ML20148T789

Forwards Notice of Issuance of Reg Guide 5.58,Summary Statement,Reg Guide 5.58, Considerations for Establishing Traceability of Special Nuclear Matl Accounting Measurements, & Value/Impact Statement

ML20148T789
Person / Time
Issue date:	11/14/1978
From:	Minogue R NRC OFFICE OF STANDARDS DEVELOPMENT
To:	Stephens C NRC OFFICE OF THE SECRETARY (SECY)
Shared Package
ML20148T793	List: ML20148T789 ML20148T833 ML20148T845
References
RTR-REGGD-05.058, RTR-REGGD-5.058, TASK-MP-801-4, TASK-OS NUDOCS 7812060080
Download: ML20148T789 (1)

Text

.

g-k-o

.. y 7 g .

3 , . .' 4 e w

• DISTRIBUTION:

LVGossick, ED0 JGDavis, IE HRDenton, NRR CVSmith, NMSS N0y i 41978 SLevine, RES RBMinogue, SD RGSmith, SD GAArlotto, SD KRGoller, SD GWRivenbark, SD RSPurple, SD RJJones, SD WBBrown, SD SPTurel, SD EHill, SD Chase R. Stephens MSparks, SD Docketing and Service Branch Office of the Secretary sch ELD REGULATORY GUIDE ,

Enclosed is a notice of issuance of Regulatory Guide 5.58 which should be published in the nctice section of.the Federal Register.

Original signed by:

ROEEM B. MINOGUE Robert B. Minogues Director l Office of Standards Development l

Enclosures:

1. Federal Register Notice i

2. Sunnary Statement

3. Regulatory Gufde 5.58 l

4. Value Impact Statement for I Regulatory Guide 5.58 a

y /

,(d 7812060o270 SD:P EHil

/ SD: ,8C GWRivenbark i

11/p'/78 SD Task No. MP 801-4 J .

ll/f/78 SD:SG6[ SQ[5GS.,BC SkORHhE,AD SD:SHSS,DIR orne. SD:00/ ~ SD R

.u.....

._.SPT[eI:ab R i![ R5kNkle KRGollerg/IP RGNh hku) g, o*v= ,. 1/3)/78 .g[j//78 /Cl((/78 /L,./f_]78 11LL/23_ .jlL3/A ..

NRC FORM 318 (9 76) NRCM 0240 'dr u. s. oovsanusNT PaiNvine orrecas sore-eae.es4 _

, L1

. v ,

7590-01 1

NUCLEAR REGULATORY COMMISSION REGULATORY GUIDE Notice of I.ssuance and Availability The Nuclear Regulatory Commission has issued a new guide in its 1 Regulatory Guide Series. This series has been developed to describe i and make available to the public methods acceptable to the NRC staff of implementing specific parts of the Commission's regulations and, in some cases, to delineate techniques used by the staff in evaluating specific problems or postulated accidents and to provide guidance to applicants concerning certain of the information needed by the staff in its review of applications for permits and licenses.

Regulatory Guide 5.58, " Considerations for Establishing Traceability of Special Nuclear Material Account'ng Measurements," presents conditions and procedural approaches acceptable to the NRC staff for establishing and maintaining traceability of special nuclear material control and acco0nting measurements. Traceability is the Ability to relate indivi-dual measurement results to national standards or nationally accepted measurement systems through an unbroken chain of comparisons. l l

Comments and suggestions in connection with (1) items for inclusion in guides. currently being developed or (2) improvements in all published l guides are encouraged at any time. Public coments on Regulatory Guide 5.58 will, howcVer, be particularly useful in evaluating the need for l an early revision if received by January 19, 1979.

Du9e_ l J g((2<t oot 7

. L.

