

nondestructive methods are most useful for the assay of low density,
heterogeneous material such as waste which it would be impractical to homogenize
for chemical assay, for finished product such as rods and assemblies which it
would be prohibitively expensive to analyze destructively, and for determining
the amount of residual hold up in equipment (i.e., material remaining after a
clean-out). As mentioned before, it can also be used to distinguish quickly be-
tween different enrichment levels, as a "go no-go" type of measurement. One of
 28

the greatest inherent advantages of NDA, the ability to provide real time or
near real time data, is not as significant in LEU plants as in those processing
HEU or Pu because the timeliness goal.for detection of diversion of LEU is so
long (one year).

The major source of inaccuracy in NBA is usually the lack of

representativeness of the calibration standards. These have to represent the
unknowns both isotopically and with respect to spatial distribution, matrix
material, and container design and composition, as well as, possibly, chemical
composition. Because of the great variety of fuel material forms in a conver-
sion-fabrication plant, it is not practicable to establish a national or
international system of NDA standards for every category, and the operator may
have to develop his own working standards to match his needs. Ultimately, these
should be traceable to national or international reference materials, by being
either prepared from or compared with then (Milner 1982).

Some NDA reference materials are available. A list of such mate-

rials available within the IAEA, the Commission of the European Communities, and
six countries of the E.E.C. has been compiled by the European Safeguards Re-
search and Development Association (EUR 6089 EN). They include UO2 pellets and
rods and UO2 and U3O3 powder of assorted enrichments. Uranium enrichment stan-
dards for the calibration of enrichment meters have also been developed (de
3ilvre 1982; Milner 1982). These are sets of isotopically characterized sam-
ples of U3Og ranging from 0.3 to 4.52 in U-235 enrichment, infinitely thick to
the low energy gamma rays from U-235, and packaged in containers with a known
attenuation.

Nondestructive assay of solid wastes is usually performed by

measuring the number of gamma rays emitted by ^ 2 3 5 (sometimes by U-238 or its
daughters), by means of a sodium iodide scintillation counter (Hatcher 1981).
Corrections for attenuation may have to be made, either by calculation or by
measuring the transmission of a source placed behind the sample. To minimize
spatial distribution effects the sample container may be rotated, and the detec-
tor may also be translated parallel to the axis of rotation (LA-5652-M). The
system is usually calibrated against specially prepared standards packaged in a
similar manner and containing a known amount of U-235. The material in these
standards, in turn, is prepared from or compared with certified reference
materials.

Methods employing active neutron interrogation to induce fission

are also sometimes used to assay waste packages. These are usually more
sensitive and more accurate than the passive gamma ray systems, but involve more
elaborate and expensive equipment.

Because of the great variety of waste packages with respect to

size, shape, matrix, amount of material, etc., a single estimate of accuracies
and precisions is not very meaningful. Typically, however, a Nal gamma ray
assay without transmission correction will have accuracies and precisions in the
range of 5-10Z, and, with transmission correction, 1-52, for samples containing
1-20 g. of U-235. Active assay systems employing isotopic neutron sources may
have accuracies as low as J"IZ and precisions of several percent (Rogers 1982).
 29

Passive gamma counting instruments may also be used to measure re-

sidual hold up in process equipment or, alternatively, they may be used to deter-
mine that such hold up is negligible. Usually they consist of portable,
collimated Nal detectors (Hatcher 1981). Again, they must be calibrated against
specially prepared standards. Accuracies and precisions, typically on the order •
of 20%, are not as good as for the other applications of NDA, but because of the
usually small amounts of material may be acceptable.

NDA methods may be used to measure or check enrichments. Accu-

rate measurement for accountability purposes usually requires the counting of so-
lution samples under carefully controlled conditions, and comparing the count
rates with those of standard solutions containing uranium of very similar
enrichment and identical concentration. Dnder these conditions, accuracies of
±0.12 and precisions of A).VZ are achievable routinely (Rogers 1982).

Enrichments of solid samples such as powder and pellets may also

be measured for "overcheck" or quality control purposes by utilizing the
enrichment meter principle (Hatcher 1981). This involves counting the 185.7 keV
U-235 gamma rays from a sample thick enough to be opaque to this radiation and
comparing the count rate to that of thick standards of known enrichment. Al-
though accuracies and precisions are good, they are not as good as those from
mass spectrometry, and so the method is rarely, if ever, used for accountability
data in LEU fabrication plants.

As an alternative to enrichment determinations based on the mea-

surement of the yield o-f 185.7 keV gamma rays from thick samples of uranium,
Hagenauer (Hagenauer 1982) has developed a method based on the measurement of
low energy gamma rays and X-rays with a high resolution coaxial germanium
detector. The peaks of interest are the 92.38 and 92.80 keV gamma rays of Th-
23A emitted after the decay of TJ-238 and the thorium K ^ a n d ^al X-rays with
energies of 89.96 and 93.35 keV respectively, produced in the decay of U-235.
The method has the advantage that the results are comparatively independent of
sample thickness or matrix, and since the gamma rays are produced directly in
the decay of U-238 it does not depend on the accumulation of U-238 descendants,
as does isotopic assay based on the 1001 keV gamma ray of U-234, and hence can
be used on materials which have been newly purified chemically. For the best
precision it was found to be necessary to fit the observed peaks with a computer
peak fitting code. The measurement uncertainties for this method over a range
of enrichments from 0.27 to 60.0" were, on the average, less than 0.3Z.

Nondestructive devices for assaying fuel rods have been developed

and are commercially available. The designs usually include provisions for both
scanning the rods for uniformity of loading and of enrichment and for measuring
total (i.e., integrated) fissile content. Isotopic neutron sources are used to
induce fission in the U-235 and the resulting fission product delayed gammas are
used to check the pellet to pellet uniformity and the fission product delayed
neutrons the overall fissile content. The precision of the latter measurement
(which takes on the order of 1 minute) is, typically, 0.6" (RSD) (Johnson 1981).
As in the case of the enrichment measurements, however, the plant is likely to
rely on pellet sampling and analysis and total pellet weight, obtained during
the loading of the rod, for its accountability data.
 30

2.6 Documentation

To ensure accuracy and a permanent record, a facility operator will nor-

mally establish a system for documenting his measurement methods and results. A
properly organized and maintained record system documenting the measurements and
calculations entering into the determination of all components of the material
balance will also facilitate the application of safeguards by the IAEA. Such a
system will enable the IAEA to achieve the following objectives with the minimum
intrusiveness and burdening of the operator;

1. Verification of the material balance and its individual components

(all strata of receipts, shipments, and inventory).

2. Evaluation of the material-unaccounted-for and its uncertainty and as-

sessment of the significance of the former in terms of the latter.

3. Resolution of anomalies.

Five important elements of a measurement documentation system may be

identified. These are a procedures manual for measurements, a procedures manual
for measurement quality, measurement logs, measurement quality logs, and a mate-
rials record system that provides traceability for all measurement results to
source data. Source data are defined as "those data, recorded during measure-
ment or calibration or used to derive empirical relationships, which identify nu-
clear material and provide batch data. 'Source data' may include, for example,
weight of compounds, conversion factors to determine weight of element, specific
gravity, element concentration, isotopic ratios, relationship between volume and
manometer readings and relationship between plutonium produced and power
generated" (IAEA/SG/INF/1, para. 127).

2.6.1. Measurement Procedures Manual

All measurement methods should be described by written procedures, These

should describe the procedures for taking and analyzing samples, bulk
measurements, and NDA measurements. Changes in procedures should be recorded,
and the manual should provide for an administrative mechanism whereby such
changes can be authorized and communicated to measurement personnel, to ensure
that current methods are being used.

The manual should also describe calibration and standardization procedures,

specifying the role of standard reference materials and other kinds of
standards. These topics may be treated more fully in the measurement quality
manua1.

2.6.2. Measurement Quality Procedures Manual

A measurement quality procedures manual should describe procedures for the

control of each component of the measurement system. It should describe calibra-
tion and standardization procedures, the preparation, use, and handling of
standards, the frequency of re-calibrations or checks on calibrations and perfor-
mance of the system, and the use of control charts or other statistical methods
of ascertaining whether measurements are in control. It should also describe,
for each measurement method, procedures for determining random and systematic
 31

errors and biases, and specify the frequency with which these determinations
should be made. It should provide criteria and procedures for the removal from
service of malfunctioning equipment, for the qualification of new equipment and
methods, and for the re-qualification of equipment being returned to service.

Since procedures may change from time to time, a convenient form for bot.h
the measurement and measurement quality procedures manuals is a loose-leaf
notebook. Appropriate controls should be established to assure that such
notebooks contain only current pages.

2.6.3. Measurement Logs

Logs provide a record of the day-to-day operations of an analytical labora-

tory and other measurement procedures. Typically, they record the date, re-
ceipt, identity, point of origin, form, and quantity of samples sent for anal-
ysis, the type of analysis requested, the actual analytical procedure used, its
results, and the signature of the analyst(s). Separate logs may be kept for
each major piece of equipment (e.g., mass spectrometers). Print outs from auto-
matic instruments should be made a part of the record. The information kept in
the logs should be detailed enough that an auditor can trace a measurement re-
sult for any specific sample to the laboratory record of its actual measurement.
Further, through the combined use of the measurement logs and the quality-
control logs (see below) it should be possible to trace the measurement to
nationally accepted standard reference materials and procedures.

Since a measurement log is a chronological record of events, it usually con-

sists of a bound notebook with numbered pages.'.

2.6.4. Quality-Control Logs

The day-to-day operation of the measurement quality control system should

be recorded in a log. This is discussed further in Section 3.5.3. Like measure-
ment logs, measurement control logs usually consist of bound notebooks with seri-
ally numbered pages and, possibly, supplementary loose records relating to main-
tenance or daily checks of individual instruments.

2.6.5. Material Records

In addition to being recorded permanently in the measurement and measure-

ment-quality logs, measurement results must be transmitted in timely fashion to
those organizational units that use the data to establish inventory change data,
batch data, or physical inventory data for each MBA, or to control the fabrica-
tion process. At these points they will be entered into the various general,
subsidiary, and transaction ledgers (see Section 4.2). The transmitted records,
more properly termed reports, should not only contain the information necessary
for the operator's accounting -and process control activities, but should facili-
tate the IAEA's record examination. In particular, they should permit both the
operator and an inspector to trace the results of measurements of material quan-
tity and chemical and isotopic composition to their source in the log books and
other documents discussed in the previous sections.

An example of a measurement report is given in Table 2.4.

 32

Table 2.4. Example of a Measurement Report

1. Header Information

1.1. MR #
1.2. Date
1.3. Sample identity (material description, batch, item #,
sample #)

2. Measurement Information

2.1. Date of measurement

2.2. Variable being measured (e.g., concentration or amount of
element, enrichment, etc.)
2.3. Measurement type
2.4. Measurement instrument (type and serial #)
2.5. Sampling technique
2.6. Results
2.7. Uncertainty
2.8. Reference document(s)
2.9. Signature of analyst
 33

Similarly, estimates of measurement uncertainty must be transmitted to

those charged with responsibility of computing the uncertainty in the material
unaccounted for (a*jup) for each MBA, comparing shipper's and receiver's values,
and maintaining process control. An example of a measurement uncertainty report
is shown in Table 2.5. As with the measurement reports, the information
supplied should be sufficient to enable an inspector to trace uncertainties back
to the source data.

Copies of both the measurement and measurement uncertainty reports are also
sent to the facility's central records department. Entry into central records
may be either manual or by computer terminal, depending on the system in use-
An additional form of importance in establishing an audit trail is the
identifying tag or label that ordinarily will accompany each item of nuclear ma-
terial. The information it bears may consist simply of an identification number
which may be used to recall additional information from the central record sys-
tem, or it may contain the information shown in Table 2.6., or a subset thereof,
or possibly even additional or other information, depending on the needs of the
operator.

The material records system is described in greater detail in the detailed

description of the Records and Reports Element.

2.7 Example

A description of the reference facility is given in Appendix A. A block di-

agram showing the. process, one possible choice for the MBA structure, and the
key measurement points for verification is presented in Figure A.I.

It should be emphasized that the MBA choice is only one of a number of pos-
sible alternatives. For example, the entire facility might be represented by a
single MBA, with rod loading, assembling, and storage being designated as KMPs;
likewise, the shipper-receiver MBA could be replaced by shipper-receiver KMPs.
On the other hand, if there are a number of parallel fabrication lines, one for
each enrichment, for example, it might be convenient to make each line a sepa-
rate MBA.

Another point to be noted is that the operator may choose to have any num-
ber of measurement points and material control areas for process control and to
satisfy domestic accounting and physical protection requirements. The KMPs
shown in Figure A.I are those that, in this example, would provide access to the
IAEA for its routine verification activities.

Possible measurement methods used at the flow and inventory KMPs are shown
in Tables A.I and A.2, respectively. Also shown are the container types and the
sampling methods. The methods shown are those that might be used by the opera-
tor. The IAEA's verification methods are not necessarily the same; usually, for
example, it will make greater use of NDA, which is cheaper, faster, anc* more con-
venient, though not as accurate as analytical methods.

Container and batch size information is presented in Table A.3. Also shown
are the locations (KMPs) at which each material type is measured.
 34

Table 2.5. Example of a Measurement Uncertainty Report

1. Date
2. Batch, item and sample identity
3. Material description
4. Variable
5. Estimate of random and systematic errors and bias
6. Standards used
7. Reference document(s)

3. Signature of analyst

Table 2.6. Example of Item Label

1. Identification #*
2. Batch #
3. Material description code
4. Gross weight
5. Tare weight
6. Element weight
7. Enrichment

*May be the only data element on the label.

 35

If the material is not to be processed further for some time after the mea-
surements at a particular KMP but is to be stored, it may be convenient for the
the IAEA to seal the container with tamper-indicating seals. At some future
time, such as a physical inventory taking or verification, the integrity of the
seal will assure the continued validity of the measurements, making re-
measurement unnecessary and reducing the burden on both the operator and the
IAEA. Sealing can also reduce the effort required to resolve anomalies.

Some typical accuracies and precisions for each type of measurement are
shown in Table 2.7. The types of standards used and the frequency of calibra-
tion, as well as the frequency of determination of accuracy and precision, are
also given. Biases are not shown, since the measurement data are assumed to be
corrected for these. Methods of determining random and systematic error and
bias are discussed in detail in the guidelines description of the Measurement
Quality Element.

To propagate uncertainties like those shown in Table 2.7. into the material
balance and material unaccounted for, it is necessary to break the flows and
inventories down into the individual measurement strata, and to know how many
measurements are made on each stratum. Random errors are then added in
quadrature (i.e., by taking the square root of the sum of the squares of the in-
dividual errors) for the stratum, while systematic errors are applied to the
stratum as a whole. This results in an uncertainty for the entire stratum,
which is combined with that for each of the other strata, taking due account of
covariances, to yield an overall uncertainty, C^OT, in the MTJF. The procedures
for accomplishing this are explained in the detailed description of the Measure-
ment Quality and Material Balance Closing Elements.
 Toi»le 2.7. Typical Operator's Measurement Methods and Accuracies at LEU Conversion and
Fuel Fabrication Plant

Accuracy Appraisal
Measurement or Errors Calibration
Sampling Standards Replicate
Method Random Systematic Frequency Standards Frequency Analysis Freq.

Analytical

Gravimetric 0.02X 0.01Z Monthly NBS U 0 2 2 per week 15 every six

standard months

Titration 2Z 0.2Z Monthly NBS U02 2 per week 15 every six

standard months

Fluorometric 25% 2% Monthly NBS UO2 2 per week 15 every six

standard months

NDA

Bucket y-ray 15* 25Ï Monthly 1,5,10,50 gra 2 per week 15 every aix
internal months
bucket
standards

Barrel y-xay 10X 25X Monthly 10,20,50,100 2 per week 15 every six
gm internal months
barrel
standards

Can Y ~ r a y 10Z 10Z Monthly 5,10,25,50 gm 2 per Week 15 every six

internal can months
Standards

Survey meter 10Z 15Z Semi-annual 5,10,20 gm 2 per week 15 every eix
internal hold- months
up standards
 Table 2.7. Typical Operator's Measurement Methods and Accuracies at LEU Conversion and
Fuel Fabrication Plant (Cont'd)

Accuracy Appraisal
Measurement or Errors Calibration
Sampling Standards Replicate
Method Random Systematic Frequency Standards Frequency Analysis Freq.

Dulk

OF, cylinder scale 0.5 kg 1-0 kg Monthly UP5 cylinder 2 per week 15 every six
artifact months
standards

Powder scale 20 gm 10 gm Monthly 10 kg, 50 kg 2 per week 15 every six

weights months

Tray scale 10 gm 5 gm Monthly 1 kg, 10 kg 2 per week 15 every six

weights months

Pellet stack scale 0.3 gm 0.2 gm Monthly 1 kg, 5 kg 2 per week 15 every B I X
weights months

Scrap scale 10 gm 5 gm Monthly 1 kg, 10 kg 2 per week 15 every six

weights mouths

Recycle powder ecale 20 gra 10 gm Monthly 10 kg, 50 kg 2 per week 15 every six
weights months

Boat scale 10 gm 5 gm Monthly 1 kg, 10 kg 2 per week 15 every six

weights months

Liquid level 10 liters 5 liters Monthly Known amounts 2 per week 15 every six
of II2O months
 Table 2.7. Typical Operator's Measurement Methods and Accuracies at LEU Conversion and
Fuel Fabrication Plant (Cont'd)

Accuracy Appraisal
Measurement or Errors Calibration
Sampling Standards Replicate
Method Random Systematic Frequency Standards Frequency Analysis Freq.

Sampling

0.2% Semi-annual Special materials Semi-annual 15 every 6 inos.

internally prepared

Sintered Pellets 0.2% Semi-annual Special materials Semi-annual 15 every 6 inos.

internally prepared

AMI 5% 3% Semi-annual Special materials Semi-annual 15 every 6 D O S .

internally prepared

Sludge 3% 2% Semi-annual Special materials Semi-annual 15 every 6 mos.

internally prepared

Dry Scrap 3% 2% Semi-annual Special materials Semi-annual 15 every 6 mos.

internally prepared

Recycle U0 2 0.3% 0 Semi-annual Special materials Semi-annual 15 every 6 mos.

internally prepared

Liquid Waste 4% 2% Semi-annual Special materials Semi-annual 15 every 6 mos.

internally prepared

Isotopic

Mass Spectrometry 0.2% relative 0.2% relative Monthly NBS standard 2 per week 15 every 6 mos.
 39

References for Part 2.

ANSI N15.7: ANSI N15.7-1972, Analytical Procedures for Accountabilitv of Ura-

nium Hexafloride, American National Standards Institute, Inc., New York,
1972.

ANSI S15.18: ANSI N15.18-1975, Mass Calibration Techniques for Nuclear Material
Control, American National Standards Institute, Inc., New York, 1975.

ANSI N15.19: ANSI N15.19-75, Volume Calibration Techniques for Nuclear Material
Control, American National Standards Institute, Inc., New York, 1975.

ANSI N15.20: ANSI N15.20-1975, Guide to Calibrating Nondestructive Assay

Systems, American National Standards Institute, Inc., New York, 1975.

ASTM 83: Annual Book of ASTM Standards: Part 45: Nuclear Standards, American
Society for Testing and Materials, Philadelphia, Pa., 1983.

Bingham 1975: C D . Bingham, J.M. Scarborough, and C E . Pietri, Methods of Sam-

ple Preparation and Analysis for Wide Variations in Materials Types - A Re-
quirement for a National or an International Safeguards Laboratory, pp.
107-155 in Vol. II of Safeguarding Nuclear Materials. Proceedings of a Sym-
posium organized by the International Atomic Energy Agency and held in
Vienna Oct. 20-24, 1975 (Vienna, 1976).

de Bièvre 1982: P. de Bièvre, H. Meyer, J. Van Audenhove, B.S. Carpenter, and

J.W. Gramlich, Progress Report on the Establishment of Uranium Isotope Abun-
dance Reference Materials for NDA, pp. 65-74, ESARDA 15.

Brouns 1978 : R.J. Brouns, F.P. Roberts, and C.L. Upson, Considerations for Sam-
pling Nuclear Materials for SNM Accounting Measurements, NUREG/CR-0087,
U.S. Nuclear Regulatory Commission, 1978.

Eberle 1970: A.R. Eberle, M.W. Lerner, C.G. Goldbeck, and C.J. Rodden, USAEC Re-
port NBL-252, 1970.

Ehinger 1979: M.H. Ehinger, J.M. Crawford, and M.L. Madeen, Redundant Measure-
ments for Controlling Errors, Nuclear Materials Management, Vol. VIII, Pro-
ceedings Issue, 504 (1979).

Ellis 1981: J. Ellis, Survey of Bulk-Measurement Technology and Measurement

Standards, in Los Alamos National Laboratory report LA-8901-C.

EUR 6089 EN: C. Bibliocca, M. Cuypers, and J. Ley, List of Reference Materials
for Nondestructive Assay of U, Th, and Pu Isotopes, CEC report EDR 6089 EN,
Brussels, 1979.

Guzzi 1982: G. Guzzi, In-Field Use of a Quadrupole Mass Spectrometer for Iso-
topic Analysis of PFs, ESARDA Bulletin No. 3, 8 (1982). European
Safeguards Research and Development Organization.
 40

Hageauer 1982: R. Hagenauer, Nondestructive Determination of Uranium Enrichment

Using, Low Energy X and Gamma Rays, Nuclear Materials Management, Vol. XI,
Proceedings Issue, 216 (1982).

Hakkila 1981: E.A. Hakkila, R.T. Gutmacher, C.C. Thomas, Jr., A Survey of Chemi-
cal and Destructive Analysis Methods, in Los Alamos National Laboratory
Report LA-8901-C.

Hatcher 1981: C. Hatcher and H. Smith, Elements of Nondestructive Assay (NDA)

Technology, in Los Alamos National Laboratory Report LA-8901-C.

Hough 1979: G. Hough, T. Shea, and D. Tolchenkov, Technical Criteria for the Ap-
plication of IAEA Safeguards, in Vol. I of the Proceedings of a Symposium
on Nuclear Safeguards Technology, Vienna, 2-6 Oct. 1978 (IAEA, Vienna,
1978).

IAEA/SG/INF/1: IAEA Safeguards Glossary, IAEA, 1980.

IAEA-174: IAEA Safeguards Technical Manual, Part S, IAEA, Vienna, 1976.

Indusi 1974: J.P. Indusi and W. Marcuse, A Method to Determine the Minimum Cost
Measurement Plan Consistent with any Feasible Limit of Error on MOF,
Nuclear Materials Management, Vol. Ill, No. Ill, Proceedings Issue, 269
(1974).

Jackson 1976: D.D. Jackson, D.J. Hodgkins, R.M. Hollen, and J.E. Rein,
Automated Spectrophotometer for Plutonium and Uranium Determination, Los
Alamos National Laboratory Report LA-6091, February, 1976.

Jaech 1973: John L. Jaech, Statistical Methods in Nuclear Material Control,

TID-26298, U.S. Atomic Energy Commission, 1973.

Johnson 1981: K. Johnson, NDA Methods Used at Plant, in Los Alamos National Lab-
oratory Report LA-8901-C.

Keisch 1980: B. Keisch and S. Suda, Temperature Effects in Dip-tube Manometrv,

Nuclear Materials Management, Vol. IX, Proceedings Issue, 148 (1980).

Keller 1967: E.L. Keller (Ed.), Uranium Hexafluoride, Handling Procedures and
Container Criteria, ORO-651 Rev. 1, Union Carbide Corp., 1967.

LA-5652-M: E. Ray Martin, David F. Jones, and L.G. Speir, Passive Segmented
Gamma Scan Operation Manual, Los Alamos National Laboratory Report LA-
5652-M, July 1974.

LA-8901-C: Proceedings of an Advanced International Training Course on State

Systems of Accounting for and Control of Nuclear Materials, Los Alamos Na-
tional Laboratory Report LA-8901-C, 1981.

Lovett 1974: James E. Lovett, Nuclear Materials: Accountability. Management,

Safeguards. American Nuclear Society, 1974.
Milner 1982: G.W.C. Milner, Certified Reference Materials for Safeguards. A Re-
view of their Preparation, Characterization, and Certification, pp. 59-63,
ESARDA 15.

NBS 260-27: E.L. Garner, L.A. Machlan, W.R. Shields, Uranium Isotopic Standard
Reference Materials, NBS Special Publication 260-27, National Bureau of
Standards, Washington, D.C. 1971.

Pietri 1978: C E . Pietri, J.S. Palier, and C D . Bingham, The Chemical and Iso-
topic Analysis of Uranium. Plutonium, and Thorium in Nuclear Fuel
Materials, pp. 1-18 of NBS-528, Analytical Methods for Safeguards and
Accountability Measurements of Special Nuclear Material, Proceedings of an
ANS Topical Meeting held May 15-17, 1978, Williamsburg, VA.

Reilly 1977: T.D. Reilly and M.L. Evans, Measurement Reliability for Nuclear Ma-
terial Assay, Los Alamos National Laboratory report LA-6574, January, 1977.

Rodden 1972: C.J. Rodden (Ed.), Selected Measurement Methods for Plutonium and
Uranium in the Nuclear Fuel Cycle, U.S.A.E.C., 1972.

Rogers 1982: D.R. Rogers, SNM Measurement Methods: The State of the Practice,
Nuclear Materials Management, Vol. XI, p. 209, Institute for Nuclear Mate-
rials Management, 1982.

Strohmeier 1974: W.O. Strohmeier, Turbine Flowmeters, Past, Present, and Fu-
ture, in Flow: Its Measurement and Control in Science and Industry, N.E.
Wendt, Jr. (Ed.), Instrument Society of America, Vol. I, Part 2, pp.
687-693, Pittsubrgh, 1974.

Wing 1981: N.S. Wing, Analytical Methods Used at Plant, in LA-8901-C.

Zook 1981: A. Creig Zook, L.H. Collins, and C E . Pietri, Determination of

Nanogram Quantities of Uranium by Pulsed-Laser Fluorometry, Mikrochimica
Acta 1981 II, pp. 457-468.
 42

Part 3. MEASUREMENT QUALITY

3.1 Introduction

An essential element in any nuclear material control and accounting system

is a program for the control of the quality of the measurements of material quan-
tity. A knowledge of measurement uncertainties, biases, sources and magnitudes
of random and systematic errors, and measures to control the quality of measure-
ment are important for process control, product quality, health and safety, and
domestic safeguards. They are also essential for international safeguards,
since the evaluation of material unaccounted for and the resolution of individ-
ual measurement anomalies may both depend upon a knowledge of the inherent mea-
surement errors. The extent and complexity of the programme will depend, inter
alia, on the type of nuclear activity and the form and quantities of nuclear ma-
terial. This section will provide information to assist the facility operator
in establishing a measurement quality programme including technical aspects as
well as administrative aspects such as organization, personnel qualification and
training, and documentation.

Measures for the evaluation of precision and accuracy of measurements and

the estimation of measurement uncertainty as well as procedures to control the
quality of measurements can be translated into three objectives of a facility
measurement quality programme as follows:

1. To provide a sound technical basis for estimation of material-

accountancy measurement precision and accuracy.

2. To ensure the control of the quality of material-accountancy

measurements.

3. To provide documented evidence that material-accountancy measurements

have met quality criteria.

Implicit in the above objective is the need for definition of the quality
criteria that the material accountancy measurments are expected to meet. One
definition of such quality criteria for international safeguards has been
stated in the Safeguards Glossary (IAEA/SG/INF/1). Under the term
"International standards of accountancy" there are identified some expected mea-
surement uncertainties in terms of the uncertainty of MUF for some bulk facility
types. These have not been set forth as requirements or as established stan-
dards but provide an indication of what is expected, or at least what is
considered achievable, for an individual material balance. The expected uncer-
tainty of MUF for a fuel fabrication plant is 0.32.

It is not the purpose of this document to establish such quality criteria

for measurements systems. Rather this document will address the Measurement
Quality Programme and the measures that can be used to determine measurement
quality in terms of individual measurements and material balance uncertainty.
The measures addressed can be used to determine whether a measurement system
meets established criteria or to determine the criteria that the system is capa-
ble of meeting.
 While the ultimate goal is the determination of the material balance uncer-
tainty, this must be done on the basis of the uncertainties of the elements of
the measurement processes at the individual KMPs. These uncertainties are
combined appropriately to produce the uncertainties of the material balance com-
ponents and the uncertainty of the material balance, O>jup* '^ne selection of in-
dividual KMPs and of corresponding measurement processes, with their associated
uncertainties, are closely interrelated. This relationship is important in de-
signing the entire measurement system to achieve an acceptable uncertainty in
the material balance with the use of the resources available to the operator.

AS pointed out above in addition to the measurements essential to the

IAEA's safeguards programme, the operators of the facility must also necessarily
carry out measurements to satisfy their own requirements for materials account-
ing, quality and process control, and equity in transactions with vendors of
feed stock or recipients of finished products. Since most such measurements are
identical or similar to those required by the safeguards programme, the design
of the entire system to achieve these two purposes with a minimum of effort
would be advantageous. The determination of and control of the quality of the
elements of the measurement process will be addressed in this Part. The treat-
ment of these uncertainties to produce the material balance uncertainty will be
addressed in detail in the description of the Material Balance Closing Element
in Part 6.

3.2. Svstem Qualification

An important factor in the initial choice of a measurement method by the op-

erator is the accuracy and precision of which the method is capable. A number
of compilations of the expected capabilities of the measurement methods used in
LED conversion fabrication plans have been published (Reilly 1977; Rodden 1972:
Pietri 1978; Hough 1982; LA 8901-C; IAEA-174), which the operator may use in
selecting the methods for his plant. Data on accuracies and precisions have
been given in Section 2.5. The IAEA has published a set of "target" values for
the systematic and random uncertainties, as relative standard deviations, for
various measurement methods and material categories (Hough 1982). Some of these
are extracted in Tables 3.1 and 3.2 for analytical measurements of some of the
major materials appearing in an LEU fuel fabrication plant. The uncertainties
shown are those arising from both sample treatment and analytical measurement,
but do not include sampling errors. The uncertainties are roughly the same as
or somewhat larger than those given in Sec. 2.5.3.

Using a computer model of the measurement system of the plant and the tar-
get values shown in the tables, it has been demonstrated (Hough 1982) that a
°MUF °f •^•22 relative or 16 kg D-235 at an enrichment of 3" can be achieved in
a plant with a throughput of 250 tonnes of LEU per year, only slightly smaller
than the model plant under consideration. Therefore, if the operator is guided
in his selection of measurement methods by the data in Tables 3.1 and 3.2 and if
the selected methods actually do perform at least as well in practice, his mea-
surement system should perform as expected in the statement of the international
standard of accountancy.

To establish the performance of the measurement system it is necessary to

test or qualify it before it is put into routine use in the plant. Such tests
are also necessary when individual components or measurement methods are
 44

Table 3.1. Target Values for Standard Deviation (7. Relative) of

Analytical Measurements of Element and Concentration in
Fuel Materials in LEU Fabrication Facility
(extracted from Hough 1982)

R = Random; S =* Systematic

Method Titrimetry Coulometry Gravimetry

Material R S R S R S

U-Oxide Powders 0.15 0.15 0.15 0.15 0.1 0.1

U0 2 Pellets 0.15 0.1 0.15 0.15 0.05 0.1

U-Oxide Scrap 0.3 0.5 0.3 0.5 — —

U-Oxide Dirty Scrap 0.3 2.0 — — — —

Table 3.2. Target Values for Standard Deviation (" Relative)

of U-235 Determinations in LEU Fabrication Facility
(extracted from Hough 1982)

R = Random; S * Systematic

Abundance Gas Mass Soec. Thermal Ion. Mass Soec. Gamma Suec.
(2) R S R S R S

U 2 3 5 , All 0.7 0 .2 0 .2 1 1 1
Materials 3 0.1 0 .15 0.5 0.5 0.5
changed. The qualification process consists of testing the equipment over the
full range of operation under the expected environmental conditions to ensure
that it meets specifications, calibrating it, determining the random and system-
atic errors and bias of each measurement method, and documenting the procedures
and results. This will demonstrate that the overall measurement goals, ex-
press-ed as the standard deviation of the material unaccounted for, can be
achieved.

In Sec. 2.5 the various categories of measurements — bulk, analytical

(both chemical and isotopic), and nondestructive — as well as the sampling pro-
cess were discussed. Qualification of these methods would determine the random
and systematic uncertainties and biases associated with them.

Thus, the sampling procedures can be qualified during the pre-start up

phase of the plant, when natural uranium is used to test the fabrication pro-
cess. A total of n replicate samples can be taken by each of the proposed-
sampling techniques and analyzed for uranium concentration. This will give the
series of results x^,X2>••.,Xn, say, with mean value x. If the random
analytical error is small compared with the random sampling error, the latter is
given by

(1)
i-1 n-1

where a r is the random sampling standard deviation. For reliable results £

Too large a o r suggests that the sampling procedure is in need of improvement.
Once the plant is operating routinely the tests should be repeated at intervals
to ensure that the sampling process is still under control.

If the analytical uncertainties in the sampling tests are not negligible,

then either they have to be determined in a separate experiment or analysis of
variance techniques must be used to separate them from the sampling errors.

The performance of the equipment and the various error components in bulk
measurements can also be determined by pre-operational qualification. For
weighings the procedure is straightforward, using accurately known standard
weights which have been related to nationally accepted standards. Repeated
weighings of the standard weight and comparison of the mean with the standard
value determines the bias of the particular scale or balance being tested. If
the weights obtained from the repeated weighings are xj, X2 S ...x^, with mean
.-alue x, and the weight of the standard is x 0 , the bias of the weighing device
is x-x 0 . The uncertainty in the bias, or the systematic error, is

n
I (2)
n(n-l)
 46

where a 0 is the uncertainty in the weight of the standard. The random error
in the weighing procedure is then

Again, for reliable results, n £10.

Volume measurements of a tank usually require a calibration over the entire

range of volumes (Roddem 1972). The calibration may be obtained by adding known
amounts of a weighed liquid to the tank and reading the corresponding liquid
level by means of a sight glass or D-tube manometer. A curve is then fitted to
the experimental points by standard mathematical techniques such as least-
sq,uares. The fitting procedure usually provides not only the parameters (e.g.,
slope and intercept for a linear relationship) but their uncertainties as well,
or else the calibration can be repeated a number of times and the observed dis-
tribution of the parameters used to compute the uncertainties. These
uncertainties then introduce a systematic error into subsequent determinations
of volumes by means of the composite calibration curve, while the distribution
of results for repeated measurements of the same volume provides the random
error.

The bias and random and systematic errors of analytical measurements can be
determined in a manner similar to that described for weighings, by the use of
nationally accepted chemical and isotopic standards and replicate measurements.
Other important pre-operational qualification activities (which could also be a
part of a continuing measurement quality control program) could include compari-
sons with the results of alternative and independent measurement methods,
interlaboratory comparisons, and measurement round robins involving several
laboratories.

In the use of analytical procedures at a given facility, a number of fac-

tors which can influence the quality of the results must be taken into considera-
tion:

Properties, such as composition and concentration, of the standard material

must be comparable to those of the material routinely analyzed. For materials
which undergo dissolution, the chemical state of the original material, for
example uranium metal, UO2, or D3O3, may be of no consequence. For certain
analytical procedures, for example, fluorometry, the presence of interfering con-
stituents may have a significant influence on the measurements. In instances
where such constituents may be present, and especially in those cases in which
the concentration of those constituents may fluctuate with time, their influence
on the measurements must be investigated. Usually the presence of trace ele-
ments which may interfere with analytical measurements is determined by emission
spectroscopy.

The occurrence of effects which would perturb a calibration should be

detected, or preferably, prevented, for example, the incomplete dissolution of
a sample, contamination of a standard, absorption of moisture leading to a
 47

weighing error, or a change in the stoichiometry of a uranium oxide sample.

Other sources of error are, for example, the evaporation of water from aliquoted
samples or uncorrected changes in density arising from changes in temperature.

These, and other possible sources of error can be minimized and the relia-
bility of calibrations and measurements assured through the use of well
documented standard procedures which have been thoroughly tested and qualified.

Determining the bias and systematic'error of nondestructive assay methods

is usually less straightforward than for analytical methods, because representa-
tive standards may not be readily available for the full range of categories of
material. In the absence of such standards, possible alternatives for the deter-
mination of these error components in the qualification phase are total recovery
of the samples (the most accurate but also the most expensive method), compari-
son of the results of a particular NDA method with those of another, more accu-
rate method, either at the facility or at another laboratory, or participation
in round-robins involving the exchange of well characterized standards between
laboratories.

The qualification program should establish the capability of the measure-

ment system to perform as expected under the conditions existing in the plant in
routine operation. This would include not only the range of expected environmen-
tal conditions (e.g., temperature, humidity, vibration, electrical voltage, fre-
quency, and noise) but the range of performance of personnel responsible for
performing the measurements. It is possible, for example, that the accuracy and
precision actually attained with a method will depend on the analyst. A
properly designed qualification program would be able to determine such
differences, if they exist. The pre-operational qualification program should
also ensure that measurement personnel have the proper technical qualification
and have been trained in the specific procedures to be used for measurements.

To summarize, a qualification program should include the following

elements:

(1) Acquisition of the necessary standards (mass standards, chemical and

isotopic standards, etc.) and preparation of working standards.

(2) Preparation of written procedures for testing measurement equipment

and methods to ensure that they meet specifications and achieve the
desired performance levels under the expected environmental
conditions.

(3) The actual testing of equipment and procedures to determine systematic

and random error and bias, and also to establish the reliability of
the equipment under the actual range of operating conditions.

(4) The documentation of results in a form that will facilitate review and
evaluation.

3.3 Standards and Calibrations

For the comparison of a measurement system with international standards it

is implicit that standards exist with which the results of facility measurements
 48

are compared or with which facility measurement processes are calibrated. There
are two general types of such standards. These are standard reference materials
and mass or weight standards. Standard reference materials are well
characterized materials that are used to calibrate a measurement system or to
produce scientific data that can be readily referred to as a common base (Rodden
1972; IAEA-174). Standard weights are those used in determining and controlling
the mass measurement processes. For each type of standard there is a hierarchy
of standards that can be used. For standard reference materials this hierarchy
goes from atomic weight standard, to ultimate standard, primary standard, and
working standard. The definition, preparation and use of these standards is
discussed in detail in Chapter 2 of Rodden 1972 and also in Chapter 2 of IAEA-
174. The hierarchy for mass calibration standards goes from primary standard
weight to secondary and tertiary standard weights, replica standards, and the
calibration balance. The definition, preparation, and use of these stanoards is
discussed in detail in Chapter 4 of Rodden 1972 and Chapter 3 of IAEA-174.
IAEA-174 also discusses the use of standard reference materials for non-
destructive analysis application.

As has been noted in Section 2.5, it is the hierarchy of these standards

that provides the traceability of a facility's measurement system results to a
national or international measurement system or standard. Each level of the
standards hierarchy is measured against the next level so that the working stan-
dards used in routine facility measurement processes can be identified with the
certified higher levels of the standards hierarchy. The means for accomplishing
this are discussed in detail in both Rodden 1972 (Chapters 2 and 4) and IAEA-174
(Chapters 2 and 3 ) .

The measurements performed on special nuclear materials in the model plant

under consideration may conveniently be divided into three categories for the
purpose of discussing their standards and calibration. These are: (1) bulk:
weight and volume; (2) analytical; and (3) nondestructive analysis (NDA). These
will be considered in turn.

3.3.1. Bulk Measurements

3.3.1.1. Weight

The materials to be weighed include the feedstock materials UFg

(cylinders: 2 to 12 tonnes), uranyl nitrate (T7NH) (tanks: up to 20 tonnes), D 0 2
powder (hoppers: 1 tonne; buckets: 25 kg; Cans: 5 k g ) , DO2 pellets (trays: 5
kg; stacks: 5 kg), and scrap (5 to 50 kg). In addition, small samples of these
materials are taken at various KMPs and weighed before they are analyzed. It is
thus necessary to establish standards and calibrations for several different ma-
terials with weights extending over many orders of magnitude.

The establishment of standards of mass is one of the oldest func-

tions of national standards laboratories. Current practices and typical
attainable accuracies are summarized, for example, in National Bureau of Stan-
dards Circular 547, "Precision Laboratory Standards of Mass and Laboratory
Weights" (NBS 547). Some typical series of standard weights tested and
calibrated by NBS with utility for safeguards measurements are: Class M
 49

(reference standards): range 0.05 mg to 25 kg, tolerance* from 0.0054 mg below

1 gn to 5 ppm at 25 kg; Class S (working standards): range 0.05 mg to 25 kg,
tolerance from 0.014 mg below 100 mg to 2.5 ppm at 25 kg; Class P (laboratory
weight): range 0.05 mg to 1000 kg, tolerance 20 ppm from 0.1 to 1000 kg. A com-
parison of these tolerances with the typical precisions for various weighing de-
vices given in Table 2.3 of the preceeding section, "Nuclear Material
Measurements," shows that with the possible exception of the top loading scale,
the accuracy attainable for the weight standard is in general at least an order
of magnitude better than the precision of the weighing device calibrated.

The specific procedures for the calibration of mass measurements on

nuclear materials, including specific procedures for UFg cylinders, are
addressed in detail in ANSI N15.18-1975, "American National Standard Mass Cali-
bration Techniques for Nuclear Material Control" (ANSI N15.18). General proce-
dures for the calibration and control of weighing devices are discussed in de-
tail in Rodden 1972 (Chapters 2 and 4) and in IAEA-174 (Chapters 2 and 3 ) . Sta-
tistical methods for the calibration of weighing devices and the determination
of the errors of calibration and of routine weighing operations are described in
detail in IAEA-TECDOC-261 (Section 2.6), in Jaech 1973 (Chapter 3) and in Jaech
1978. These descriptions include discussion, with examples, of a variety of con-
ditions that might be encountered by the facility operator. Such conditions as
single standard and multiple standard calibrations, rounding effects and time ef-
fects are addressed. The descriptions also include discussion of the determina-
tion of random errors using standards and under routine operating conditions.
These discussions are not repeated here.

3.3.1-.2. Volume

Two types of volume measurement may be performed: static, in which

the volume of liquid in a calibrated tank is determined from a measurement of
height or pressure, and transfer, in which the flow of liquid past a given point
is measured. In these determinations and in the calibration of the measuring
devices, measurements of mass, pressure, temperature, length, and density may be
combined in order to relate the results to primary standards.

Procedures for the calibration of volume measurement.'- are given in

IAEA-174 (Chap. 3 ) , in Rodden 1972 (Chap.4), and ANSI N15.19-1975, "American Na-
tional Standard Volume Calibration Techniques for Nuclear Material Control"
(ANSI N15.19). When storage vessels are calibrated by measuring directly the
volume of liquid added, either certified test measures, whose volume has been
calibrated by a national standards laboratory, or positive displacement piston
provers, which deliver a measured amount of liquid at each stroke, may be used.
The positive displacement piston is calibrated either with the use of certified
test measures, or alternatively, by disassembly and dimensional metrology
techniques, where the area of the piston and the length of the piston stroke are
accurately determined. The development of such a device was been described by
Shuler (Shuler 1977).

•The tolerance of a weight is the maximum allowable departure of the weight

from its nominal value.
 50

The calibration of flowmeters is similar in principle to other vol-

ume calibrations in that the delivered volume is determined by either
gravimetric or volumetric measurements. Since the response of a flowmeter is
sensitive to the operating conditions, e.g. pressure, flow rate, temperature,
viscosity, and density, as well as the mechanical condition of the meter, these
conditions must be controlled during the calibration and should reproduce the
range of conditions expected during actual operation. (ANSI N15.19)

Statistical methods for calibration of measurement systems such as

volume determination processes are described in detail in IAEA-TECDOC 261
(Section 2.6.2) and in Jaech 1973 (Chapter 3 ) . These decriptions include discus-
sion, with examples, of a variety of conditions such as linear and non-linear
calibrations and single point calibrations. Also discussed are the determina-
tion of random errors using standards and under routine operating conditions. A
number of statistical tests of the validity of linear calibrations carried out
in various ways are discussed in detail by Suda and Shephard (Suda 1976) and in
ANSI N15.19. These discussions will not be repeated here.

Methods for the calibration of volume measurements apply in general

to any instance (not only volume measurements) where a linear calibration is
required; a common example in nuclear materials control is the calibration of ra-
diation detection instruments for the measurement of special nuclear materials.

3.3.2. Analytical Measurements

The principal methods employed for the analysis of uranium at a low

enriched uranium fuel fabrication plant have been presented in the preceding
Part on Nuclear Material Measurements. These methods can be calibrated and
standardized" and their uncertainties estimated through the use of standard
and/or certified reference material. The definition, preparation and use of
standard reference materials is described in detail in Rodden 1972 (Chapter 2)
and IAEA-174 (Chapter 2 ) . Certified reference materials are defined by the In-
ternational Organization for Standardization as follows: "Certified reference
materials are reference materials accompanied by, or traceable to a certificate
stating the property value(s) concerned (and its associated uncertainty) issued
by an organization, public or private, which is generally accepted as techni-
cally competent." (ISO-1977). The hierarchy of standard reference materials and
the use of working standard traceable to national certified standards also are
discussed in these references.

The principal organizations currently supplying certified reference mate-

rials with applicability to safeguards measurements are: The U.S. National Bu-
reau of Standards (NBS), Washington, D.C.; the New Brunswick Laboratory (NBL),
Argonne, Illinois; British Nuclear Fuels Limited (3NFL), Capenhurst, Cheshire,
England; the Central Bureau for Nuclear Measurements (CBNM), Geel, Belgium; the
Commissariat a L'Energie Atomique (CEA), Grenoble and Pierrelatte, France; the
Japan Atomic Energy Research Institute (JAERI), Tokai, Ibaraki, Japan; and the
International Atomic Energy IAEA (IAEA), Seibersdorf, Austria. These organiza-
tions furnish more than 100 certified reference materials suitable for
safeguards measurements. In addition to these suppliers, other organizations
produce a number of secondary standards representative of most materials which
are present in the nuclear fuel cycle.
 51

The standard reference materials currently available are described in a num-

ber of catalogues and publications, for example, the National Bureau of
Standards' Standard Reference Materials Catalog (NBS 260), the New Brunswick
Laboratory's Certified Reference Materials Catalog (NBL 1982), report EUR 6937:
Catalogue of Chemical and Isotopic Nuclear Reference Materials, issued by the
Central Bureau for Nuclear Measurement of the Commission of the European
Communities (EUR 6937), and report ISO/REMCO 89: New Certified Reference Mate-
rials in Current and Planned Production, issued by the International Organiza-
tion for Standardization in Geneva (ISO/REMCO 89). Detailed information on the
preparation and standardization of NBS standard reference materials is given in
NBS Special Publication 260-27 (NBS 260-27).

In Table 3.3 a number of the standard reference materials for the calibra-
tion of analytical procedures used for the determination of uranium or of trace
elements in uranium are listed.

Because the certified reference materials provided by national standards

laboratories and other sources do not reproduce the entire range of materials
which are encountered in most production facilities, and are often expensive or
of limited availability, it is often necessary for a facility operator to de-
velop additional, facility reference materials or working calibration and test
materials which are traceable through a set of comparisons to a standard refer-
ence material. A number of considerations are important in the choice of an ap-
propriate material and in the design of the measurements necessary for its
standardization, for example, the desired uncertainty, the stability of the mate-
rial under various conditions, its comparability with the material routinely
analyzed and suitability for the analytical method employed, and considerations
of convenience and economy.

Examples of the production and standardization of facility reference mate-

rials have been given by Clark and Jackson (Clark 528) and Wong, Cate, and
Pickles (Wong 528). In the work of Clark and Jackson the working calibration ma-
terial was a uranyl nitrate solution, which was compared with a similar refer-
ence material prepared from NBS standard reference material 960 uranium metal.
Weighed aliquots of both materials were prepared and analyzed by both
gravimetric and the modified Davies-Gray methods at the facility and at two inde-
pendent- laboratories as well. The relative limit of error established for the
concentration of uranium in the working standard was 0.032. In the work of
Wong, Cate, and Pickles, very small uranium standards for x-ray fluorescence
analysis (4 to 1000 p gm) were produced from UO2 powder or NBS Standard Refer-
ence Material 960 uranium metal. The samples were produced by dissolving the ma-
terial in nitric acid and weighing aliquots of the solution. Sample masses were
determined with accuracies of JO.2Z.

Standard reference materials specifically applicable to mass spectrometric

analysis consisting of uranium samples of known isotopic composition are avail-
able from a number of sources; for example, CEA, BNFL, NBS, NBL, JAERI, and CBNM
(EUR 6937, NBS 260, NBL 1982). The U-235 enrichment ranges from highly depleted
material to material containing essentially pure (>992) U-235, with a range of
20 to 60 different enrichments available from the major standards institutes.
Reference materials are ordinarily available as uranium metal or oxide, but cer-
tain standards are available as uranyl nitrate or UF4 and UFg. In addition to
samples containing major amounts of masses 235 and 238, and small amounts of
 52

Table 3.3. Standard reference materials available for the determination

of uranium or of impurities in uranium.

Supplier Reference No. Material Comment

CBNM EC-101 D metal CRM

EC-110 U0 2 CRM
EC UO2 pellets WSRM

CEA MUI U metal CRM

- UO2 pellets WSRM
Floralies (several U metal trace elements
standards)
Uranium divers U metal trace elements
Mickey U3O8 trace elements
Etoile UF4 trace elements

IAEA SR-54 D3O8 trace elements

JAERI Ul U3Og trace elements

D2 uU 3 ometal
8
trace elements
U4 CRM
JMC 765 D3O8 trace elements

NBL 17-B 0F4 WSRM

18 D03 WSRM •
112 0 metal trace elements
114 WSRM
121 U(depleted) non-volatile impurities
123(1-7) D3O8 trace elements
124(1-7) U3O8 trace elements

NBS 950b u3o8 CRM

960 U metal CRM
136c Potassium Oxidizing agent
Bichromate

CBNM - Central Bureau for Nuclear Measurements

CEA - Commissariat à L'Energie Atomique
CRM - Certified Reference' Material
IAEA - International Atomic Energy Agency
JAERI - Japan Atomic Energy Research Institute
JMC - Johnson Matthey Co.
NBL - New Brunswick Laboratory
NBS - National Bureau of Standards
WSRM - Working Standard Référença Material
 53

nasses 234 and 236, standards are available which contain tJ-233 and U-235 in a
known ratio for measurements which involve "spiking" or isotopic
dilution with D-233.

The methods employed for the preparation of isotopic standards are

described in NBS Special Publication 260-27 (NBS 260-27). The present standards
were prepared by mixing, in appropriate ratios, known amounts of two batches of
material, one consisting of essentially pure U-235, and the other essentially
pure 0-238. The ratios of the major isotopes were determined with the use of
thermal ionization mass spectrometers, for which sources of bias, chiefly
preferential ionization of the lighter isotope, were thoroughly investigated and
carefully controlled. The minor isotopes present, T3-234 and U-236, were deter-
mined in a separate measurement which employed a "spike" of D-233. The final
uncertainties in the U-235/U-238 ratio are less than 0.1" at the 952 confidence
limit.

There are two general areas of use for standard reference materials in the
determination and control of analytical measurements. One area is routine use
such as using potassium dichromate titrant in a determination or using isotopic
working standards in mass spectrometric analysis. Another area is the use of
standard reference materials to calibrate the methods in the sense of
determining biases and systematic errors. Routine use normally is specified in
the written procedures for the analytical method. This can be seen in the
methods described by Rodden (Rodden 1972). For the determination and control of
bias and systematic error the standard reference materials are anal"^ed by a
given method and the results evaluated on the basis of the measured values
compared to the assigned values. The procedures and statistical methods for cal-
ibration of measurement methods and determination of biases and systematic
errors are described in detail in IAEA-TECDOC-261 (Section 2.6), in Jaech 1973
(Chapter 3 ) , and in KUREG/CR-1284). The descriptions also include discussion of
the determination of random errors using standard reference material and using
non-standard material, i.e. production or process material.

3.3.3. Nondestructive Analysis (NDA)

Because the methods of nondestructive analysis are diverse and involve a va-
riety of different nuclear radiation detection systems, and because the mate-
rials investigated are also diverse, ranging from products such as DO2 pellets
and fuel elements to in-process holdup and waste streams, the task of calibra-
tion and standardization is far more difficult and complex than for more tradi-
tional measurement methods. Despite these difficulties, considerable progress
has been achieved in this area. Recommendation1! for calibration procedures have
been promulgated in the American National Standard "Guide to Calibrating
Nondestructive Assay Systems" (ANSI N15.20).

The requirements for physical standards for NDA techniques were the subject
of a study conducted by an advisory group of the IAEA in 1977 (AG-112). With
respect to facilities processing low-enriched uranium, the recommendation of the
group calls for primary standards of l^Og to calibrate measurements on uranium
fluoride or oxide feedstock materials, secondary standards consisting of pellets
 54

and fuel pins for each relevant enrichment and reactor type, and mock-ups of
fuel assemblies serving as normalization standards.*

A list of reference materials av-ailable in 1978 for the nondestructive

assay of special nuclear materials has been compiled by Bibliocca, Cuypers, and
Ley (EUR 6089 EN). The principal items with relevance to a fuel-fabrication
facility are uranium oxide or fluoride powder in various enrichments, ÏÏO2
pellets, and entire fuel rods.

One of the most useful and widely employed NDA techniques is the determina-
tion of the enrichment of a uranium sample by the measurement of the number of
185.7 keV gamma rays per unit surface area and time emitted in the decay of
U-235 to Th-231. The techique is especially useful because for a "thick" sam-
ple, the 185.7 keV gamma-ray intensity is directly proportional to the
enrichment, and the presence of other materials of low atomic number has very
little effect on the accuracy of the measurement (BNL 50414). The method can be
used either for a coarse attributes measurement, to distinguish natural or
depleted uranium from enriched material, or, with appropriate instrumentation
and methodology, to determine enrichment with high precision. The thickness of
material required to yield 99Z of the gamma rays from a source of infinite thick-
ness ranges from J2im for uranium metal to <^1 cm for

To provide for the calibration of these measurements the United States Na-
tional Bureau of Standards (NBS) and the Central Bureau for Nuclear Measurements
(BNM) at Geel, Belgium, have undertaken a cooperative effort for the production
of certified reference materials for this purpose. (Harry 1979, de Bièvre
1982). The certification will include the isotopie abundance of U-235 and other
sample characteristics which are important to its accurate determination by
gamma-ray spectrometry. Sets of five samples of <^200 gm of* U3O8 sealed in alumi-
num cans with enrichments ranging from 0.3 to 4.52 have been fabricated by
blending materials with known isotopic ratios. The goal of the effort is to pro-
duce standards which are characterized such that the isotopic abundance can be -
determined by high resolution gamma-ray counting with a total uncertainty <0.252
and a precison of <0.12. Measurements on the isotopic abundance of U-235 in the
samples by both mass spectrometry and gamma-ray spectrometry have been performed
and are in agreement to better than 0.12 (de Bièvre 1982). After an initial pe-
riod of evaluation and comparison at a number of laboratories these reference
materials should constitute the standard for the calibration of enrichment mea-
surements by the above technique for the forseeable future.

The problem of providing reference materials for a finished product, i.e.,

fuel rods, has been addressed in a collaboration between Euratom, IÂEÀ, and the
research centers Euratom Ispra, and CEN/SCK, Mol (BLG 522). The objective was
to provide working standards for the calibration and testing of a rod scanner
used to determine the homogeneity of the enrichment of the UOj pellets in fin-
ished fuel rods. The rod scanner is capable of detecting, in about one minute,
pellets which deviate from the nominal enrichment by 10 or 15%, with a 52 false

*A normalization standard is defined as "a sample selected from production

material with a nominal value for the parameter of interest assigned from
production data."
 55

rejection rate. To produce the standards a number of production rods containing

pellets with several different enrichments were dismantled, reference rods
fabricated, and a number of pellets sampled for analysis. In addition, to test
the performance of the rod scanner, additional fuel pellets were fabricated from
UO2 powder samples of different enrichments and introduced into production rods
to simulate rods with off specification pellets. A series of destructive and
non-destructive measurements were carried out on the materials, including
elemental assay, mass spectrometry, gamma-ray enrichment measurements with Nal
and Ge(Li) detectors and a SAM-2 instrument, and determination of the weight of
UOo in the assembled reference fuel rods. The final uncertainties in the U-235
enrichment were *t>.lZ and the total uncertainty in contained U-235 A).15".

At many facilities, a major source of uncertainty in the materials

accountancy is associated with measurements on scrap and waste. This arises
because of the great diversity of nuclear materials, enrichments and
concentrations, the matrix of extraneous materials, and the packaging. These
sources of error can be somewhat reduced through concentration-compaction,
ashing, or chemical separation, and the standardization of materials and
containers. For the calibration of NBA systems dedicated to measurements on
such materials, the reference materials employed should resemble in as many re-
spects as possible the material analyzed and in particular, they should span the
entire range of properties, for example, concentration or isotopic enrichment
possessed by the materials analyzed. The calibration should be checked periodi-
cally by the random selection of an item for verification. Replicate NDA mea-
surements should be performed on this item a number of times sufficient to give
the desired precison of the NDA measurement, and the nuclear material in the
item then recovered for analysis by an independent technique. To provide for
the calibration of NDA systems which monitor these materials, the New Brunswick
Laboratory has initiated the preparation of prototype reference materials (Voeks
582). A total of 19 reference standards containing 932 enriched uranium in vari-
ous concentrations in a matrix of ion exchange resin, cellulose fiber, or syn-
thetic calcined ash were prepared and packaged in appropriate containers. The
materials were characterized by various analytical procedures in comparison with
NBS standard reference materials. The standards have been sent to 14
laboratories in the United States and Europe in an interlaboratory evaluation
program.

IAEA-174 (Chapter 6) describes a number of methods for the nondestructive

assay of nuclear materials. These descriptions include identification of the
calibration mechanisms. Statistical methods for calibration of measurement sys-
tems such as NDA methods are described in detail in IAEA-TECDOC 261 (Section
2.6) and in Jaech 1973 (Chapter 3 ) . These descriptions include discussion,
including examples, of a variety of conditions such as linear and nonlinear
calibrations and single point calibrations. The descriptions also include
discussion of the determination of random errors using standards and under rou-
tine operating conditions.

3.4. Sampling

In an effective measurement control program, particular attention must be

given to the errors, both random and systematic, that originate with the sam-
pling of material for analysis. Sampling errors, unlike the errors of various
 56

measurement techniques, cannot readily be determined and controlled by measure-

ments and calibrations employing standard reference materials.

There are many possible sources of sampling errors: In solid powders or

aggregates, particles in containers or process streams may become segregated by
size. In liquids the concentration of materials may not be uniform throughout
the volume; suspended solids or colloids or multiple phases may exist. The com-
position of a sample that has been taken may change with time through oxidation,
the absorption or evaporation of water, contamination or leakage.

The evaluation of sampling errors and the establishment of procedures for

their minimization and control should be carried out to the fullest extent possi-
ble during the measurement qualification phase before routine operation of the
facility. Thereafter, periodic monitoring of the sampling process is necessary,
as well as a re-evaluation in the event of the introduction of changes in the
system which might lead to new sampling errors. The conditions and factors
which influence the sampling process, such as the type of material, its lot or
batch structure and variability, and accessibility and capability for mixing
should be evaluated. For each item to be sampled, a sampling plan should be
formulated which takes into account the type of material treated and the preci-
sion and accuracy required, and which specifies the sampling technique, the
timing of the sample taking, and the number and size of samples required. For
a given sampling procedure the associated random and systematic errors must be
determined.

Four sources of error contribute to the overall error in the determination

of the properties of a material through sampling and analysis. These are the
systematic and random errors of the sampling process and the systematic and ran-
dom errors associated with the analysis. Each of these sources of error should
be determined independently.

To determine the systematic error of the sampling process, it is necessary

to compare the results obtained with those from an independent sampling method
known to be unbiased. Recommended procedures for obtaining an unbiased sample
include the taking of very large samples> the compositing of many small samples
taken randomly over the entire quantity of material being sampled, and the
homogenization and multiple sampling of all the material (Brouns 1976). The com-
parison of samples taken under different conditions of sampling and treatment
may disclose previously undetected sources of sampling bias.

Random errors in both sampling and analysis can be determined by replicate

measurements. In the simplest example (IAEA-TECDOC-261), if a total of n paired
measurements are performed, and the differences between the results of each pair
of measurements are denoted by d^, then the total random variance for the pro-
cess is given by

n n 2
 57

In this simple example only the total random variance can be determined.
The sampling and analysis procedures can be designed to determine independently
the random variances of the sampling and analysis by taking a certain number,
say m samples and performing n replicate analysis on each sample for a total of
mn analyses. This procedure can be extended further to take account of, for
example, possible differences among several lots of material, among several
analytical procedures used, or among analyses performed at different
laboratories. The standard statistical methods employed to extract the individ-
ual variances are described in detail in ref. IAEA-TECDOC-261, p. 25, et seq.
and in Jaech 1973, p. 101 et seq.

Since a given sampling plan may involve a number of batches or items, the
taking of a number of samples, and replicate analyses, it is advantageous to op-
timize the sampling plan, i.e., for a given total number of analyses, to mini-
mize the total variance or conversely, for a desired variance, to perform a mini-
mum number of analyses. One approach to this problem involves the compositing
of a number of samples in order to permit a smaller number of analyses to be
made. In this procedure a certain amount of information concerning the varia-
tion from sample to sample may be discarded, but if sufficient historical infor-
mation on the variances has been accumulated this information may not be needed.
The general problem of optimizing the sampling plan has been treated in detail
by Gutmacher, et al. (Vlash-1335). The even more general problem of optimizing
the entire measurement system, including the sampling plan, s? as to produce the
minimum cost for any feasible limit of error has been treated by Indusi and
Marcuse, using a non-linear programming technique (Indusi 1974).

3.5 Control Programme

Two objectives can be stated for the control aspect of a facility measure-
ment system: to ensure the continued validity and quality of the measurements,
and to provide documented evidence that the measurements have met established
quality criteria. These objectives can be attained by a properly administered
programme of measurement monitoring and control data evaluation.

3.5.1 Measurement Monitoring

The initial steps in establishing a measurement programme are to determine

the precision and accuracy needed for each measurement, select the methods that
will provide that precision and accuracy and calibrate the methods to provide as-
surance that the methods do perform as expected. When the needed measurement
procedures, including sampling, have been established they should be documented
in complete detail, and should include the detailed steps to be performed, condi-
tions to be avoided, and indications of anomalies to be aware of. Adherence to
the established procedures in routine operations can be assisted by the initial
training and qualification of personnel and by subsequent audits of the measure-
ment processes. However, to provide assurance that established procedures art
being followed and that the measurement processes are performing as expected
monitoring is needed. This monitoring can be accomplished when routine measure-
ments are performed by the remeasurement of standards or known materials. These
measurements may be performed at infrequent intervals using certified or primary
standard reference materials but should be made at more frequent intervals using
secondary or working standards, or on selected production items such as fuel pel-
lets or rods which are reserved for periodic remeasurements. In addition, peri-
 58

odic analysis of a second independent production sample by a second analyst may

be used as a monitoring means. The use of an independent method, i.e. different
from the routine method, may be used as a means of monitoring or investigating
relative measurement bias.

The frequency of measurements on each type of material depends on the indi-

vidual measurement process and the requirements of the facility. Examples of
typical schedules for standards measurements are found in Table 2.7 of Part 2,
on the Measurement System Element and in the presentation by R.A.
Schneider in LA-8901-C (Schneider 1981). The examples given in Sec. 3.7 illus-
trate this control process.

A useful means for investigating the relative measurement biases which may
exist between different facilities, and which in some instances may lead to ship-
per-receiver differences, consists of the exchange of facility working standards
or other materials for analysis by two or more laboratories. Also, an indepen-
dent agency may prepare samples for analysis by a number of facilities and thus
make a comparative evaluation of the results. One example of this approach is
the SALE (Safeguards Analytical Laboratory Evaluation) program carried out by
the New Brunswick Laboratory (NBL-305). In this program four known sample mate-
rials (uranyl nitrate solution, UO2 powder, PuÛ2 powder, and (Pu,U)02 pellets)
are distributed to about 40 participating Laboratories for measurement on a bi-
monthly basis. After analysis and reporting of results the reference values are
given to the laboratories, the results evaluated statistically, with codes
assigned to protect the anonymity of the Laboratories, and reports are issued bi-
monthly.

3.5.2. Control Data Evaluation

Evaluation of data generated by measurements on both routine production ma-

terials and standards can serve control programme objectives. Evaluation of the
data assures continued validity and quality of measurements, and recording of
the data in the evaluation process provides the documented evidence that the mea-
surements have met established criteria. Statistical control limits can be
established for each of the measurement processes over which control needs to be
maintained. A measurement process is considered to be in statistical control
when repeated measurements from the process, or a sample of such measurements,
behave in a random manner in a stable probability distribution, making it possi-
ble to draw conclusions about the process in a statistical sense.

To ensure that a given process remains in control, it is useful to display

the data on a control chart as a function of time. Various statistical parame-
ters may be displayed, typically the mean x, the standard deviation O, or the
range R. For a frequently occurring measurement the value of a measured parame-
ter, or the mean of several measurements, x, may be plotted. The data plotted
usually will be the result of a measurement on a standard. However, measure-
ments on process materials also may be used. Usually two control limits are
drawn on the chart: lines above and below the expected mean value deviating by
±2a and ±3a respectively. The inner limits are at the 95Z confidence level, and
the outer at the 99.7% level, i.e., 5 out of 100 points are expected to fall
outside the 2a con£rol limit but only 3 out of 1000 ouside of the 3a limit. One
common control philosophy is to repeat a measurement if a result falls beyond
the 2cr limit but within the 3a limit. If several subsequent measurements fall
 59

within the 2a limit no action is required; if not the method is declared out
of control, the cause investigated, and remedial action initiated. If one
measurement falls outside the 3o limit the method is declared out of
control, the cause investigated, and remedial action initiated. When isolated
points on the control chart occur far out of the expected range of variation,
i.e., beyond 3o or 4o, the probable cause often may not be a statistical
fluctuation or change in bias but rather human error, for example a gross
mistake in the reading of an instrument or in recording a result. In such a
case, it is useful first to investigate and eliminate or confirm the
possibility of a mistake before addressing the potentially more serious
possibilities of a shift in bias or in the calibration of the system.

Beyond their principal purpose in maintaining the measurement system in a

state of control, control charts also serve to show certain statistical
properties and trends, for example the magnitude of the standard deviation or &
slow shift in the average value of a parameter,

Another useful control chart is the cumulative sum (Cusum) chart (Ewan
1963). Here, starting with some particular measurement, the sum of the devia-
tions of the measurements from a reference value, for example, the accepted
value y of a standard, is plotted versus time:

Cusum = Sj

The series of values plotted would be

C*l-u)
(x2-y)

An important feature of the Cusum chart is the ability to visualize at any

time the apparent mean of an arbitrary group of points. If the two ends of a
group of r points, S N and S N _ r , are connected by a straight line, the mean of
the r values is (Su-Sj^_r)/r, the slope of the line, with the application of an
appropriate scale factor. Thus a change in the mean value of the original param-
eter will be indicated by a change in the slope of the cusum chart, and the num-
ber of points affected will be evident. If the change in slope results in a
persistent trend away from the expected horizontal line it is evidence of a per-
sistent measurement bias. Short term fluctuations arise either from random
errors or short term changes in bias. If the system is in statistical control
and there is no bias relative to the reference value, the Cusun values will be
dispersed about a horizontal line with fluctuations corresponding to the
expected random errors.
 60

Many other examples of the application of control charts in special nuclear

materials control may be found in the literature. (NUREG/CR-1284, NBS-91,
NBS-300).

3.5.3. Documentation

When the measurement control procedures have been evaluated and approved
they should be documented in complete detail including: acquisition,
preparation, maintenance and use of standards; calibration of individual
systems; monitoring and control of measurement performance; recording and treat-
ment of data; actions to be taken when anomalies occur; and procedures for mak-
ing changes in the programme.

The day-to-day operation of the measurement quality control system should

be recorded in a log. Activities that should be covered include, among others,
instrument calibrations, check weighings, stability checks, specifications of
standard materials and other kind of standards, statistical tests of data, re-
pair and maintenance records, and operating status of instruments. Determina-
tions of precision, accuracy, and bias should also be recorded. Entries should
be dated and signed. Documentation should provide time data, i.e. dates, so
that the uncertainties and biases can be applied to the appropriate periods in
which measurements were made. The documentation also should be sufficiently
detailed that an auditor can determine the operating status of all instruments
used in the measurement of any specific material, together with the relevant
error data and the identity of the standards used to establish bias and
calibrations. Such logs usually consist of bound notebooks with serially
numbered pages. Supplementary loose records consisting of control charts and
other evaluation data also would be kept and related to appropriate log entries.

3.6. Administration

The organization and administration of the measurement control program

should be specified in detail in writing. Individual program functions should
be identified and assigned, with their interrelationships specified, and the
responsibility and authority of individuals clearly defined. Along with individ-
ual responsibility, checks and balances should be provided, where appropriate
and feasible, so that the actions of one individual are verified by others. One
individual should have overall responsibility for and be accountable for measure-
ment control and quality.

Facility management should be involved in the establishment, operation, and

modification of the measurement control programme. The criteria that the mea-
surement programme would be expected to meet should be established by facility
management on the basis of requirements that might be imposed on the facility by
its customers or its government authority. Facility management should assure
that the programme established will meet the criteria established and that any
changes will not cause the programme to fail to meet such criteria. Procedures
should be established for management review and approval of changes to the
programme. Management also should conduct periodic reviews of the programme to
assure that it is being operated according to written approved procedures and
that these procedures remain appropriate for current facility operations.
 61

For each individual measurement process in the program, the qualifications

of the operators are specified, in particular the education, training, and work
experience necessary to perform the required operations.

A training and qualification program for operators should be instituted.

The training should include instruction in the fundamentals of the measurement
process and functioning of the equipment and the actual methods and procedures
employed. There should be frequent written tests and oral examinations. After
qualification in the fundamental principles, extensive experience in the actual
operation of "the equipment should be acquired, with emphasis on the methods and
procedures necessary to maintain control, and on the recognition of situations
or malfunctions which are indicative of loss of control. Finally, operating ex-
aminations should be held in which the trainees analyze samples of unknown con-
stitution. For qualification, minimum standards should be set for both the pre-
cision achieved and the efficiency, or throughput, of the measurements. Peri-
odic testing and requalification of operators should be carried out and
retraining and requalification, in the event of the introduction of new measure-
ment techniques.

3.7 Example

3.7.1. Introduction

A description of the reference facility is given in Appendix A. A block di-

agram showing the process, one possible choice for the MBA structure, and the
key measurement points is presented in Figure A.I. Possible measurement methods
used at the flow and inventory KMPs are shown in Tables A.I and A.2 respec-
tively. Also shown are the container types and the sampling methods. Some typi-
cal accuracies and precisions for typical LEU facility measurements are shown in
Table 2.7 in Part 2. These are not specific to the model plant, and not all
thdse listed would necessarily be used for material-control and accounting
purposes. Table 6.2, however, lists those measurement methods and their
associated precisions and accuracies to be identified with the model facility
and used in the material balance closing calculations shown in Pert 6. The esti-
mation of these precisions and accuracies and the control of the quality of
these measurements could be accomplished using a program described in the follow-
ing paragraphs. The statements, procedures, and instructions that follow are
stated in the imperative mode because they are an example of what the facility
would issue. In the present context, however, they should be interpreted as ex-
emplary rather than prescriptive.

The example is taken, with minor editing, from Schneider 1979 and Schneider
1981.

3.7.2. Technical Requirements.

3.7.2.1. Reference Standards

For safeguards measurements there are three general criteria for

reference standards. First, reference standards must be traceable to a national
or international system of measurement. Second the reference standards must
meet the quality objectives of the measurement program. The tolerance or uncer-
tainty associated with a reference standard shall be smaller than the uncer-
 62

tainty objective of the measurement. A desirable, but not always attainable,

goal would be for the uncertainty of the reference standard to be of the order
of five to ten times smaller than the uncertainty goal for the measurement
method.

Third, reference standards must be representative of the measure-

ment process and range of application; or, in the case where the reference stan-
dard is not used directly, representative secondary standards or working stan-
dards shall be derived from the reference standard.

3.7.2.2. Standards Replication Program

The frequency with which standards are measured shall be sufficient

to detect out of control situations in a timely manner in order to minimize the
number of items which may have to be remeasured. Where standards are measured
for the purposes of detecting small biases and to provide estimates of measure-
ment bias, the frequency shall be sufficient to provide a precise estimate of
any possible bias.

3.7.2.3. Calibration Standards

Calibration standards should be traceable to primary standards,

cover the range of application of the method, and be representative of the mate-
rials and items undergoing measurement. The last requirement is particularly im-
portant when the measurement process is affected by the material or object
undergoing measurement. A case in point is the passive gamma counting of solid
waste for 0-235 where it is necessary to duplicate both the composition of the
waste and the container by the calibration standards.

3.7.3. Mass Measurements

3.7.3.1. Reference Standards

The mass standards used in the measurement control program are

shown in Table 3.4. Also included in the program are sets of NBS Class S and
S-l standard weights (NBS-547) which are used to recertify working weights and
to calibrate the analytical balances.

3.7.3.2. Measurement of Mass i -ndards

Each scale is zeroed at first use on each operating shift and a con-
trol standard weighed. Once each week, at a random time, a quality control tech-
nician weighs a known standard in the working range on each scale using the same
standard each time on all scales to the extent possible. The standard is
weighed while the scale is in use and the scale is not zeroed just prior to
checking the standard.

Twenty-six observations per scale result during the six month mate-
rial balance period. The frequency is selected to produce sufficient data to ob-
tain good estimates of systematic error limits.
 63

3.7.3.3. Mass Calibrations

Prior to use, newly acquired scales and balances shall be evaluated

with respect to accuracy and precision. After zeroing the balance and setting
it up according to the manufacturer's instructions, known standards are weighed
at approximately 50Z and 100Z of full scale and at a point within the expected
working range. Standard weights listed in Table 3.4 are used. A series of 15
runs using these three weights is made. A second series of 15 runs is repeated
at a later time (at least 24 hours later), with a second person performing the
weighing. A third series of 15 runs is repeated by either of these first two
persons or by a third person, again at a later time; that is, at least 24 hours
after the second series of runs. These data form the basis for initial certifi-
cation of the scale.

On a monthly basis, the quality control gauge technician calibrates

each scale, documents the calibration records, and updates the calibration
stickers. Calibration must be performed prior to use of a scale that has been
out of service if the date is beyond that indicated by the due date sticker.
Recalibration also takes place whenever the scale in question undergoes major re-
pair that could affect overall performance in the opinion of the quality control
gauge technician. On an annual basis, each key accountability scale is cali-
brated by the Plant Instrument Group.

3.7.4. Analytical Measurements

3.7.4.1. Reference Standards

The reference standards for the analytical measurements are shown

in Table 3.5. With the exception of the sintered pellet standards and the
grinder-sludge standard, the analytical standards are either NBS standard refer-
ence materials (primary standards) or are prepared directly from those
materials. The sintered pellet standards are secondary standards based on mea-
surements made by an outside laboratory. The pellets were assayed by a titra-
tion method which was traceable to NBS standards.

Isotopic standards are used for determining the multiplier correc-

tion factor (mass-discrimination factor) for the mass spectrometer and also as
routine control standards for the mass spectrometer.

3.7.4.2. Measurement of Analytical Standards

Standards shall be run using each analytical technique with a mini-

mum frequency of two per week, except that during those periods when a given
analytical technique is not in use, the standards need not be run. Under
continued operation, this produces 52 results for each analytical technique
employed in measuring uranium and U-235 during each six-month material balance
period. The schedule provides bias estimates with a level of precision suffi-
cient to detect biases that are very small relative to random-error limits. By
running standards on a weekly schedule, timely indication of lack of control is
given.

Laboratory technicians shall be instructed to exercise the same

care when running standards that is used when running production samples. Only
 64

Table 3.4 Example of Mass Standards

Scale Type Standards Certification

1. Powder, Boat, Sets of 5, 10, 20, 25 kg NBC Class C. Certified

and Tray Scales Metal Weights. by Metrology Lab.
Recertify 18 Months.

2. Rod Loading Set of 0.6, 0.7, 2.4, Certified by Metrology

Scales 3.6, 4.9, 5.7 kg Metal Lab. Recertify
Rods 18 Months

3. UF, Scales Set of (5) 50, 250, 500, Certified by Metrology

o
and (3) 1000 pound Metal Lab. Recertify
Blocks. 18 Months.

Replica Mass Standards Certified at Diffusion

Plants versus NBS
2 Cylinders Depleted U Weights

Tare Approximately Recertify 2 Years.

1650 Pounds

Full Approximately
6400 Pounds.
 65

Table 3.5 Analytical Standards

Measurement Standard Certification

Z U-Gravimetric Sintered U0 2 Depleted UO2 pellets

pellets certified by measurements
made by New Brunswick and
Ledoux Laboratories.

Z D-Low-Level Solutions Parts per million Aliquots of NBS 950 A,

solution 960, or their successors
standards are weighed, dissolved
with HNO3, and diluted
to volume and/or weight.

" D-Grinder Water Grinder water- Characterized, weighed

suspended quantities of sintered
UO 9 U 0 2 are diluted to
volume with water.

X U-General NBS-950 A or 960 NBS Standard Reference

Materials

7. U-235 by Mass NBS UOIO, TO15, NBS certified isotopic

Spectrometer U020, U050, U500, standards of standard
D100, U150, U200 reference material
 66

those technicians authorized to run production samples using a given analytical

technique may produce standards data employing that technique. Standards should
be measured in rough proportion to- the number of production samples analyzed.

3.7.4.3. Calibrations of Analytical Measurements

Fluorometric. Each sample analyzed has a unique calibration which

is derived by spiking the sample with a known quantity of dissolved NBS l^Og or
NBS metal. A calibration curve is not required.

Titration (Davies—Gray Method). When the Davies-Gray method is

used for accountability analyses, the titrant and automatic recording titrator
shall be calibrated against aliquots of the 0 Low Level standards. The relation-
ship between the volume of potassium dichromate titrant and uranium amounts is
determined over the range of uranium amounts expected in sample aliquots. A
minimum of four standard values within this range is used and each is assayed a
minimum of three times. Recalibration shall be performed annually or whenever
the titrant is changed or standards data indicate a need for recalibration.

Gravimetric Analysis. The accuracy of the gravimetric uranium anal-

ysis is directly related to weights and ignition to stoichiometric t^Og. Bal-
ances shall be serviced annually by the manufacturer, calibrated monthly by
plant personnel, and checked daily with a Class S-l weight to assure proper oper-
ation.

U-235 Analysis. Multiplier correction factors shall be determined

using NBS isotopic standards (NBS U010, U015, etc.). No other calibration is
required. The factors shall be re-estimated on an annual basis. More frequent
estimation is not required because biases in the mass spectrometer are
controlled through the running of standards.

3.7.5 NDA Measurements

3.7.5.1 Calibration Standards

General. The NDA calibration standards for the waste assay counter
are shown in Table 3.6.

P-235 in 55-Gallon Barrels. The 30 one-gallon jugs described in

Table 3.6 are used in preparing each of the five calibration standards. A given
barrel calibration standard is created by uniformly dispersing known quantities
of U-235 in 30 gallons of sawdust. The sawdust and uranium mixture is placed in
the 30 one-gallon jugs. There are 3 layers of jugs in a barrel, with each layer
comprising 10 jugs so that the barrel holds 30 jugs total. For each standard,
three 500-second net counts are taken.

An upper limit of about 200 grams U-235 is chosen to correspond to

the maximum amount of uranium within a process barrel. Barrels that exceed the
upper limit are down loaded and the contents dispersed into other barrels so
that the calibration curve limit is not exceeded.
 67

Table 3.6 Waste-Assay Calibration Standards

Measurement Standard Certification

Total D-235 in Five 55-gallon barrels each The U-235 amounts in

55-gallon barrels containing 30 one-gallon jugs the barrels are based
containing a mixture of low on measurements made -
enriched uranium and sawdust with using the gravimetric
from 2.0 to 200 g U-235 per barrel and mass spectrometer
approximately equally divided analytical methods.
among the 30 one-gallon jugs. They are traceable to
NBS through the NBS
analytical standards.
 68

Calibration Frequency

The system calibration shall be repeated annually, or whenever mea-

surement-system changes are made that might affect the calibration. An annual
calibration frequency is judged to be adequate because the system is carefully
controlled through frequent running of the standards. Further, replicate mea-
surements made monthly on four production barrels that span the range of 0-235
provide assurance that the calibration remains fixed throughout the entire
range.

Measurement of NDA Standards

For the NDA measurement of uranium solid-waste containers, two of

the standard barrels are run at the beginning of each time period during which
waste barrels are to be measured. Once the system meets documented criteria for
operation based on the results of these runs, the same two standards are run
thereafter at a minimum frequency of once per hour until the measurement opera-
tion is completed. This frequency is chosen because it is sufficient to control
expected instrument drift3. Further, if an out-of-control situation is
detected, the number of items requiring remeasurement since the last acceptable
check point is not excessive. The two standards are run using the same proce-
dures as for the process waste, and only those persons authorized to measure the
process waste containers measure the standard barrels.

3.7.6. Sampling

3.7.6.1. Homogeneous Materials

The process produces uniform lots of OO2 powder and pellets. The
principal sampling error affecting those materials is the lot-to-lot variation
in uranium composition. That variation creates, in effect, a random sampling
error when Average uranium-element factors are used for inventory accounting.
For these homogeneous materials, systematic errors of sampling are negligible.

3.7.6.2. Heterogeneous Materials

In the conversion-fabrication process three types of heterogeneous

materials are typically encountered. These are heterogeneous scrap such as ADU,
grinder sludge, and "dirty" powder; liquid wastes containing ADD solids; and
solid wastes.

To estimate the random and systematic sampling error of those

materials, special experimental programs shall be carried out for each type of
heterogeneous material. These are described in the next section.

3.7.7. Special Experiments

3.7.7.1. Heterogeneous Scrap

For heterogeneous scrap items - ADU, grinder sludge and dirty pow-
der - two experimental programs are carried out to estimate sampling errors.
 69

For the estimation of random sampling errors a routine resampling

program is carried out. During each six-month accounting period, 15 containers
of each type of scrap are resampled and assayed for percent uranium to provide
a basis for estimating random sampling errors.

Small scale oxidation experiments have been carried out in which

the entire contents of scrap containers are converted to U3O8 in a quantitative
manner. The experimental approach is similar to that of doing a gravimetric
assay of the entire contents of a 5-gallon bucket of scrap.

The contents of the scrap buckets were quantitatively transferred

into tared oxidation trays such that each tray represented a gravimetric assay
sample size of about five kilograms of scrap. The between tray percent uranium
values for trays from the same scrap container provide an estimate of the
heterogeneity of the material (or inherent random sampling error). The differ-
ence between the original sample result for percent uranium and the percent ura-
nium found by the oxidation assay approach for the whole container provides an
estimate of the systematic sampling error. Obviously, the need for such experi-
ments would be eliminated if all difficult to measure scrap were routinely
converted to readily measurable

3.7.7.2. Solid Waste

Solid waste, which consists of contaminated gloves, rags, plastic

sheets, etc., presents a potential matrix problem for the passive gamma counter.
If the uranium in the waste is not uniformly distributed or if it is shielded by
high-density materials, the counter gives low results.