7590-01 1

- 1 l

Coments should be sent to the Secretary of the Comission, U.S.  ;

Nuclear Regulatory Comission, Washington, D.C. 20555, Attention:

Docketing and Service Branch. l l

Regulatory guides are available for inspection at the Comission's Public Document Room, 1717 H Street NW., Washington, D.C. Requests for single copies of the latest revision of issued guides (which may be reproduced) or for placement on an automatic distribution list for I single copies of future guides in specific divisions should be made in writing to the U.S. Nuclear Regulatory Commission, Washington, D.C.

20555, Attention: Director, Division of Technical Infonnation and Docu-  ;

ment Control. Telephone requests cannot be accamodated. Regulatory guides are not copyrighted, and Comission approval is not required to reproduce them.

(5 U.S.C. 552(a))

Dated at Rockville, Marvland this 13th day of November 1978. l l

For the Nuclear Regulatory Comission.

l

/

,$ $ kwas l RobertB.Minogue,DI' rector )

Office of Standards Development l l

l 1

i

' ' ' ~ , .

. +. .

t i

SUMMARY

STATEMENT Notice Regulatory Guides - The Nuclear Regulatory Comission has issued i a new guide in Division 5, " Materials and Plant Protection," of the Regulatory Guide Series. The guide is:

Regulatory Guide 5.58 - Considerations for Establishing Traceability of Special Nuclear '

i Material Accounting Measurements l

l l

l 1

l Q= ,. l v ,

l o

v_ *e*e*

, OFFICE OF STANDARDS DEVELOPMENT REGU' LATORY GUIDE 5.58 1

CONSIDERATIONS FOR ESTABLISHING TRACEABILITY OF SPECIAL l NUCLEAR MATERIAL ACCOUNTING MEASUREMENTS A. INTRODUCTION assigned value) is known relative to national stand-ards or nationally accepted measurement systems.

Part 70, " Domestic Licensing of Spec.ial Nuclear Material," of Title 10 of the Code of Federal Regu- This guide presents conditions and procedural ap-lations requires that for approval to possess and use preaches acceptable to the NRC staff for establishing more than one effective kilogram of special nuclear and maintaining traceability of SNM c3ntrol and material (SNM)8 the licensee must provide proper accounting measurements. No specific: Methods will physical security and an adequate material control be presented herein since the methodMoghto be used and accounting system. Section 70.51, " Material Bal- for any given measurement musNhe taitokd to the ance, Inventory, and Records Requirements," re- needs and peculiarities of tinAmicvakt ^

proech mate- l quires licensees to calculate material unaccounted for rial, reference standarda$ inAunEsintation, and cir- i (MUF) and the limit of error of the MUF value cumstances. Rationale (and, pertinent' analytical fac-(LEMUF) following each physical inventory and to tors will be presantedJfVconsideration as to their compare the LEMUF with prescribed standards. Sec- applicability to the heas t at hand.

tion 70.58, " Fundamental Nuclear Material Con- 4 trols," requires licensees to maintain a program for R. OfSCUSSION the continuing determination of systematic and ran -

b

'k 9 dom measurement errors and for maintaining control ,

• 5I of such errors within prescribed limits. Section 3 niehsurements for control and accounting are l 70.57, " Measurement Control Program for Spe " ' d on a great variety of material types and

~

Nuclear Materials Control and Accounting,"{ ro- [ c" entrations, with a diversity of measurement pro-cedures, by a large number of licensees at all the vides criteria acceptable for establishing measurement and control systehW and maintaininban[i

% various industrial, research and development, and Implicit in the criteria stated in ih.57 ih the academic facilities involved. A way of 1 inking all requirement of traceability of all SNM "e$lgueT and these measurements and their uncertainties to the t