To estimate the potential systematic error which arises from such

a matrix effect, special chemical leaching experiments have been carried out en
process-waste barrels. Those experiments show that a significant bias (passive
gamma counting 30-40Z lower than leaching) can result when the actual waste
matrix is more dense and heterogeneous than the calibration standards. From the
special leaching experiments, a matrix bias-correction equation «as developed
which can be used for "after-the-fact" bias corrections. Because of the
practical difficulty of controlling the solid waste matrix in an exact manner,
periodic quantitative leaching experiments are included in the measurement-
control program.

3.7.8. Statistics

3.7.8.1. General

The scope of the statistical aspects of the measurement-control pro-

gram is shown in Table 3.7. Statistical techniques are applied to all elements
of the measurement processes to estimate measurement errors, set control limits,
derive calibration relationships, and estimate biases. In addition, comprehen-
sive statistical evaluations are made periodically of the overall program and
error propagation methods using current error estimates to calculate the plant
OyqjY' The more important details of the statistical application are described
next.
 70

Table 3.7. Scope of Statistical Program

Weighing Sampling Overall

Control limits Scrap resampling program Comprehens ive annua1

data for error estimation measurement review

Use statistical control to Lot-to-lot variation of Update error parame-

keep scales within error powder and pellets ters each six months
limits-no bias corrections

Estimate error parameters U3O3 experiments for MDF calculations

from routine data and sampling error each six months
experiments

Use S/R data for UFg Liquid waste experiments Monitor all program
weighing errors -LE and bias estimation data

Analytical

Control limits

Mass spectrometer
calibration

Bias estimation and

test of significance

Random error from routine

data

OTA

Control limits

Calibration equation

Standards data for

instrument drift

Chemical recovery
versus counting
for matrix bias
 71

3.7.8.2. Statistical Applications

Calibrations

For the determination of U-235 in solid waste? a quadratic calibra-

tion curve is developed. For the multiplier correction factor for the mass
spectrometer, an average correction factor is developed after a statistical eval-
uation of the data for non-random effects.

Control Limits

From the calibration and standards data, statistical control and ac-
tion limits are developed for the waste assay system, analytical methods, and
the routine weighing of standards by the quality control technicians.

The general control philosphy for standards measurements is to

remeasure the standard if the result falls outside the 0.05 level of signifi-
cance but within the 0.001 level of significance. If repeated measurements fall
within the 0.05 level of significance, no further action is required.
Otherwise, the method is declared out of control. If a standard measurement
falls outside the 0.001 level of significance , the method is declared out of
control and remedial action initiated.

3.7.8.3. Error Estimation

3.7.8.3.1. Random Errors

Replicate measurement data are analyzed by the method of

paired differences to provide estimates of the random error variance. The vari-
ance among the paired differences is an estimate of twice -he random measure-
ment-error variance.
3.7.8.3.2. Systematic Errors

Weighing. In the cases of scales and balances, the pri-

mary purpose of standard weighing is to detect when scale adjvistments are needed
and not to generate data that form the basis for making bias corrections. Gener-
ally, it is difficult to use the data generated by the standards program to ob-
tain realisti estimates of systematic error limits. Generally speaking, the
best statement that can be made is that all weighings of the standards were
within, say, one scale division (or the rounding interval of digital-readout
scales.)

To obtain more realistic estimates of systematic error

limits for scales, special weighing experiments are conducted using unknown
weights. These are weighed singly and in combinations on the various scales.
Standard weights are then assigned these standards on the basis of the consensus
data. A systematic error component, described by the variance of the population
from which the systematic error for a given scale was selected at random, is
then estimated by the square of the average dif'. «renee betwen the observed
weights for the standards (weighed singly and it combinations) and the
corresponding assigned weights based on the conseusus data (Jaech 1978).
 72

For the balances used in rod loading, the sensitivity is

such that the systematic error variance can be estimated from the quality-
control data on the weighing of known standards. An analysis of accumulated
standards data is used to estimate the systematic error variance. Down loading
experiments are also carried out which provide additional data for error estima-
tion.

For the UFg scale, shipper-receiver data are used to ob-

tain a realistic estimate of the systematic error. The statistical techniques
for estimating both the long term and short term systematic error variances from
shipper-receiver data are described by Jaech (Jaech 1975a).

Analytical. The primary function of the control program

it to give an early signal of possible problems with the analytical technique in
question rather than to create data that will form the basis for making bias cor-
rections after the fact.

Recognizing that small biases will occur, the standards

data for each analytical technique are analyzed at the end of a material balance
period to estimate the bias that existed for that analytical technique during
the material balance period. The estimated bias is the average difference be-
tween the observed measurements on the standard and the assigned standard value.
Alternatively, the statistic may be the difference in logarithms of the raw data
or, equivalently, the logarithm of the ratio. The latter approach is allowed
when more than one standard is used and where biases are consistent on a rela-
tive rather than absolute basis. This approach is followed with the mass-
spectrometer standards data.

The standard deviation of the estimated bias is

calculated as the standard deviation among the differences divided by the square
root of the number of differences comprising the average. The uncertainty in
the standard is also taken into account. If the appropriate data are
subsequently corrected for bias, the standard deviation of the estimated bias is
regarded as a systematic error standard deviation (Jaech 1975b).

_, NBA. Systematic errors in the solid waste measurements

are estimated from the calibration data ar.d from the leaching experiments
described earlier.

Sampling. Systematic errors in sampling heterogeneous ma-

terials are estimated from the special experiments described earlier.

3.7.8.3.3. Application to Overall Program

Regarding statistical applications to the overall measure-

ment control program, the following activities are carried out:

1. Annually, a comprehensive measurement review is

performed in which all the standards data and applica-
ble shipper-receiver data are evaluated statisti-
cally. Error estimates are developed for key
accountability measurements and yearly trends
 73

analyses are made. The measurement review is

documented and a detailed report issued.

2. Error parameters for the a*fup calulations are updated

each six months and used in the a^up calculation
which is performed at the end of the six-month
inventory.

3. All data generated in the measurement control program

are monitored routinely and appropriate statistical
techniques applied to detect trends and identify pos-
sible p'Mblem areas.

3.7.9. Documentation

All data generated in the program are recorded and a formal system of re-
cord-keeping and documentation is followed. Examples of the various records and
reports associated with the program are shown in Table 3.8.
 74

Table 3.8 Records and Reports

Records Reports

CUSDM and control charts G

MDF
Out of control documentation Measurement reviews

Calibration data & calculations Bias adjusted MUF

a._._ data & calculations Standard & referee reports

MUr
Q.C. weekly standard Measurement review results
weighing data

Standards preparation data Special experiment report

Standards program data Error parameter reports

Mass standards calibration and

certification

Measurement review
 75

References to Part 3

AG-112: IAEA Report AG-112, Advisory Group Meeting on the Use of Physical Stan-
dards in Inspection and Measurement of Nuclear Materials by Non-Destructive
Techniques, 22-26 August 1977. International Atomic Energy IAEA, Vienna,
1977.

ANSI N15.18: American National Standard Mass Calibration Techniques for Nuclear
Material Control, ANSI Standard N15.18, Institute for Nuclear Materials Man-
agement and American National Standards Institute, New York, N.Y. 1975.

ANSI N15.19: American National Standard Volume Calibration Techniques for Nu-
clear Material Control, ANSI Standard N15.19, Institute of Nuclear Mate-
rials Management and American National Standards Institute, Nev York, N.Y.,
1975.

ANSI N15.20: American National Standard Guide to Calibrating Nondestructive

Assay Systems, ANSI Standard N15.20-1975, Institute of Nuclear Materials
Management and American National Standards Institute, New York, N.Y. 1975.

BLG 522: C. Beets, G. Busca, J. Colard, M. Curbellini, M. Cuypers, B.

Fontaine, F. Franssen, D. Reilly, F. von Craenendonck, and P. van Loo, Ac-
quisition and Verification of Reference L.W.R. Rods, Report BLG 522, FBFC,
CEN-SCK, Euratom, Luxembourg, 1980.

BNL 50414: L.A. Kull and R.O. Ginavin, Guidelines for Gamma-Rav Spectroscopy
Measurements of 2-35y Enrichment, BNL Report No. 50414, Brookhaven National
Laboratory, Upton, N.Y. 1974.

Brouns 1976: R.J. Brouns and F.P. Roberts, Control of Sampling Errors, Nu-
clear Materials Management, Vol. V, No. 3, 622 (1976).

Clark 528: J.P. Clark and C.H. Jackson, Establishing the Traceability of a
Pranyl Nitrate Solution to a Standard Reference Material, Proc. of American
Nuclear Society Topical Meeting on Analytical Methods for Safeguards and
Accountability Measurements of Special Nuclear Materials, Williamsburgh,
May 15-17, 1978. NBS Special Publication 528, National Bureau of
Standards, Washington, D.C. 1978.

de Biêvre 1982? P. de Bièvre, H. Meyer, J. von Audenhove, B.S. Carpenter, and

J.W. Gramlich, Progress Report on the Establishment of Pranium Isotope Abun-
dance Reference Materials for NDA. Specialist Meeting on Harmonization and
Standardization in Nuclear Safeguards, Petten, 27-29 April, 1982. European
Safeguards Research and Development Association, 1982.

EUR 6089 EN: C. Bibliocca, M. Cuypers, and J. Ley, List of Reference Materials
for Non-Destructive Assay of P, Th, and Pu Isotopes, Commission of the
European Communities Report EUR 6089 EN, Luxembourg, 1979.

EUR 6937: Y. Le Digou, Catalogue of Chemical and Isotopic Nuclear Reference

Materials, Central Bureau for Nuclear Measurement: Commission of the
European Communities report EUR 6937, Luxembourg, 1980.
 76

Ewan 1963: W.D. Ewan, When and How to Use Cu-Sum Charts, Technometrics, Vol.
5, No. 1, 1 (1963).

Harry 1979: R.J.S. Harry and H.T. Yolken, Development of Uranium Oxide Refer-
ence Materials for G-amma-Ray Measurements of Enrichment, Nuclear Materials
Measurement, Vol. VIII, Proceedings Issue, 54 (1979).

Hough 1982: G. Hough, «££!•» Qualification of Measured Data for Safeguards

Use, pp. 13-21 in ESARDA-15 (1982).

IAEA-TECDOC-261: IAEA Safeguards Technical Manual, Part F: Statistical

Concepts and Techniques, Volume 3, IAEA, Vienna, 1982.

IAEA-174: IAEA 174, IAEA Safeguards Technical Manual, Part E: Methods and Tech-
niques (IAEA, Vienna, 1975).

IAEA/SG/INF/1: IAEA Safeguards Glossary, IAEA/SG/INF/1, IAEA, 1980.

Indusi 1974: J.P. Indusi and W. Marcuse, A Method to Determine the Minimum Cost
Measurement Plan Consistent with any Feasible Limit of Error on MUF, Nu-
clear Materials Management, Vol. Ill, No. Ill, Procedings Issue, 269
(1974).

ISO 1977: ISO Guide 6-1977(E), Mention of Reference Materials in International

Standards, section 2.2. International Organization for Standardization,
Geneva, 1977.

ISO/REMCO 89: P.D. Risdale, New Certified Reference Materials in Current and
Planned Production, International Organization for Standardization report
ISO/RISMCO 89, Geneva, 1982.

Jaech 1973: J.L. Jaech, Statistical Methods in Nuclear Material Control, TID-
26298, Superintendent of Documents, U.S. Government Printing Office,
Washington, D.C. 20402, 1973.

Jaech 1975a: J.L. Jaech, Case Studies on the Statistical Analysis of Safeguards
Data, Paper IAEA-SM-201/14, Vol. 1., Safeguarding Nuclear Materials, IAEA
Safeguards Symposium, Vienna, 1975.

Jaech 1975b: J.L. Jaech, Some Thoughts on Bias Corrections, Nuclear Materials
Management Vol. IV, No. 2, pp. 40-44, 1975.

Jaech 1978: J.L. Jaech, Estimation of Scale Accuracy and Precision: A Case His-
tory, Nuclear Materials Management, Vol. VII, No. 3, 81 (1978).

LA-8901-C: E.A. Hakkila, K..G. Gutmacher, and C.C. Thomas, Jr., Survey of Chemi-
cal and Destructive Analysis Methods, Advanced International Training
Course on State Systems of Accounting for and Control of Nuclear Materials,
Santa Fe and Richland, April 27-May 12, 1981, Report LA-8901-C, Los Alamos
National Laboratory, 1981.

NBL 1982: NBL Certified Reference Materials Catalog, 1982, New Brunswick Labora-
tory, Argonne, Illinois, 1982.
 77

NBL-305: Safeguards Analytical Laboratory Evaluation (SALE) 1981 Annual Report,

NBL-305, U.S. Dept. of Energv, New Brunswick Laboratory, Argonne, 111.,
1983.

NBS-91: M.G. Natrella, Experimental Statistics, National Bureau of Standards

Handbook 91, National Bureau of Standards, Washington, D.C., 1963.

NBS 260: NBS Special Publication 260, NBS Standard Reference Materials Catalog,
1981-83 Edition, National Bureau of Standards, Washington, D.C., 1981.

NBS 260-27: E.L. Garner, L.A. Machlan, W.R. Shields, NBS Special Publication
260-27, Uranium Isotopic Standard Reference Materials, National Bureau of
Standards, Washington, D.C. 1971.

NBS-300: Precision Measurement and Calibration, NBS Special Publication 300,

Vol. 1, Harry H. Ku (éd.), National Bureau of Standards, Washington, D.C.,
1969.

NBS 547: T.W. Lashof and L.B. Macurdy, Precision Laboratory Standards of Mass
and Laboratory Weights, National Bureau of Standards Circular 547, Section
1, National Bureau of Standards, Washington, D.C. 1954.

NUREG/CR-1284: R.J. Brouns, J.A. Merrill, and F.P. Roberts, Methods of

Determining and Controlling Bias in Nuclear Material Accounting
Measurements, NUREG/CR-1284, Pacific Northwest Laboratory, Richland, Wash.,
99352, 1980.

Pietri 1978: C.E. Pietri, J.S. Palier, and C D . Bingham, The Chemical and
Isotopic Analysis of Uranium. Plutonium, and Thorium in Nuclear Fuel
Materials, pp. 1-18 of NBS-528, Analytical Methods for Safeguards and
Accountability Measurements of Special Nuclear Material, Proceedings of
an ANS Topical Meeting held May 15-17, 1978, Williamsburg, VA.

Reilly 1977: T.D. Reilly and M.L. Evans, Measurement Reliability for Nuclear
Material Assay, Los Alamos National Laboratory report LA-6574, January,
1977.

Rodden 1972: C.J. Rodden, Selected Measurement Methods for Plutonium and Ura-
nium in the Nuclear Fuel Cycle, TID-7029 (2nd éd.), 1972, National Techni-
cal Information Service, U.S. Dept. of Commerce, Springfield, Va. 22151.

Schneider 1979: R.A. Schneider, A Measurement Control Program for a Conversion

Fabrication Plant, Nuclear Materials Management, Vol. VIII (Proceedings
Issue), pp. 573-592, 1979.

Schneider 1981: R.A. Schneider, Measurement Control Program, Advanced Interna-

tional Training Course on State Systems of Accounting for and Control of Nu-
clear Materials, Santa Fe and Richland, April 17-May 12, 1981, Report LA-
8901-C, Los Alamos National Laboratory, 1981.

Shuler 1977: L.E. Shuler, L.W. Doher, and L.W. Morrow, "A Displacement Tech-
nique for Calibration of Special Nuclear Material Tankage Volumes," Nuclear
Materials Management, Vol. VI, No. 3, 472 (1977).
 78

Suda 1976: S.C. Suda and D.F. Shepard, Effects of Measurement Errors on Linear
Calibration Equations, BNL-50432, Brookhaven National Laboratory, Upton,
N.Y., 1976.

Voeks 582: A.M. Voeks and N.M. Trahey, Preparation of Prototype NDA Reference
Materials: A Progress Report, Proc. of American Nuclear Society Topical
Meeting on Measurement Technology for Safeguards and Materials Control,
Kiawah Island, South Carolina, Nov. 26-30, 1979. NBS Special Publication
582, National Bureau of Standards, Washington, D.C. 1980.

Wash-1335: R.G. Gutmacher, F. Stephens, K. Ernst, S.P. Turel, and T.E. Shea,
Methods for the Accountability of Plutonium Dioxide, Wash-1335, Lawrence
Livermore Laboratory, Livermore, Calif., 1974.

Wong 528: C M . Wong, J.L. Câte, and W.L. Pickles, Proc. of American Nuclear
Society Topical Meeting on Analytical Methods for Safeguards and
Accountability Measurements of Special Nuclear Materials, Williamsburgh,
May 15-17, 1978. NBS Special Publication 528, National Bureau of
Standards, Washington, D.C. 1978.
 79

Part 4. RECORDS AND REPORTS

4 .1 Introduction
Reports based on the records are the results of the nuclear material ac-
counting system. The verification and evaluation by a safeguards authority of
these reports and their supporting records and accounting system make possible
statements of the effectiveness of safeguards. The consistency of the data
appearing in the various related levels of the records and reports provides the
basis for identification of anomalies which may indicate a loss or diversion of
nuclear material. Depending upon the national regulations and the local situa-
tion, the data to be transmitted to the IAEA may be prepared at various places.
In some cases it will be the facility itself, where the accounting reports would
be compiled and completed in the very form in which they ultimately reach the
IAEA. In other cases, the facility personnel will only prepare the data and
transmit them to some other body - e.g., a national authority - which will then
complete the reports to the IAEA. In any case it is the facility data that are
basic to the final reports to the IAEA. The following description provides in-
formation regarding a system of records and reports which includes the necessary
data elements for a nuclear material accounting system that will permit verifica-
tion and evaluation. The general discussion in Sees. 4.2-4.5 is followed by a
detailed example in Sec. 4.6 showing how, in a typical case, the various record
and report forms would be filled out.

4.2 Accounting Records

Accounting Records are defined in the IAEA Safeguards Glossary

(IAEA/SG/INF/1) as a set of documents kept at a nuclear facility showing the
quantity of each type of nuclear material present, its distribution within the
facility, and any changes affecting it. To provide an effective nuclear mate-
rial accountancy system a primary requirement is that the accounting records
accurately show the nuclear material inventory and all inventory changes that af-
fect it. Another requirement is that the accuracy and consistency of such re-
cords are verifiable. Accounting records consisting of the following three com-
ponents can meet these requirements:

a) Ledgers. These are the records in which the inventory changes are
summarized and totaled for specified time periods.

b) Inventory change journals (sometimes called transaction journals) or

iournal entry forms. These are the records of original entry where
the data from various supporting documents are recorded as inventory
changes occur.

c) Supporting documents. These provide the primary data to be entered

into the records for each inventory change. Supporting documents con-
tain source data which are the result of the operation of the facility
measurement and measurement quality system. Source data are defined
in the Safeguards Glossary as those data, recorded during measurement
or calibration or used to derive empirical relationships, which iden-
tify nuclear material and provide batch data. "Source data" may in-
clude, for example, weight of compounds, conversion factors to deter-
mine weight of element, specific gravity, element concentration,
 80

isotopic ratios, relationship between volume and manometer readings

and relationship between plutonium produced and power generated.
These source data are to be found in the Operating Records which are
discussed in section 4.3 following.

It is possible, where there are a limited number of inventory changes, to

eliminate the journals. In this case the entries would be made directly to the
ledgers and summarized periodically. However, in the case of a low enriched ura-
nium conversion and fuel fabrication facility the number of inventory changes
will undoubtedly be large enough to warrant the use of both journals and ledgers
to assure efficient and effective recording of the inventory changes. If a fa-
cility uses electronic data processing for its nuclear material accounting sys-
tem then the journal accounts may not be needed, even though there are a large
number of transactions. Each inventory change could be coded and entered di-
rectly to the ledger accounts. Computer programmes could then be used to re-
trieve data from the ledgers in any categories needed for subsequent report
preparation or records audit.

4.2.1 Ledgers

A ledger is the record of final entry which provides a summary of a l l in-

ventory changes to determine the book inventory, i . e . the quantity of nuclear ma-
terial that should be present at the f a c i l i t y . Entries to the ledgers are made
from the totaled journal accounts and individual journal entry forms. Separate
ledgers should be kept for different material types. At a low enriched uranium
facility there may be a separate ledger or ledger account for natural, depleted
and enriched uranium. Accounts for different enrichments also may be kept or a
single unified account may be kept for all uranium. A general ledger for each
material type could be kept for the entire facility with subsidiary le.dgers for
each area for which nuclear material accounting was needed. Such accounts for
IAEA designated material balance areas would significantly f a c i l i t a t e
preparation of accounting reports required under safeguards agreements and would
facilitate subsequent IAEA verification a c t i v i t i e s . The example given in Sec-
tion 4.6 shows ledgers only for 37. enriched uranium for each of the three
IAEA-designated MBAs. Similar records could be kept for other materials and
other areas of the facility and for the entire facility if the facility operator
should choose to do so.

4.2.2 Inventory Change Journals.

Inventory change journals are chronological records of the various types

of inventory changes that occur at a facility. They are periodically totaled
and entered into the ledger accounts. Entries to the journal accounts are made
on the basis of the supporting documents for the respective accounts, e.g.
receipts, shipments, discards, etc. When there are only a few inventory changes
the journals may not be needed and the inventory changes could be entered di-
rectly into the ledger accounts. However, when there are many changes, as would
be the case in a low enriched uranium conversion and fuel fabrication facility,
the journal accounts provide an efficient and convenient way to obtain both
detailed and summary data for particular types of changes that would be needed
for subsequent report preparation and records audit. Typically journal accounts
would be maintained for those inventory changes which were most frequent, such
as receipts, shipments, internal transfers and measured discards, as well as for
 SI

the different material types consistent with the ledger material-type accounts.
Journal-entry foras or other supporting documents would be used for other inven-
tory changes such as accidental losses, exemptions or de-exemptions or MUF ad-
justments which would be entered directly to the ledger accounts. If there are
only a moderate number of inventory changes the journal accounts could be kept
in the same document with separate columns for the different accounts. An exam-
ple of such a journal is shown in Figure 4 . 1 . If there are many changes then i t
may be more efficient and convenient to maintain the journal accounts in sepa-
rate documents such as Figures 4 . 2 . and 4 . 3 , for example, for the receipts ac-
count and removals account, respectively. As noted above, if the f a c i l i t y uses
electronic data processing the inventory change journals may not be needed.

4.2.3 Supporting Documents.

One requirement of an accurate, effective and verifiable accounting system

is that the data for each inventory change be recorded at the time of the change
and at its source. This can be done by using special forms for each type of
change which occurs frequently, such as receipts and shipments, and general ,
forms for less frequent changes. Such documents, collectively, are called
supporting documents because they are the records which support the entries of
the inventory change data in the journals and ledgers. It is important to note
that for each entry in the journal and/or ledger accounts there should be
supporting documents which include the data necessary to describe fully the in-
ventory change. Such documents provide the audit trail between the data in the
journals and ledgers and the original source data, i.e. measurement, calibra-
tion, identity or other data applicable to the inventory change. To provide
this connection the supporting documents should provide a unique identification
for each inventory change. This can be provided by the title of the form, e.g.
Receiving Notice, and a unique number on the form for each inventory change.
The use of pre-numbered forms is an effective way to assure accurate inventory
change identity. A facility operator may choose to identify inventory changes
by various terms. It seems logical, however, to use IAEA terminology in this
discussion because eventually the facility data must be collected and reported
in IAEA terms and format. Code 10 of the General Part of Subsidiary
Arrangements for NPT Safeguards Agreements identifies and defines the various
types of inventory changes and the IAEA reporting codes for each. Some of
these, of course, would not be appropriate for a conversion and fuel
fabrication plant, e.g. nuclear production or nuclear loss. While the form of
the supporting documents and the terminology used for the inventory changes
are important, most important is that all of the necessary data are recorded
on the supporting documents as follows:
a) Inventory Change Identity. Some mecns should be provided to differen-
tiate a specific inventory change from all others so that it can be
referenced in the records and traced during audit. This normally is
done by consecutive inventory change numbers with perhaps a different
series for each type, e.g. receipts, shipments, discards, etc.

b) Date of Change. This should be the date the inventory change actually
occurred, not the date the supporting document was prepared, although
the supporting document should be prepared on the date of the change
if possible.
 Figure /\.l Inventory Change Journal

Faoilit.ys MBAt Material Descriptions Document Hot

Iootopo Codel Element Coda» Unit!

o Data
of from To
1C KMP ICD Kumbor Mtl MOBB. Heoeipta Shipments Discards
Glinngo
Code Code Ho. of fteao.
Items Code Code
U U-2Î5 V U-2 35 0 U-2 35

lomarkn t Slgnnturos I'D go Hoi

 Klguro i|.2 Inventory Change Journal - ilcceiplo

Kaoilitv: Material Description: HDA: Iuolopo Codeij Element Codo: Unit!

Duto IC KM1' ICI) No. MU. Heau. Ifeceiptu S/li difforenceo Other Additionu
(1 o or Prom To Code Numlior of Codo Codo
Code
Oliitngu ItOlllB U-215

Huiimi'ka: Signaturo: Doc. No. l'Q/;o No.

 Figure Inventory Change Journal — llcinov.tln

Knellltv: Material Doocription: MM: Iiinlopo Code: Element Coda: Unit

Rate 1C KHP icn| Ho. Htl. Hoao. Shipments - Other Removnls

or From To Code Code Numbel* of Codo Codo
Change Items «-235 U-2

flcmnrk n : !ii(;nntnro: l)oc. Uo. l'a(;o No.

 85

c) Type-of Change. If this i s a change that w i l l eventually be reported

to the IAEA i t would be useful to include the IAEA code as well as any
name the f a c i l i t y chooses to u s e . This would make preparation of the
IAEA Inventory Change Report e a s i e r .

d) Material Description. This should include i d e n t i f i c a t i o n of material

type and isotopic enrichment, as appropriate. As above, if the change
i s to be reported to the IAEA, IAEA codes should be included.

e) Material Movement. This would be the information which i d e n t i f i e s a

change in location of the material, e . g . r e c e i p t , shipment, discard,
or internal transfer. If the f a c i l i t y control areas are the same as
IAEA MBAs then internal transfers would need to be reported to the
IAEA as changes in MBA inventories and the IAEA codes should be
included. If a f a c i l i t y operator chose to have internal control areas
within the IAEA MBAs then transfers between such areas would not need
to be reported to the IAEA. The omission of IAEA codes from such
transfer documents could be used to indicate that such transfers were
not to be included in IAEA Inventory Change Reports.
f) Batch Identity. A batch is defined in the Safeguards Glossary as a
portion of nuclear material handled as a unit for accounting
purposes at a key measurement point and for which the composition
and quantity are defined by a single set of specifications or
measurements. A batch normally is identified by a number. See
Table A.3 in Appendix A. When batch follow-up is called for the
supporting documents should identify a l l batches involved in an
inventory change so that increases and decreases can be assigned to
the appropriate batches. See Code 10 of the General Part of
Subsidiary Arrangements for KPT Safeguards Agreements for rebatching
reporting procedures.
g) Number of Items. The number of items in a batch or in each of Several
batches, especially if rebatching is involved, should be recorded.

h) Batch Data. These are defined in the Safeguards Glossary as the total
weight of each element of nuclear material and in the case of pluto-
nium and uranium the isotopic composition when appropriate. For a low
enriched fuel f a c i l i t y , of course, only uranium element and the U-235
isotope would be appropriate. IAEA element and isotopic codes should
be included for changes that will be reported to the IAEA.

i) Batch Data Basis. This should provide the basis from which the batch
data entries were derived. Normally this should refer to some measure-
ment by the f a c i l i t y . It could, however, be shippers data for re-
ceipts or perhaps an estimated quantity for an accidental loss. Refer-
ence to other source data such as laboratory notebooks or analytical
. reports should be included. Reference to the appropriate KMP also
should be included along with the IAEA measurement code, for future
ease in preparing reports.

99
 86

4.2.4 Inventory Changes, Adjustments and Corrections.

For inventory changes which are expected to occur often, data handling and
controls can be made easier when inventory change forms are designed s p e c i f i -
cally to record the data that would be associated with each type of change. One
such example would be that which recorded the actual movement of material from
one material control area to another, i . e . from a shipper to a receiver. This
would include off s i t e receipts and shipments and internal f a c i l i t y transfers be-
tween MBAs. Figure 4.4 is an example of such a form. Another inventory change
which might be expected to occur frequently and warrant a special form would be
measured discards. Figure 4 . 4 . could be used for discards by considering the re-
ceiver as being the place of discard. However, to assure control of discards,
a separate form such as Figure 4.5 could be used, which c a l l s for an approved
signature before material is discarded. This form, Figure 4 . 5 , also could be
used for other types of inventory changes which do not involve a shipper and re-
ceiver, e . g . accidental losses or gains or exemptions or de-exemptions, but
which should involve approvals for the changes.

In addition to inventory changes involving actual changes or movements of

material there are adjustments and corrections which result in changes in the
recorded inventory q u a n t i t i e s . Shipper-receiver differences are one type of ad-
justment. These result from measurements performed at the f a c i l i t y on nuclear
material previously recorded at the shipper's value. The receiving form docu-
ment, e . g . Figure 4 . 4 , should provide for recording the f a c i l i t y measurement re-
sults and the shipper-receiver difference, to serve as the supporting document
for making the shipper-receiver adjustments to the records. If a significant
shipper-receiver difference is found i t should, of course, be investigated.
This could result in a corrected document being issued by the shipper which
would serve as the supporting document for a correction of the shipper's data
that had been recorded in the f a c i l i t y records.

Another type of adjustment i s material unaccounted for, which results from

a physical inventory taking. The supporting document for such an adjustment
would be the physical inventory summary and reconciliation supported by the
final physical inventory l i s t i n g , which would record the measurement results of
the physical inventory taking. See Part 5, Physical Inventory Taking.

Corrections of mistakes in the data should be supported by appropriate doc-

umentation. This could be done by issuing a corrected support document such as
a receiving report, shipping report, e t c . or by using a journal entry from,
e . g . , Figure 4 . 5 , to record the data supporting the correction.

4.3 Operating Records

From a nuclear material accountancy standpoint operating records consist

of the documentation of the r e s u l t s of f a c i l i t y systems and procedures which pro-
vide data relevant to accounting for nuclear material. Such systems and proce-
dures are the Nuclear Material Measurements System, the Measurement Quality
Programme and Physical Inventory Taking procedures. Operating records are
defined s p e c i f i c a l l y in the Safeguards Glossary as:

a) those operating data which are used to establish changes in the

quantities and composition of nuclear material;
 Figura 4-4 Inventory Change Document - Material Transfer

Material Doocriptioiu Material Code: Iuotope Codet Element Codei Unitt

Snipper Name and Addreaai ' MBA: IJeoeiver Name and Addreaa MDAi

Change Codai Change Code:

KHP CodeI KHP Codai

Shlppor Data neceivor Data

Batch Ho
rt c Identity of Net Element Isotope Net 5t Element * Iaotopo
Items Height Ele. loo. Weight Ele. Weight loo. Noteo
Weight Height Weight

CO

Doo. Hoi Hotoui Shipper-Deceiver Dlfforonoet Date HeoBurod!

HT-
Shipping Datot Shipper St(,natural Deceiving Dates- Hooeiver Signature 1
 Figure 4.5 Inventory Change Documont - Journal Entry Form

Facilityt HM: Change Code: KHP Code: Date:

latorial Description! Haterial Codo: Tootopo Code: Element Codei Unlti

Number
Document Batoh Element.' iBotopo
& of Hotoe
Heforonoe Identity Items Height Height
3

00
oo

Prepared By» Approved By: rôt!.'Ho.

j&-
 89

b) the data obtained from the calibration of tanks and instruments and
from sampling analysesf the procedures employed to control the quality
of measurements, and the derived estimates of random and systematic
error;

c) a description of the sequence of the actions taken in preparing for,

and in raking, a physical inventory, to ensure that it is correct and
complete; and

d) a description of the actions taken to ascertain the magnitude and

cause of any accidental or unmeasured loss that might occur.

4.3.1 Nuclear Material Measurements

Item a) in the Glossary definition of operating records refers to the re-

sults that are obtained from the facility measurement of nuclear material.
These data probably would be recorded at the location where the measurement was
made, e.g. weigh station or analytical laboratory. Various means of recording
the data could be used, e.g. manual, print weight scales, on-line computer termi-
nals, etc. Whatever the means for data recording a primary concern from a re-
cords and reporting standpoint is to identify the data record so that it can be
accurately identified in the appropriate supporting documents and other
accounting records and reports. The details of recording measurement data are
addressed in the documentation section of Fart 2, which deals vith Nuclear Mate-
rials Measurements.

A.3.2 Measurement Quality

Item b) in the Glossary definition of operating records refers to the re-

sults obtained from the facility programme for control of measurement quality
including standardization and calibration of various measurement systems. These
data would be recorded so as to be available for use in making facility measure-
ments and for facility evaluation of measurement results, e.g. shipper-receiver
difference evaluation, MUF evaluation etc. These data also are needed by IAEA
inspectors in their verification and evaluation of facility nuclear material
accountancy. The details of recording measurement quality programme data are
addressed in the documentation section of Part 3, which deals with Measurement
Quality.

4.3.3 Physical Inventory Taking

Item c) in the Glossary definition of operating records deals not only

with results of the physical inventory taking procedures but includes the proce-
dures themselves. Part 5 addresses Physical Inventory Taking, including the use
of written physical inventory procedures. Part 5 also addresses the documenta-
tion of physical inventory results in terms of the physical inventory item lists
which provide source data for the physical inventory summary and reconciliation,
which can be considered the supporting document for entering a MUF adjustment
into the records.
 90

A.3.A Losses

Item d) in the Glossary definition of operating records refers to actions

that would be non-routine, occurring only when sn accidental or unmeasured loss
occurs. The action to be taken following an accidental or unmeasured loss of nu-
clear material will vary depending on the degree of seriousness of the incident
with regard to its effect on such things as facility operations, personnel
health and safety, and nuclear material accountancy. From a nuclear material
accountancy standpoint the primary concern is to determine the quantity of mate-
rial lost. Documentation of the actions taken and procedures followed in
determining the quantity lost will permit evaluation of the incident in terms of
reliability of the quantity statement and of the cause of the loss, i.e. was it
an actual accident that would account for the loss or could some material have
been diverted? The following information would be useful to include in documen-
tation of the results of a loss investigation:

a) to the extent possible the identity, material type and quantity of ma-
terial missing;

b) basis for establishing the quantity missing;

c) relation of missing quantity to total batch or inventory from which it

is missing;

d) the circumstances and time of discovery of the loss; and

e) the most probable loss mechanism.

The information recorded for items a) and b) would provide source data to
adjust the accounting records for the loss and to provide some estimate of the
uncertainty in the quantity missing. Such losses would be entered into the re-
cords by use of an I CD such as that shown in Figure 4.5.

4.4 Accounting Reports

There are three primary types of reports associated with a nuclear mate-
rial accountancy system: Inventory Change Reports, Material Balance Reports,
and Inventory Reports (Physical Inventory Lists). Other reports may be called
for by facility management or a State Authority, such as a MUF evaluation report
or a measured discard report but only the three primary types are addressed
here. While safeguards agreements call for reports of these types to be
submitted by or through a State Authority to the IAEA, it is the facility nu-
clear material accountancy system that provides the data for the reports whether
they are prepared by the facility operator or by a State Authority. For pur-
poses of discussion it will be assumed that the facility operator prepares these
reports essentially as called for in applicable safeguards agreements.

4.4.1 Inventory Change Reports.

In one sense the documents which contain the data for individual inventory
changes can be considered inventory change reports. For example, copies of an
inventory change document for the receipt of nuclear material at a facility
could be considered a report back to the shipper and possibly to a State

104

Figure 4.10
Data Flow and Document Relationships
 91

Authority that a certain amount of material had been received. In many

countries the State Authority has specified the forms to be used and their dis-
tribution for such inventory changes, i.e., transactions involving transfers of
material between facilities. In the context of the present discussion, however,
these individual inventory change documents are considered to be supporting docu-
ments and a part of the Accounting Records as described in section 4.2.3.

Inventory change reports, as being considered here, are the periodic sum-
mary reports of all of the inventory changes that occurred during a given period
(usually monthly). The data for these reports could be taken directly from the
individual inventory change documents. However, if there are numerous inventory
changes it may be more efficient and convenient to maintain inventory change
journals or ledgers as discussed in section 4.2.2. The data requirements for in-
ventory change reports that may be specified in safeguards agreements by State
Authority or by facility management should be taken into consideration «hen the
format and content of inventory change journal accounts and journal entry forms
are established. Report preparation can be much easier if all of the required
data elements are in the ledger accounts and search of supporting documents is
not needed to complete a report. For infrequent transactions where no journal
accounts are maintained the journal entry forms should be designed to contain
the required data elements. It would further facilitate report preparation if
the journal accounts and forms had essentially the same format as that used in
the inventory change reports. Subsequent examination and verification also
would be easier if there were consistency in format and content of the accounts
and the reports. For NPT Safeguards Agreements reporting, the IAEA has
established the Inventory Change Report formats in Forms R.01.1/c and R.01.2
shown in Figures 4.6 and 4.7. Note that the latter calls for additional iso-
topic information. The Subsidiary Arrangements will stipulate which form is to
be used. Instructions for completion of these forms are given in Code 10 of
the General Part of Subsidiary Arrangements for NPT Safeguards Agreements and
will not be repeated here. Different formats have been established for
inventory change reporting under other safeguards agreements, such as those
pursuant to INFCIRC/66, but essentially the same data elements would be needed
regardless of format.

4.4.2 Material Balance Reports.

A material balance report is a statement of the balance of material on

hand taking into account all inventory changes, adjustments and corrections, and
a physical inventory. One of the adjustmen • would, of course, be the adjust-
ment for the difference between the book inventory and the physical inventory,
i.e. MUF. The data for the material balance report should be available from the
ledger accounts, including the physical inventory data, which are obtained from
the physical inventory taking and become the ending inventory in the ledger when
the MUF adjustment is made. The data requirements for material balance reports
that may be specified in safeguards agreements, by State Authority or by facil-
ity management, should be taken into consideration when the format and content
of the ledger accounts are established. Report preparation can be made easier
if all of the required data elements are in the ledger accounts and search of
journal accounts or supporting documents is not needed to complete a report.
Subsequent examination and verification also will be easier if there is consis-
tency between the reports and the ledgers and between the ledgers and the
journals. For NPT Safeguards Agreements reporting, the IAEA has established the

105
 92

This page is intentionally left blank

i1 1
 1
~ - * ;

1
1 ! 1 •

HOIl
i : ; ! : i i - •
m
! - : ; 1
1
tz
Si — ; ' [ |
B ! ' i : ! ! 1 i

~ "
$t$*e enç*3» 1 J
= ice: t»c.«v
; I
i : i i i 1 ,
i !
1 ? J • ! ; ; i ! ( |

111
1
c : ]
> • ; ! . i i
£
301
; |Ï|I1
,'S
5
!
. , . : :
i , ; • , i ]
! i

i ; . . i
i

; s
: ï • <

— !

; -

C
k~

r
mal
:

MCi.eit;S3C
Tries.»»
I i
J'JVd

F-
LU

—
•t „ -
_
§c:"2 i
•

i i
—
i 1
j i : ! ' : ' i' i
l* C —
ettr

I sl i - ( 1

1
i ;

Q
LLJ (

1
LU
AU

O
IINT

J5
< D ™ j j
ï
• ;

1 '
• [ j
1
, • i :
ti !
: : ; ; : ' ! >
: j • • i , .
i

K *- -
r N0li"WHNC3 C -
Ë • 1
•
- —

i i : 1 i 1 1 : • 1 ' ) 1 i \

u i* 3 |s
! i ! : i ' i ;
i i ' 1 : !

107
l|l!J<ll'f Clltllllt)

I- 2 Comiliy Cmli!
5 I) F.u:ililv Cull»
9 1) MIIACmln
1.1- III ll(Mjinniiiij Dale of round Covorc.l jYYMMDD)
10 24 D.ilf ul HcpoiU'd Evcnl or tiiiliinj Oilo of Period (YYMMOOJ
À'!>*2/ Niiiiil)i'i nl line tntrici. incieajcdbv 1-
70-31 Foi l A t A u w only

70-73 llepoit Nmnlicr

74 ?'J Alii .l.uh- (culled i l Hie IAEAI
 '..7 INVENTORY CHANGE REPORT (ICR) FORM R.01.2
COUNTHY tERIOD COVERED BY RErORT: FROM
FACILITY REPORT No.
MATERIAL BAL «MCE AREA
PAGES I SIGNATURE

QL JL.l I I I
MBA/FACILITY/
I I I I
A C C O U N T A N C Y DATA
I I I !
CORRECTION
TO
COUNTRY NUMBER 1 Ul
DAIE OF lYrs OF NAME OR OF
INVENTORY KMP NUMBER OF ITEMS MATERIAL «EIGHT OF
INVENTORY CODE OE SCRIP IION
CHANGE IN FISSILE
CHANGE BATCH
BATCH
WEIGH 7 O F
ELEMENT ISOTOPES 32
(URANIUM ONLYI
Is)

.1 LJ I I l-l I-I .I.I 11. _ . . I.J_I_LJ_L l-l-l.l L.

j"/i I . T _ _ : L UJ..1.LI.J I I LJ J

i: I I I I I i I I I I I I \.
J-LI-1.1. _ l . LJL i.l I -t-J. I I-I-1J . I I I J I I I I I I I I I I I I I I I

I I I I I I I I I I I I I I I I I I I I I I I I

1. I L.I .1 I. I I L I . I- l-L . I I I I I I I I I I I I I I

» I I I I I I I 1 I 1 t i i i t t t I I I i I I I I I I I I I

I S OT O P I C D A T A CORRECTION
TO

SUM) REPORT
fto.

i J5)
 110

The following sections describe the use of the above documents at the exam-
ple f a c i l i t y . For convenience, a l l of the Heures for the
 97

Material Balance Report format in Form R.03 shown in Figure 4.8. Instructions
for completion of this form are given in Code 10 of the General Part of
Subsidiary Arrangements for NPT Safeguards Agreements and will not be repeated
here. Other safeguards agreements, such as those pursuant to INFCIEC/66, may
not specifically call for material balance reports based on physical
inventories, but many of the same data elements would be required for the
periodic reports called for by such agreements.

4.4.3 Invencory Report•

An inventory report may be a statement of ft book or a physical inventory

at the end of a period. A book inventory report would be based on the data from
the ledger accounts. A physical inventory report would be based on the data
generated during a physical inventory taking. Various levels of detail may be
called for in an inventory report by safeguards agreements, State Authority, or
facility management. For example, facility management may require only a sum-
mary report of physical inventory and MUF, a State Authority may require only a
report of MUF, while an NPT Safeguards Agreement calls for a physical inventory
listing which lists separate batches including batch identity and batch data for
each. The data elements needed to provide the required levels of detail in re-
ports should be taken into consideration in establishing physical inventory tak-
ing procedures. For KPT Safeguards Agreements reporting, the IAEA has
established the Physical Inventory Listing format in Form R.02/c shown in Figure
4.9. Instructions for completion of this form are given in Code 10 of the
General Part of Subsidiary Arrangements for NPT Safeguards Agreements and will
not be repeated here. Procedures for physical inventory taking which will
generate the necessary data for KPT reporting are discussed in detail in Part
5. Other safeguards agreements differ in that some call for book inventory
reports and others for physical inventory reports. The procedures discussed
here should assist in either situation.

4.5. Data Handling

4.5.1 MBA and Facility Records.

Because material accounting reports are made on an MBA basis, it would be

efficient and convenient for each MBA to have its own separate set of nuclear ma-
terial accountancy records. A set of central records also may be kept for the
entire facility as an internal control on the correctness of the individual MBA
records and reports.

For purposes of this discussion and in the interest of simplification, it

is assumed that the facility material control areas are the same as the IAEA-
designated MBAs. The facility operator may choose to divide the facility into
a number of material control areas different from those used for IAEA
safeguards. He also may choose to have separate sets of records for each such
area. If so, the same principles generally would apply to the facility areas as
are discussed here for the IAEA MBAs. However, procedures would have to be
established to select and collect the data from the several facility areas that
would constitute an IAEA MBA to permit material accountability on the IAEA MBA
basis.
 (ANSI and ISO TEST CHART No 2)

98

This page is intentionally left blank

 99

BALANCE 1REPORT IMBR) FORM R.03

F i g u r e 4.8
- MATERIAL
REPORTING PERIOD. FROk. TO
CO LJNTHY . . . . . .
FA CILITT
REPORT N . .
MA T E 8 I A L BALANCE AREA PAGE N>. OF PAGES ! SIGNATURE
t • <• *• >9 ai 11 t» >* «>

1 I ! i 1 1 ' t
, . , 1 Ml 1. . . i : ' (
6

COMCISE V OIE
CORRECTION
COHIIHUA1IOH
A C C O U N T A N C Y D A T A
TO
Z
)-
§

ELEMENT
•c
ECfTRY NAME «EIGHT OF
WEIGHT OF FISSILE ISOTOPES REPORT >

i ELEMENT
1
(URANIUM ONLY)
(!)
i
ac
z
w
i 1 if » »» M
— M ** »• M

, DI i l ' 1
7

! j !

, PI ) ! • • i J 1 1 ! ] 1
:

, a 1 1 1 i ) ' ' i 1 ! i I 1 i i
! 7

, DI > • i • 1 i . . ! | ' • |
J_ « • '

, ia • i 1 • • ' < • i i ; i : 1
7

, DI 1 i 1 ! 1 . :
1
7

DI ' ! ! ! j ' 1 , , i •
7

, DI _] 1 • 1 i i f , 1 > '
7

,
, a
?
• • i : f i ! (

DI I ! : \ i i • ! ! 1 1 1 t ! .
7

M , ! ) • ]
7

a ! 1 1 '
t
1
7

, a
7

, DI i l l ' ' ; 1
1 ' I 1 ! | ' 1 ' '
?

IP ! t t 1 1 • I 1
' ' I ' 11 ! f 1 1

r 1 1 ! 1 1 ! ! f 1 1 | | J 1 ' ! • | 7

, DI 1 1 ' | i 1 ' 1 1 1 1 I I ' I 1 •

7

Fp-HD/Hw.1 (Me 77)

 100

Header Coding:

1- 2 Country Code
5- 8 Facility Code
9-12 MBA Code
13-18 Beginning 0 « e of Period Covered (YYMMDDI
19-24 Ending Date of Period Covered (YYMMDD)
25-27 Number of line Entries, increased by 1.
28-31 For IAEA uie only

70-73 Report Number

74-79 Attribute (coded at the IAEA)

Codes of line Entries (to be used, if desired, as appropriate)

PB RA LN
RA RF LN
RF RALD
RARD LD
RD RATW
RA RS TW
RS RAEU
RA RN EU
RN RAEQ
RANP EQ
NP RATU
RADU TU
DU RALA
RADQ LA
DQ RAED(for decreasei in E)
RA PW ED (for decreases in E)
FW RANE(for decreases in N)
RAED (for increases in D) NE (for decreases in N)
ED (for increases in D) RADE(for decreases in D)
RA NE (for increases in E) DE (for decreases m D)
NE (for increases in E) RANO (for decreases in N)
RA OE (for increases in E) NO (for decreases in N)
OE (for increase) in E) RABE
RA NO (for increases in D) BE
NO «or increases in Dl RAOI
RASF Dl
SF RABA
RASD BA
SD RAPE
RASS PE
SS RAMF
RASN MF
5N
 101

Figure 4.9 PHYSICAL INVENTORY LISTING (PIL) FORM R 02/c

c DATE
F «EPOHTNa.
M fcTEBIAL BALANCE AREA PACE Ha. . OF . PACES SlCHATUilE

, , , 1, , , 1 1 ,,111! 1 1i J , ,, , , i , ,
4

CCSRECTION

CONÇUE NOIE
Z A C C O U N T A N C T D A T A
.,§ TO
KMP CODE

NUUSEK
i o NAME 0 * OF m
i

ELEMENT
NUM8EB OF ITEMS «EICHT OF

s z
BATCH IN
(ATCH
11 «EICHTOF
ELEMENT
FISSILE
ISOTOPES
(URANIUM ONLV)
r I «EPOUT
Na. et
tu z
o a
IMI,9 1 H < MM 1 III 1 1 1II1 i It 111 111 1 1 1 1 1i I I I 1 1 11 I I t |l III ! I H I H i l l
lu.
3

9
9
9

3
i
9
i

1 9

| 9

i
i
s
 102

Header Coding:
1-2 Country Code
S- 8 Facility Code
9-12 MBA Code

19-24 Date IYYMMDD)

25-27 Number of line Entries, increased bv 1.
2S-31 For IAEA use only

45-50 Received •<•> SGIT Icoded at the IAEA!

52-57 Puncning date Icooed ai me IAEA)

70-73 Reoori Number

74-79 Attrc-jte (coded at me I A E A )
 103

4.5.2 Source Data and Operating Records.

Regardless of where the final nuclear material accountability reports are

prepared, the source of the data is at the operating level of the facility.
Quantity data are generated by the operation of the Nuclear Material Measure-
ments System discussed in Part 2. Quantity uncertainty data are generated by
the operation of the Measurement Quality Programme discussed in Part 3. Inven-
tory data are generated by the use of Physical Inventory Taking Procedures
discussed in Part 5. As discussed in these respective parts, various means are
used to record these data where they are generated as they are generated, for ex-
ample, a weigh station with a manual log or a print weight scale, the analytical
laboratory with laboratory notebooks and laboratory-analysis reports, and pro-
cess measurement points where in line NDA devices may be tied directly to a com-
puter print out or where material is weighed and sampled for laboratory anal-
ysis. These data are recorded as the nuclear material accountancy operating re-
cords for the respective areas of the facility to which they are applicable.

The means for recording source data in the operating records will vary as
discussed in the other parts. Regardless of how recorded, it is most important
that the data are identified with specific ite "5 and batches of material and
that this identity can be recorded and referenced in the accounting records,
i.e. Inventory Change Documents, Inventory Change Journal etc. so that the data
can be traced to their origin.

4.5.3. Data Flow.

Data are generated at each KMP for the items and batches moving through or
stored at the KMP. Each item i s identified by a label of some type, physically
attached to the item. The label may be no more than a number etched or stamped
onto a container. For convenience or perhaps for process control the label also
may contain other information such as batch number, material description and
quantity data. The label information is used to identify the item with the mea-
surement quantity and quality data in the operating records and supporting docu-
ments applicable to the item. This i s the f i r s t link in the path from the mea-
surement source data to the final accounting reports. Figure 4.10 shows the
data flow and relationships among source data in operating records, supporting
documents, accounting records, and accounting reports.

When an inventory change occurs in an MBA, e . g . a transfer in or out, a ma-

t e r i a l discard, or an adjustment for a shipper-receiver difference, an Inventory
Change Document (ICD) such as that shown in Figure 4.4 or Figure 4.5 i s
prepared. Much of the information needed to complete these documents w i l l be
available from the source data in the operating records. Other information w i l l
be determined on the basis of each occurrence, e . g . change code, various dates,
shipper and receiver i d e n t i t y , e t c . Batch identity probably would have been
established in the operating records so that item identity within a given batch
could be found there. However, i t may be convenient to include batch/item iden-
t i t y in the "Notes" space of these forms. This might be particularly useful
when items such as fuel rods are being transferred. If a receiver changes a
batch designation this could be noted in the receiver's "Notes" space to main-
tain batch follow-up, if appropriate. Batch and/or item identity is included in
the data on the forms to provide the link from supporting documents to source
data in operating records.
 104

Figure 4.10
Data Flow and Document Relationships

Source Measurement Quality

Data in and Quantity Data Identified
Operating and Item and Batch Items
Records Identity

Inventory Change Facility List of

Documents Inventory Items

Supporting
Documents

Physical Inventory
Summary and
Reconciliation

MUF and.
Adjusted ) Ending I
Book > Physical (
Inventory ) Inventory '

if

Accounting
Records Inventory Change Ledger
Journal and Accounts
ICD Files

/
/

Inventory Material Physical

Accounting Change Balance Inventory
Reports Report Report List (PIU
 105

Distribution of completed ICDs may vary but as a minimum the MBA records
office «ould receive a copy as would f a c i l i t y central records if such are
-maintained. In case of material transfe. , at least one copy would have to be
sent to the shipper'. Copies probably would be retained by the originating
operating unit, e.g. shipping unit, receiving unit, or waste discard u n i t . The
State Authority also may require that they receive copies. If Inventory Change
Journals (ICJ) are maintained, such as shown in Figure 4.1 or Figures 4.2 and
4 . 3 , the ICD is posted to such journals upon receipt by MBA and/or central
records. Those ICDs for which journals are not maintained should be held in a
current file until they are posted to the ledger accounts. Such ICDs may be
posted to the ledger as they are received or may be retained until the end of a
material balance interval, totaled and posted along with the journal accounts.
At the end of a material balance interval the journal accounts and journal entry
ICD forms are totaled and posted to the appropriate ledger accounts. Cross ref-
erences to the ICDs and ICJ accounts are included in the ledger account postings
to permit tracing the data back through the accounting records to the supporting
documents and source data. Figure 4.11 is an example of a General Ledger format
that could be used.

If a material balance interval is closed with a physical inventory taking

then the Facility Fhysical Inventory List and the Physical Inventory Summary are
prepared as described in Fart 5. The MUF which results from the physical inven-
tory reconciliation and the final ending physical inventory values are posted to
the ledger accounts. The ending physical inventory value posted co the ledger
becomes the beginning inventory for the next material balance interval. The
physical inventory source data are obtained from operating records for material
that had been measured prior to the inventory or from measurements made during
the inventory. Item identity information is used to identify the item with the
measurement source data as was done for inventory changes discussed above. If
the material balance interval is not closed with a physical inventory then no
MUF or physical inventory values are generated to be posted to the ledger. A
book inventory for the material balance interval is calculated by totaling the
ledger accounts, including any adjustments such as rounding or shipper-receiver
differences.

If data are generated and handled as described here then:

a) The data contained in the Inventory Change Documents, Inventory Change

Journal accounts and the Inventory Change Journal Entry Form files
should be sufficient to prepare the Inventory Change Report shown as
Figures 4.6 and 4 . 7 .

b) The data contained in the Ledger Accounts should be sufficient to pre-

pare the Material Balance Report as shown in Figu' 4 . 8 . Some data
from the Inventory Change Journal accounts may be needed for adjust-
ments for rounding and for shipper-receiver differences.

c) The data in the final Facility Physical Inventory List should be suffi-
cient to prepare the Physical Inventory Listing as shown in
Figure 4 . 9 .
 Figuro General Ledger

Facility: I.EU - Country ABC HUA: 1 Kleinunl Codo: E

Material Description Unit: G Isotope Code: C,
3% Enrlclietl Uranium
Un Docoiptu Ollioi Addiliuna Sir ;»nonts Olliur Ilcinavala Inwf korjr
Date 1CJ or ICD
lo. H 11-215 U W-215 U U-2VÏ 11 0-215 II ^ 11-2 «
1
2
3
t,
5
6
7
CO

9
in
H
12
n
15
K>
17
18
19
20
21
22
U
25 .
26
27
28
29
30
31
r..-.- li.i.
 X07

4.5.4 Shipper-Receiver Differences.

When nuclear traterial is transferred between MBAs and is measu/ed by both

the shipper and recei >• some differences in quantity may occur. If the differ-
ence is too large to t-a a tributed to measurement uncertainty then investigation
should be undertaken to determine the cause of the difference and determine the
correct quantity. In such cases revised Inventory Change Documents would be
issued containing the correct quantity. If the shipper-receiver difference is
not statistically significant, i . e . , attributable to measurement uncertainty,
but there s t i l l is some difference some mechanism needs to be provided to ac-
count for this difference in the nuclear material accounting records and
reports. The instructions given in Code 10 of the General Part of Subsidiary
Arrangements for NPT Safeguards Agreements for the Inventory Change Report and
the Material Balance Report provide for reporting shipper-receiver
differences. Some means should be included in the nuclear materiel
accountancy system to provide the data for these reports. Because the
Inventory Change Document for Material Transfers, Figure 4.4, includes space
for recording shipper-receiver differences, one means for obtaining
shipper-receiver data would be to go back through the f i l e of the receipts
documents and extract the data. If there are numerous receipts, however, i t
may be more convenient to have journal accounts for shipper-receiver
differences to which would be posted the data as they were determined. The
journal accounts would then provide the shipper-receiver difference data for
each transfer to be reported in the Inventory Change Report and could be
totaled to provide the shipper-receiver data to be included in the ledger
accounts and in the Material Balance Report as adjustments to the book
inventory. A convenient way to accomplish t h i s could be to include a column
in the Receipts Inventory Change Journal as shown in Figure 4.2. Using this
means would assure that a l l transfers were accounted for and that the
shipper-receiver data were associated with the correct transfer.

When material is received into an MBA the shipper's quantity is posted to

the journal account, increasing the MBA inventory. When the material is
measured by the receiver the receiver's quantity is entered on the ICD transfer
form but is not posted. Only the difference is posted to the journal shipper-
receiver difference column in the journal for the appropriate transaction. The
MBA inventory also is adjusted for the amount of the difference. IAEA practice
is always to subtract the shipper-receiver difference. Thus, if the receiver's
measured quantity were larger than the shipper's, the shipper-receiver differ-
ence would be considered negative and when subtracted as an adjustment to inven-
tory would increase the receiving MBA inventory. The reverse would be true if
the receiver measured less than the shipper. If there is no difference then
zero should be posted for that transaction so that when the accounts are totaled
there is assurance that the data for all the transactions have been posted.

4.5.5 Satch Follow-Up.

When transfers of material are made between f a c i l i t i e s and in particular

between facilities in different countries, i . e . international transfers, i t
often is necessary for the receiver to use different batch identity information
to comply with the particular requirements of his facility or State nuclear mate-
r i a l accountancy system. When this is the case there shouH be some means for
identifying the batch identity of the shipper in terms of the batch identity of
the receiver. One easy way to do this is to use the "Motes" column on the
 108

receiver's side of the Material Transfer ICD, Figure 4 . 4 , to record the new
batch identity. Distribution of such forms so annotated should preclude
confusion as to batch t r a c e a b i l i t y . When a f a c i l i t y attachment c a l l s for
batch follow-up, batch changes are required to be reported to the IAEA. The
instructions for this reporting are given in Code 10 of the General Part of
Subsidiary Arrangements for NPT Safeguards Agreements for the Inventory Change
Reports. The information needed for such reports can be entered into the
f a c i l i t y records using ICDs or ICJs for batch changes within MBAs or between
MBAs. at the same f a c i l i t y . F a c i l i t y process records should provide the
necessary data to permit this follow-up.

A.5.6 Internal Control.

The accuracy and integrity of the accounting data in a nuclear material

accountancy system will depend on the control used in the generation and trans-
mission of the data through the system. Some considerations in establishing
such control ares

a) A plan of organization should establish clear lines of authority and

responsibility and separate the operating, recording and custodial
functions. For example, a central nuclear material accounting group
should be organizationally independent from custodians «ho have respon-
s i b i l i t y for the material. The responsibilities and authority of the
two groups should be detailed in a f a c i l i t y procedures manual.

b) Forms, records, and procedures should provide for authorization of in-

-"•- ventory changes and their recording. For example, the ICD for mate-
rial transfers should be completed and signed by the shipping material
custodian. The receiving custodian should verify the data and acknowl-
edge receipt by signing Che form. The transaction should be entered
into the accounting records without delay and the transfer form
properly f i l e d .

c) Duties should be divided so that r.o one person handles a transaction

from beginning to end.

d) Document, should be controlled by pre-numbered serial numbers and a l l

such docuu ts periodically accounted for.

4.6 Example Records and Reports

4.6.1 Introduction.

The facility used in this example is the conversion-fabrication plant

described in Appendix A. Figure A.I shows the process scheme and IAEA MBAs for
the f a c i l i t y . As noted in the Appendix i t i s assumed that the f a c i l i t y control
areas and IAEA MBAs are the same. The records and reports example i s given for
the six month material balance interval 1 August 1982 through 31 January 1983.
General Ledgers are shown in Figures 4.12, 4.13 and 4.14 for the three MBAs. A
fourth General Ledger for the entire f a c i l i t y might also be maintained but is
not shown in the example. The quantities in the Facility General Ledger should
be the sum of the quantities in the MBA ledgers for the respective accounts. The
 109

complete sec of records for Che six-month interval is not shown in the example
because this would greatly increase the size of this document for no real advan-
tage. Examples of various records are shown Sufficient to describe the system
and information flow without showing a l l data for the entire interval.

The records and reports system for the example f a c i l i t y consists of the

following:

1. Inventory Change Document - Material Transfer shown as Figure 4 . 4 .

Each MBA has pre-nunibered forms w*th the prefix MT for material trans-
fer and the MBA number. The forms for MBA-1 thus would be numbered
MT1-0001, MT1-0002, e t c . , for MBA-2 they would be MT2-0001, e t c .

2. Inventory Change Document - Journal Entry Form shown as Figure 4 . 5 .

These are pre-numbered forms with the prefix JE but no MBA prefix.
Blocks of numbered forms are assigned ty the Central Control unit to
MBAs as needed.

3. Inventory Change Journal - Receipts shown as Figure 4 . 2 . These docu-

mencs are coded:

ICJ-1 for MBA-1

ICJ-3 for MBA-2

ICJ-5 for MBA-3.

Pages are numbered consecutively in each MBA.

4. Inventory Change Journal - Removals shown as Figure 4 . 3 - These docu-

ments are coded:

ICJ-2 for MBA-1

ICJ-4 for MBA-2

ICJ-6 for MBA-3.

Pages are numbered consecutively in each MBA.

5. General Ledgers shown as Figure 4 . 1 1 . These documents are identified

by the MBA number and the consecutively numbered pages.

6. IAEA Inventory Change Reports, Material Balance Heports, and Physical

Inventory Liscing shown as Figures 4 . 6 , 4 . 7 , 4.8 and 4 . 9 .

7. Various operating records from which source data are obtained to docu-
ment inventory changes using the above identified forms. Documenta-
tion of source -\ ;a is discussed in those parts addressing i t s genera-
tion, i . e . Part 2 , Nuclear Material Measurements, Part 3 , Measurement
Quality, and Part 5, Physical Inventory Taking.
 110

The following sections describe the use of the above documents at the exam-
ple f a c i l i t y . For convenience, a l l of the figures for the example are at the
end of the example section.

4.6.2 Inventory and Material Flow Assumptions.

The example given in Fart 5 describes the physical inventory taking as of

31 January 1983 at the end of a six-Don th material balance interval. It is
assumed that physical inventories of the three MBAs also were taken as of 31
July 1982. These physical inventory values are shown on line 1 of the General
Ledgers, Figures 4.12, 4.13 and 4.14 in the Inventory account. They are dated
as of 31 July 1982 with reference to the PILs of 4 August 1982, the date when
the final PILs were completed and the inventories reconciled. These are, of
course, assumed values with no support being shown in the example. They are the
ending inventory values for Che interval ended 31 July 1962 and beginning inven-
tory values for the interval 1 August 1982 through 31 January 1983.

During the example material balance interval the following material flows
are assumed, based on a throughput of twenty to twenty-two 1 MTU batches per
month :
a) Receipts from off s i t e into MBA-1 :

98 cylinders UFg (98 batches) at about 1 MTU per cylinder.

1280 buckets U02 (32 batches) at about 25 kgU per bucket.

b) Transfers from MBA-1 to MBA-2:

98 cylinders JFg (9'J batches) at about 1 MTU per cylinder.

1280 buckets UO2 (32 batches) at about 25 kgU per bucket.

c) UFg heels returned from MBA-2 to MBA-1 and shipped o f f - s i t e from

MBA-1.

d) Transfer of loaded rods from MBA-2 to MBA-3 assuming bbout twenty

200-rod batches per month containing 1 MTU per batch.

e) Discard from MBA-2 of about 350 batches of liquid waste and 260
batches of solid waste containing a total of about 2 MTU.

f) Transfer of reject rods from MBA-3 to MBA-2 assuming about a 12 reject

rate or about 260 rods containing 1.3 MTU.
g) Shipment off s i t e from MBA-3 of 247 assemblies containing 123.5 MTU.

4.6.3 Example Transaction Recording.

Using the material flows assumed in section 4.6.2 above, example transac-
tion documents are described in the following paragraphs to show how each would
be entered into the f a c i l i t y records.

124
 Ill

4.6.3.1 Off Site Receipts.

It is assumed that the off s i t e receipts are received uniformly-

over the material balance period. It is further assumed that about 10 MTU are
received every other week. Figure 4.15 is an example of the transfer document
for the first shipment of UF5 for the example material balance i n t e r v a l . The
shipment is assumed to have been made from an enrichment plant in the same coun-
try as the fuel fabrication f a c i l i t y , thus the shipping change code SD (shipment
domestic). The various codes on the form are those as given in the Code 10 part
of NPT Safeguards Agreements. It is assumed that the shipping f a c i l i t y operates
under the same system as the receiving f a c i l i t y using the same forms, e t c . , be-
cause they are in the same country. The shipper's data has been entered on the
form including the cylinder i d e n t i f i c a t i o n numbers as the batch i d e n t i t y . In
the example f a c i l i t y these also are used as batch i d e n t i t y . If this were not
the case a new batch identity could be entered in the notes column on the re-
c e i v e r ' s side of the form upon receipt of the shipment. The shipper's data are
posted to the receipts journal, Figure 4 . 1 6 . These data appear on line 1 of Fig-
ure 4.16 dated the date of r e c e i p t . The change code i s , of course, RD (receipts
domestic) for the example f a c i l i t y and is received in MBA 1 at KMP 1A. The
shipper-receiver data would not have been entered at t h i s time. When the
receiver's measurements have been completed the r e c e i v e r ' s side of the transfer
document, Figure 4 . 1 5 , is completed, including the shipper-receiver difference
data. These data are then posted to the receipts journal for the appropriate
transfer as indicated by ICD No WT2-0271. Each of the numbers on the receiver's
side of the material transfer document, i . e . net weight, percent element, and
percent isotopic, should be traceable to f a c i l i t y source documents such as
weight records and laboratory reports and notebooks and identified with the cyl-
inder batch number. Similar shipments are received every two weeks and are
posted to the receipts journal with the appropriate document reference to the
subsequent transfer documents. These postings would appear on subsequent l i n e s
in Figure 4.16 to represent the nine additional UF5 receipts ( l i n e s
4,8,11,15,18,22,25,28 and 3 0 ) . Examples of the material transfer documents for
these are not shown but would be e s s e n t i a l l y the same as Figure 4 . 1 5 . Periodi-
c a l l y , as may be called for by the f a c i l i t y attachment, the data from the r e -
ceipts journal and material transfer documents would be used to prepare the In
ventory Change Report R.01.1/c or R.0.2, Figures 4.6 and 4 . 7 - Examples of these
w i l l be shown l a t e r in this description.

Later during the material balance interval, receipts were changed

from UFg to UO2. Figure 4.17 is an example of the transfer document for the
f i r s t shipnment of UO2. It is assumed to be a domestic shipment from a conver-
sion f a c i l i t y in the same country. In the case of the UFg each line on the
transfer document represented a batch of 1 item, i . e . a cylinder. In the case
of the UO2 each line on the transfer document represents a batch which consists
of 40 items, i . e . buckets. Separate l i n e s on the transfer document could be
used for each bucket, but i t may be more convenient to attach a separate sheet
which would shown the weight data, gross, tare and n e t , for each bucket. It is
assumed that the shipper's measurement programme i s the same as the receiver's
in this case, i . e . three buckets are sampled and the samples analyzed. The ura-
nium and isotopic factors resulting from these analyses are applied to the net
weight of each bucket and these data recorded on the labels or in some document
identifiable with each bucket. These source data would need to be sem: with the
 112

material transfer; they are not shown in the example. Similar data would be
generated and recorded in the source documents for MBA-1 for the receiver.

Upon receipt the shipper's data are posted to the receipts journal
as shown on line 31 of Figure 4.16 in the same way as the UFg data were posted.
When the receiver's measurements are completed) the receiver's side of the trans-
fer document is completed and the shipper-receiver difference posted to the jour-
nal on the appropriate l i n e , i . e . in this case for ICD number MT2-0324. Two ad-
ditional UC>2 shipments are received during the material balance period. These
are shown on lines 33 and 34 of Figure 4 . 1 6 . The material transfer documents
are not shown in the example but would be similar to the example of Figure 4 . 1 5 .

4.6.3.2 Transfers from MBA-] to MBA-2.

It is assumed that transfers of material are made several batches

at a time, perhaps once each week, i . e . five or six 1 MTU batches. Thus, the
f i r s t transfer for the material balance interval would have been immediately
after the inventory on 4 August 1982. It is assumed this transfer consisted of
six batches. A material transfer ICD for this transfer i s shown in Figure 4 . 1 8 .
MBA-1 completes the shipper's side based on the data generated when those spe-
c i f i c cylinders were received. MBA-2 completes the receiver's side based on the
net weight of material removed from the cylinder and put into process. The e l e -
ment and isotope factors are those determined at KMP-1A in MT.A-1. The transfer
is posted to the removals journal of MBA-1 and to the receipts journal of MBA-
2 using the shipper's data (MBA-1) when the transfer i s made. This i s shown on
line 1 of Figures 4.19 and 4.20 respectively. Other such transfers are shown on
subsequent lines ( l i n e s 2 , 3 , 5 , 7 , 9 , 1 0 , 1 2 , 1 3 ) 1 6 , 1 7 , 1 9 , 2 0 , 2 3 , 2 4 and 26 of Figure
4.19 and 2 , 3 , 4 , 7 , 8 , 9 , 1 0 , 1 1 , 1 4 , 1 5 , 1 6 , 1 7 , 2 0 , 2 1 and 22 of Figure 4 . 2 0 ) , but no exam-
ple transfer documents are shown.

It appears that there are shipper-receiver differences on the six

batches but they actually are the heels l e f t in the cylinders. The heels are
not posted to the journal as were the shipper-receiver differences because the
heels quantities are included in the quantities received from MBA-1 and posted
in the receipts journal, Figure 4 . 2 0 . The heels w i l l be returned to MBA-1 and
sent back to the enrichment plant. Transfer document examples are not shown for
these heels transactions from MBA-2 to MBA-1 because they would be similar to
Figures 4.15 and 4 . 1 8 . Journal postings for these heel transfers are shown on
lines 3 , 4 , 5 , 8 , 1 1 , 1 2 , 1 3 , 1 4 , 1 9 , 2 0 , 2 1 , 2 2 , 2 5 , 2 8 , 2 9 and 30 of Figure 4.21 and l i n e s
2 , 3 , 5 , 7 , 9 , 1 0 , 1 2 , 1 3 , 1 6 , 1 7 , 1 9 , 2 0 , 2 3 , 2 4 , 2 6 and 27 of Figure 4 . 1 6 . I t is assumed
that heels are returned from MBA-1 to the enrichment plant when a shipment of
UFg is received. These transactions are shown in the MBA-1 removals journal on
lines 4,8,11,15,18,22,25 and 27 of Figure 4 . 1 9 .

Transactions similar to these for UFg would occur when batches of

U02 were transferrred from MBA-1 to MBA-2 except that there would be no heel r e -
turns to MBA-1. The material transfer document would be similar to that for UFg
transfers shown in Figure 4.18 except the receiver (MBA-2) would use shipper
(MBA-1) data as measured at KMP-1C. These transactions would be posted tc the re-
spective MBA removals and receipts journals as they occurred. They are shown on
lines 29,30,31,33,34 and 35 of Figure 4.19 and lines 25,26,27,30,31 and 32 of
Figure 4 . 2 0 .
 113

For batch follow-up purposes the original batch i d e n t i t y , i . e . c y l -

inder number for UFg and batch numbers for UO2, would be used in the transfers
from MBA-1 to MBA-2. The cylinder numbers also could be used for the transfers
of h e e l s . This would not cause confusion because i t would be impractical to use
the cylinder number for batches of UO2 produced in the conversion part of the fa-
c i l i t y . Records of the batches of UFg put into the process would be kept but it
would not be practical or necessary to try to identify U02 batches produced with
UFg batches introduced. As U02 batches are produced they would be assigned
batch i d e n t i t i e s and measured at KMP-4 either for process control or for subse-
quent inventory at KMP-H. Records of these measurements should be maintained as
described in Part 2 , Nuclear Material Measurements. See Table A.3 in Appendix
A for batch identification for the example f a c i l i t y .

4.6.3.3 Transfers from MBA-2. There are three types of transfers

out of MBA-2:

1) Heels to MBA-1 as described above.

2) Transfers of loaded rods to MBA-3 and

3) Discard of both liquid and s o l i d waste.

As described in Appendix A, UO2 batches are processed into sintered

p e l l e t s and stored in ground p e l l e t storage (KMP-J) on the basis of measurements
made at KMP-5. The results of these measurements would be recorded as described
in Part 2 , Nuclear Material Measurements and be used at KMP-6 to record the quan-
tity loaded into the rods. Process records would show the batch number and item
identity of each rod as i t moved through the process. When the loading of each
batch of 200 rods is completed ( t h i s would occur approximately daily) an Inven-
tory Change Document - Material Transfer i s prepared for the transfer and the
rods transferred to MBA-3. An example of such a document is shown as Figure
4 . 2 2 . While the individual rods could be l i s t e d on the lines of the transfer
document i t probably would be more convenient to attach the rod-loading protocol
( i . e . , l i s t i n g of rod loadings) for each batch. This would give the individual
rod identity and loading data. These data would be obtained from the source
data records from KMP-5 a n d KMP-6 as noted in Figure A.I in Appendix A. Only
the total batch data would be entered on the front of the transfer document.
There would be l i t t l e or no advantage in posting each batch transaction to the
journals as they occur although this could be done to assure that a l l were
recorded. Each period (probably monthly) when the Inventory Change Report (ICR)
is prepared for the IAEA, the rod transfer documents (there would be 25—30 of
them) could be t o t a l l e d . These totals would then be postsd to the respective re-
ceipts and removals journals of MBA-2 and MBA-3. Figure 4.22 shows one batch
transfer. The rod protocols are not shown but they would be a l i s t of the rods
with their associated identity and quantity data. Totals for the monthly trans-
fers are shown on lines 6 , 1 5 , 2 3 , 3 1 , 3 5 , and 39 of Figure 4.21 for shipment from
MBA-2 and on lines 1 through 6 of Figure 4.23 for MBA-3 r e c e i p t s . Note, how-
ever, that because each transfer represents a batch, they would have to be
listed separately on the ICR, as shown later in Figure 4 . 2 8 .

Haste is generated at various places in the process in MBA-2 and is

collected and processed to KMP-E and KMP- F as shown in Figure A.I in Appendix
A. The quantities measured at KMP-E and KMP-F are recorded as described in
 11*

Part 2 , Nuclear Material Measurements. Periodically the wastes are discarded to

burial or waste liquid lagoons. While a material transfer document could be
used for such discards the transaction is not truly a transfer in that the mate-
r i a l is no longer carried in any MBA inventory. A more appropriate form would
be one such as shown as Figure 4 . 2 4 . This form provides space for a l l the data
elements needed to record the removal of the discards from inventory along with
signatures showing approval of the transaction. It has been assumed that a log
or other process record is maintained by the organizational unit responsible for
discards in which the identity and quantity source data are recorded as they are
generated. This log is referenced in the journal entry form to provide
traceability to the source data. Figure 4.24 shows the solid waste discards
made at one time during the example material-balance interval. The individual
solid waste batches SW-0102 through SW-0222 shown in Figure 4.24 could have been
listed on the I CD - Journal Entry Form but i t is convenient and reduces
paperwork to record the t o t a l s and reference the Waste Discard Log for individ-
ual barrel and drum data. In this case the PIL of 1-8-82 is referenced because
the discard consisted of the 120 barrels of solid waste waste on inventory at
the beginning of the material balance interval. The journal entry form is
posted to the MBA-2 inventory change journal for removals. The posting to the
removals journal is shown on line 1 of Figure 4 . 2 1 . Other such transactions are
shown on subsequent l i n e s of Figure 4.21 ( l i n e s 2,9,10,17,18,26,27,33,34,37 and
38). Similar journal entry forms and journal e n t r i e s would be made for the l i q -
uid waste discards. Note that the waste discards are reported on the ICR (Fig.
4.28) as a single batch for each type of waste to avoid l i s t i n g each item sepa-
r a t e l y , resulting in 220 l i n e s on the ICR. The journal entry reference numbers
would permit tracing back to individual item data i f needed.

4.6.3.4 Shipments and Transfers from MBA-3.

There are two types of transfers out of MBA-3:

1) Shipments of finished fuel assemblies and

2) Return of reject rods to MBA-2.

Shipments of completed assemblies to various reactors w i l l vary

depending on reactor reload schedules and the p o l i c i e s of the reactor operators
regarding load times and fresh fuel inventories. Figure 4.25 i s an example
transfer document for a shipment of 10 assemblies to a reactor. The receiver's
side has been completed assuming that the reactor operator accepts the fuel fa-
c i l i t y data and has returned a copy of the transfer document to the shipper. The
quantity data would be obtained from the rod protocols sent with the rods from
MBA-2. These data in turn can be traced back to the source data records of
KMP-5 and 6. Each shipment i s posted to the removal journal as shown on line 1
of Figure 4 . 2 6 . Other such shipments are shown on subsequent l i n e s of Figure
4.26 ( l i n e s 2 , 3 , 6 , 7 , 1 0 , 1 1 , 1 4 , 1 5 , 1 6 , 1 9 , 2 2 and 2 3 ) .

Reject rods are returned to MBA-2 from MBA-3 for rework or recov-
ery. This transfer is documented in e s s e n t i a l l y the same way as the original
transfer from MBA-2 to MBA-3 except that the rods are no longer necessarily
associated with the batches they were in o r i g i n a l l y . A group of reject rods
transferred back to MBA-2 could be considered as a batch i f their composition
and quantity can be defined by a s i n g l e set of s p e c i f i c a t i o n s . It is assumed

128
 115

that reject rods are accumulated in MBA-3 and returned to MBA-2 at about one-
month intervals (probably just before the IC? i s made to the IAEA). Figure 4.27
is an example of the transfer document for the return of rods to MBA-2. A s i n -
gle line item on the front of the document i s supported by the reject rod proto-
col attached (not shown in the example). The data for the protocol would come
from the record of the original transfer of the particular rod from MBA-2 to
MBA-3. A reject batch number prefixed R is assigned to successive transfers.
Each transfer is posted to the respective removals and receipts journals for
MBA-3 and MBA-2 as shown on lines 4,8,12,17,20 and 24 of Figure 4.26 for the
MBA-3 reject rod removals and on l i n e s 5,12,18,23,28 and 33 of Figure 4.20 for
the MBA-2 reject rod returns.

4.6.4 Example Inventory Change Journals and General Ledgers.

Figures 4.16, 4.20 and 4.23 are example Inventory Change Journals Receipts
for the three example f a c i l i t y MBAs respectively for the material balance inter-
val 1 August 1982 through 31 January 1983. Figures 4 . 1 9 , 4.21 and 4.26 are the
example Inventory Change Journals - Removals for the three MBAs for the same i n -
terval. The posting of these journals records the transactions described in the
preceding sections 4 . 6 . 3 . 1 through 4 . 6 . 3 . 4 . All of the transactions have been
posted but only typical example transaction documents such as material transfer
I CDs have been shown (Figures 4 . I S , 4 . 1 7 , 4 . 1 8 , 4 . 2 2 , 4.25 and 4 . 2 7 ) .

At the end of each month during the material balance interval a l l of the
journal accounts are totalled and posted to the General Ledger for the three
MBAs, Figures 4 . 1 2 , 4.13 and 4 . 1 4 . The beginning inventories as discussed in
section 4 . 6 . 2 are shown on line 1 of each ledger. The f i r s t month's receipts
and shiptr;nts are shown on the next two lines and other removals are shown on
the next l i n e s . There are no "other receipts" in the example. The other r e -
movals in the example are shipper receiver difference for MBA-1 and discards for
MBA-2. Shipper-receiver differences are considered other removals because the
IAEA practice is always to subtract shipper-receiver differences. See Section
4 . 5 . 4 , Shipper-Receiver Differences, for further discussion. For each line a
running book inventory balance i s calculated either adding or subtracting as ap-
propriate the line entry from the preceding line inventory balance. This gives
a book inventory at the end of each month. At the end of the material balance
interval an adjustment for MUF from the physical inventory summary and a final
entry of the ending physical inventory for the material-balance interval are
posted. In the example only MBA-2 has a MDF because MBA-1 and MBA-3 are item
control areas and should not be expected to have MOFs.

4.6.5 Example Accounting Reports

It is assumed that Inventory Change Reports are submitted to the IAEA each
month and a Material Balance Report and Physical Inventory Listing are submitted
at the close of each material balance interval. Figure 4.28 i s an example Inven-
tory Change Report for the f i r s t month of the material balance interval, i . e .
August 1982 for MBA-2. Only the f i r s t page of the report is shown for Figure
4.28. A second page would be needed to l i s t the remaining fuel rod batches
transferred to MBA-3. Similar reports would be made for the other MBAs. The
data to complete this report can be taken from the Inventory Change Journal -
Receipts ( l i n e s 1 through 5 of Figure 4.20) and the Inventory Change Journal -
Removals ( l i n e s 1 through 5 of Figure 4 . 2 1 ) . While each of these lines repre-
 116

sents Che transfer of several f a c i l i t y batches, a l l such batches can be

considered together as a batch for IAEA reporting purposes because they a l l have
composition and quantity defined by a single set oi specifications or
measurements. This combination w i l l considerably reduce the amount of paper
involved in these reports. For example, the f i r s t line on Figure 4.20 is the
summation of the data on the ICD-Material Transfer No. MT1-0100 (Figure 4.18)
which l i s t s six separate batches of UFg transferred from MBA-1 to MBA-2. If
each of these is l i s t e d on the ICR chat means six lines instead of one. Thus,
if each f a c i l i t y batch is l i s t e d the report would require more'than 50 lines in-
stead of only ten. If i t is necessary to l i s t a l l f a c i l i t y batches on the ICR
then the data would have to be obtained from the ICDs, such as MT1-0100,
referenced in the column headed "1CD No." in the journals. The exception to the
individual batch grouping in the example is the fuel rod batch l i s t i n g «here
each batch is transferred on a separate transfer document. The f a c i l i t y may
wish to use different batch transfer procedures to permit combining batches on
the ICR.

The Material Balance Report to Che IAEA i s shown in Part 6, Figure 6 . 1 . A

comparison of the data in Figure 6.1 shows that the General Ledger for MBA-2,
Figure 4 . 1 3 , provides the data needed for the report. Similar reports can be
prepared using the ledgers for the other MBAs. In Part 6, Table 6.1 is a
summary of f a c i l i t y record data. Comparison of these data with the documents
described in this section w i l l show the source of the data in Table 6 . 1 .

Figure 5.7 in Part 5 is an example Physical Inventory L i s t i n g . I t s

preparation is described in Part 5. The verified f a c i l i t y inventory l i s t i n g and
the f a c i l i t y Physical Inventory Summary provide the data needed to prepare the
Physical Inventory Listing sent to the IAEA and to make Che final MUF and inven-
tory entries in Che ledgers.

4.6.6 Batch Follow-up

In some cases i t is required to maintain batch follow-up in an MBA. The

following paragraphs describe how t h i s could be done in the example f a c i l i t y .

In MBA-1 i t is quite simple because the batch identity and quantity as

received from off s i t e can be used to transfer material to MBA-1. The shipper's
batch i d e n t i t i e s can be used or new identity assigned and noted on the transfer
document recording the receiver's data.