NMS.is necessary to achieve valid overall accounta-accounting measuremen/s 4the National bility. To this end, all measurement systems must be Measure-ment System (NMS) J means of reference stand- C mPatible with the NMS, and all measurement i the(nbility to relate ind -

ards. Traceability mean(ltg to nhtional standards results must orbe traceable to the appropriate national vidual measuremenhesu (Primary) reference standards or Primary Certified nationally accepted thei surgment systems through an unbrokeoqsRainq Reference Materials (PCRMs). To obtain this neces-ard mda~ts asm'ql con '

device, ~ arisons, or instrument andwhosereferencesary stand-compatibility for any given SNM measurement

$t7 task, secondary (intermediate, waking) reference 9 standards or Secondary Certified Reference Materials ,

'For definithns, see paragraphs 70.4(m) and (O of 10 CFR (SCRMs) appropriate for each SNM type and meas- 1 Part 70. urement system are nearly always required. Table 1 2The listed regulations do not apply to special nuclear defines the various types of reference materials, materials involved in the operation of a nuclear reactor. in waste disposal cpa.tions, or' as sealed sources. See paragraphs 'The term "value" includes instrumental response and other 70.51(c),70.57(b), and 70.58(a) cf 10 CFR Part 70. pertinent factors.

USNRC REGUL TORY GUIDES commnts .houso tm .ent to the secret.ry of me comm on U.S. Nuc4.,

oe. e e. de.c,.e . .n. e mm are'"7.,3-~~ ***"* x. 2mneo,, .o s

aeo.%.

method co e. .. .ccept.t.We

.e. t n o to the NR.C

,0.te tech.,

.t.f'meof implementing e.- t,, , . ofe,so.

the rh..pecific..p.,,t.e

.e eo, me ,et. no ,en t,m . n.

u-- a-

.c L'.nc.M. m.!"2.oLil'.',".*uethod e em "n"ote,e. 5 :T. .O. .,o,"e *,or,'!,.*,en.ato"'t .".

e on, o t ent ,,om the 2. as ch .no re.t ae.cio,.

1 Fuem .nd M. ten.e Facihte.

,,,ooec.

7.1,.n.oe,t. n

8. Occupetson.e H

,.m.t

, out in the ounde.

tne u. nee . cont wdl ncet>.

o, .accoot.tne

,_t . .nif they prr.vedt i,, me gnygt,.,gsg

. t e fo, the finding. g,t .~ ,e.fth .n- .me..w it "Tr,"'..'."~"

nt o '"=" '",".a.".'r=',',L,':".3"?"co a"r%T ",:=" '" T*.

• in .Decific dnn n.e,=#e ment. .nd to ..nect new ,nfo,m.tlon o, cuDene,1ce. Howeve,. Comment. On et on should t,e m.de an wnting toaegui.to the U.S. Nucie P.,1 th eve to, ty , nec. ,ho, t .nd Docu nf t,o 7 8 1 2 0 6 0 0 M _.

/' '.$

, Traceability is a property of the'overall measure- ordinarily possible, and alternative means for ment, including all Certified Reference Materials achieving traceability must be employed. This neces-

,~ (CRMs), instruments, procedures, measurement con- .sary linkage of measurement results and their uncer-

. ditions, techniques, and calculations employed, Each tainties to the NMS can be achieved by:

component of a measurement contributes to the un-i

certainty of the measurement result relative to the ' a. Periodic measurements by the licensee of SRMs

NMS. The NMS itself comprises a number of com-

' or PCRMs whose assigned values and uncertainties ponents, including Standard Reference Materials have been certified by the National Bureau of Stand.

(SRMs) or PCRMs, national laboratories, calibration ards (NBS).. These measurements may include inter- I facilities, and standard-writing groups. If the NMS is national reference materials whose assigned values

viewed as an entity capable of making measurements have been approved and accepted by the NBS. This without error, traceability can be defined as the abil- option applies only if the materials to be measured 1

ity to relate any measurement made by a local station have a substantially identical effect upon the meas.