In MBA-2 the matter is more complex. In the conversion process i t is un-

likely that i t would be possible to maintain batch identity through the process.
Batches of UFg or UNH are put into the process but i t would be d i f f i c u l t and
probably of l i t t l e value to attempt to identify which UO2 and scrap batches come
from which UFg or UNH batches. It would suffice chat process records show the
batch data for batches added to the process and new batch data for batches of
UOo, scrap and waste generated. For customer specification purposes i t may be
necessary to separate different materials or enrichments. In such cases a s e -
ries of process input batches could be identified with a s e r i e s of process
output batches.

In the p e l l e t i z i n g process the U02 batch identity could be carried through

the process as indicated in Appendix A, Table A.3. As each batch of UO2 i s
 117

processed Che p e l l e t s produced retain the UO2 batch number. Any scrap and waste
generated could be identified with the UOj batch from which i t came so that the
KM/RP code of Code 10 may be used. Rods prepared from p e l l e t s could s t i l l r e -
tain the UOj batch identity so that when transferred to MBA-3 the RM/KP code of
Code 10 could be applied.

In MBA-3 the situation is again simple. Rods are identified in batches

that go into assemblies or are rejected and returned to MBA-2. Process records
w i l l show which assemblies were prepared from which batches and which rods were
rejected. These data w i l l allow application of the Code 10 procedures.
As with UFg and UNH input to conversion, rework of p e l l e t s and scrap recov-
ery make it d i f f i c u l t to maintain batch identity except on a customer or con-
tract campaign b a s i s . In these cases that may be a l l that is needed or p r a c t i -
cal.

References

IAEA/SG/INF/1: IAEA Safeguards Glossary, IAEA/SG/INF/1, International Atomic

Energy Agency, Vienna, 1980.
 118

This page is intentionally left blank

132

I I I I I I I I I I I I I II1I II I . I
 Flgilre 4.12 General Ledger

Facility: t£tt - Country ABC MBA: I Element Code: T.

Haterial Description: 3Z Enriched Uranium Unit: G I notope Code: G

Line Rrcoiptft Other Additions Shipment** Other Peranvals Inven'.ory

No. Date ICJ or ICD 11 U-235 U U-235 U U-235 H 0-235 U U-235

1 31- 7-82 PIL of 4-8-82 30000000 900000

2 31- 8-82 ICJ-1 Page 23 20115920 604160 -9560 80 501254BO I5Ô4ÎOO
3 31- 8-e2 ICJ-2 Page 18 23972730 719410 26Ï52750 7B469O
4 30- 9-82 1MJ-I Page 23 20196550 606200 1660 210 46347640 1390680
5 30- 9-82 ICJ-2 Psge 18 30291990 910140 16055650 4B054Ô
31-10-82 ICJ-1 >R' 23 20161100 604840 362I646C 1085160
31-10-82 ICJ-2 «8° 1» 24151290 724970 !206517( 360215
8 30-11-82 ICJ-1 Pane 23 30157840 905380 -650 -1010 4222366(1 1266600
9 30-11-82 ICJ-2 Page 18 20225330 606430 2199813(1 660170
10 31-12-82 iCJ-1 Page 23 19986500 599200 -20320 50 42005150 1259320
11 31-12-82 ICJ-2 Page 18 17001960 506820 Î5OO3I9J 752500
12 31-1-83 ICJ-1 Page 23 2000/360 599650 12920 -50 44997631) 1352200
13 31- 1-83 ICJ-2 Page 18 _I499763O 452200 300000CO 900000
14 31- 1-83 Totals 130625270' 3919455* 130640930* 391997O r ~ 3OO000('O 900000
15 31- 1-83 MUF 300000110 900000
J6_ 31- 1-83 PIL of 4-2-83 30001)0'JO 900001
17
18 Explanatory Notes: ,
19 ~~ ~~ _
20 'Running book inventory is calculated for each entry, tine I is the ending physical inventory for the prior material-balance
2J interval. Line 16 ia the ending physical inventory for the current Material-balance interval.
22 " "
23 Total reccipta for the Material balance interval as shown on line 36 in Fig. 4.16t Inventory Change JournaI-Rcceipta for HBA-
24
2_5 Total shipaents as shown on ÎTnc 37 in F I R . 4 . l 9 t Inventory Change Journal-Removal a, for H B A - I .
26 ________^^_»_______»___^_____
27 Shipper-receiver difference entry as shown on line 36 in Fin. 4*16.
28 ~
29 'Physical inventory and book inventory are tiic same, hence no HUT.
30 ~ ~
31 ' "
Page Ho. 0075
 Figure 4.13 General Ledger

Facility) I-EU - Country ABC HUA: 2 Elèvent Code: E

Material Description: 31 Enriched Uranium Unit: G Isotope Code: G

Line Receipts Other Addition! Shi paents Other Reaovals Inventory

No. Date 1CJ or ICD II U-235 U U-235 U O-2JS U U-235 U U-235

1 31- 7-82 PIL of 4-8-82 5820000 175000

2 31- 8-82 ICJ-3 PaRe 65 24113190 723640 24II3190 898640
3 31- 8-82 ICJ-4 Page 35 20143J40 604990 800000 2400(1 8989850 269650
4 30- 9-82 ICJ-3 Page 65 30351780 911940 39341630 1181590
5 30- 9-82 ICJ-4 Page 35 21202020 635830 225000 67'0 17914610 539010
6 31-10-82 ICJ-3 Page 65 24213480 726880 42128090 1265890
7 31-10-82 ICJ-4 Pane 35 21162760 635560 250000 7'O0 20715330 622830
B 30-11-82 ICJ-3 Pane 65 2O2O5I5O 605820 40920480 1228650
9 30-11-82 ICJ-4 PaRe 35 2IU424O 634S6O 250000 -500 19536240 586590
10 31-12-82 ICJ-3 PaRe 65 1722IB8O 513410 36758120 HOODOO
11 31-12-82 ICJ-4 Page 35 20996250 629930 250000 7500 15511870 462570
12 31- 1-83 ICJ-3 Page 65 15194510 458110 30706380 920680
13 31- 1-83 ICJ-4 PaRo 35 21002470 610270 225000 6750 9478910 283660
14 31- 1-83 Total 9 131299990' 3939800' 125641080' 3771140' 2000000'' 60000' 9478910 283660
15 31- 1-83 HUF 354050* 8670' 9124860 274990
16 31- 1-83 PIL of 4-2-B3 9124860 274990
17
18 Explanatory Motcai
19
20 'Running book inventory ia calculated for each entry. Line I i» the ending pliytiça.1 inventory for the prior ••terijii balance
21 interval. Line 16 1» the ending nhyatcal inventory for the current Material balance interval.
22
23_ 'Total receipts for the Material bajnnce interval «a shown on line 35 in Fig. 4.20. Inventory Change Journal - Receipts, for HPA-2.
24_
25 jhipmenta aa ahuwu on liiuj M in .Fig. A.jli Inventory Change Journal-Rcaoyalt. for HPA-2.
26_
27 other removals aa ahown on line M in ' $;?li Inventory change Journal-Rcwovala. for Httft-2»
28
29 the fhyaical Inventory Sntrnnry. Table 5>2, prepared in the ewaaipje for Physical Inventory*
30_
31 *
Page No.
 Figure 4.14 General Ledger

Facility: I£U - Country ABC HBA: 3 E!caent Code: E

Material Deacri ption: 3X Enriched Ut un i un

a Un i t: C Inotnpe Code: G

Line Receipts Other Addi tions Shipments Other Removals Inventory1

Ho. Date ICJ or ICI) U U-235 II U-235 U 11-215 U 11-235 U U-215

1 31- 7-82 PH. of 4-8-82 90100000 2709000

2 31- 8-02 ICJ-5 Page 52 20021750 601340 IIO32175O 13IO34C
3 31- 8-82 1CJ-5 Page 47 20232940 60B470 90088810 27OI87C
4 30- 9-82 ICJ-5 Page 52 21002860 629870 IIIO9167O 3311760
5 30- 9-82 1CJ-6 Poge 47 29220090 606600 90S 71580 2725140
6 31-10-82 ICJ-5 Page 52 21001920 610760 III873500 33559O0
7 31-10-82 ICJ-6 Page 47 20228920 607050 91644580 274885C
B 30-11-82 ICJ-5 Page 52 20974890 629780 112619470 3378630
9 30-11-82 ICJ-6 Page 47 30220380 906900 82399090 2471730
10 31-12-82 ICJ-5 Page 52 20996250 629930 . 103395340 3101660
11 31-12-82 ICJ-6 Page 47 II198240 3365S0 92197100 2765110
12 31- 1-83 ICJ-5 Page 52 210024/0 630270 113199570 3395380
13 31- 1-83 ICJ-6 Page 47 22699620 680400 90499950 2714980
14 31- 1-83 Tot-ls 1250001401 3751950' 12'. 800190" 3745970» 90499950 2714980
15 31- 1-83 HUF - 0 - - 0 - 90499950 2714980
16 31- 1-83 PIL of 4-2-83 90499950 2714980
1?
ta Expjdnator y Noteal
1<J
20 'Running book inventory is calcul tot! for ench entry. Line 1 is the endingjthysical inventory for the prior material balance

22
23 'Total receipts for tlic «sterfrtl balance interval as shown on line 13 in Fig. 4.23« Inventory Change Journal-Receipts for HBA-3.
24
25 •Total •h i puent» aa shown on line 26 in Fix. 4.26, Inventory Change Journal-Rénovai a, for HBA-3*
26
27
28
29
30
31
Page Ho. 0225
 Figure 4.15 Inventory Changs Document - Material T r a m fer

Material Description: UT, - 3X enriched Material Code: 0CRB Iiotope Codai G Element Codes E Unitt C

Shipper flame and Addrees: Enrichment Facility (UEF) MBA: 2 Receiver Name and Addreaa: Facility LE'J MBA; 1
Country ABC Country ABC
Change Codel SD Change Codel RD

KHP Code: 2 KMP Codel 1A

Shipper Dtta: Receiver Datai

Line Batch No of Net I Element X laotope Net Z Element X laotope

No. Identity Items Weight Ele. Weight Iso. Height Weight Ele. Weight lao* Height Notea

1 C42608 1 1486740 67.56 1002560 3.011 30240 1486540 67.57 1004460 3.011 30240
2 C49641 1 1478330 67.57 998910 2.986 29830 1478620 67.56 998830 2.986 29830
3 C35OU 1 1474320 67.67 9971/70 2.993 29860 1474620 67.66 997730 2.993 29860
4 C59 741 1 1481300 67.66 1000370 3.011 30180 1481400 67.63 1001870 3.011 30170
5 C60497 1 1484700 67.64 1002370 2.996 30090 1484500 67.64 1004120 2.991 30030
6 Clt>939 1 1475960 67.65 998490 2.991 29860 1475460 67.64 998000 2.989 29830
7 CEO352 1 1486780 67.53 1002140 3.014 30260 1486700 67.55 1004270 3.011 30240
8 «0'<2l 1 1485790 67.57 1002070 3.000 30120 1485290 67.60 1004060 2.990 30020
9 C322J5 1 1474010 67.61 996580 2.991 ' 29810 1474000 67.63 996870 2.992 29830
10 C7328O 1 1472880 67.65 996400 2.993 29820 1472570 67.63 995900 2.991 29790
11 Total 10 14B00810 - 9997560 l - 300070' 14799510 - 10006110 - 299840
12
13 Explanatory Note:
14
15 'Entered to Line 1. F I R . 4 .16.

Doc. No. Notes: Shipper-Receiver Difference! Date Keaeuredi

HT2-O2J1 U -8J50 V U-235 »230 »' JO-8-82

Shipping Datei Shipper Signature I Receiving Datei Rocalver Signât»*!

5-8-82 9-8-82
 Figura 4.16 Inventory Change Journal * Receipt!

Facility: Facility LEU - Country ABC Material D e s c r i p t i o n : 31 E n r i c h e d Uranium KBAI 1 lac tpe Code: G Element Code: E Unit! G

Line Date of IC I CD Ho. o f Htl. Meaa « Receipts S/S differences Other Additions
!!o. Change From To Code Code Number Items Code Code U U-235 U U-235 U U-235

1 9-8-82 UÎK LEU RD 1A MT2-0271 10 4.CRB H 9997560 300070 -8550 •230

1J-8-82 HÎIA2 HDA1 II D 0 KT2-0125 6 ,hCIUl tl «0560 1220 - 0 - - 0 -
3 18-a-a; MUA 2 MUAI Rl) H MT2-0126 6 N 39640 1190 - 0 - - 0 -
4 2J-3-82 UEF LEU RD 1A MT2-O272 10 ,hCim M 9996770 300460 -1010 -150
5 25-8-32 K11A2 MIU1 Rl) B MT2-O127 f, ,fCRB N 41390 1240 - 0 - - 0 -
6 Jl-3-32 Monthly T o t a l 3 20115920 604180 -9560 80
7 1-9-8 2 NIIA2 MIIA1 RD n MT2-O128 6 ,fCRB N 40460 1200 - 0 - - 0 -
a 6-9-82 UtT u:u RD lA MT2-0273 10 itiCKB M 10004290 300390 •920 • 90
9 S-9-S2 M3A2 HBA1 RD B MT2-O129 6 .jCRB N 39730 1210 - 0 - - 0 -
10 I5-9-H2 MI1A2 HUAI RD 8 MT2-O13O 6 4.CKD N 40S20 1220 - 0 - - 0 -
11 2O-'J-82 UCF Lf.U KD lA MT3-0274 10 ,|,CRU H 999310 299850 + 740 • 120
12 22-9-82 MBA 2 MI1A1 RD 13 MT2-O131 6 ,t,CRB N 391C0 1180 - 0 - - 0 -
13 2S-9-32 MBA 2 MBA 1 RD 8 MT2-O132 6 *CHD N 39290 1150 - 0 - - 0 -
14 30-9-82 MonChly T o t a l s 20196550 606200 1660 210
15 (,-10-82 Ut F LEU RD lA MT2-O275 10 (tiCRD H 9997840 300240 -620 -110
16 6-10-82 M11A2 MBA.1- RD 8 MT2-O133 6 N 40110 1200 - 0 - - 0 -
17 13-10-32 MBA 2 MliAl RD 8 MT2-O134 6 ,fCUIl N 40010 1180 - 0 - - 0 -
IS 18-10-82 Uf.F LEU Kl) )A MTÎ-0276 10 ^.CRD M 10002420 299800 • 910 • 310
19 20-10-82 MUA2 HUAI RD' R MT2-OI35 6 _;),CHU N 39700 1190 - 0 - - 0 -
:o 27-10-82 MBA2 MBA I RU 8 MT2-O136 6 ,t>CHB N 41020 1230 - 0 - - 0 -
21 31-10-32 Monthly Total] 20161100 604840 290 200
22 1-11-82 UEF LEU RD lA MTJ-0277 10 <\crm H 10004100 300420 -990 -400
23 3-11-82 MBA 2 MBA 1 RD 8 MT2-O137 6 $CRD N 38930 1170 - 0 - - 0 -
10-11-82 M3A2 HBAl MT2-0138 39720 1200 - 0 - - 0 -
15-11-82 UEF LEU RD lA MTJ-O27B 10 10004250 300380 -860 -50
HDA2 M1IA1 0 MTÎ-0I39 40410 1220 - 0 - - 0 -
MUA2 MBA! RD B MTJ-0140 j.GRll 40290
29-11-82 UEF LEU lA HTJ-0279 10 Q.CRB 9990140 299800 • 1200 -560
-11-»1 Hnnthly Totnli 30157840 905180 -1010
DO 13-12-lt» ULF uhU RU 1A MT2-028O B ,t>CKU M 7994270 239030 -800 -200
Jl 29-12-82 CON LEU RD IC MT2-O324 480 KQbB M 11992230 359370 -19520 • 250
32 31-12-82 Monthly T o t a l ) 1998'500 599200 -20320 50
33 10-1-33 CON LEU KD 1C MT2-O333 400 FQjB M 9999120 299700 +8630 • 150
34 24-1-83 CON 1£U RD JC MT2-0347 400 M 10008240 299950 +4290 -200
35 31-I-8J Monthly Totals 20007360 599650 12920 -50
36 SIX MO.TOTAL- - - - - 1476 - - Î30625270 3919450 -15660 -520 .

Remarks: Signature! Doe. No. Fige No.

23
!

Figure 4.17 Inventory Changa Document - Material Tranafer

Material Description: U> 2 - 31 enriched Material Code: FQ$B Isotope Code: C Clement Code: Uniti C

Shipper Name and Addresa: Converiion Facility (CON) MBA: 2 Receiver Name and Addreaai Facility LEU MBA: 1
Country ADC Country ABC
Change Codel SD Change Code: RD

KMP Code: 2 KMP Code: lc

Shipper Data: Receiver Data:

Line Batch No of Net X Element Z Iaotope Net Z Element Z laotope

Ho. Identity Items Weight Ele. Height Iso. Weight Weight Ele. Weight Iso. Weight Notea

1 113 40 1139350 87.85 1000920 3.09 30930 1139400 87.25 994130 3.07 30520
2 119 40 1138270 88.29 1004980 2.93 29450 1138170 B8.69 1009140 3.01 3O3BO
3 120 40 1134400 88.06 998950 3.07 30670 1134390 BB.46 10034B0 2.98 29900
4 121 40 1133920 87.85 996150 3.01 29980 1133600 88.15 999450 3.00 29980
5 122 40 1136520 87.55 995020 2.86 28460 1136600 88.15 1001910 2.91 29160
6 123 40 I135VJ0 87.75 996390 2.96 29490 1135450 87.85 997490 3.01 30020
7 124 40 1134610 87.80 996190 3.14 31380 1134710 87.93 998320 3.03 30250
8 125 40 1132000 88.37 1000350 2.91 29110 1132010 88.64 1003410 2.90 29100
9 126 40 1139'. 10 ' 00.12 1004450 2.86 28730 1139390 68.33 1006420 2.68 28960
10 127 40 1133040 88.10 998910 3.11 31070 1133710 86.10 998800 3.10 30960
11 I I'D 40 1133460 1111.02 997670 3.10 30930 1133650 88.14 999200 3.09 30880
12 129 40 1131970 B8.54 1002250 2.9Î 29270 K319I0 B8.32 999700 2.90 26990
13 TOTAL 480 13623240 - 11992230' - 359370' 13623910 - 12011750 - 359120
14
15 Explanatory Note:

'Entered to Line n , FIR. 4.16

Doc. Hoi Hoteil ttvickatt ara numb trad 1 through HO t.itti Shippar-Racaivar Dlf'irancei Data Maaiuredt
tha batch number aa a prêtIK. See
«12-0324 attached lint Eor individual itea data. U -19520 g1 U-Î35 • 250 » l 7-1-83

Shipping Dace: Shipper Signature: Receiving Date: Receiver Signature:

23-12-82 Ï7-12-82

< i
 Figure 4.18 Inventory Change Document * Material Transfer

Material Description: (IF - 3X enriched Material Code; $CRB laotope Code) G Element Codei E Unitl G

Shipper Naae and Addreaa: Facility LEU MBA: 1 Receiver Name and Addreaa: Facility LEU KB A: 2
Country ABC Country ABC
Change Code: SD Change Code: RD

KHP Code: 1* KHP Codet 4

eiver'a e. Receiver Data:

Line Batch Ho of Het Z Elemont X laotope Het X Element X Isotope

Ho. Identity Items Weight Ele. Weight lno. Weight Height Ele. Weight Iao. Weight Note»

I C10544 1 1487640 67.57 1005200 2.986 30020 1477640 67.57 998440 2.986 29810
2 C76873 1 1483730 67.66 1003890 3.011 30230 1473730 67.66 997130 3.011 30020
3 C35042 1 1473240 67.65 996650 2.991 29810 1463240 67.65 989880 2.991 29610
4 C57329 1 1480310 67.57 1000250 3.001 30020 1470310 67.57 993490 3.001 29820
5 C65853 1 1486740 67.53 1003990 2.993 30150 1476740 67.53 997240 2.993 29850
6 C92359 1 1472390 67.65 996070 3.014 10020 1462390 67.65 989310 3.014 29820
7 SIX HO.TOTAL 6 8884050 - 6006050 - 180150 8824050 - S965490 - 178930
8
9
10 Explanatory Notes:
11
12 Entered to Line 1. FiR. 4. 19, for MBA-1, shipment. and Line 1, F I R . 4.20. for MDA-2, receipt.
13
14 Entered to Line 3, Fig. 4. 21, for MBA-J, shipment*. and Line 2* F i R . 4.16, for MBA-1. receipt.
15

Doc. Ho: Note*I Shipper-Receiver Differencet Date Heaauredi

KT1-0100 lia»la U 40560 |' U-23S 1220 |* U-8-B2

Shipping Date: Receiving Datai Receiver Slgnaturel

4-8-82 4-8-82 •Ct»»*~e<7

 Figure 4.19 Inventory Change Journal - Removals

Facility: Material Description: MBA) Isotope Codai Element Codel Unttl

Facility LCD - Country ABC 31 Enriched Uranium 1 G

Line Date of 1C KMP I CD Ho. of Htl. Heas. - Shipments Other Removals

Ko. Change Proea To Code Code Number Items Code Code V U-235 U U-235

1 4-8-82 MBAl HBA2 SD 1A MTl-0100 6 $GRB M 6006050 180150

2 ll-B-82 HBA1 MBA 2 SD 1A HT1-0101 6 $GRB H 6000120 180130
3 13-3-82 MBAl MBA2 SD 1A MT1-0102 6 iJCRB H 5960820 178950
4 23-d-32 HBAI UEF SD E MT1-0103 12 IJICRB H 80200 2410
5 25-3-82 MBAl MBA2 SD 1A HT 1-0104 6 <)>GRB M 5925540 177770
6 31-8-32 Monthly Totals 23972730 719410
7 1-9-82 HBA1 MBA2 SD 1A HT1-0105 6 ifGRB M 6133140 184700
8 6-9-32 MBAl UEF SD 8 HT1-0106 12 ,JiGltB H 81850 24&0
9 3-9-82 MBAl MBA2 SD 1A MT1-0107 6 4>GKB M 6000060 180000
10 13-9-82 MB.M HI1A2 SO 1A MT1-D108 6 eJCRB M 5997560 180050
11 20-9-82 MBAl •t'EF SD H HT1-0109 12 <t>CRB N 80250 2430
12 22-9-82 MBAl HIIA 2 SD lA MT1-0110 6 IJIGRB M 6000640 174440
13 29-9-82 MBAl MBA2 SD 1A MT1-01U 6 ifGRB H 5998490 1860B0
14 30-9-82 Monthly T o t a l s 30291990 910140
15 4-10-82 HaAl UEF SD 8 MTI-0U2 12 4>GRB N 78450 2330
16 6-10-32 MBAl HBA2 SD 1A MTl-0113 6 _J.CRB M ' 5997390 1B0040
17 13-10-82 HRAl MBA 2 SD 1A HT 1-0114 6 *CRB H 5998410 180070
18 18-10-82 HBAI UEF SD 8 MT1-0U5 12 <iCRB N 80120 23B0
19 20-10-82 HBAI H11A2 SD 1A HT1-0U6 6 .J.GR11 H 5998430 180070
20 27-10-82 HBAI Mil A 2 SD 1A MT1-0117 6 $GRB M 5998490 180080
21 31-10-82 Monthly Total» 24151290 724970
22 1-U-S2 MBAl ' UEF SD 8 MTl-0118 12 iJ>CRB N 80720 2420
23 3-11-82 MBAl MUA2 SD lA HT1-0119 6 H 5996910 179810
24 io-ti-a: HU.U HBA2 SD 1A MT1-0120 6 it>CRB H S9Ï5O1O 179850
25 15-11-82 HBAI UEF SD 8 HT1-0121 12 $CRB H 78650 2370
Î6 17-11-82 HBAI H11A2 SD lA MTl-0122 8 «GHB M 7993340 239570
J7 29-11-82 MOM UEF SD 11 MT1-O123 14 OIGRU H 80700 . 2410
ÎS 3O-U-8Î Monthly Totals 20223330 6OC43O
39 6-12-82 HBAI HHA2 SD ' 1C HTI-0124 240 F0>B H 5996120 179880
30 13-12-82 HDA1 MISA2 SD 1C HT1-O125 240 F<WB H 5996110 176890
31 20-12-82 MBAl MBA2 SD 1C HT 1-0126 200 FWB M 5009730 150050
32 31-12-32 Monthly Totals 17001960 500820
33 10-1-83 MDAl MBA2 SD 1C MT1-0127 200 F<J."|B H 4993220 149920
34 17-1-83 HBAI MI1A2 SD 1C MT1-0128 200 FQ-tiB H 4995210 151760
v
35 24-1-83 HBAI MBA2 SD 1C Ml1-0129 200 FWB H 5009200 150520
36 31-1-Ï2 Monthly Totals 14997630 452200
37 S I X MO.TOTAL - - - - - 1476 - - 130640930 3919970

keaacka: Signât ..Doc. Ho. Paga Ho.

,J¥ ICJ-2 18
 Figure 4.20 Inventory Change Journal - Receipt!

Facility: Facility LEU - Country ABC Material D e s c r i p t i o n ! 3Ï Enriched Uranium MBA4 2 Iaotope Codet G Elenent Codai E Uniti C

Line Date of IC KMP ICD Ho. of Htl. Heaa . Shipments S/R differences Other Additions
Ho. Change From To Code Code Number Items Coda Code U U-235 U U-233 D U-235

1 4-8-82 KBA1 MBA2 RD 2 HT1-O10O 6 <{PCRB M 6006050 180150

2 11-8-82 H3.M HBA2 RD ? MT1-0101 6 <)>citn M 6000120 180130
3 18-8-32 MBA1 MBA2 RD 7 MT1-0102 6 >JGHI) M 5960820 178950
4 25-8-32 H3A1 MBA2 RD 2 MT1-0104 6 <|<CRB M 5925540 177770
5 31-6-82 MBA3 MBA2 RD 6 MT3-OO51 44 DQ1F N 220660 6640 - 0 - - 0 -
6 31-8-32 Monthly Totals 24113190 723640 - 0 - - 0 -
7 1-9-32 MIAl KBA2 RD 2 MT1-0105 6 QCRB M 6133140 184700
8 8-9-82 HBA1 KBA2 RD 2 MT1-0107 6 QGRI1 M 6000060 180000
9 15-9-32 USAI MUA2 RD 2 MT1-0108 6 qCRB M 5997560 190050
10 22-9-82 MBA1 MBA2 RD 2 MT1-0110 6 QCRB M 6000640 174440
11 20-9-32 KBA1 HBA2 RD 2 MT1-01U 6 QGRB M 5998490 186080
12 30-9-82 MBA3 MBA 2 RD 6 MT3-OO54 44 DQ1F N 221890 6670 - 0 - - 0 -
13 30-0-82 Monthly Totals 30351780 911940 - 0 - - 0 -
14 6-10-82 MBA I MBA 2 RD 2 MT1-0113 6 QCRB M 5997390 180040
15 13-10-82 MJ1A1 MBA2 RD 2 MT1-0U4 6 QGRB M 5998410 180070
16 20-10-82 HBA1 MDA2- RD 2 MT1-0U6 6 ({GRB H 5998430 180070
17 27-10-62 HUA I HSA2 RO 2 HTl-0117 6 QGRB H 5998490 180080
IB 29-10-82 MBA3 HBA2 KD 6 MT3-OO57 44 DQ1F N 220760 6620 - 0 - - 0 -
19 31-10-82 Monthly Totals 2421340 726880 - 0 - - 0 -
20 3-11-82 MBA1 HBA2 RD 2 MT1-0119 6 QCRB M 5996910 179810
21 10-11-82 MBA1 MBA2 RD 2 MT1-0120 6 QCRB M 5995010 179050
22 17-11-82 MCAt HBA2 RJ> 2 MT1-0122 8 QCRn M 7993340 239570
23 30-11-32 KB A 3 MBA2 IID 6 KT3-OO61 44 tlQlF N 219890 6590 - 0 - - 0 -
24 30-11-62 Monthly T o t a l » 20205150 605820 - 0 - - 0 -
25 6-12-82 M0A1 MBA 2 RD 4 MT1-O124 240 FQfB H 5996120 179880
26 13-12-82 NBAS MUA 2 RD U MT1-0125 240 F()^B N 5996110 - 176890
27 20-12-32 MBA1 MBA2 RD 4 MT1-0126 200 FQl>n H 5009730 150050
28 30-12-82 MBA3 MDA2 RD 6 MT3-OO63 44 DQIF H 219920 6590 - 0 - - 0 -
29 31-12-32 M o n t h l 'f Totals 172218B0 513410 - 0 - - 0 -
MBA1 MUA2 RD 4 MT1-O127 200 N 4993220 149920
30
31
10-1-83
M3A1 MUA2 1U> t, MT1-0128 200
Fom
FQt>B N 4995210 151760
17-1-83
32 24-1-83 HBA1 MDA2 RD I, KM-0)29 J00 TQ>0 N SOO92OO 150520
33 31-1-83 MBA3 MUA2 RD 6 HT3-0066 40 DQIF 1968S0 5910 - 0 - - 0 -
34 31-1-83 Monthly T o t a l ) 15194510 458110 - 0 - - 0 -
35 SIX »O. TOTAL - - - - - - - - 131299990 3939800 - 0 - - 0 -

Reaarkg: Signature! Doc. N o . Ptge No.

ûtf*mA3'd&<>*~*-' ICJ-3 65
 Figure 4.21 Inventory Change Journal - Removal*

Ficilicy! Facility LEU - Country ABC M a t e r i a l Description! 31 Enriched Uraniua MBA: 2 Isotope Codel G Eleaent Codet E Unitl G

Lin: Dace of IC KW ICD No. of Mtl. Heas. Shipments Other Kemovala

No. Change F rota To Code Code Number Items Code Code U U-235 U U-235

1 6-8-32 MBA 2 LD 4 JE-00325 120 T.l>HE M 600000 18000

2 6-8-32 MBA2 LD 9 JE-OO326 100 US HE M 200000 6000
3 13-8-32 MDA2 MBA1 SO 11 MT2-0125 6 4.CRB M 40560 1220
4 1S-8-S2 HEA2 MBA1 SD 8 MT2-0I26 6 4CK» H 39640 1190
5 25-8-32 MBA2 MBA1 SD 8 MT2-O127 6 j.CKB M 41390 1240
6 31-8-82 HBA2 MBA3 SO 6 See Rroks. 4000 DQIF M 20021750 601340
7 31-8-82 Monthly r Totals 20143340 604990 800000 24000
3 1-9-82 MDA2 MBA I SD B MT2-0128 6 4CK0 M 40460 1200
9 3-9-32 MBA2 LD 9 JE-00327 25 T.|,NE M 125000 3750
10 3-9-82 MBA2 LD JE-0032B 50 U.VNK M 100000 3000
11 8-9-32 MISA 2 MUAI SD 8 MT2-O129 6 c].CKU M 39730 1210
12 15-9-82 H3A2 MBA1 SD B MT2-O13O 6 «CRO M 40520 1220
13 22-9-32 MBA 2 MBA1 SD 8 MT2-013I 6 «CltB M 39160 1180
K 39-9-82 HiU2 MUA I SD 8 MT2-O132 6 M 39290 1150
15 30-9-82 HBA2 HBA3 SD 6 Sec Rinks. 4200 DQIF M 21002860 629870
16 30-9-82 Honchlj T o o l3 21202020 635830 225000 6750
17 1-10-82 MDA2 LD JE-00329 30 T.>NE M 150000 4500
13 1-10-82 MBA2 LD JE-0O33O 50 U>NE H 100000 3000
19 6-10-32 U3,\2 MBAl SD' 8 MT2-0133 6 ,|iCRn M 40110 1200
20 13-10-82 I1HA2 MBA1 SD 8 HT2-O13'. 6 «CRU H 40010 1180
21 20-10-82 MBA 2 HDA1 SD 8 MT2-O135 6 4.CKD M 39700 1190
22 27-10-82 MBA2 MDA1 SD B MT2-0136 6 fCKB M 41020 1230
23 31-10-82 MUA2 MUA3 SD 6 See Rmks. 4200 DqtF M 21001920 430760
24 31-10-82 Monthly Totals 21162760 635560 250000 7500
25 3-11-82 MUA 2 HKA1 SD 8 MT2-0137 • 6 ^CRII H 38930 1170
26 i-11-82 MBA2 LD t JE-00331 30 TiJME M 150000 4500
27 5-11-82 HBA2 L0 9 JE-00332 50 M 100000 3000
23 10-11-82 M11A2 HUM SD B MT2-OI3S 6 +GKB H 39720 1200
29 17-11-82 MI1A2 MBA1 SD B MT2-0139 icun H 40410 1220
30 «-II-B2 MIIA2 KUAl SD H M'h»0Wi0 a 4CUD M 40290 1190
31 30-11-82 MBA 2 HI) A3 SO 6 Sec llmk», 4200 DQ1K M 2O974B9O 629780
32 30-11-82 Monthly Total! 2U3424O 634560 250000 7500
33 3-12-62 MBA2 LD 9 JE-QO333 30 T«tlE M 150000 4500
3% 3-12-82 MBA2 ID 9 JE-0033'i 50 UfNK H 100000 3000
35 31-12-82 MI1A2 HBA3 SD fl See llwka. 4200 DQ1F M 20996250 629930
36 31-12-82 Monthly T o t a l s 20996250 629930 250000 7500
37 7-1-83 nn.\2 LD 9 JE-00335 25 T]>NE M 125000 3750
38 7-1-83 MDA2 LD 9 JE-00336 50 ^NE M 100000 3000
39 31-1-83 M11A2 MBA3 SD b See Rinks. 4200 DQIF M 21002470 430270
40 31-1-83 Monthly Totals 21002470 630270 225000 6750
41 S I X HO.TOTAL - - - - - 25808 - - 125641080 3771140 2000000 60000

fceaurkt: Line 6 - KT2-030O*MT2-0319 Lins 31 - MT2-O36VO385 Signtturei ~~. — Doc. Ho. ?ag« No.
Line IS - HT2-032O»MT2-0341 Line 35 - MT2-O3B&.-O4O7
Line 23 - MT2-0342»MT2-0363 Line 39 - HT2-040&>0429 ICJ-4 35
 Figure 4.22 Inventory Change Document - Material Transfer

Hiterial Description: Fuel Rod» - 3Z enriched Material Code: DQ1F Iiotope Codet C Element Codei E Unit: G

Shipper Name and Address: Pacilicy LEU MBA: 2 Receiver Name and Addresat Facility LEU HBAi 3
Councry ABC Country ABC
Change Code) SD Change Codei RD

KMP Code: 6 KMP Code) «

Shipper Data: Receiver Data:

Line Batch Ho of Net X Element X Iaotope Net X Element t laotope

No. Identity Items Weight Ele. Weight Iao. Height Height Ele. «eight Iao. Weight Notes

010 200 1135630 86.32 1002430' 3.01 30170

10

13 Explanatory Note:
14
15 One of 20 batches entered on Line 1. Fig, 4.23 for HB7V-3 receipt and on Line 6. Fit. 4.21 for HOA-2 ihiprocnt.

D o c . Noi Notes: See attached rod loading protocola For Shipper-Receiver Difference! Data Heaturedt
individual rod da£a»
HT2-0324 Ghlppar Data Accepted -

Shipping Date: Shipper Signatures,— Receiving Datai Receiver Signature

5-8-82 5-8-82
 Figure 4.23 Inventory Change Journal - Receipt»

Facility: Facility LEU - Country ABC Material Deacription: 3Z Enriched Uranium MBA! :3 Isotope Codei C El euent Codet E Uniti C

Line Dace of IC KMP ICD Ho. of Mtl. Meas. Receipts S/R Differences Other Additions
Ho. Change Froa To Code Code Number Items Code Code U U-235 U U-235 U U-235

1 31-8-82 MBA2 HBA3 RD ù Sec Raiko. A, 000 DplF N 20021750 601340 - 0 - - 0 -

t 31-3-82 Monthly Total 20021750 601340 - 0 - - 0 -
3 30-9-82 MDA2 MBA3 RD 6 See Rmka. A, 200 DQ1F M 21002B60 629870 - 0 - - 0 -
A 30-9-82 Monthly Total 21002660 629870 - 0 - - 0 -
5 31-10-82 MBA 2 MBA 3 RD 6 See Rick». A, 200 DQ1F N 21001920 630760 - 0 - - 0 -
6 31-10-82 Monthly Total 21001920 630760 - 0 - - 0 -
7 30-11-82 f!BA2 MUA3 RD 6 See Rmka. 4,200 DQ1F N 20974890 62978C - 0 - - 0 -
8 3Û-U-62 Monthly Total 20974890 629780 - 0 - - 0-
9 31-12-82 MBA2 MBA3 RD 6 See Rmka. 4,200 DQ1F N 20996250 629930 - 0 - - 0 -
10 31-12-82 Monthly Total 20996250 629930 - 0 - - 0 -
11 31-1-62 MBA2 HBA3 RD 6 See Rmka. 4,200 DQ1F N 21002470 630270 - 0 - - 0 -
12 31-1-82 Monthly Total 21002470 630270 - 0 - - 0 -
13 SIX HO.TOTAL- - - - 25,000 - - 125000140 3751950 - 0 - - 0 -
1A
15
16
17
18
19
20
21
22
23
24
25
26 •

27
28
29
30

Remarks) Line 1 - HT2-0300*0319 Line 7 - HT2-0364*03B5 Signaturel Doc. Ho, Fagt Ho.
Line 3 - MT2-032O>-0341 Lin« 9 - MT2-0386+0407
Line 5 - KT2-0342*0J63 Linell - MT2-0408+0429 ICJ-5 32
 Figure 4.24 Inventory Change Document - Journal Entry Fora

Facility! Facility LEU - Country ABC KBAt Z Change Codei LD KMF Codai Date: 6-B-82

Material Description: Material Code: Isotope Code: Element Code! Unit:

Solid Waste containing 31 enr. U T^NE

Line Document Batch Number Element laotope

No. Reference Identity of Item Weight Weight Notei

1_ See Waste Discard SV-0102 through SW-0222 120 600000' 18000>

2_ L O R >!BA2 1-7-82
3_ chrouch 4-8-82
4_ And
3 PIL as of 1-8-82
6_
7
B Explanatory Note:
9
10 'Entered to Line I, Fig. 4.21.
"ll

13

IS
16
17
18
19
20
21
22

Approvtd Byl Doc. No.

JB-0O32J
 Figure 4.25 Inventory Change Document - Material Transfer

Material Description: 32 enriched fuel aaieobly Material Code! BQ2F Isotope Code: G Element Code: E Unit: C

Shipper Name and Addresa: Facility LEU MBA: 3 Receiver Name and Address: HBAi 1
Country ABC
Change Code: 3D Fover Reactor One (PR-I) Change Codel RD
Country ABC
KMP Code: 7 KMP Codei 1

Shipper Data: Receiver Datai

Line Batch Ho of Net Z Element z Isotope Net Z Element Z Isotope

Ho. Identity Items Weight Ele. Weight ISO. Weight Weight Ele. Weight Iso. Weight Notes

1 E3-O778 ; 569680 87 85 500460 3.09 15*60

2 E3-O779 569130 88 29 502480 2.93 14720
3 E3-O7SO 567200 88 0b 499480 3.07 15330
4 E3-0781 566960 87 85 498070 3.01 14990
5 E3-O782 568260 87.75 498650 2.86 14260
6 E3-0783 567740 88.37 501710 2.96 14850
7 E3-0784 567310 88. 12 499910 3.14 15700
8 E3-O785 569710 88. 10 501910 3. 11 15610
9 E3-O786 566920 88.02 499000 3.10 15470
10 E3-O787 567660 88. 54 502600 2.92 14680
11 TOTAL 10 56BO57O - 5004270 - 1S1O70'
12
13 Explanatory Note;
14
15 'Entered to Line 1, PiR. 4.26 ,

Doc. Ko: Note»! See attached iheett tor fuel rod Shipper-Receiver Differences Data Measured:
protocols for each aasenbly. Shipper Dtta Accepted
HT3-0048

Shipping Date: Shipper Signature! Receiving Datai Receiver Slgnatursiz-N , <^T^ r

Ï-8-82
dAuX 5-8-B2
 Figure 4.26 Inventory Chang* Journal - Removal»

Facility: Material Deicript'.om MBA: 3 Iaotope Codel G Element Code: E Unit! C

Facility LEU - Country ABC 3Z Enriched Uranium

Line Dace of IC KMP ICO N o . of Htl. Meal. Shipments Other Rénovai»

Ho. Change From To Code Code Number Items Code Code U U-235 U U-235

1 2-8-82 MBA3 PR1 SD 7 MT3-0048 10 BQ2F M 5004270 151070

2 9-3-82 MBA 3 Pltl SD 7 MT3-0049 10 BQ2F H 5008930 150790
3 23-8-82 MBA3 PR1 SD 7 MT3-OO5O 20 BQ2P N 9999080 299970
4 31-8-32 MBA 3 MBA 2 SO 6 MT3-OO51 44 DQ1F H 220660 6640
5 31-8-82 Hontlil!t Totals 20232940 608470
6 6-9-82 HBA3 PR1 SD 7 MT3-OO52 20 BQ2F N 9989740 299730
7 20-9-82 HBA3 PR2 SD 7 MT3-OO53 30 BQ2F M 10008460 300200
8 30-9-82 UIIA3 MBA 2 SD (j MT3-OO5A 44 DQIF N 221890 6670
9 30-9-62 Monthlv • Totals 20220090 606600
10 4-IU-32 KBA3 PR2 SD 7 MT3-OO55 20 BQ2F H 10010240 300450
II 18-10-82 MJ1A3 PR2 SD 7 MT3-OO56 20 BQ2F H 9997920 299980
12 29-10-82 MBA 3 ME A 2 SD 6 MT3-OO57 44 Dt)lF M 220760 6620
13 31-10-62 Monthly Totals 20228920 607050
14 1-11-82 t:BA3 PR3 SU 7 MT3-OO58 20 BQ2F N 9988*140 300210
15 15-11-82 MUA3 PR3 SD 7 MT3-OO59 20 BQ2F N 10009730 300100
16 29-11-82 MUA3 PR3 SD 7 MT3-OO6O 20 . BQ2F N 10001820 300000
17 30-11-62 MUA3 MBA2 SD 6 MT3-0061 44 BQ1F N 219B90 6590
IK 30-11-82 Monthly T o t a l ! 30220380 906900
19 13-12-82 MBA3 PR*. SD 7 MT3-0062 22 BQ2F N 10978320 329960
20 30-12-82 MBA3 HHA2 SD 6 HT3-0063 44 DQ1F N 219920 6590
21 31-12-82 Monthly T o t a l * 11198240 336550
t 1
3-1-83 HBA3 PR4 SD 7 MT3-0064 BQ2F. H 10989460 329510
23 17-1-83 MBA3 PR4 SD 7 HT3-0065 23 BQ2F N 11513280 344980
24 31-1-83 M1A3 MBA2 SO 6 MT3-0066 40 DQ1F N 196880 5910
31-1-8J Monthly Total»
rs
26 TOTAL - - - - - 507 - N
22699620
124800190
680400
J745970
27
28
29
30

RemarkJ: Page Ha.

«7
 Figure 4.27 Inventory Change Document - Material Transfer

Material Description: 31 enriched reject rods Material Code: DQ1F Isotope Code: C Eleaent Codet E Unit! G

Shipper Name and Address: Facility LEU MBA: 3 Receiver Name and Address: MBA: 2
Country ADC
Change Codei SD Facility LEU Change Code: ED
Country ABC
KMP Code: 6 KMP Code: 6

Shipper Data: Receiver Datai

Line Batch No of Net Z Element Z Isotope Net Z Element Z Isotope

No. Identity Items Weight Ele. Weight Iso Weight Weight Ele. Weight Iso. Weight Notes
1
1 R-0012 44 249840 220660' 6640
2
et

Explanatory N u te:
4
5 'Entered to Line 4, F I R . 4.26, for MBA-3 shipment and Line 5, Fix. 4.20, for MBA-2 receipt.
6
7
B
9
10
11
12
13
14
15

Doc Ho: Notai! Shipper-Receiver Difference! Date Measured!

HT3-OO51 Shlppir's Uaight Aeoaptsd

Shipping Dacet Shipper Signature! Receiving Datât Rocslvsr

%
31-8-82 3J-8-82
 Figure 4.28 INVENTORY CHANGE REPORT (ICR) FORM R.01.1/c
CO UNTRY ***" PERIOD COVERED BY REPORT FROM 1—8—8? TO 31—8— 8 ?
FA CILITY LEU REPORT No. 1
MA TERIAL BALANCE AREA 2 PAGE Ho. 1 OF 2 PAGES SIGNATURE
S 9 13 11 25 28 31 1 ïo DO

I Ljll I l l 1 1 1 1 1 1 1 1 1 1 1 1 (1 II
COBREf TIOM

CONCISE NOTH
MEASUR. S A S I S
MBA/COUNTRY A C C O U N T A N C Y O A T A
CONTINUATION

TO

INVENTORY
NUMBER

T Y P E OF
0

CHANCE
DATC O F NAMC OR OF

UNIT V 9 / 9
EUT Î Y 0 v a. WEIGHT OF
No
INVENTORY u NUMBER OF ITEMS z FISSILE
CHANGE a. BATCH IN WEIGHT OF REPORT EHTRY
FROM TO ELEMENT ISOTOPES • No. No.
BATCH u (URANIUM ONLY)
0 t

111 1 IJ.IJ.Lli . JJ.LLL. ..LI I L L . I I I II . ...I'll. JJ.I LLL_. -WILL 31 .. 1 1 1 1 1 iJ.J.J_LIJLL_-. III .IIJJ.LLLIL.- l U -LU.
j
to
IB 70
79
u
n «6 41
-
inx. Il
in
1 04088? MBAI MBA? RD 2 MT1-0100 6 0GRB E 6OO6O5O g 180150 G M
—
2 110882 MBAI MBA? RD 2 Mia-0101 6 0GHB E 6000120 g I8OI3O G M

3 180882 MBAI MBA2 RD 2 MT1-010? 6 0a RB E 596O82O g 178950 G M

4 250882 MBAI MBA? RD 2 MTl-0104 6 0GHB E 5925540 g 177770 G M

5 310882 MBA3 MBA2 RD 6 MT3-OO5I 44 DQIP E 2?O66O g 6640 G M

6 060882 MBA2 LD 9 JE-OO325 120 E 200000 g 6000 G M 3

7 O6O882 MBA? LD
9 JE-OO326 100 U0NE E 600000 g 18Ô00 G
V J
8 130882 MBA? MBAI SD 8 MT2-O125 6 E 4O56O g l?20 G
M 3
-
9 I80882 MBA? MBAI SD 8 MT?-01?6 6 0a RB E 39640 g 1190 G M
3
iô 2508O2 MBÀ2 "MBAI SB 8 wv?-dï2î 6 pm E 41390^ g G TU
—
3

li Ô5Ô88?T """ MBA? " MBA3 SD 6 MT2-O3OO 2Ô0 DQIP" Ë IOO2436 g 3OÏ7Ô G M 3

l? 06088? MBA2 MBA3 SD 6 MT2-O3O1 200 DQIP E 1OO?56O g 30240 G M 3

- -- —
100882 MBÀ? ÏJBA3 SD 6 MT2-O3O2 200 DQIP " E 998916 g 2983O G M 3

14 11088? MBA? MBA3 SD 6 MT?-(>303 200 DQIF E 997870 g 29860 GM 3

15 120882 MBA? MBA3 SD 6 ~MT?-<Î3D4 200 DQIP E 1OOO37O g 3OI8O 0 M 3

i6 "T3Ô882 MBA? MBA3 SD "S MT2-O305 ?00 DQIP E 1002370 g 3OO9O "M - 3

17 998490

a| a
160882 MBA? MBA 3 SD 6 MT?-O3O6 POO . DQIP E g 2986O 3
...
Ï8 170882 MBA? MBÂ3 SD 6 MT2-O3O7 ?00 DQIP E 1002140 g 3O?6O 3
n
19 18088? MBA2 MBA3 SD 6 MT2-O3O8 200 DQIP E 100?070 g 3OI2O J M 7
—
26 190882 MBA? MBA3 SD 6 MT2-O3O9 POO DQIP 5 99658O g 29810 G K 7

200882 MBA2 MBA3 SD É MT2-O31O 200 DQIP 3 ?98?e J M

3
2? ?10882 MBA? MBA3 SD MT?-0311 200 DQIF 1004450 g 28730 G H
 136

This page is intentionally left blank

 137

Parr 5. PHYSICAL INVENTORY TAKING

5.1 Introduction

The physical inventory is the bench mark of nuclear material accounting and
control. Various means of control, accountancy and containment and surveillance
can be employed to account for and control nuclear materials. However, a mate-
rial balance based on a measured physical inventory that provides conclusive evi-
dence of the physical presence of the material is the only means for the facility
operator to assure that no significant losses or diversions have gone undetected.
Of course, the other components of the material balance, receipts, shipments,
discards, etc., also must be accurately measured for the balance to be meaning-
ful.

The following description provides information regarding inventory

procedures, including pre-inventory preparation procedures, the inventory taking
procedures, and post inventory procedures, to assure a complete and accurate phys-
ical inventory including assurance of data handling accuracy and integrity. This
description also addresses the relationship of this element of an SSAC to other
SSAC elements, descriptions of which are provided in other parts.

5.2 Inventory Measurement

As a significant component of the material balance, the quantity of material

on inventory must be measured or assurance provided that prior measurements re-
main valid so that the balance has meaning in assuring that material has not been
lost or diverted. Also, the precision and accuracy of the physical .inventory mea-
surements can have a significant effect on the quality of the material balance in
terms of the

5.2.1 Precision and Accuracy.

The uncertainties of the respective components of a material balance combine

to produce the total material balance uncertainty, i.e. the uncertainty of mate-
rial unaccounted for (OMtfpO. If the inventory uncertainties are large in rela-
tion to those of the other components, their effecr on OJUJF for the total balance
will be greater. Consequently, to meet goals chat may have been established for
<?MU7 o r t 0 reduce a^UF even if no specific goals have bsen set, some oi the ran-
dom and systematic errors associated with the inventory measurements may need to
be relatively smaller for inventory quantities which make up relatively large por-
tion of the total inventory. Similarly, the measurement of a portion of the in-
ventory with a relatively small nuclear material content may not need to be as
precise as that of a portion having a relatively large nuclear material content.
For example, the inventory measurement of the residual material in a piece of
cleaned out equipment can be less precise than the inventory measurement of the
material in that same piece of equipment that has not been cleaned out for inven-
tory and that still contains a relatively lar&. quantity of nuclear material.

The bias in inventory measurements may not affect the Oi^up of a material bal-
ance but can affect the size of the MBF, *.e. how well a balance is achieved.
 138

The effect of Che bias ia invencory measurements will depend on Che quantity
involved and Che magnitude of the bias.

The detailed description of Che Nuclear Materials Measurement Element of an

SSAC addresses Che selection of KMPs to assure that all material flows and
inventories are measured with appropriate precision and accuracy to provide an ac-
ceptable uncertainty for the material balance. A significant aspect of this as-
surance is the selection of measurement methods for inventory KMPs thac have pre-
cision and accuracy appropriate to the significance of the inventory quantities
CO the material balance.

5.2.2 Invencory Measurement Quality.

Control of the quality of measurements made for invencory purposes is neces-

sary Co assure valid measurements and to permit determination of the
uncertainties associated with such measurements. The quality of many inventory
measurements will be so controlled because they will have been made prior to the
inventory as routine floe measurements. Such measurements could be controlled as
addressed in the detailed description of the Measurement Quality Element of an
SSAC. The use of such prior measurements is discussed in section 5.2.4. Spe-
cific inventory KMP measurements also could be controlled in accordance with fa-
cility procedures established on the basis of the Measurement Quality descrip-
tion.

5.2.3 Factors, Nominal Values, and Calculated Valves.

By their nature, factors such as the nuclear material content oL a process

intermediate, nominal values such as the stoichiometric value for a compound or
che Carget isotopic composition of a process material, and calculated values such
as by-difference values or values based on mixing ;alculac5.ons depend on steady
state or normal operations. Perturbations such a» unknown losses, substitution
of materials, or diversion invalidate the use of factors, nominal values- and
calculated values as accurate representations of nuclear material elemsnt and iso-
ropic values for purposes of a measured physical inventory.

A factor is a measured value that is applied to one or swre items of the

same kind. The continued validity of factors should be monitored through a mea-
surement quality control programme, and the uncertainty to be associated with the
factor determined t-hrough the measurements made in such a programme. Such a
programme is discussed in the detailed description of the Measurement Quality Ele-
ment of an SSAC. Supporting data for factors should be documented as would any
nuclear material measurement source data. ?he uncertainties of each factor
should be included in the calculation of a;£jp - o r the material balance in which
such factors vere used to obtain iavencory data.

Because nominal values are not the result of a current measurement they do
not provide assurance of an accurate measure of nuclear material and their use in
physical inventory taking should be avoided. If -î» is necessary to use a nominal
value, the basis for the value should be clearly established so that any systema-
tic error introduced can be treated appropriately in the calculation of

While calculated values may result from one or more measured values, there
often is an unmeasured component which invalidates the calculated value as an

152
 139

assurance of the presence of nuclear material. For example, if a measured quan-

tity of material were introduced into a process vessel and a measured quantity
removed from the vessel, calculation of the material remaining in the vessel pro-
vides no assurance that the material actually is in the vessel and has not been
lost or diverted. The use of such by-difference values in physical inventory tak-
ing should be avoided. On the other hand, calculated values derived froa a repre-
sentative sample of a weighed quantity of material do provide assurance of the
presence of the material because there is no unmeasured component in the calcula-
tion.

5.2.4 Validity of Prior Measurements.

Measured values used in physical inventory taking come from many sources,
including flow and process measurements made prior to the inventory taking. The
use of such prior measurement data can reduce physical inventory effort by
reducing the need for measurements to be made during the inventory. For such
data to provide assurance of the presence of the material it is necessary that
the validity of the data be assured. That is, because the measurement was made
at some prior time some means should be used to assure against undetected occur-
rence of such things as unrecorded removal of portions of material from a con-
tainer, removal of containers or items from a vault, or substitution of items or
quantities of different materials. À primary means that a facility operator may
use for providing this assurance is tamper indicating C/S measures such as seals
(see Sec. 2.7). For IAEA verification it would be necessary for the IAEA to seal
the material with IAEA seals.

To be effective the tamper indicating measures should-provide means of

identifying the material with the measurement data. There should be means for
tracing material identity) i.e. batches, lots, samples, etc., to laboratory log
books or other records of measurement résulte. These quantity and identification
data also should be recorded so that there would be a clear indication of any
changes made in quantity or identity. To be effective tamper indicating measures
must be applied immediately upon completion of the operations which establish the
nuclear material content of an item. Such operations may include the
nondestructive assay of fuel elements or rods that have been identified and
sealed or of a series of rods or pellets which are then stored in containers or
in a vault under seal. They also may include sampling, packaging, and weighing
a lot of material into a number of containers, each of which is sealed with a tam-
per indicating seal. The integrity of the sample also must be protected until it
is analyzed.

5.2.5 In-Process Measurements.

The extent to which a process must be shut down, drained down, or cleaned
out to measure inventory quantities of nuclear material will depend on the quan-
tity of nuclear material in the equipment at any stage of operation, cleanout,
shutdown, etc. and the precision and accuracy with which such material can be
measured.

5.2.5.1 Process Draindown and Cleanout.

Process draindown and cleanout are relative terms which indicate de-
grees of removal of nuclear material from process equipmen'. Draindown implies
 uo
moving the bulk of the nuclear material to measurable points in the process but
not necessarily cleaning out the process equipment. Cleanout implies a more rig-
orous effort to remove the material from the process equipment so that it may be
considered to contain no residual nuclear material. Whatever the degree of re-
moval, the procedure is essentially the same. The nuclear material is moved from
its normal process location -to A location and form in which it can be measured.
The location may be a calibrated process vessel or it may be bottles or con-
tainers separate from the process.

Process equipment from which material has been removed should not be
considered "clean", i.e., to contain no nuclear material, unless measurements are
made to verify this. Measurements should either verify that the residual quan-
tity is not great enough to affect significantly the material balance or should
result in a material quantity to be included in the inventory. If experience
with the cleanout procedures actually used indicates that the equipment will be
clean, it may not be necessary to verify the cleanout.

While prior measurements and cleanout procedures are acceptable bases

for concluding that a process residue contains no nuclear material, such is not
the case with draindown procedures. Draindown procedures and prior measurements
of residues should be used to establish expected quantities, inventory factors,
variations in holdup, and limits of uncertainty. The residual quantities should
be measured for each inventory or measurements made to verify the validity of any
factors used. Such measurements of residual quantities may be accomplished using
NDA techniques which have been calibrated to the equipment by measurement and
cleanout tests at a prior time. Further, the quantity of nuclear material
remaining is equipment should not be calculated as the difference between the
amount put into the equipment and the amount taken out.

5.2.5.2 Dynamic Inventories.

To eliminate or minimize the disruptive effect of shutdown and

cleanout or draindown inventory procedures, dynamic inventory techniques could be
considered. Such techniques include any inventory procedures which permit the
process to continue to run during inventory, i.e., remain in a dynamic state.
Some techniques may require changes in processing procedures which will result in
reduction of process throughput but not to the extent of cleanout or draindown.
Two techniques are described here that may be appropriate for an L£U conversion
and fuel fabrication facility. Specific application will depend on the facility
and process.

Strict material handling controls and cutoff procedures for material

movements and transfers are necessary for dynamic inventories to ensure that:
(1) material does not move through the process without being inventoried, (2) ma-
terial is not recycled during the inventory in such a way as to cause it to be
inventoried more than once, (3) material is not removed from the process after in-
ventory in such a way that it might be inventoried a second time in a storage lo-
cation, and (4) material which has already been inventoried as a raw material or
process intermediate in storage is not introduced into the process.

(a) Process Blank. This technique is particularly applica-

ble to batch-type processes where it is feasible to separate batch flow and in-
sert a blank, i.e., a batch containing no nuclear material (batch blank), or to
 141

clean out the equipment stepwise as the process proceeds (cleanout blank). The
technique also could be applied to a continuous flow process, but the introduc-
tion of the blank and maintaining segregation of the blank from the normal pro-
cess flow are more difficult. •

(b) Counter Current Inventory. The basic principle of this

technique involves movement of the inventory taking team counter current to the
process flow. As the material moves past the inventory team, or vice versa, the
inventory is measured, counted, recorded, etc. This technique is best applied to
processes or psrts of processes in which the nuclear material is contained in dis-
crete items, such as fuel pins, pellets, or containers of material such as fur-
nace trays or bottles or process intermediates.

Caution should be observed in this type of inventory

taking that the inventory teams do not bypass material moving past them or that
they do not inventory material more than once. Tagging is not always possible in
this kind of operation but some procedures should be employed to assure inventory
accuracy. Such procedures may involve physical segregation or area tagging
rather than item or container tagging.

5.3 Inventory Organization and Planning

Physical inventories planned, organised, and carried out according to

written inventory procedures and instructions can be conducted with better assur-
ance of being complete and accurate as. well as being accomplished with a minimum
of interference with facility operations.
<-*
5.3.1 Assignment of Responsibilities.

Clearly defined responsibilities, duties, and authorizations together with

proper orientation of personnel and assurance that each person understands hi*
place in the inventory programme will materially assist in more rapid, trouble-
free conduct of physical inventories. Responsibilities and authorizations at the
different organizational levels involved in the inventory should be clearly set
forth in writing.

5.3.1.1 Facility Inventory Supervisor.

One person who is familiar with the areas to be inventoried and with
the principles and procedures of conducting physical inventories of nuclear mate-
rials should be assigned primary responsibility for planning, organizing, and
conducting the physical inventory. Specific statements should be included for
the inventory supervisor's authority with regard to process shutdown, startup, in-
terruption, and control preceding, during, and following the inventory. The in-
ventory supervisor should not be a member of an inventory team but should be
available at all times during the inventory to handle problems that might arise
and to assure that the inventory is proceeding satisfactorily.

5.3.1.2 Area Inventory Supervisor.

One person in each area should be assigned the responsibility for the
inventory in that area. This person should be familiar with the operations and
material in the area and may be the material custodian for the area. He should
 142

be responsible for all aspects of the inventory within the assigned araa as
directed by the written inventory instructions for the area. His authority
should extend to any matters relevant to preparation for and conduct of the inven-
tory within his area within the framework of the written inventory instructions.
He should not have the authority to deviate from the written instructions without
approval of the Facility Inventory Supervisor.

5.3.1.3 Inventory Teams.

Teams for conducting the inventory, i.e., the actual listing,

tagging, measuring, etc., should consist of at least two persons. More people
may be required on teams where material handling or nondestructive measurements
are to be carried out during the inventory. Alternatively, nondestructive-
measurement teams or material handling teams may be established to assist other
inventory teams as required and specified in the inventory instructions. Each in-
ventory team should contain one person who is familiar with the areas assigned
for inventory and one person from another area or organizational unit to serve as
a controller to assure accuracy and compliance with instructions. The inventory
teams should have no authority to deviate from specific instructions without ap-
proval of the Facility Inventory Supervisor.

5.3.2 Cutoff Procedures.

À major consideration in planning a physical inventory is establishing the

timing of and controlling the various cutoffs necessary for accurate inventories.
Cutoffs or cutoff procedures are so called because, at the specified time,
activities such as movement of material or posting to the records are stopped or
cut off. The time of the inventory should be specified for each material balance
so that a well defined material balance interval can be established. Cutoff pro-
cedures should b? established to assure that the quantity of material which re-
sults from the physical inventory taking accurately represents the material physi-
cally present at the specified inventory time and that this quantity is accu-
rately reflected in the records at the same specified time. Material cutoff pro-
cedures should be coordinated with records cutoff procedures to assure that the
same material balance interval is used for both.

5.3.2.1 Receipt , Shipments, and Other Removals.

Inventory instructions should specify a time at which material

receipts, shipments, and other removals such as waste discards are cut off. Ho
more material should be received into or shipped or removed from the facility
after that time. When it is necessary to make a shipment or take a receipt after
the cutoff time and before the inventory is finished, such receipts and shipments
should be made with the approval of and under the control of the Facility Inven-
tory Supervisor. He should assure that such items are properly reflected in the
records and in the physical inventory data. Any receipt after the cutoff should
be identified and segregated so that it will not be included in the physical in-
ventory. Addition of such receipts to the records should be made for the mate-
rial balance interval following the inventory. If it is necessary to make a ship-
ment after cutoff but before the inventory is completed, such shipments should be
inventoried before shipment at the measured shipment value for the material and
such quantities included in the physical inventory. The records for the shipment
should be shown in the following material balance interval. Material discards
 143

should be coordinated with inventory schedules so that all discards have been
made and properly recorded before the inventory cutoff time.

5.3.2.2 Internal Transfers.

For a given facility there may be more than one material control area
established by the facility operator which taken together will provide a total fa-
cility material balance. In such cases inventory instructions should specify the
cutoff times and procedures for each such area. Not all areas need be
inventoried at the same time, but they may be taken in sequence with or counter
to the process or merely in sequence of assignment to the inventory team. Mate-
rial movements between areas should be controlled prior to, during, and after the
inventory to assure that all material has been inventoried and that none is
inventoried more than once. Area records cutoff should be coordinated with the
material transfer cutoff so that the records will accurately reflect the inven-
tory of the area. Internal transfer cutoff times do not have to be the same for
all areas. They may progress with the inventory. When the inventory in an area
is complete, the transfer cutoff should remain in effecf until the entire inven-
tory is complete.

If it is necessary to make internal transfers before completion of

the inventory, these should be made on an exception basis with the approval of
and under the control of the Facility Inventory Supervisor. He should assure
that the proper adjustments to the inventory are made for such transfers.

5.3.2.3 Process.

Process cutoff does not necessarily mean shutdown of the process.

The process may continue to operate during inventory. Cutoff controls should b<=
established to assure accuracy of in-process inventory. The techniques for dy-
namic inventory require carefully controlled cutoff procedures to provide an accu-
rate in-process inventory. Transfers to or from the process should be handled in
a manner similar to that for area internal transfers. Material and records cut-
off for the transfers to and from the process should be carefully controlled. In-
ventory instructions should specify the manner in which the in-process inventory
is to be taken and the process cutoff controls to be used. Such controls could
include:

a) No raw material added to the process after a specified

time;

b) No recycle permitted after a specified time;

c) Transfers to and from intermediate storage cut

off at a specified time;

d) Scrap and waste generated after a specified time

or between specified times segregated from all
other scrap and waste; and

e) Product produced after a specified time or between

specified times segregated from all other product.
 144

5.3.2.4 Records.

The records for receipts, shipments, and ocher removals should be cut
off after the last receipt, shipment, or removal prior to the specified cutoff
time and the books closed for the material balance interval. For individual
areas, records for internal transfers should be cut off after the last transfers
prior to the specified cutoff tim«. The material cutoffs should be controlling,
with the records brought up-to-date to the final material transactions.

5.3.3 Facility Inventory Listing and Stratification.

Preparation of facility inventory item lists which include batch data and
identification and stratum information may make data processing easier and more
rapid. Preliminary facility inventory item lists which include batch and stratum
information can assist in inventory taking, inventory verification and subsequent
data processing by both the facility operator and the IAEA.

Batch is defined in the Safeguards Glossary (IAEA/SG/INF/1) as a portion of

nuclear material handled as a unit for accounting purposes at a key measurement
point and for which the composition and quantity are defined by a single set of
specifications or measurements. If one considers the "set of specifications and
measurements" for material accounting purposes to mean the element factor for the
items in the batch, then a batch can be defined as those items related through a
common element factor. This is the definition used in IAEA-TECDOC-261. Batches
having the same set of measurements, i.e. measurement error values, and the same
physical and chemical characteristics can then be combined into strata for pur-
poses of calculation of the uncertainty of the measurements of the strata and sub-
sequent propagation of the errors for the respective material balance components
to obtain Oygjp. The procedures for such propagation are discussed in IAEA-
TECDOC-261, Chapter 3.

Some facility attachments call for the facility operator to provide the IAEA
inspectors with a stratified inventory list for use in IAEA inventory verifica-
tion. Facility attachments also call for the FIL which accompanies each material
balance report to list all batches separately and to include identification and
batch data for each. These two requirements can be met readily if the facility
operator uses facility inventory item lists in the physical inventory taking
which include batch and stratum information. The important thing is to have for
each item on inventory a means to identify the batch to which the item belongs
and measurement error values for the batch. This will permit collection of the
item data into proper batches and strata. Stratum is defined in the Safeguards
Glossary as a grouping of items/batches having similar physical and chemical char-
acteristics (e.g. volume, weight, isotopic composition, location) for the purpose
of facilitating statistical sampling for measurements needed to establish and ver-
ify the material balance and its uncertainty ( )

For statistical computation purposes a stratum should consist only of

batches of approximately the same quantity and type of material, i.e. physical
and chemical characteristics, and the same measurement error values (i.e., mate-
rial measured by the same methods). For the purpose of facilitating inventory
taking and inventory verification the additional information on location is
needed. Facility inventory item lists either preliminary or final should be
prepared to show for each item the batch and stratum to which the item belongs
and its location. Normally, facility physical inventory would be taken by loca-
tion with lists being prepared for each location. Stratification of lists for
each location would facilitate subsequent verification. Preliminary facility
item lists such as shown in Figures 5.1 through 5.5 for each location will
greatly assist IAEA inspectors to develop their sampling plans without having to
wait for the final inventory item lists to be completed. A batch code can be
established sc that each unique batch of a given size, material type and measure-
ment error values has a separate identifier. The stratum code could then be
established so that all like batches having the same measurement error values
would be included in a stratum. Proper selection and designation of KMPs could
permit the use of the KMP designations as the stratum identifier. This also
could include the location identity so that all like batches at a given location
would constitute a stratum. It is logical to consider KMPs as a basis for stra-
tum identification because it is the KMP designation that identifies that certain
measurements are made and it is the like measurement error values that determine
the grouping of batches into strata. If the KMPs are established as shown in the
example in Appendix A, then the KMP designation would be sufficient for stratum
identification because each KMP, including the inventory KMPs, represents a mate-
rial type. If. however, different enrichments were involved and separate strata
for enrichments were needed then an enrichment code would be added to the KMP
designations of the example such as KMP-1A(3) and KHP-IA(N) for UFg receipts
of 3% enriched and natural uranium, respectively. On the other hand, although
KMP-1 is for receipts of all material types, the A, B, and C designations have
been added to identify the strata in coordination with the inventory KMPs-A,
B, and C. The majority of the present report formats for reports to the IAEA
in accordance with Code 10 of the General Part of Subsidiary Arrangements for
NPT Safeguards Agreements permit only single digit KMP designations. In
making such reports the A, B, and C designations for KKP-1 would not be used.
Also, if an inventory KMP represented several material types such as in a
storage vault then material type identifiers would be needed there too. An
illustration of this variation is shown in the example for KMP-L in scrap
storage where there are five different kinds, i.e. strata, of scrap. A
computer based system called NUMSAS (Nuclear Material Statistical Accounting
System) has been developed by the Joint Research Center of the Commission of
the European Communities which provides for the collection and treatment of
data along the general lines discussed here. Details of NUMSAS are described
in reference EUR 6471 EN.

Location code is not necessary for Ojjuy computational purposes because loca-
tion does not affect the material balance uncertainty unless the location is an
indication of static inventory. Strata at different locations could be combined
for computation of O^jjp. However, static inventory, i.e. material that has been
inactive during the material balance period, has no effect on Ojnjp and should not
be included in the calculations. Facility inventory item lists should identify
material that has been static during the period so that such material can be
eliminated from the Oj^jy calculations. Location, e.g. in a sealed vault, may be
one means for accomplishing this. The time at which tamper indicating seals were
applied to containers would be another. If the material were measured and sealed
in a prior material balance period and the seal^as still intact then the date-
rial would not effect the OMUf calculation. Note, however, that unless IAEA
seals have been placed on static material, the IAEA would need to verify the mate-
rial during their PTV activities.

Another possible means to identify material that would not effect the Ojfljy,
i.e. static material that had not been measured during a material-balance inter-
val, would be a measurement number and time code. For each time or batch

159
 146

measured there could be a code for the measurement used, the measurement number
and the time of measurement.. For example the fortieth weighing on scale number
3 during the eleventh week of 1983 could be coded W03-040-113. Similar codes
could be used for element concentration and isocopic measurements using such
prefixes AS E for element and I for isotopic. These code numbers associated with
items or batches on the inventory item lists would show whether another measure-
ment had been made during the material balance interval and thus would affect the
OMijp- For example a batch on both beginning and ending inventory having the same
measurement code number would be identified as static and not affecting

Strata in different material balance components, however, need to be

considered separately because some will be positive and some negative, i.e.
beginning inventory and receipts will be positive and shipment and ending inven-
tory will be negative. Thus the item lists resulting from the physical inventory
taking will represent a negative material balance component when combined with
the other components, as discussed in the detailed description of the Material
3alance Closing Element.

5.3.4 General Inventory Instructions.

Inventory instructions should include a detailed description of the proce-

dures to be followed in taking the inventory to assure that all items and mate-
rials are inventoried once and not more than once. Such description should in-
clude instructions for listing the inventory and tagging inventoried items and.
for control of inventory lists and tags. Inventory teams should be assigned spe-
cific blocks of numbered sheets and tags, and all such items should be accounted
for by the team to the Facility Inventory Supervisor. The inventory sheets
should have a column for the tag number along with other columns for the inven-
tory data sufficient to identify each item, record the associated quantity of nu-
clear material element and isotope, and any other data pertinent for the inven-
tory such as sample numbers, NBA tests results, log book references and batch and
strata identification. When an item is inventoried it should be tagged and
listed. The tag number should appear in the column for tag numbers on the line
for the properly identified item. Every tag number assigned to a team should
either appear on one of the lists assigned to that team or be returned to the Fa-
cility Inventory Supervisor. All inventory sheets should be returned whether
used or not. When a sheet is full, it should be initialed by both team members
and any auditors or other observer, as appropriate. Completed sheets should be
collected by the Facility Inventory Supervisor at frequent intervals during list-
ing. Sheets that are voided for some reason should not be destroyed but should
be marked void, initialed by the team members and any auditors or observers as ap-
propriate. Multiple copy listing can be used to provide the Facility Inventory
Supervisor with an original control copy of the lists to check against tag num-
bers and list numbers and subsequently to maintain control of any changes made to
the sheets during reconciliation.

The inventory instructions should make provision for anomalies and

discrepancies such as the discovery that a facility tamper indicating seal has
been opened. The inventory teams should have no discretion to deviate from the
written instructions; therefore, the instructions should provide that the Area or
Facility Inventory Supervisor be called to resolve any problems.
 147

5.3.5 Area Inventory Instructions.

Inventory instructions should be prepared for each area to specify in deta'l

how the inventory is to be conducted in each.

(a) The type of inventory should be specified, i.e., shutdown, dy-

namic, counter current, etc.

(b) The extent of shutdown and cleanout should be specified with ap-
propriate cutoff for processing and material transfers. When only a portion of
the process is to be shut down, instructions should be given as to the equipment
to remain operative, that to be shut down, and how the interface between the
static and dynamic portion of the area is to be controlled.

(c) For equipment that is shut down, instructions for cleanout,

flushout, dismantling, etc., should be given. These instructions should ''nclude
the treatment, handling, and measurement of material removed from such equipment.

(d) For areas of the process not shut down, specific operating in-
structions should be included as to how the process will be operated to permit
inventorying the material. Cutoff procedures should be included to provide for
the interface between inventoried and non-inventoried material associated with
the operating process. Procedures for measuring the content of the operating pro-
cess should be specified.

(e) Where item control is used in an area, instructions shouid be

included for identifying and locating all such items in relation to the records
for the items. The inventory team could prepare a list of items as they locate,
identify, and tag them. This list then could be checked against the identity and
location records for the items. Conversely, the inventory team could use a copy
of the record and check each item as it is located, identified, and tagged. The
team should assure that all items physically present are tagged and checked
against the list and that all items on Che list are located, identified, and
tagged.

(f) Where item control includes use of facility tamper indicating

seals for containers or vaults, inventory instructions should include procedures
for verifying the integrity of the tampar indicating devices. Instructions
should include procedures for the team to follow if they find a tamper indicating
device that has been compromised.

(g) When facility tamper indicating measures have been used so that
previous measurements of the nuclear material content may be used for inventory
purposes, inventory instructions should identify the measurement data that are to
be used

(h) When facility tamper indicating seals have not been used or have
been compromised, when nuclear material has not yet been sealed, e.g., trays or
fuel pellets, or when sealed items are not uniquely identified, e.g., unnumbered
fuel rods or pins, the validity of previous measurements should be verified or
the nuclear material content of the items remeasured. Inventory instructions
should specify the extent of such verification or remeasurement.

161
 148

5.3.6 Preliminary Inspection and Review

Prior Co the conduct of a physical inventory the Facility Inventory

Supervisor along with each Area Inventory Supervisor should conduct a preliminary
inspection of the facility areas to be inventoried and review inventory instruc-
tions and procedures with the responsible personnel. Such inspections and
reviews should be made sufficiently in advance to allow time for corrective ac-
tion, if needed.

5.3.6.1 Process Conditions

The Facility and Area Inventory Supervisors should review process con-
ditions and status with operating supervisors of each area to be inventoried. In-
ventory instructions should be reviewed in relation to production schedules to as-
sure that they are compatible and that any areas to be shut down are properly
scheduled for shutdown.

5.3.6.2 Area Preparation*

Preparation in each area for inventory should be reviewed with area

custodians and Area Inventory Supervisors to assure that each undertands his
instructions, duties, and responsibilities during the inventory. The areas
should be inspected to assure that material is measured and sealed, packaged,
labeled, scored, or otherwise prepared for the inventory. Arrangements should be
made for the measurement and sealing of any unmeasured material in the area that
will be present during the inventory.

5.3.6.3 Records

Facility and area records should be reviewed to assure that they are
current and that record clerks and accountants understand the records cutoff pro-
cedures and times applicable to each set of records.

5.3.6.4 Measurements

Sampling and measurement procedures to be used during the inventory

should be reviewed with the responsible persons to assure that they understand
their instructions, duties, and responsibilities. Any equipment to be used in
such procedures should be inspected and calibrated or calibration records checked
to assure that the instrumentation is ready for accurate inventory measurement.
The analytical sample schedule should be reviewed with the laboratory supervisor
to assure prompt analytical results.

5.3.6.5 Inventory Teams

Inventory instructions should be reviewed in detail with each inven-

tory team for each area it is to inventory to assure that each team member under-
stands his duties and responsibilities. A tour of the area in which each team
will work should be used to orient the inventory teams.

162
 149

5.4 Conduct of Inventory

The inventory should proceed according to instructions and plans. If proper

planning) inspection, and personnel orientation have been carried out, there
should be a minimum of problems. Nevertheless, the Facility Inventory Supervisor
should not be assigned to any specific activities, such as being a team member,
so that he can be available at any time to take care of anomalies and to approve
deviations from planned procedures. During the inventory it may be useful for
the Facility Inventory Supervisor to move from area to area to maintain cogni-
zance of the progress of the inventory. When touring the area, he should keep in
touch so that he is available for problems and to maintain control of tags and in-
ventory lists.

5 .5 Post Inventory Activities

5.5.1 List and Tag Accuracy Check

Upon completion of the inventory in each area, before the area is released
from cutoff the Facility Inventory Supervisor should inspect the area with the
Area Inventory Supervisor to assure that all material in the area has been tagged
with current inventory tags. A random sample of the items in each should be
checked against the inventory lists, and a random sample of the items on the
lists should be checked against the items is the areas to assure that items have
been tagged and have been recorded accurately on the inventory sheets. It is nei-
ther practical nor necessary that these verification checks be 100%. A valid sta-
tistical sampling plan should.be used for the two populations, i.e., the popula-
tion of tagged items and the population of listed items. Statistical sampling
plans for various situations are discussed by Jaech (Jaech 1973). If these tests
of the inventory accuracy do not show a high level of confidence, on the order of
95Z or better, additional checks should be made or the ares reinventoried. In ad-
dition to the list and tag accuracy checks for an area, the tags and lists
assigned to the area team also should be accounted for. Upon completion of the
entire inventory all tags and inventory item lists should be accounted for by the
Facility Inventory Supervisor.

5.5.2 Cutoff Verification

Upon completion of each area inventory and of the entire facility inventory,
the facility and area cutoff procedures should be verified to assure that all in-
ternal transfers were recorded in more than one area; that material was
inventoried in the proper area; and that all receipts and shipments were recorded
properly with the respect to the cutoff times for the respective material
movements. All transfers into or out of each area should be checked for a short
period (perhaps a day) prior to and after the cutoff time to assure that the
transfers were recorded in the records of the area in which the material was
inventoried. Documentation of receipts and shipments should be checked in a simi-
lar manner to assure that only those receipts included in the material balance
interval ended by the inventory just taken are included in the plant records for
that interval and that all shipments made before the cutoff time have been
removed from the records. Documentation of measured discards also should be
checked in a similar manner.
 150

S.5.3 Inventory Summary and Reconciliation

The raw data from the inventory item lists should be summarized as soon as
possible after listings are completed. Some area summaries may be prepared
before the total facility inventory is completed. It also may be that completed
summaries will have to wait on analytical results to provide element and/or iso-
topic data for some line items on the inventory sheets. Analytical scheduling
should be planned to make this delay as short as possible. The original inven-
tory item lists and their summaries should be maintained under the control of the
Facility Inventory Supervisor. Reconciliation of area summaries to area records
should be done under the control of the Facility Invencory Supervisor. The area
inventories should be combined under the control of the Facility Inventory
Supervisor to provide the total facility inventory.

Inventory reconciliation involves comparing the results of the physical in-

ventory to the inventory as stated in the records and resolving any differences
to the extent possible by correction of errors in either set of data. A record
should be made of all adjustments made during reconciliation to either the physi-
cal inventory data or to the records. Such adjustments to the inventory data
should be approved by the Facility Inventory Supervisor. • Adjustments to the re-
cords should be approved according to established facility control procedures for
such adjustments. Detailed descriptions of such procedures are included in the
document addressing the Records and Reports Element of an SSAC. The final adjust-
ment should be the material unaccounted for (MUF) which will bring the records
for the respective areas and the total facility into agreement with the physical
inventory. This MUF adjustment should be documented and approved for each area
and for the total facility according to the established facility procedures for
recording MUF.

The final step to permit the evaluation of the effectiveness of the nuclear
material accountancy system is the calculation of the uncertainty of the MU7
(OHUP) for each area and/or the entire facility. Using the physical inventory
data properly stratified, other facility record data, and measurement quality
data, UMUF can be calculated and evaluated. Detailed descriptions of the statis-
tical techniques and methodology for this are presented in Part 6, on the Mate-
rial Balance Closing Element of an SSAC.

A material balance report for each MBA is called for in most facility
attachments. The physical inventory results for each MBA are combined with data
from other facility records to produce such a report for each MBA. This is
addressed in the detailed descriptions of the Records and Reports and Material
Balance Closing Elements of an SSAC. Each material balance report is required to
have attached a physical inventory listing (PH.) which lists all batches,
specifying material identification and batch data for each batch. This listing
should be easily prepared from the lists that result from the physical inventory
taking if they are prepared and coded as described herein. In addition, final
stratified inventory item lists can easily be prepared for IAEA inspector verifi-
cation purposes. If facility control areas and IAEA MBAs are not the same then
separate summaries and reports would need to be prepared for the IAEA MBAs,
rearranging the data to fit the different MBA structure.

164
 151

5.6 Example of ?hyscial Inventory Taking Procedures, Instructions and Data

Handling

5.6.1 Introduction

The facility used in this example is the conversion fabrication plant

described in Appendix A. Figure A.I shows the process scheme and IAEA MBAs for
the facility. As noted in the Appendix it is assumed that the facility control
areas and IAEA MBAs are the same. Tables A.I and A.2 describe the key measure-
ment points appropriate for IAEA safeguards. The inventory to be used in the ex-
ample for the reference facility is given in Table 5.1. Tne procedures and in-
structions that follow in Sections 5.6.2 through 5.6.5 were prepared for the fa-
cility described in Appendix A. The procedures and instructions are expressed is
the imperative mode because they are examples of what would be issued st a facil-
ity. They are not offered as universal requirements. Variations ars, of course,
quite possible. The facility operator might choose t- have different material
control areas for internal control purposes, e.g. conversion a separate control
area from pelletizing. The facility operator would undoubtedly make some measure-
ments intermediate between the KMPs for process and quality control purposes. He
might also decide not to shut the plant down and clean it out for inventory but
to use some of the dynamic -Inventory techniques discussed in section 5.2.5.2.
Also, all materials may not be processed to the somewhat ideal foras assumed in
the example. In such cases different instructions would need to be prepared and
different items, batches, and strata of inventory would be produced.

The reference facility and the example have been kept simple to permit
easier understanding of the basic approach. More complex procedures and instruc-
tions would be needed for a larger, more complex facility where there might be a
higher throughput or more than one process line, either conversion or
pelletizing. The reference facility is assumed to process only one enrichment,
which would be unusual. Normally a facility would be expected to process several
different enrichments including some input of natural uranium. In such cases ad-
ditional measurements for isotopic accounting would be needed and additional
strata differing in enrichment would be called for. Parallel control areas might
be called for which would not all shut down at the sane time but perheps in
sequence. It may also be necessary in order to maintain production schedules not
to shut down the rod assembly line. In such case the counter current inventory
technique could be used. This technique also could be used in the pelletizing
line but with more difficulty because of the need to make intermediate measure-
ments of pressed but unsintered pellets, sintered but unground pellets, and scrap
material generated in pressing and grinding. The batch blank or cleanout blank
might be used in the conversion line to avoid shutdown.

5.6.2 General Inventory Procedures.

Physical inventory of the total facility shall be taken twice each year at
approximately six-month intervals. Scheduling shall be the responsiblity of the
Head of the Nuclear Material Control Group in consultation with the Facility Man-
ager and the Head of Production. The approximate schedule shall be provided to
the IAEA six months in advance with exact dates provided not later than 30 days
prior to the start of the inventory.

165

o
c
o
 152

Table 5.1 Example Reference F a c i l i t y Inventory

MATERIAL TYFE NUMBER OF ITEMS TOTAL URANIUM (kg)

28 C y l i n d e r s 28,000

UO, P o w d e r 2S0 Buckets 7,000

Sintered P e l l e t s 1000 Trays 5,000

Dirty Powder 90 Drums 2,250

ADU Scrap 15 Drums 375

Green Scrap 20 Drums 500

Sintered Scrap 20 Drums 500

Grinder Sludge 100 Buckets 500

Rods 50 Bins (10,000 rods) 50,000

Reject Rods 100 Rods 500

Liquid Wastes 100.Drums 200

Solid Wastes 80 Barrels 400

Assemblies 80 Assemblies 40,000

1913 Items plus 135,225

50 Bins of 10,000 rods

166
m
o

o o
z —
 153

The process shall be shut down for physical inventory taking in accordance
with cutoff times and procedures specified in the detailed inventory
instructions. All process vessels, lines, and equipment shall be cleaned out so
that gemma-ray survey meter measurements at KM?s D, G, I and K shall show negligi-
ble quantities of nuclear material hold up as specified in detailed inventory
instructions. All nuclear material shall be retained in or collected in standard
containers appropriate to the material and the associated KM? as shown in the In-
ventory Key Measurement Point Table (Table A . 2 ) . Items and material not
previously measured shall be reduced to measurable form to the extent possible
and measured. Items which were previously measured but which have not been
maintained under tamper indicating sealed conditions must be remeasured. Sam-
pling and sealing of containers should be started as soon as possible to provide
time for the laboratory to complete the sample analyses before the inventory
closing.

Process and inventory schedules shall be coordinated such that:

(1) nuclear material in the conversion area is processed through KMP-4

to UO2 powder storage in MBA-2:

(2) nuclear material in the pelletizing area is processed through KMP-5

to ground pellet storage in MBA-2;

(3) all scrap is processed to ncrap storage in MBA-2 and if possible

through scrap processing and KMP4 to UO2 powder storage in MBA-2;

(4) all H q u i d and solid wastes are processed to storage, KMPs E and F,
and if possible removed to discard and burial; and

(5) rod loading and assembly processing are completed such that all rods
are either in rod storage or assembly storage.

Deviations from the preceding process inventory status mist be approved by

both the Head of the Nuclear Material Control Group and the Head of Production.

5.6.3 Responsibility and Authority

5.6.3.1 Facility Inventory Supervisor.

The Head of the Nuclear Material Control Group is designated as the

Facility Inventory Supervisor. As such he is responsible for planning,
organizing and conducting the physical inventory. Specifically his responsibil-
ity and authority shall be as follows:

(1) He shall be responsible for scheduling the inventory and

coordinating the process shutdown with Production. His author-
ity for process shutdown shall require coordination and concur-
rence of both the Facility Manager and the Bead of Production.

(2) Within the established inventory schedule and process shutdown

and startup schedule he shall be responsible for and have sole
 154

authority to establish and enforce all cutoff times and

procedures.

(3) He shall be responsible for assuring that the physical inventory

taking is conducted in accordance with established written proce-
dures and instructions and shall have the authority to cause
such actions as are necessary to attain such assurance. He also
shall have the authority to authorize deviation from established
procedures and instructions, if, in his opinion, such deviation
is necessary to the satisfactory conduct of the inventory
taking.

(4) He shall not be a member of any inventory team but will be avail-
able at all time3 during the inventory to handle problems to as-
sure satisfactory conduct of the inventory taking.

(5) He shall be responsible for assigning personnel to inventory

teams in consultation with the respective MBA custodians and
other appropriate facility management persons.

(6) He shall be responsible for authorizing startup of the process

after the physical inventory taking is complete. He sV^ll be re-
sponsible for assuring that the process is not started until the
inventory data have been processed to a point where it can be
assured that there are no additional actions necessary for a
satisfactory inventory that might: be precluded by process
startup. If process startup is delayed beyond the originally
scheduled time he shall coordinate the decision for such delay
with the Facility Manager and the Head of Production.

5.6.3.2 Area Inventory Supervisor.

The material custodian in each material balance area is designated as

the Area Inventory Supervisor for the MBA in which he is custodian. As such he
is responsible for preparation for and conduct of the physical inventory taking
in his MBA within Che framework of the facility general inventory procedures and
the specific inventory instructions applicable to his MBA. Specifically, his
responsiblity and authority shall be as follows:

(1) He shall be responsible for assuring that his MBA is prepared

for the physical inventory taking in accordance with the facil-
ity inventory procedures and instructions. He shall have the au-
thority to cause such actions to be taken as are necessary to at-
tain such assurance.

(2) He shall be responsible for assuring that the physical inventory

taking in his MBA is conducted in accordance with estab-lished
written procedures and instructions applicable to his MBA and
shall have the authority to cause such actions as are necessary
to attain such assurance. He shall have no authority to autho-
rize deviation from established procedures and instruct-ions. If
he deems such a deviation necessary he shall obtain authoriza-
tion for it from the Facility Inventory Supervisor.
 155

(3) He may be a member of an inventory team in his MBA.

5.6.3.3 Inventory Teams.

Inventory teams shall be designated by the Facility Inventory

Supervisor in consultation with other appropriate managerial staff. The teams
shall include one person familiar with the area being inventoried and one person
from the Nuclear Material Control Group. Each inventory team shall be responsi-
ble for carrying out the physical inventory taking instructions applicable to its
assigned ares and shall have no authority to deviate from such instructions
without the approval of the Facility Inventory Supervisor. There will be one in-
ventory team assigned to each MBA.

5.6 .A Inventory Instructions

5.6.4.1 General Schedule and Cutoffs.

A physical inventory of the entire facility will be taken for the

six-month material balance period ending 2400, January 31, 1983*. Preparations
for process shutdown and cleanout should be scheduled so that inventory teams can
begin listing and tagging no latter than 0800, February 2, 1983*. Process
startup is scheduled for 0800 February 4, 1983, provided that inventory taking,
summarizing and reconciliation are complete and that IAEA verification has been
completed. The following cutoff times and procedures shall be observed.

5.6.4.1.1 Receipts, Shipments and Other Removals. There

shall be no receipts into or shipments or other removals from the facility after
2400, January 31, 1983 and before 0800, February 4, 1983. If a shipment must be
made after this time, for example, to meet a customer commitment, the Facility In-
ventory Supervisor shall be notified 60 that appropriate records can be made to
assure proper handling with regard to the inventory and the material balance
of the material quantity involved. If material arrives at the facility for re-
ceipt it shall be held in quarantine until completion of the inventory aad the Fa-
cility Inventory Supervisor notified of its presence. If other removals are nec-
essary, such as waste discard accumulating in excess of storage capacity, the Fa-
cility Inventory Supervisor shall be notified so that appropriate record can be
made of the removal. Every effort should be made to* make such removal and ship-
ments of product and DF 6 heels prior to the 2400 cutoff.

5.6.4.1.2 Internal Transfers. There shall be no transfers

between MSAs after 2400, January 31, 1983 and before 0800, February 4, 1983.
Every effort should be made to process reject pellets through KMP-6 to
storage in MBA-2 prior to 2400, January 31, 1983. If this cannot be done reject
rods shculd not be transferred to M3A-2 but sent to rod storage in MBA-3
appropriately marked as rejects and segregated from acceptable rods.

*These times and dates and all others that follow are meant to be illustrative
only.

169
 156

5.6.4.1.3 Process. The process shall be shut down for

cleanout as soon after 2400, January 31, 1983 as all material then in process
has been processed either to U0 2 powder or to sintered pellets. The process
may be shut down on the shift prior to the 2400 cutoff if it is expedient to
do so to assure that all sintered pellets are processed through KMP-5 to
storage. No material shall be added to the process after the 2400 cutoff.
The waste process shall continue to operate as long as necessary to process
any waste generated by the scrap conversion and pelletizing process cleanout
to waste storage. Scrap from the cleanout shall be processed to scrap
storage. The rod loading and assembly process shall continue to operate until
all rods are in rod storage or in assembly storage. All material in HBA-1
shall be processed to KMPs A, B, or C as appropriate and stored under
tamper-indicating sealed conditions until needed for transfer to MBA-2 after
the inventory. UFg heels shall be processed through KMP-8 and if possible
transferred to HBA-1 and returned to the vendor prior to the 2400 cutoff.
5.6.4.1.4 Records. All transaction documents for
transactions prior to 2400 January 31, 1983 shall be made available to MBA and
Central Records no later than 0800, February 1, 1983. All transactions up to
2400 January 31, 1983 shall be posted to the facility central records and to
the HBA records and the books closed as of the 2400 cutoff time. The MBA and
total facility book inventories as of the 2400 cutoff time shall be calculated.
5.6.4.2 MBA-1 Shipper Receiver Area.
All containers in this area must be sealed. Items which ar not
sealed or which have broken seals must be remeasured. For UFg cylinders
this is done by reweighing and using the original factor determined for that
cylinder at KMP-1C. For the UNH tank after the cutoff time when no further
material is to be added to the process, the tank is weighed, sealed, and the
original factor from KMP-1B applied. For any unsealed U0 2 buckets the
procedures for KMP-1C shall be applied except that each unsealed bucket shall
be sampled to verify its content. The bucket shall remain a part of its
original batch. Facility Inventory Supervisor review of this MBA will begin
at 1800, February 1, 1983. Inventory Team listing and tagging will begin at
0800, February 2, 1983.
5.6.4.3 MBA-2 Process Area.
All material in this area shall be processed to one of the six
inventory KMPs, E, F, H, J, L, or M.
(a) All duct work with potential uranium accumulations shall be
disassembled and any material removed, put in standard
containers appropriate for the material as specified for
KMP-L, measured, sealed and put in scrap storage.
(b) All process equipment with potential uranium shall be
completely cleaned to assure that no holdup is present.
Equipment handling liquids shall be flushed to remove any
solid buildup. Material removed shall be treated either as
scrap and processed to scrap storage (KMP-L) or processed
 157

through the waste treatment process to either solid or liquid

waste storage (KMP-E and F ) .

(c> Trays of pellets should be processed through grinding and

KMP-5 to storage. Trays of pellets shall be retained in the
sealed pellet storage cabinets in MBA-2, (KMP-J) or processed
through rod loading to rod storage in MBA-3 (KMP-N).

(d) Buckets of U 0 2 powder shall be measured, sealed, and stored

in U 0 2 powder storage (KMP-H). Any buckets found unsealed
shall be sampled and measured as indicated for KMP-4. The
verified buckets shall remain with their original batches.

(e) Upon completion of duct and equipment cleanout the equipment

shall be gamma-scanned as indicated for KMPs D, G, 1, and K.
Any indication of remaining material holdup shall call for addi-
tional cleanout.

(f) Pellet storage areas and the furnace should be inspected to as-
sure that there are no loose pellets. Any found shall be
treated as scrap and processed accordingly to KMP-L.

(g) Filters and other solid wastes shall be processed

through waste treatment to KMP-E.

(h) Facility Inventory Supervisor review of this MBA

will begin at 1900, February 1, 1983. Inventory Team listing
and tagging will begin at 0800, February 2, 1983.

5.6.4.4 MBA-3 Assembly Area.

All material in this area shall be processed to one of two inventory

locations, rod storage (KMP-N), or assembly storage (KMP-4.'.

(a) Roda shall not be transferred from MBA-2 unless they can be
processed to KMP-N.

(b) P.eject rods not transferred to MBA-2 shall be segregated in a

separately identified sealed bin in rod storage until the inven-
tory is completed.

(c) Assemblies shall be f " >leced and placed in assembly storage

(KMP-O).

(d) Only a small amount of solid waste containing negligible

quantities of uranium should be generated in this area. Any
generated after cutoff should be packaged, gamma scanned and
held for the inventory.

(e) Facility Inventory Supervisor review of this M3A will begin at

0600, February 2, 1983. Inventory Team listing and tagging will
begin at 0800, February 2, 1983.
 158

5.6.4.5 Physical Inventory Item List and Stratification

Upon completion of the inventory preparation as directed in the pre-

ceding sections each Area Inventory Supervisor shall prepare a preliminary inven- .
tory item list based on the process and material accounting records and the re-
sults of the inventory preparations for his MBA. This listing shall be prepared
using the pre-numbered Physical Inventory Item List forms obtained from the Facil-
ity Facility Inventory Supervisor. Table A.3, Inventory Material Description,
shall be used to define the entries on the Physical Inventory Item List. A sepa-
rate sheet shall be prepared for each material type and stratum. If there is ma-
terial on inventory other than that identified in Table A.3, a separate Physical
Inventory Item List shall be made for it. Such material should be batched or
stratified. The minimum information to be provided for such material shall be de-
scription or identification, quantity and location. If a KMP designation is not
appropriate or sufficient to identify the material location an additional loca-
tion identification should be entered along with the KMP/Stratum identification.
The persons preparing the preliminary listing shall sign each sheet in the block
designated "Listed by". The original lists shall be sent to the Facility Inven-
tory Supervisor no later than 1700, February 1, 1983.

Figures 5.1, 5.2, 5.3, 5.4 and 5.5 are examples of typical prelimi-
nary inventory item lists for some different material types, i.e. material in
storage, loaded rods, reject rods, and scrap. Noce that the inventory tag number
column is not filled in. This is to be done during the inventory talcing by the
inventory teams.

5.6.4.6 Inventory Team Instructions.

Each inventory team will be provided with the preliminary inventory

lists for their assigned MBAs, pre-numbered blank Physical Inventory Item List
forms, pre-numbered inventory tags to affix to items as they are inventoried and
Inventory Tag Reconciliation forms (figure 5.6). Inventory teams shall proceed
as follows:

(a) In MBAs 1 and 2 one member of the team:

(1) Identifies each container and tray storage cabinet by

number.

(2) Determines that each container or cabinet is sealed and

that seal is not broken.

(3) Identifies pellet trays not in sealed storage cabinets by

tray numbers. The Area Inventory Supervisor or the Facil-
ity Inventory Supervisor must be notified of these to ar-
range for them to be remeasured.

(4) Identifies cabinets that contain no trays.

(5) Calls out all pertinent data identified in (1), (2), (3),
and (4) including tray numbers in sealed cabinets as shown
on the card attached to the cabinet and data shown on the
 159

container label, tag, or record card, including seal

identification.

(6) Affixes inventory tags to containers, trays, or cabinets

calling out the numbers. Empty cabinets are also tagged.

(b) The second team member:

(1) Records the inventory tag number on the inventory item list
in the proper column and in the line corresponding to the
identified container, tray or cabinet. Note: An inventory
tag number for a s-ealed cabinet will correspond to all of
the trays in that cabinet that are listed on the inventory
item list.

(2) Verifies that the data on the list are as called out by the
first team member.
(3) Makes proper notations on the list whenever any data differ
from those called out. A.n entry on the list should not be
changed or crossed out but corrected data should be entered
next to it on the list.

(4) If a seal is missing or broken, notes this on the list and

circles the seal number. The Area Inventory Supervisor or
the Facility Inventory Supervisor shall be notified so that
material in that container can be remeasured.

(5) Any item or batch with a seal number having a prefix letter
A, B, or C shall be noted in the remarks column as "static"
inventory.

(6) Completes write-in inventory item listing form should there

be a container, tray, or cabinet found not to be on the in-
ventory list, e.g. empty cabinet.

(7) Records data for remeasured items.

(c) In MBA-3 one member of the team:

(1) Identifies sealed rod storage bins by number.

(2) Identifies individual rods that are not stored in sealed

bins by aerial number.

(3) Identifies assemblies by serial member.

(4) Identifies bins that contain no rods.

(5) Calls out pertinent data identified in (1), (2), (3), and
(4), including serial numbers of rods in sealed bins as
shown on the card attached to the bin.
 160

(6) Affixes inventory tags to rods, bins, and assemblies as

identified, calling out the numbers. Empty bins are also
tagged.

(d) The second team member:

(1) Records the inventory tag number on the inventory item list
in the proper column and in the line corresponding to the
identified item. Note: An inventory tag number for a
sealed bin vill correspond to all of the rods in that bin
that are listed on the inventory item list.

(2) Verifies that the data on the list are as called out by the
first team .lumber.

(3) Any item or batch with a seal number having a prefix letter
A, B, or C shall be noted in the remarks column as "static"
inventory.

(4) Completes write-in inventory item listing form should there

be a rod, bin, or assembly found not to be listed, e.g.
empty bin.

(e) Both team members in each MBA shall sign each of the inventory
item lists used in their MBA.

(f) Both team members in each MBA shall complete and sign the tag
reconciliation form.

(g) All inventory item lists (used, voided, or unused), unused inven-
tory tags, and the inventory tag reconciliation sheets are given
to the Facility Inventory Supervisor.

5.6.5 Post Inventory Activities.

5.6.5.1 List and Tag Check.

Upon completion of the listing by the inventory teams in each MBA the
Facility Inventory Supervisor shall:

(a) Account for all pre-numbered forms and tags assigned to that MBA
team.

(b) Using an attributes sampling plan, randomly select a sufficient

number of listed items from the inventory item lists to verify
with 952 certainty that items listed were tagged and listed
correctly.

(c) Inspect the area with the Area Inventory Supervisor and randomly
select a sample of the items in the area to assure that the
items had been tagged and listed correctly.
 161

(d) Cause such actions as he may deem appropriate Co correct any

discrepancies found.

5.6.5.2 Cutoff Verification.

Upon completion of the inventory taking and list and tag check, the
Facility Inventory Supervisor shall, for the 24 hour period prior to the inven-
tory cutoff time:

(a) review the records for all internal transfers into and out of
each MBA to assure that all trecsfers were recorded properly in
the records of both the receiving and shipping MBA and tha* no
transfers occurred or were recorded after the cutoff time;

(b) in MBA-1, review the records for all receipts to assure that all
receipts prior to cutoff were properly recorded and that no re-
ceipts occurred or were recorded after the cutoff time;

(c) in M5A-2, review the records for all waste discards to assure
that such removals prior to cutoff were properly recorded and
that none occurred or were recorded after the cutoff time; and

(d) in MBA-3, review the records for all shipments to assure that
all shipments prior to cutoff were properly recorded and that
none occurred or were recorded after the cutoff time.

Any discrepancies found in any of the preceding shall be corrected in

the records so that the material involved in any transfers, receipts, shipments,
or discards is correctly documented in relation to the material-balance interval
ended 2400, January 31, 1983. Physical inventory listings also may need to be
corrected for improper material movement after cutoff; for example, material
received after cutoff which is mistakenly listed on the physical inventory but is
not (and should not be) included in the records.

5.6.5.3 Inventory Summary and Reconciliation. Upon completion of

the preceding checks, the Facility Inventory Supervisor shall prepare from the in-
ventory item lists physical inventory summaries for each MBA and for the entire
facility. Table 5.2 is an example of such a summary. These summaries are
compared to the book inventories for each MBA and the total facility. Any differ-
ences are investigated and resolved to the extent possible by correcting mistakes
in each set of data. Any corrections or other adjustments shall be clearly noted
in the records that are changed. Changes shall be approved by the Facility Inven-
tory Supervisor with the concurrence of the appropriate Area Inventory
Supervisor. The final adjustment will be material unaccounted for (MOF) made to
the records to bring the MBA and facility records into agreement with the physi-
cal inventory. These adjustments must be approved by the Facility Inventory
Supervisor and the Facility Manager and each MBA Material Custodian for his MBA.

A material balance report for each MBA shall be prepared for submis-
sion to the IAEA along with a FIL for each MBA using the appropriate IAEA forms.
Figures 5.7 is an example of a Physical Inventory Listing on IAEA Form R.02/c for
some of the material on inventory in MBA-2.
 162

Table 5.2

Physical Inventory Sutnnary

MBA Material Items Quantity

Type (kg 0)

MBA-1 28 Cylinders 28,000

UO2 Powder 80 Buckets 2,000
UNH None 0

Total PI 30,000
Book Inventory 30,0CQ
MDF - MBA-1 0

MBA-2 DDj Powder 200 Buckets 5,000

Sintered Pelists 200 Trays 1,000
Dirty Powder 90 Drums 2,250
ADU Scrap 15 Drums 375
Green Scrap 20 Drums 500
Sintered Scrap 20 Drums 500
Grinder Sludge 100 Buckets 500

Total PI 9,125
Book Inventory 9,479
MÏÏF - MBA-2 354

MBA-3 Rods 10,000 Rods 50,000

Reject Rods 100 Rods 500
Assemblies 80 Assemblies 40,000

Total PI 90,500
Book Inventory 90,500
MDF - MBA-3 0
 163

References
EUR 6471 EN: F. Argentesi, T. Casilli, and M. Franklin, Nuclear Material Statisti-
cal Accounting System, CEC report EUR 6471 EN, 1979.
IAEA/SG/INF/1: IAEA/SG/INF/1, IAEA Safeguards Glossary, IAEA, Vienna, 1980.
IAEA-TECDOC-261: IAEA Safeguards Technical Manual, Part F: Statistical Concepts
and Techniques. Vol. 3, IAEA, Vienna, 1982.
Jaech 1973: John L. Jaech, Statistical Methods in Nuclear Material Control,
TID-26298, U.S. Atomic Energy Commission, 1973, p. 319 ff.
 164

This page is intentionally left blank

 Figure 5.1

PHYSICAL INVENTORY ITEM LIST

Inventory Date: 31-1-83 Item: Cylinder natch Identification: Cylinder Number

MBA Number: 1 Batch: 1 Cylinder Stratum Identification: KMl'-A

Material Type: UF Location: UF, Storage (KMP-A)

Inventory Item Seal Batch Net Weight % Element X Isotope

Tag No. Numbe r Numbe r Identity (gins) Element wt. (gms) Isotope wt. (gms) Remarks

C42608 E2536 C426O8 1486738 67.56 1004440 3.011 30244

C49641 E2537 C49641 1478334 67.57 998910 2.986 29827
C35O11 E2538 C35O11 1474317 67.67 997670 2.993 29860
C59741 E2539 C59741 1481304 67.66 1002250 2.930 29366
C69497 E2540 C69497 1484698 67.64 1004250 3.011 30238
C18939 E2541 C18939 1475965 67.65 998490 2.996 29915
C45177 E2542 C45177 1486776 67.53 1004020 2.991 30030
C80352 E2543 C80352 1485793 67.57 1003950 3.014 30259
C90421 E2544 C90421 1474013 67.61 996580 3.000 29897
C32235 E2545 C32235 1472875 67.65 996400 2.991 29802
C7328 E2546 C7328 1484103 67.56 1002660 2.993 30010
C6734 E2547 C673/i 1485433 67.62 1004450 3.004 30174
C92885 E2548 C92885 1476874 67.63 998810 3.011 30074
C10120 E2549 C10120 1471505 67.66 995620 3.010 29968
C46366 E2550 C46366 1479805 67.59 1000200 2.997 29976
C73442 E8997 C73442 1483925 67.62 1003430 2.999 30093
C00112 E8998 C8W1J2 1477375 67.58 998410 2.992 29872
C56641 E8999 C56641 1485649 67.59 1004150 3.013 30255
C52986 E9000 C52986 1473182 67.64 996460 3.011 30003
C44380 E6200 C4438O 1480332 67.52 999520 3.000 29986
C15455 E6201 C15455 1474064 67.55 995730 2.997 29842
C85453 E6203 C85453 1467874 67.67 993310 2.996 29760
C22451 E6204 C22451 1475987 67.63 998210 3.010 30046
C77468 E62O5 C77468 1481124 67.60 1001240 2.995 29987
C11066 E6206 CI 1066 1480352 67.59 IOOO57O 3.001 30027
C37183 E62O7 C37S33 1485034 67 .62 1004180 2.999 30115

Te9ted by: Inventoried by: Verified by: Ho.: 1000

 Figure 5.2

PHYSICAL INVENTORY ITEM LIST

Inventory Date: 31-1-83 Item: Bucket Batch Identification: Bucket Number Prefix

MSA Number: 2 Da tell : 40 Buckets Stratum Identification: KM'-II

Material Type: UO Location: U0 2 Storage MBA-2 (KMP-II)

Inventory- Item Seal Batch Net Weight Z Element X Isotope

Tag No. Number Number Identity (gms) Element wt. (gens) Isotope wt. (gms) Remarks

150 2100 150 28503 88.06 25100 3.01 756

150 2101 150 27901 88.06 24570 3.01 740
150 2102 150 28242 88.06 24870 3.01 749
150 2103 150 28140 88.06 24780 3.01 746
150 2104 150 27947 88.06 24610 3.01 741
150 2105 150 28026 88.06 24680 3.01 743
150 2106 150 28708 88.06 25280 3.01 761
150 2107 150 28685 88.06 25260 3.01 760
150 2108 150 28083 88.06 24730 3.01 744
150 10 E- 2109 150 28503 88.06 25100 3.01 756
ISO 11 E- 2110 150 28060 88.06 24710 3.01 744
150 12 E- 2111 150 28776 88.06 25340 3.01 763
150 13 E- 2112 150 28776 88.06 25340 3.01 763
150 14 E 2113 150 28333 88.06 24950 3.01 751
150 15 E- 2114 150 28151 88.06 746
24790 3.01
150 16 E- 2115 150 28764 88.06 25330 3.01 762
150 17 E- 2116 150 29003 08.06 25540 3.01 769
150 150 28571
18 15- 2117 88.06 25160 3.01 757
150 150 28492
150 19 E- 2118 28662
88.06 25090 3.01 755
20 E- 2119 150 88.06 25240 3.01 760
150 283)0
150 21 E- 2120 150 88.06 24930 3.01 750
28776
150 22 E- 2121 150 28810 88.06 25340 3.01 763
150 23 E- 2122 150 28072 88.06 25370 3.01 764
150 2'» E- 2123 150 28640 88.06 24720 3.01 744
150 25 E- 2124 150 28208 88.06 25220 3.01 759
26 E- 2125 150 88.06 24840 3.01 748 Continued to Sheet No. 1051

Tested by: Inventoried by: Veri fled by: Ho.: 1050

 Figure 5.3

PHYSICAL INVENTORY ITEM LIST

Inventory Date: 31-1-83 Item: Drum Dated Identification: Drum Number

MBA Number: 2 Batch: 1 Drum Stratum Identification: KMP-L-4

Material Type: Green Scrap Location: Scrap Storage (KMP-L)

Inventory Item Seal Batch Net Weight % Element X Isotope

Tag No. Nuinbe r Numbe r Identity (gms) Element wt. (gms) Isotope wt. (gms) Remarks

2-6330 E-9970 2-6330 32164 77.85 25040 3.11 779

2-6331 E-9971 2-6331 31383 78.29 24570 3.10 762
2-6332 E-9972 2-6332 31822 78.06 24840 2.99 743
2-6333 E-9973 2-6333 31805 77.85 24760 3.03 750
2-6334 E-9974 2-6334 31734 77.55 24610 2.89 711
2-6335 E-9975 2-6335 31730 77.75 24670 2.97 733
2-6336 E-9976 2-6336 32404 77.80 25210 3.04 766
2-6337 E-9977 2-6337 32142 78.37 25190 2.96 746
2-6338 E-9978 2-6338 31644 78.12 24720 2.92 722
2-6339 E-9979 2-6339 32074 78.10 25050 2.99 749
2-6340 E-9980 2-6340 31659 78.02 24700 3.01 743
2-6341 E-9981 2-634] 32162 78.54 25260 3.04 768
2-6342 E-9982 2-6342 32535 77.64 25260 2.89 730
2-6344 E-9984 2-6344 31671 78.21 24770 3.09 765
2-6345 E-9985 2-6345 32518 77.68 . 25260 3.01 760
2-6345 E-9986 2-6346 32645 77.93 25440 2.86 728
2-6347 E-9987 2-6347 32175 78.01 25100 3.14 788
2-6348 E-9988 2-6348 32057 78.11 25040 2.91 729
2-6349 E-9989 2-6349 32469 77.52 25170 3.03 763

Tested by: Inventoried by: Verified by: No.: 1200

 Figure 5.4

PHYSICAL INVENTORY ITEM LIST

Inventory Date: 31-1-83 Item: Fuel Rods Batch Identification: Rod Serial Number Prefix

MBA Number: 3 Batch: 200 Rods Stratum Identification: KMl'-N

Material Type: Location: Rod Storage Bins (KMP-N)

Inventory Item Seal Batch Net Weight X Element X Isotope

Tag No. Number Number Identity (gms) Element wt. (gms) Isotope wt. (gms) Remarks

130-001 E-5001 !30 5636 88.01 4960 3.02 150 Stored in Bin #8
130-002 E-5001 130 5636 88.0< 4960 3.02 150 with Seal No. E 5001
130-003 E-5001 130 5624 4950 3.02 149
130-004 E-5001 130 5636 88.01 4960 3.02 150
130-005 E-5001 130 5715 88.01 5030 3.02 152
130-006 E-5001 130 5658 88.01 4980 3.02 i50
130-007 E-5001 130 5658 88.01 4980 3.02 150
130-008 E-5001 130 5670 88.01 4990 3.02 151
130-009 E-5001 130 5670 88.01 4990 3.02 151
130-010 E-5001 130 5727 88.01 5040 3.02 152
130-011 E-yiOl 130 5681 88.01 5000 3.02 151
130-012 E-5001 130 565d 88.01 4980 3,02 150
130-013 E-5001 130 5727 88.01 5040 3.02 152
130-014 E-5001 130 5636 88.01 49bO 3.02 150
130-015 E-5001 130 5647 88.01 4970 3.02 150
130-016 E-5001 130 5693 88.01 5010 3.02 151
130-017 E-5001 130 5704 88.01 5020 3.02 152
130-018 E-5001 130 5636 88.01 4960 3.02 150
130-019 E-5001 130 5693 88.01 5010 3.02 151
130-020 E-5001 130 5693 88.01 5010 3.02 151
130-021 E-5001 130 5693 88.01 5010 3.02 151
130-021 E-5001 130 5693 88.01 5010 3.02 151
130-022 E-5001 130 5681 88.01 5000 3.02 151
130-023 E-5001 130 5670 88.01 4990 3.02 151
130-024 E-5001 130 5727 88.01 5040 3.02 152
130-025 E-5001 130 5681 88.01 4990 3.02 151
130-026 E-5001 130 5693 88.01 5010 3.02 151 Continued on Sheet #1501

Tested by: Inventoried by: Verified by: No.: 1500

 Figure 5.5

PHYSICAL INVENTORY ITEM LIST

Inventory Date: 31-1-83 Item: Reject Rods Batch Identification: Rod Number

MBA Number: 3 Batch: 1 Rod Strata Identification: KMP-N

Material Type: UO in Rods Location: Rod Storage Bins (KMP-N)

Inventory Item Seal Batch Net Height % Element % Isotope

Tag No. Number Number Identi ty (gms) Element wt. (gms) Isotope wt. (gms) Remarks

120-012 N/A 120-012 5726 87.85 5030 3.09 155 Stored in Bin #10
120-016 N/A 120-016 57J4 87.85 5020 3.09 155
120-175 N/A 120-175 5737 87.85 5040 3.09 156
120-193 N/A 120-193 5646 87.85 4960 3.09 153
122-061 N/A 122-061 5667 88.06 4990 3.07 153
122-07'4 N/A 122-074 5712 88.06 5030 3.07 154
122-091 N/A 122-091 5655 88.06 4980 3.07 153
122-122 N/A 122-122 5667 88.06 4990 3.07 153
122-147 N/A 122-147 5723 88.06 5040 3.07 155
125-081 N/A 125-081 5697 88.12 5020 2.99 150
125-075 N/A 125-075 5708 88.12 5030 2.99 no
125-020 N/A 125-020 5663 88.12 4990 2.99 149
125-121 N/A 125-121 5651 88.12 4980 2.99 149
125-140 N/A 125-140 5719 88.12 5040 2.99 151
125-178 N/A 125-178 5663 88.12 4990 2.99 149
130-111 N/A 130-111 5624 88.01 4950 2.98 148
130-112 N/A 130-112 5658 88.01 4980 2.98 148
130-113 N/A 130-113 5693 88.01 5010 2.98 149
130-114 N/A 130-114 5647 88.01 4970 2.98 148
130-120 N/A 130-120 5704 88.01 5020 2.98 150
130-121 N/A 130-121 5647 88.01 4970 2.98 148
130-122 N/A 130-122 5670 88.01 4990 2.98 149
140-001 N/A 140-001 5679 88.22 5010 3.01 151
140-002 N/A 140-002 5679 88.22 5010 3.01 151
140-003 N/A 140-003 5724 88.22 5050 3.01 152
140-004 N/A 140-004 5702 88.22 5030 3.01 l'51 Continued on Sheet No. 15

Tested by: Inventoried by: Verified by: No.: 1550

1
 170

Figure 5.6
Inventory Tag Reconciliation

Number of Items
1. Listed on preliminary inventory list
2. Listed on write-in inventory lise
Total Items Tagged

Inventory Tags
1. Number assigned (from thru )
2. Deduct numbers not used
( from thru )
3. Deduct voided tags (Nos )
Tags used
Tags used should equal items tagged
Explain variances below:

Prepared by: Date:

 171
Figure 5.7 - PHYSICAL INVENTORY LISTING (PIL) FORM R.02/C
c 3UNTRY A B C OATE 310163
F ACILITY LEV REPORTS.. 11
MATERIAL BALANCE AREA 2 PACE HA. OF PACES 1 SIGNATURE

11 ! Mil !..:
CORRECTION

DESCRIPTION
A C C O U N T A N C Y D A T A
§
Î1
TO

MATERIAL
i NUMBER
NAME OR Of
I «is

ELEMENT
ITEMS WEIGHT OF
U
NUMBER OF
CATCH IN • E I C H T Of } FISSILE REPORT
z
Z BATCH ELEMENT ISOTOPES s!S H». 0;
(URANIUM ONLY)
I SIS z
. • . i : 1 ji ' ,1 1• ' i , • i
: 1-Û'J. J 1 ' •1 L LwL L L -
1 Hl 150 40 FQSB r 999675 29990 IGIMI .
2 Hl 151 40 ' FQ«B E i 999152 g 29975 IG:M' •••

3 fil 152 40 FQtB |E i 999540 R 29986 IGIM: ; !

4 JJ. 153 4Q FQ*3 k 1000834 g 30025 JG!MI i
5 Hl 154 40 FQ4>B lE 1000024 g 30000 i G J Ml S

6 L 2669 1 ROND JE 25051 g 752 ÎGI M! ji

S
7 L 2670 1 RQND te 24915 g 747 |GiM'
!
8 L 2671 1 RÇND k 24926 g 748 |Gl» ;

9 2672 1 ROND £ 25048 j> 1 751 '.G » i !

m Ti 7671 , RQND E' 25029 g 751 IGiM i

11 L 2674 1 RQND E' 25087 £ 753 IGiM ! f

12 1 2675 1 ROND 25099 _E 753 ici M i J

13
14
TU
~v 2676
2677
1
1
ROND
RQND
E'
JE
24926
24995
'< g
ig !
748
750
GIM
iG^K i
i }

! :
15 1, 2678 1 RQNp E 25034 B ' 751 1 G! M i
:
16 L 2795 1 ROND E 24993 750 ,G:K

17 L 2796 1 RQND El 24972

r6- 749 'GiK !

18 L 2797 1 RQND E 24959 g 749 Gi^ ;

2798 1 24907 iE 5
19 ROND E 747 i G!M
20 L 2800 1 RQND El 24953 g ! 749 GlK .'

21 L 2001 1 RQND E 25072 E i 752 G) K ! i !

22 L 2002 1 RQND E! 24942 _ 748 -G: M ' 5

r, ! ! :
?3 2003 1 KQKD 25051 g 7S? T." •
£
24 L 2004 ! 1 ROND fei 25014 'E 750 G.Ù • i i
;
25 L 2005 1 RQND JE ' 24990 g '• 750 G! M, ' Is
26 jj 149 200 JO7A lE 999782 'E i 29993 G: M : : s
5
27 II 2592 RKND JE '. 24901 ' E "47 :G'K
28
29
TL 2593
2594
1
1
KKND
RKND JE
fe 24945
24917
•E !
E
748
748
IGlK
i G[ Mi
i 5

30 L 2596 1 RKND F. 25012 ie 750 \cj« i

i
31 L 2597 1 RKND E 25063 E 752 IG'MJ !
5
32 Ti ?SQR 1 RKND r 24985 p 750 IGJM! '
33 L 2599 1 S
RKND E 24908 E 747 IGIÎC
34 L 2600 1 î
RKND E 25018 g 751 ici»
35 L 2601 1 RKND E 25090 E 753 GIMI s
36 L 2602 1 RKND E 24979 ig 749 G! Mi s
37 L 2603 1 RKND E 25047 ig l 751 JG|«i s

M-56C/c/ftav.d (Oci 7S)

 172

Header Coding:
1- 2 Country Code
5- 3 Facihtv Code
9-12 MBA Code

19-24 Date (YYMMOOI

25—27 Numoer ot line Entries, increased bv 1.
28-31 For IAEA use only

45-60 deceived in SGIT (csdefl at me IAEA

52-57 Punching Bate Icodeo at the IAEA!

70-73 Report Number

74—79 Attribute (coded at the IAEA'
 173

Part 6. MATERIAL BALANCE CLOSING

6.1 Introduction

While it is the physical inventory that is the bench mark of nuclear mate-
rial accounting and control, in that it provides evidence of the physical pres-
ence of the material, it is the material balance based on the physical inventory
that permits the determination of whether significant losses or diversions have
occurred undetected. Data, not only from the physical inventory, but from all
components of the material balance must be based on measurements of known
uncertainties so that the material balance and MUF resulting from the data can
be evaluated and meaningful conclusions reached. While safeguards agreements
and facility attachments do not call for the State or the facility to report or
even to calculate Ofjuy, or to otherwise evaluate MUF, it would be expected that
either the State Authority or facility management, or both, would choose to do
so. The descriptions in this part provide technical information regarding the
methodology for the treatment of the material balance component data, including
error propagation for calculation of the variance of the MUF and procedures for
testing the significance of the MUF.

The IAEA has established procedures by which its inspectors perform inde-
pendent evaluations of facility WSF using both IAEA derived data and facility
data. These procedures are described in IAEA-TECDOC-261. In order for the IAEA
to carry out its evaluation procedures effectively and efficiently, facility mea-
surement uncertainty data must be available. If a facility operator has
generated data as described in Part 3 for the Measurement Quality Element suffi-
cient to calculate Omjp as described in this part, then he will have the data
needed by the IAEA for its evaluation.

6.2 The Material Balance Equation

Closing a material balance for a time period spanned by two successive

physical inventories involves the use of the material balance equation to calcu-
late the balance or imbalance of material in terms of material unaccounted for
(MUF). This basic equation is as follows:

MUF B + I - 0 - E

where:

MUF •= Material Unaccounted For

B « Beginning Inventory

I «• Input

0 « Output

E « Ending Inventory

The four basic components of the material balance equation, i.e. inputs
and outputs and beginning and ending inventories, each consist of a hierarchy of
items, batches, and strata. An item is the primary unit which has an element
 174

weight associated with it. A number of items collectively form a batch. A

batch consists of items related by a common element factor. If an item has a
factor (i.e., measured value, for example of the element) uniquely associated
with it, then the corresponding batch consists of only that one item. A number
of batches collectively form a stratum. A stratum consists of all batches of
like material, of like sign and with the same measurement error values. Fi-
nally, appropriate strata are combined to form each of the four components of
the material balance equation. There may be several strata of items and batches
of beginning inventory and of inputs, such as receipts. These strata will have
plus signs. There also may be several strata of items and batches of outputs,
such as shipments or measured discards, and of ending inventory. These strata
will have negative signs.

As just stated, the primary unit having element weight associated with it
is the item. The element weight data is obtained from measurements made in ac-
cordance with the facility measurement system. Detailed information for
establishing a measurement system is given in Part 2 dealing with the Nuclear Ma-
terial Measurements Element of an SSAC. In addition, for each item a measure-
ment uncertainty record is prepared. This will consist of the identification of
the bulk, sampling and analysis methods, including NDA methods, used in
obtaining the item quantity and the uncertainties associated with each such
method. Detailed information for determination of the sampling and measurement
uncertainties is given in Fart 3 dealing with the Measurement Quality Element of
an SSAC. Finally, the measurement results and their uncertainties are recorded
accurately and in a way such that the proper data can be accumulated for the re-
spective components of the material-balance equation. Detailed information for
recording and reporting measurement data are given in the parts dealing with the
Nuclear Material Measurements, Measurement Quality, and Physical Inventory Tak-
ing Elements of an SSAC, as well as in the part dealing with the Records and Re-
ports Element.

Upon completion of the physical inventory taking, the physical inventory

data are sorted into the various strata and totaled to establish the value for
the ending inventory component of the equation. The data for the various items,
batches, and strata are obtained from the records to establish the o*her
commponents of the equation, i.e. beginning inventory, inputs and outputs. From
these three components the book inventory (B + I - 0 in the above equation) is
established. Adjustments to this value due to rounding differences or ship-
per-receiver differences may be needed. Adjustments may also be necessary for
conditions, such as mistakes, that may be discovered during the inventory recon-
ciliation. These adjustments are made, as appropriate, and the ending physical
inventory is compared with this adjusted book inventory to determine MCTF.
Detailed information for data handling procedures to provide the appropriate
data for use in the material balance calculation are given in Fart 4 addressing
the Records and Reports Element of an SSAC.

6.3 Evaluation of tiDF

The MUF for a given material balance period is a measure of the perfor-
mance of the facility in maintaining control of its nuclear material. In the
evaluation of MUF it is helpful to begin by making a distinction between a
"true" MUF (MUF-j) and an "observed" h'OF (MUFQ) . The observed MDF is the quan-
tity that provides an estimate of the true MUF, and is algebraically equal to
 175

the true MUF plus the combined errors due to measurement. The true MUF, on the
other hand, is the sum of amounts of material removed from the plant without
being recorded (including diversions), amounts in inventory but not measured
(hidden inventories), and amounts not accounted for because of mistakes in
recording and/or processing the data. Note that a distinction is made between
errors of measurement and mistakes.

In using M U F Q t 0 provide an estimate of MUF-j» the combined effect of the

errors of measurement is described by the standard deviation of MUF, a^jp. This
standard deviation is estimated on the basis of measurement errors, random and
systematic, which can be derived in accordance with the information included in
Fart 3 for the Measurement Quality Element of an SSAC.

In the evaluation of MUF consider first the situation in which there are
no measurement errors. In this case M U F Q and MUFj are the same, i.e. the true
MUF is exactly known. There are two questions to consider. The first is: Is
MUFf different from zero? This would indicate that some material is actually
"missing", i.e. removed from the facility or not inventoried, or there has been
a mistake in data. Note that a mistake could cause MUF^ to be less than zero,
i.e. negative. The other sources of non-zero MUF? would result in a positive
MUF T . The second question is: Is MUFf greater than some limit quantity (Q)
which has been established by the facility management or the State Authority?
This can be shown schematically with four different values of MUF^ on a real num-
ber line.

MUF, MUF, MUF, MUF,

kg D

i and MUF2 are close to zero so may not be of major concern. MUF3 is
approaching the limit quantity, Q, and MUF4 has exceeded it. Both would be of
concern. However, because the MUF is the result of a series of measurements,
the effect of measurement error on the evaluation must be considered. This can
be shown by replacing MUF2 and MUF3, for example, with an interval equal to ±
°MUF o r s o m e multiple of C^up as follows:

MUF, MUF,

± a.MUF ± a.MUF

The same two questions are asked but now with regard to the appropriate
ends of the intervals. For the MUF2 interval, because the zero value is
included there is no evidence that MUF^ is different from zero, and facility ma-
terial-accounting performance could be considered satisfactory.
 176

On the other hand, because the MUF3 interval includes the limit quantity,
Q, there is no evidence that MUF^ is different from the limit quantity although
it may exceed the limit quantity within the limits of the uncertainty interval.
There are six possible ways that the MUF uncertainty interval could be related
to 0 and Q. If U represents the upper limit of the interval and L the louer
limit, the relationships are as follows:

a) 0 < L < U < Q - The true MUF is greater than zero, but is less than the
limit quantity Q.

b) 0 < L < Q < D - The true MUF is greater than zero, but there is no evidence
that it is different from the limit quantity.
c) 0 < Q < L < U - The true MUF is greater than both zero and the limit
quantity.
d) L < O < Q < U - The uncertainty in the estimate of MUF is large and no defi-
nitive statement can be made about the true MDF relative to
zero or the limit quantity.

e) L < O < D < Q - The true MUF is less than the limit quantity but there is no
evidence that it is different from zero.

f) L < D < O < Q - The true WUF is less than both zero and the limit quantity.

The standard deviation of MtJF (tfffljp) is estimated by propagating the mea-

surement errors applicable to the measurements for all the components of the ma-
terial balance equation for a given material balance interval." If adjustments
have been made for shipper-receiver differences to either the input or output
components of the equation the uncertainties associated with these adjustments
should be included in the error propagation calculations. Such uncertainties
would need to take into account the measurement errors of both the shipper's and
receiver's measurements. There are many statistical techniques for estimating
CT
MUF* T h e m o s t practical of these for material accounting purposes are
discussed in detail in IAEA-TECDOC-261. Chapter 3 of this document, Error
Propagation, provides detailed step-by-step procedures for estimating Ojfljp and
includes several examples for different techniques. References to other
statistical methodology are also included. There would be little or no value in
repeating this information here.

A large Cjfljy will result in a wide interval around an observed MDF, thus
providing a relatively insensitive indicator as to whether or not material is ac-
tually missing. It may be worthwhile to consider establishing a maximum limit
for ajfljp. In IAEA/SG/INF/1, a table of expected values for Oyjjp for various fa-
cility types is given. These values are not presented as requirements but as
values which experience indicates are achievable in practice. For a uranium fab-
rication facility the expected Ujfljj1 is 0.3% of inventory or throughput for the
plant, whichever is larger.

6.4 Other Parameters Affecting MDF

As stated in the previous section the true MUF is the sum of amounts of ma-
terial removed from the plant without being recorded, unmeasured inventory, and
mistakes. Such things will affect the MUF but not the Cjnjj. as defined herein be-
cause CjfljF includes only uncertainties arising from the measurement process.
Thus, there are some limitations on the conclusions that can be drawn about di-
version of nuclear materials on the basis of only the MUF and Ojfljp. That is, it
does not necessarily follow that if an observed MUF differs significantly from
zero based on a hypothesis test and using the Ojjup as calculated herein, there
has been a diversion. Before such a conclusion is drawn, the possible effects
of other parameters on MUF and o^up should be taken into consideration.

Consider the parameters that may affect MÙF and/or o^uf* Th e s e are such
things as unmeasured or hidden inventories, improperly modeled measurement
biases, misstatements of error variances, and improperly recorded data.
"Unmeasured inventories will increase the size of MUF but will not affect the
calculated DjfljF1 Improperly modeled measurement biases may effect both the MUF
and dxfljp, depending on the nature of the improper modeling. For example, biases
that exist but are not corrected for, or corrections that do not properly re-
flect the actual biases, will affect the value for MUF. Misstatements of the
size of measurement error variances, either under- or overstatements, will
clearly effect Ojfljy. Finally, mistakes committed in the recording of data will
affect the MUF, but not the calculated ajnjp. Such mistakes are realistically im-
possible to eliminate completely, and are quite difficult if not impossible to
model properly. Internal controls, audits, and cross checks can assist in
detecting them. Information regarding these controls is included in the sepa-
rate parts which address data generation and handling.

One approach to MUF evaluation that has received some attention starts
with the premise that there is a tolerable level of lack of material control
that 11 not be included in the standard MUF- Otfup analysis. This tolerable
level is then modeled in some way. Such modeling may be done synthetically, by
identifying the various parameters that might contribute to MDF and/or OMUF»
assigning tolerable ranges of values to such parameters, and then combining
their effects, either analytically or possibly through simulation methods. This
does not provide a simple solution; it is difficult to be objective and realis-
tic in the modeling process. One study of this approach is discussed in NUSAC
1982.

Another approach is to evaluate past MUF data from typical plants or from
a given facility considered to have an acceptable level of control. Future MUF
performance may then be judged against this past acceptable performance. It is
difficult to be objective with this approach also. There is no simple solution
to the problem of MUF evaluation; objectivity can only proceed so far in the
evaluation process, and judgemental evaluation must complete any given
evaluation, aided by whatever modeling is available. A study which includes
this evaluation approach is discussed in NUSAC 1981.

6.5 Example of Material Balance Closing and MUF Evaluation

6.5.1 Introduction.

The facility used in this example is the conversion fabrication plant

described in Appendix A. Tables A.I and A.2 in the Appendix describe the key
measurement points for the facility, Table A.3 describes inventory material and
strata and Table 5.1 in Part 5 shows example inventory quantities.
 178

The example has been kept simple in two respects to permit easier under-
standing. Only one systematic error value is used rather than two, i.e. long
term systematic and short term systematic. This systematic error is treated as
long term systematic error in- the calculation. The example calculations are
only for the element balance and not for the isotopic. The examples given in
IAEA-TECDOC-261 show how these additional aspects would need to be handled.

6.5.2 Facility Material Balance.

Upon completion of the physical inventory taking an inventory summary is

prepared. This is shown in Table 5.2 in Part 5. While this table is a material
balance in that MUF was calculated, it is not complete in that the basis for the
book inventory is not shown. Records of beginning inventory, inputs, outputs,
and other inventory changes will be in the facility accounting records. The doc-
umentation of these data is addressed in the detailed description of the Records
and Reports Element of an SSAC. Table 6.1 summarizes the beginning inventory
and transaction data for the example material balance interval for all three
MEAs. Using these data, material balance reports for all three MBAs and for the
entire facility can be prepared. The format and detail of facility reports
would depend on the information called for by facility management and/or the
State Authority. For purposes of the example, only KBA-2 will be used because
the other two are not significant in the present case. This may not always be
the case in reality; for example, if there were a number of significant
shipper-receiver differences in MBA-1 that fact would have to be taken into ac-
count in calculating the material balance. Figure 6.1 is an example material
balance report for MSA-2 in the IAEA format. The detailed procedures and
criteria for preparation of the material balance report are addressed in the
detailed description of the Records and Reports Element of an SSAC as well as in
Code 10 of the General Part of Subsidiary Arrangements for NPT Safeguards
Agreements. These documents provide the guidance and instructions necessary
for the preparation of more complex material balance reports than that of the
example, including such things as treatment of shipper-receiver differences
and rounding adjustments.

6.5.3 Facility MUF Evaluation.

The MUF resulting from the example material balance for MBA-2 is 354 kg U.
Consider the first question of Section 6.3: is the true MUF (MUF-j) different
from zero, i.e. the null hypothesis? To answer the question Ojnjp must be
estimated. Actually the variance of MUF, (Oxj[jp) , is estimated by propagation
of the variances of the measurements used to determine the quantities in the ma-
terial balance. These data should be in the facility operating records. The
generation and documentation of these data, i.e. measurement variances, are
addressed in the descriptions of the Measurement Quality and Records and Reports
Elements of an SSAC. Table 6.2 summaries these data for the example material,
balance for MBA-2.

Using the information in Tables 5.2 and 6.1 it can be determined that for
the example material balance interval there are 23 strata. These are shown in
Table 6.3. Strata 1 through 9 are beginning inventory strata, 10 through 12 are
 179

Table 6.1. Facility Record Data Summary

Material Quantity Number Number

Item Type (Kg U) of Batches of Items

MBA-1
Beginning
Inventory OT
6 28,000 28 28
UC>2 Powder 2,000 2 80
Input ^6 98,000 98 98
UF$ Heels 641 98 98
1302 Powder 32,000 32 1280
Output 98,000 98 98
UFg Heels 641 98 98
OO2 Powder 32,000 32 1280

MBA-2
Beginning
Inventory UO2 Powder 1,000 1 40
Sintered Pellets 1,000 1 200
Dirty Powder 1,120 45 45
ADU Scrap 450 18 18
Green Scrap 450 18 18
Sintered Scrap 600 24 24
Grinder Sludge 400 80 80
Liquid Waste 200 100 100
Solid Waste 600 120 120
Input UF6 98,000 98 98
UO2 Pouder 32,000 32 1280
Reject Rods 1,300 260 260
Output Rods 125,000 125 25,000
Liquid Waste 700 350 350
Solid Waste 1,300 260 260
UF6 Heels 641 98 98

MBA-3
Beginning
Inventory Rods 60,000 60 12,000
Assemblies 30,000 60 60
Reject Rods 300 60 60
Input Rods 125,000 125 25,000
Output Reject Rods 1,300 6 260
Assemblies 123,500 247 247
 180

This page is intentionally left blank

 161
Figure 6.1 - MATERIAL BALANCE REPORT (MBR) FORM R.03
COUNTRY ^3C REPOiriNCPERIOO. F R O M 0 1 0 8 S 2 TO 3 1 0 1 8 3
FACILITY LEI! REPDRT N . . 10
MATERIAL BALANCE ARE> 2 PASEN». 1 OF 1 PAGEsJsiSNATURE
* i » 11
&
zo
i . • . . •

A C C O U N T AN C Y
1! 0 AT
!
A
IW CORRECTION
O T0
Z

ENTRY
ENTRY NAME
WEIGHT OF
WEIGHT OF 1 tu
3 FISSILE ISOTOPES D REPORT
u ELEMENT
\
(URANIUM ONLY) £ Ne.
S II)
s w £
II f°
Beeinnine Inventorv
0,1 PSDI E 5 S , 2 0.0,O'O € 1 : 5 0: 0 0 • G , :
iReceirts ;
0,2 R.D : i £ 3 .1 ,2 9-9. 9 9 1 •G O 3 9 i 9 i IGI
7

0,3 i , l| 9 , 9 . 9 9.9. 9, 9 9 ,G 3 0 0 0. 6. 0, 9 G 1
I i
r

Shipments
0,4 f 2 -5'6 4.1.0 SI ' iG 7 6 9 62 2 IGI

—
0,5 c T3 l] Q O 0 99 9 99 ,G 3 o.o 15 1 8 IG
I -

Measured Discards
Oi6 LDOI E 2 0 0 0 0 0 0 •Gl6 0 0 oo , IG!
1
ïndinc book Inventorv
:
0" B EL~IL E| '9 .4 7 8- 9i I Q •G 2 8 3 6 6 0 G 1
Ad -jnçtpd £nd*i np Book ITIV i
! i i

HiF
*Endine
JdllPhysical Inv.
IG IA ' 7 R' O; 1 n if- E 1 A *. n \c
1
I
|

'
0,9 P.E
•I E 1 9 , 1 . 2 4 8, 6 0 •C 2 7 , 4 9 9 0 Ici

liO
MB?
M. F I D z 3,5.4 0. 5, 0 , •c Is 6 7
1
o . IGI
!
•

Oi •
* In this case the adiusted
T" ending book jjnvei tory is the sane 2£ the '
(»-ifH}np_ book inventory because there wer« no shipper-receiver ] diff erej-.ces ! "
or other adjustments f o r .MBA-2 for this reporting period.
! , 1-
•

j i i

o , , ,, ; ; "

[
D , , : 1
' 1 '

a *

» H O / > * , 1 iras 77)

A-l
 182

H*ad*r Coding:

1- 2 Country Code
s- a Facility Code
9-12 MBA Cod»
13-18 Beginning Oatt of Period Covered (YYMMDO)
19-24 Ending Oatt of Ptriod Covered (YYMMDO)
25-!7 Numbtr of lint Entries, increased by 1.
28-31 For IAEA use only

70-73 Report Numbtr

74-79 Attribute (coded at the IAEA)

Codti of line Entries (to be uwd, if desired, as appropriate)

PB RA LN
RA RF LN
RF RA LD
RA RD LD
RO HA TW
RA RS TW
RS RAEU
RA RN EU
RN RAEQ
RA NP EQ
NP RATU
RA OU TU
OU RA LA
RA OQ LA
DQ RAED I for decreases in E)
RA FW ED (for decreases in E)
FW RANE (for decreases in N)
RA (for increases in D)
ED NE (for decreases in N)
ED (for increases in 0) RACE (for decreases in 0)
RA NE (for increases in E) OE (for decreases in 0)
NE (for increases in E) RAND (for dtcrttHs m N)
RA OE (for increases in E) NO (for decreases in NI
OE (for increases in E) RASE
RA NO (for increases in D) BE
NO (for increes» in Of RAOI
RASF Dl
SF RABA
RASO BA
SO RAPE
RASS PE
S3 RAMF
RA SN MF
SN
 183

Table 6.2. Measurement Error Data - MBA-2*

Error
Measurement Set Measurement
Material Type Stratum/KMP (Method No) per Batch Random Systematic

DO» Powder KMP-H Powder Scale (1) 40 20.0 g 10.0 g

KMP-4 Sample (1) 3 0.0020 0
Gravimetric (1) 3 0.0006 0.0002

Sintered KMP-J Pellet Tray Scale(2) 200 10.0 g 5.0 g

Pellets KMP-5 Sample (2) 5 0.0021 0
Gravimetric (2) 5 0.00009 0.0002

Dirty Powder KMP-L-2 Scrap Scale (3) 1 10.0 g 5.0 g

Sample (3) 1 0.0330 0
Titration (3) 1 0.0230 0.0014

ADU Scrap KMP-L-1 Scrap Scale (3) 1 10.0 g 5.0 g

Sample (4) 1 0.0540 0.0280
Titration (3) 1 0.0230 0.0014

Green Scrap KMP-L-4 Scrap Scale (3) 1 10.0 g 5.0 g

Calculation (4) 1 0.00144 0.0006

Sintered KMP-L-5 Scrap'Scale (3) 1 10.0 g 5.0 g

Scrap Sample (5) 5 0.0330 0.0190
Gravimetric (5) 5 0,00061 0.0002

Grinder KMP-L-3 Scrap Scale (3) 1 10.0 g 5.0 g

Sludge Sample (6) 1 0.0350 0.0200
Titration (6) 1 0.0300 0.0100

Liquid Waste KMP-F Drum Scale (4) 1 50.0 g 25.0 g

KMP-9 Sample (7) 1 0.0400 0.0200
Fluorimetric (7) 1 0.2510 0.0200

Solid Waste KMP-E NDA gamma (8) 1 0.1000 0.2500

KMP-9

UF6 KMP-2 DF 6 Scale (5) 1 0.5 kg 1.0 kg

Sample (8) 1 0.0003 0.0005
Gravimetric(9) 1 0.0002 0.00003

Reject Rods KMP-6 Wt from KMP-6 (6) 1 0.3 g 0.2 g

Factor From KMP-5 (10) 1 .00094 0.0002

Rods KMP-6 Pellet Stack Scale (6) 200 0.3 g 0.2 g

Factor From IMP-^ (10) 200 .00094 0.0002

•Errors are stated as relative standard deviation except where kg or g indicate

an absolute error; the errors are those for a single measurement.

A-3
 184

Table 6.3. MBA-2, Material Balance Interval Strata

Stratum Number Designation Description

KMP-fl VO2 Powder

KMP-J Sintered Pellets

KMP-L-2 Dirty Powder

KMP-L-1 ADU Scrap

KMP-L-4 Green Scrap

KMP-L-5 Sintered Scrap

KMP-L-3 Grinder Sludge

KMP-F Liquid Waste

KMP-E Solid Haste

10 KMP-2 DFg Cylinder

11 00. Powder

12 KMP-6 Reject Rods

13 KMP-6 Rods

14 KMP-8 UF, Heels

15 KMP-9 Liquid Haste

16 KMP-9 Solid Waste

17 KMP-H U0_ Powder

18 KMP-J Sintered Pellets

19 KMP-L-2 Dirty Powder

20 KMP-L-1 ADU Scrap

21 KMP-L-4 Green Scrap

22 KMP-L-5 Sintered Scrap

23 KMP-L-3 Grinder Sludge

A-4
 185

input strata, 13 through 16 are output strata, and 17 through 23 are ending
inventory strata. It is assumed that none of the material in these strata is
static. For example, the U O 2 powder on ending inventory is not that which
was on beginning inventory nor that received during the period but is that
recovered from scrap. It also is assumed that the liquid and solid wastes are
remeasured before discard thus contributing to the o MUF both as beginning
inventory items and as discard items. If this were not the case then strata 8
and 9 would be eliminated as static material and the quantities for strata IS
and 16 would be reduced by the amounts of strata 8 and 9. In a similar manner
if, for example, the sintered pellets or some of the scrap on ending inventory
is the same material as that on beginning inventory, that material, also,
would need to be eliminated from the a MUF calculations. Using the
statistical notation given in IAEA-TECDOC-261, shown in Table 6.4, and
following the examples in IAEA/TECDOC-261, the data for the material-balance
strata can be tabulated as shown in Table 6.S for easier application of the
equations for the variance calculations.

The random error variance of the total element weight of each stratum,
V r ( x ^ q p t ) , can be calculated using equation 3.A.2 in XAEA-TECDOC-261 as follows:

V (x, ) = x2 (<S2 In m + 6 2 IT m, + fi2 / c , r, n O

r kqpt kqpt r q . . k k r.p. k k r..t k k K (1)

To find the random variance of the MUF, V r (MUF), V r is summed

over all the strata:

(2)

The systematic error variance of the MDF, V g (MDF), can be calculated from
equation 3.A.11 in IAEA-TECDOC-261 as follows:

(3)
V (MDF) = I yf 62
s q q. sq. .p. s.p' ..t s . .t

(Note: This equation is given in TECDOC-261 for long-term system-

atic error. The example does not consider short-term system-
atic error.)

The statistics M _ 4 , M # p # , and M t are calculated as follows:

For each value of q, calculate

I Vkqpt
I, (A)
k=l

A-=
 186

Table 6.4. Statistical Notation from IAEA-TECDOC-261

» total element weight in stratum k, where the element weight is found '
with the use of bulk measurement method q, sampling is from material
type p, and analytical technique t is used. If measurement is by
NBA, regard the NBA instrument as an analytical method.

6 » a relative standard deviation; subscripts identify a specific one

s,g,r » first subscript on 6: s refers to a long term systematic error; g

to a short term systematic error; r to a random error.

q,p,t = second, third and fourth subscripts on 6; defined as for the

subscripts on x; if the measurement method in question is a bulk
method, replace p and t by dots; for example,

or.p. = random error standard deviation in sampling of material

type p.

n. • number of items per batch in stratum k.

» number of batches in stratum k.

"k
r. » number of samples drawn per batch in stratum k to estimate the batch
element concentration factor.

c. » number of analyses per sample in stratum k

k = total number of strata

V(...) = variance of quantity within parentheses; for example,

V(x. « » variance of element weight in stratum k;

V(MUF) = variance of MIXF

Note: s, g, and r subscripted on V are defined as above. If V has no subscript

it denotes a total variance.

A-6
 187

Table 6.5. Data Tabulation - MEA-2

Stratum (K) q P t
*kqpt (kg U)

1 40 1 3 1 î 1 1 1,000

2 200 1 5 1 2 2 2 1,000

3 1 45 1 1 3 3 3 1,120

4 1 18 1 1 3 4 3 450

5 1 18 - 1 3 - 4 450

6 1 24 1 1 3 5 5 600

7 1 80 1 1 3 6 6 400

8 1 100 1 1 4 7 7 200

9 1 120 - 1 - - 8 600

10 1 98 1 1 5 8 9 98,000

11 40 32 3 1 1 1 1 32,000

12 1 260 _ 1 6 _ 10 1.300

13 200 125 - 1 6 - 10 125,000

14 1 98 1 1 5 8 9 641

15 1 260 - 1 - - 8 1,300

16 1 350 1 1 A 7 7 700

17 40 5 3 1 1 1 1 5,000

18 200 1 5 1 2 2 2 1,000

19 1 90 1 1 3 3 3 2,250

20 1 15 1 1 3 4 3 375

21 1 20 - 1 3 - 4 500

22 1 20 1 1 3 5 5 500

23 1 100 1 1 3 6 6 500

A-7
 188

where A^ =» +1 for input and beginning inventory strata and = -1 for output
and ending inventory strata.

For each value of p, calculate

(5)
k-1

where A^ is defined as above.

For each value of t , calculate

C6)
k-1
Vkqpt

where is defined as above.

Before proceeding with the calculations it should be noted that the equa-
tions are based on relative errors, with subscripts being used to designate a
relative standard deviation. In Table 6.2, however, the errors for the bulk-
weight methods are given on an absolute rather than a relative basis. These
would need to be converted to relative values for each stratum using the rela-
tionship:

rq.
Vk rq.
'Tcqpt

where 6 is the relative standard deviation and a is in absolute units. Using

this relationship the basic equations are modified such that, effectively,
amounts of material weighed are replaced by numbers of weighings. In the random-
error equation (1) the first term,

/n.nL , is replaced by
rq.

In the systematic error equation (3) for the statistic Mq_ x

kqpt
replaced by n^m^ and ôgq in the V S (MUF) equation is replaced by n. i.e.
the square of the error in absolute terms.

Using the equation (1) given above for V r (X ) but replacing the first
component of the equation (x|q pt • 5pq../iicn^) with

A-8
 189

as discussed above we have for the random variance (r) of stratum 1, using
bulk method 1, sampling method 1, and analytical technique 1:

V
r(l) « W 2 r q . . +
^qpt '5r.p./rk\+ 4qpt * ( . t V r t
where!

according to the notation in Table 6.4:

stratum k is 1

bulk method q is 1

sampling method p is 1 and

analytical technique t is 1

and according to Table 6.5 for stratum 1:

n k = 40 ck • 1

1 x k q p t - 1000

and according to Table 6.2 for;

bulk method 1, cT(, « 0.02

sampling method 1, 6 r , p , • 0.002 and

analytical technique 1, 5 r - > t * 0.0006

Therefore:

(40)
vr(1)

1.4693 kg'U

Proceeding in the same manner for the other strata,

V
r(2) °- 904 V
r(13) ' 110
'451
V
r(3) ' 45
-107 91.316

V
r(4) " 38
'758 V 65.000
r(15)
- 0.025 V
r(16)

A-9

Table A.3 inventory Material Description

Material Item Batch Batch Inventory

Triant- ; P! /»^ Sf m bum
 190

V 16.343 7.347
r(6) r(17) *

V 4.258 V 0.904
r(7) r(18) -

V 26.090 Vr
(19) " 91.022
r(8)
V 30.000 V 32.298
r(9) r(20) *
37.240 V 0.028
'r(lO) r(21) *
47.087 V * 13.619

0.016 V ,„„* » 5.323

Note: For convenience only the stratum number has been included
in the parenthetical subscript to V. This should be
interpreted
p as beingg (x^
^ qqpp CC ) for the identified stratum,
i
The information to determine xfcqpt f any stratum is
given in Table 6.5 for all strata.

The total random variance of the MUF, V r (MDF), then is the sum of the vari-
ances of the individual strata, according to Eq. (2):

V (MUF)

Proceeding with the calculations for the systematic error the statistics
ttq..' M .p.' a a d M i i t are calculated for each value of q, p, and t, i.e. for each
bulk, sampling, and analytical method, according to Eqs. (4), (5), and (6).
Note that for the q values equation (4) has (x^qpc) replaced by (n^m^). Thus,
from Table 6.2

M, - 40 + 1280 - 200 - +1120

200 - 200
«2..
M3 = 45 + 18 + 18 + 24 + 80 - 90 - 15 - 20 - 20 - 100 - - 60

M4 = 100 - 350 = -250

Me = 98 - 98 » 0

M, - 260 - 25000 = - 24740

M 4 - 450 - 375 - +75

M 5 = 600 - 500 - +100

M , - 400 - 500 - -100

A-10
 191

M 7. 200 - 700 « -500

M o - 98,000 - 641 - 97359
« o.
M , - 1000 + 32,000 - 5,000 = +28,000
M ..2 1000 - 1000 - 0
M 3 * 1120 + 450 - 2250 - 375 - -105a
M 4 - 450 - 500 - -50

\.5 600 - 500 » +100

M , - 400 - 500 « -100
• •0

\.7 200 - 700 " -500

J 600 - 1300 « -700
..8
\.9 98,000 - 641 - 97,359
M 10 - 1300 - 125,000 - -123,700
Using these values and the systematic error values from Table 6.2 calcu-
late V s (MUP) using equation (3) given above, but replacing 6§q., with ff§q..:
(1120)2(0.01)2 125.440
2 2
+ (-60) (0.005) .090
+ (-250)2(0.025)2 39.062
+ (-24740)2(0.0002)2 24.48
+ (75)2(0.0280)2 4.41
2 2
+ (100) (0.0190) 3.61
+ (-100)2(0.0200)2 4.00
+ (-500)2(0.0200)2 100.00
+ (97,359)2(0.0005)2 2369.69
? 2
+ (28,000) (0.0002) 31.36
+ (-1055)2(0.0014)2 2.182
+ (-50)2(0.0006>2 0.C01
+ (100)2(0.0002)2 0.0004
 192

+ (-100)2(0.0100)2

+ (-500)2(0.0200)2

+ (-700)2(0.2500)2

+ (97,359) 2 (0.00003) 2

+ (-123,7OO)2(O.OOO2)2

V (MUF) = 34050.92 kg 2 U

The total variance of MDF, V(MUF) is the sum of the random and systematic
variances:

V(MUF) = 698.105 + 34050.92 - 34740.025

= /V(MUF) - /34740.025 = 186.387 kg U

The CTvjuf of 186 kg uranium for the example material balance interval is
0.146" of the output quantity for that interval. This is less than the 0.32
value given in the IAEA Safeguards Glossary as the expected value for a fabrica-
tion facility. However, if it were not, or if the facility operator decided a
smaller value was desired, he could review the various components of the calcula-
tions to find the largest contributors to Cjfljp. It is clear that the system-
atic error for the NDA measurement of solid waste is the major contributor in
this interval. Improvement in this systematic error factor, perhaps by improved
standards or calibration, would have a significant effect on Ctyup. For example,
if the systematic error were reduced to 10Z (from 25%), O M U F would change from
186 kg U to 95 kg U or 0.072 of output.

Now consider the questions discussed in section 6.3 regarding the relation
of the true MDF (MUF T ) to zero and some limit quantity. In this case the
observed MUF (MDF 0 ) is 354 kg D. Using a 952 confidence level the MUF interval
would be MtJF0 ± 1.96 Oyjj-p. In this case this would be 354 ± (1.96) (186) or 354
± 365. The MUF interval therefore would be - 11 kg to + 719 kg. If one uses
the glossary definition of a significant quantity for LEU, i.e. 75 kg contained
0-235, as the limit quantity Q it would be 2500 kg of 32 enriched uranium for
the reference facility. Thus, 0 = +719 kg, L = -11 kg and Q = 2500 kg. The
ULOQ relationship discussed in section 6.3 therefore would be in this case L <
0 < ïï < Q, which means that the true MUF is less than the limit quantity but
there is no evidence that it is different from zero.
 193

References for Part 6

IAEA/SG/INF/1: IAEA Safeguards Glossary, LAEA/SG/INF/1, International Atomic
Energy Agency, Vienna, 1980.
IAEA-TECDOC-261: IAEA Safeguards Technical Manual, Part F: Statistical
Concepts and Techniques, Vol. 3, IAEA, Vienna, 1982.
NUSAC 1981: Mathematical Derivation of the Automated Material Accounting Statis-
tics Svstem, NUSAC Report No. 591 prepared for the U.S. Nuclear Regulatory
Commission, September 15, 1981.
NUSAC 1982: F.E. Tingey, C.J. Barnhart and R.F. Lumb, Resolving Components of
Process Variability and Estimating the Uncertainty of the LEID, Task 2 Re-
port, NUSAC Report No. 752, prepared for the U.S. 1S.R.C., September 2, 1982.
 194

This page is intentionally left blank

 A-l

APPENDIX A

REFERENCE FACILITY

A.I. General Characteristics

For the purpose of preparing examples for a conversion fabrication plant,

the process at the reference facility is assumed to have the characteristics
listed below:

a) LEU annual throughput up to 300 tons TK^/year with nominal enrichment

of 3.0 weight percent of U-235.

b) Physical Inventory Taking performed by the plant operator — twice a

year with clean-outs of the process equipment.

c) The feed materials are low enriched uranium hexafluoride (UFg) and
uranyl nitrate solution (UNH) received from off site and DO2 powder
received from off site and from scrap recovery.

d) The product of the plant is Light Water Reactor (LWR) assemblies (PWR
and/or BWR type).

A.2. Process Description

A.2.1. Conversion Process

Solid 0F(j is received in large transport cylinders with

accompanying samples for analytical determination. Steam is used to sublime the
HF5. The gaseous UFg is bubbled through water to produce a DO2F2 solution. Gas-
eous ammonia (NH3) is added to this solution to precipitate ammonium diuranate
(ADD). The resulting slurry is centrifuged, dried, calcinated and reduced to
UO2, milled and blended. UNH solution received from off site is fed directly to
precipitation while UO2 powder is fed directly to the pelletizing process. Liq-
uid wastes are transferred to holding tanks where they are mixed and sampled for
analytical determination of U before discard. Clean scrap is sampled for
analytical chemistry and transferred to scrap recovery. Random containers of
blended UO2 products are thief sampled for analytical U determination. UO2 is
transferred to storage in the fabrication area.

A.2.2. Fabrication and Assembly Process

Previously analyzed UO2 powder is blended, milled, granulated,

pressed, and sintered into UO2 pellets. These are ground to dimensional toler-
ances and dried. Pellets from boats selected at random are collected for anal-
ysis. Scrap is collected in containers and sampled for chemical determination
prior to transfer to scrap recovery. Pellets are loaded into rods on stack
length basis. However, total pellet weight is recorded and used for accounta-
bility values for each individually identified rod. The rods are eventually
assembled into bundles which are the ultimate product.
 A-2

A.2.3. Solid Waste Treatment Process

Solid wastes are sorted, packaged and assayed. Burnable materials
are incinerated. The ash is nondestructively assayed and either packaged for
burial or returned to scrap recovery depending on the assay. Nonburnable mate-
rials are assayed nondestructively, compacted and packaged for burial.
A.2.4. Scrap Recycle Process
Three processes are used to handle scrap recycle. The first is
for scrap pellets. This involves milling, oxidation to U3O3, and reduction back
to UO2 powder. The second process is for clean scrap other than pellets. This
material is milled, dissolved in nitric acid and precipitated with ammonia to
produce ABU. This is filtered, dried and calcined to UO2. The third process
handles dirty scrap. Here material is oxidized, leached and dissolved in nitric
acid. The solution is purified by a solvent extraction and subsequently
stripped back into an aqueous phase. ATCT is precipitated with ammonia and this
material is converted to UO2 as described above. The UO2 from these three
processes is milled, blended and thief sampled for chemical assay prior to send-
ing it to storage.

A.3. Nuclear Material Flow and Key Measurement Points

Figure A-l is a diagram of the process showing the three MBAs, the thir-
teen flow key measurement points and the fifteen (A-0) inventory key measurement
points.
 SAMPLES MBA - 2
SHIPPER/RECEIVER
AREA SOLIDS
Uf6

LIQUID
PRECIPITATIONS. DISSOLVING

UO?
POWDER

ASSEMBLIES, ( "00 "\ /o.

\ LOADING Y*\*r

i. UNLOADING Y ^ 6 ^
MBA 3

MAIN FLOW OF NM
SCRAP
WASTE

Figure A.1 Process Scheme of LEU

Conversion and Fuel Fabrication
 Table A.l Description of Flow Key Measurement Points Shown on Figure 2.1

1 Key
| Measurement Description Form Container Sampling Methods Measurement Method*
| Point
|

1 1A UFfc flow from vendor UF6 Cylinder From Shipper b-UFg cylinder scale
recorded in MBA-1 e-Gravimetric !
i-Mass spectrometry I

! IB ONH flow from vendor UNH Tank One thief sample from b-UNH tank platform scale |
recorded in MBA-1 each tank e-Gravimetric |
i-Mass spectrometry |

1 ic UO2 flow from vendor uo 2 Buckets One thief sample from b-Powder scale |
measured in MBA-1 Powder each of 3 buckets e-Gravimetric |
from batch of 40 i-Mass spectrometry i
buckets

2 UF 6 flow to UF6 Cylinder None b-Mass difference from |

conversion measured full cylinder weight j
in MBA-2 and empty cylinder |
weight on UFg cylinder |
scale |
i&e-Factor from KMP-1A j
1
|
1 3 UNH flow to UNH Tank None b-Flow meter during |
conversion measured UNH piping j
in MBA-2 i&e-Factor from KMP-1B |
1
* b = bulk method, e = element analysis method, i = isotopic analysis method
 Table A.l Description of Flow Key Measurement Points Shown on Figure 2.1 (continued)

1
1 Key
Description Form Container Sampling Methods Measurement Method* |
| Measurement
| Point

U 0 2 flow to U02 Buckets One thief sample from b-Powder scale |

pelletizing measured Powder each of 3 buckets e-Gravimetric |
I in MBA-2 from batch of 40 i-Mass spectrometry or |
buckets Values from KMP-1C |

| 5 Sintered pellets uo 2 Tray 5 pellets/batch b-Pellet tray scale |

flow to storage Pellets selected at random e-Graviinetric j
in MBA-2

! 6 Rod Loading in Clad and Clad N/A - use pellet b-Pellet stack scale j
r
MBA-2 welded fuel factor e-Gravimetric on |
Transfer to M8A-3 rods pellets from KMP-5 j

6 Reject sintered U02 Rods N/A Values from j

pellets flow from Pellets road loading |
MBA-3 to MBA-2

7 Assembly flow from Assemblies Clad N/A - Use rod values Sum rod |
MBA-3 for from KMP-5,6 values from KMP-5,6 j
shipment 1
1
1
8 Heels measured UF6 Cylinder None b-Mass difference |
in MBA-2 e-Factor from KMP-1A |
j
* b = bulk method, e = element analysis method, i = isotopic analysis method
 Table A.l Description of Flow Key Measurement Points Shown on Figure 2.1 (continued)
1

1 Key
j Measurement Description Form Container Sampling Methods Measurement Method* |
| Point
1
1 Solid waste Miscel- Barrels N/A NDA gamma-ray count |
| 9E measured in MBA-2 laneous
I
1
| 9F Liquid wastes Liquid Drums Each drum mixed b-Drum Scale |
measured in MBA-2 Waste and sampled c-Fluorometric |

* b = bulk method, e = element analysis method, i = isotopic analysis method

I
 Table A.2 Description of Inventory Key Measurement Points Shown on Figure 2.1

Key
Measurement Description Form Container Sampling Methods Measurement Method*
Point

A UFt storage in Solid in Cylinder Item Seal Verification Value from KMP-1A
MBA-1 cylinder

B UNH Storage in UNH Tank N/A Factor from KMP-1B

MBA-1 b-UNH Tank
platform scale

C UOo Powder Storage uo 2 Buckets Item Seal Verification Values from KMP-1C
in MBA-1 Powaer

D Residual hold-up in ADU and Process N/A

UFt to UO? conver- uranium equipment Gamma-ray survey meter
sion in MBA-2 oxides

E Miscellan- Barrels N/A NDA gamma-ray count

£
eous solids
Waste storage in
MBA-2
F Liquid Drums Each drum mixed b-Drum Scale
and sampled c-Fluorometric

G Residual hold-up in Miscel- Equipment N/A Gamma-ray survey meter

waste process MBA-2 laneous

H UOo powder storage UOj powder Buckets Item Seal Verification Values from KMP-4 or
£
in MBA-2 or b-Powder scale
Thief sample for e-Gravimetric
analysis i-Mass spectrometry

I Residual hold-up in UOo powder, Equipment N/A Gamma-ray survey meter

pellet fabrication pellets

* b = bulk method, e = element analysis method, i = isotopic analysis method

 Table A.2 Description of Inventory Key Measurement Points Shown on Figure 2.1 (continued)

Key
Measurement Description Form Container Sampling Methods Measurement Method*
Point

J Sintered pellet Pellets Trays Item verification Values from KMP-7

storage in MBA-2

K Residual hold-up in ADU and Process N/A

scrap recovery in uranium equipment Gamma-ray survey meter
MBA-2 oxides

L-1 ADU Drums Each drum stirred b-Scrap scale

and sampled e-Titration

L-2 Dirty Drums Bach drum tumble b-Scrap scale

Scrap storage
In MBA-2
powder mixed and sampled e-Titration
r
oo
L-3 Grinder Bucket Each bucket stirred b-Scrap scale
sludge and sampled e-Titration

L-A Green Drums N/A b-Scrap Scale

scrap e-Calculated from UOo
and lubricant weights

L-5 Sintered Drums 5 pellets per drum b-Scrap scale

Scrap at random e-Gravimetric

H Lab inventory in Various Vials N/A Various b, e, and i

MBA-2 envelopes

N Rod inventory in Rods Clad Item verification Values from KMP-6

MBA-3

0 Assembly inventory Assemblies Clad Item verification Values from KMP-6

in MBA-3

* b = bulk method, e = element factor method, i = isotopic analysis method

 A-9

A
Table -3 Inventory Material Description

Material Item Batch Batch Inventory

Type Size Identification* Stratum
(CODE) Identity/Location
Cylinder 1 Cylinder Stamped KMP-A
(J2GR3) 1MTU Serial Number

UNH Tank 1 Tank Stamped KKP-B

(QJTVB) Serial Number
2KTU

uo 2 Bucket 40 Buckets Numbered Label KMP-C

(FOOL) on Buckets or
1MTU . KMP-H

Pellets Tray of 200 Trays Numbered Label KMP-J

(J07A) Pellets on Trays
1MTU

Fuel Rod 200 Rods Bin KMP-N

Rods Storage 1MTU Number
(DQ2F) Bin
or or or or
(DQ1P) Single Rod 1 Rod Rod Serial Number KMP-N
5 kg U

Assembly Assembly 1 Assembly Stamped KMP-0

(BQ2F) 0.5 MTU Serial Number
Solid Barrels 1 Barrel Numbered Label KMP-E
Waste 5 kg U on Barrel
(TONE)

Liquid Drum 1 Drum Drum KMP-F

Waste 2 kg U Number
(U0N3)

ADU Drum 1 Drum Drum KKP-L-1

Scrap 25 kg U Number
(RKND)

Dirty Drum 1 Drum Drum KHP-L-2

Powder 25 kg U Number
(RQNE)

Grinder Bucket 1 Bucket Bucket KMP-L-3

Sludge 5 kg U Number
(ROOD)

Green Drums 1 Drum Drum KMP-L-4

Scrap 25 kg U Number
(RQND)

Sintered Drums 1 Drum Drum KMP-L-5

Scrap 25 kg U Number
(RQND)
Samples Bottles N/A N/A KMP-M
(VOAE)
 A-10

Table A.3 Inventory Material Description (Cont'd)

*Satch Identification:
OF, Cylinder - A serial number stamped on. each cylinder by th& •manufacturer.
tJNH Tank - A serial number stamped on each tank by the manufacturer.
U0- - A three digit prefix number identifying the batch plus a 2
digit serial number identifying the bucket. 3atch number
obtained from Nuclear Material Control group.
Pellet Trays - The three digit prefix batch number from Che UO2 batch plus a
three digit serial number for the trays assigned by the Produc-
tion Unit.

Rods - The three digit prefix batch number from the DOj batch plus a
three digit serial namber assigned by the Production Unit and
stamped 00 rod.
Assembly - Serial number assigned by the Production Unit and stamped on
as sembly.
All Other - Serial number on tag or drum, bucket, or barrel assigned by
MBA with MBA number prefix.

If multiple enrichments are involved a suffix would be added so idertify

different enrichments.