(e.g., licensee) to the " correct" value as measured urement process as do the reference materials (RMs)

. by the NMS. If it were possible for the NMS to make or if the difference is relatively small and easily -

measurements on the same item or material as the correctable by means of the known effects of all i- '

local station, this relationship, and hence interfering parameters. Also, of course, the

' traceability, could be directly obtained. Since the urement of the RMs must be performed in a anner m, meas-NMS is largely an intangible reference system, not a identical to that employed for the SNM measurements j functioning entity, such direct comparisons are not (see Section B.3.1 of this guide).

1 Table 1 4

TYPES OF REFERENCE MATERIALS i'

l RM Type and . \

Abbreviation Definition Examples Reference Material (RM) A general term that is recom- Any or all of the materials listed mended as a substitute for that below, which previously has been re-ferred to as a standard or standard material.

Certified Reference Material (CRM) A general term for any PCRM or Any PCRM or SCRM or these SCRM or these materials as a materials as a group. See ex-group, amples below.

Primary Certified Reference A stable material characterized, Standard Reference Materials of Material (PCRM) certified, and distributed by a the National Bureau of Standards national or international standards (NBS SRMs) and Standard Mate-body. rials of the International Atomic Energy Agency (IAEA) bearing the IAEA classification, S.

Secondary Certified Reference An RM characterized against Reference Materials available Material (SCRM) PCRMs, usually by several lab- from New Brunswick Laboratory oratories. Unlike PCRMs, (NBL) or from IAEA. Those SCRMs can be typical, somewhat from the latter bear the IAEA less stable materials. classification, R.

Working Reference Material (WRM) An RM derived from CRMs or Process stream raaterials and any characterized against CRMs, used RM prepared according to this to monitor measurement methods, and related reports.

to calibrate and test methods and equipment, and to train and test personnel.

5.58-2

__ - - . - . _ . .- - - - . . - - . - - _-- =

l i bL Periodic measurements of well-characterized change or hint of change in the measurement charac.

process materials or synthesized artifacts that have teristics of the process material.

m been shown_to be substantially stable and either (a) It is doubtful that the WRMs can ever be exact

' homogeneous or (b) having small variability of representations of the material under measurement in known limits. The uncertainties (relative to the NMS) any given instance, even for highly controlled proc-associated with the values assigned to such process ess matenals, such as formed fuel pieces or uniform

. materials or artifacts are obtained by direct or indirect powdered oxide, shown to be substantially uniform in comparisons with PCRMs or NBS SRMs. both composition and measurement-affecting physical 1

characteristics (e.g., density or shape for nonde-

c. Periodic submission of samples for comparative structive assay (NDA) measurements). However, in measurement by a recognized facility having estab- most cases RMs that yield measurement uncertainties fished traceability in the measurement involved, within the selected limits for the material in question employing one or both of the above procedures, and can be achieved. Obviously, the errors resulting from

, involving only samples not subject to change in their misma'ch of the RM with the meas.aed material.will measured values during storage or transit. be leagest in heterogeneous mat.er such as waste

('?Round robin" sample exchanges between facilities materials, but in these cases the SNM concentrations can be useful in confirming or denying compatibility normally will be low and the allowable limits of of results, but such exchanges do not of themselves uncertainty correspondingly less stringent.

constitute the establishment or maintenance of The important truth being stressed here is that

- tuceability.)

every measurement must be considered, m all as-Valid assignment of an uncertainty valuc to any Pects, as an individual determination subject to error measurement result demands a thorough knowledge from a variety of sources, none of which may be of all the observed or assigned uncertainties in the safely ignored. The all-too-natural tendency to treat )

measurement system, including an understanding of successive measurements as routine must be rigor- l the nature of the sources of these uncertainties, not ously avoided. Physical RMs, in particular, tend to just a statistical measure of their existence. It is not be mistakenly accepted as true and unvarying; but sufficient, for example, to derive a root mean-square they may well be subject to changes in effective value for a succession of observed or assigned un. value (measured response), as well as unrepresenta-certainties (CRM, instrumental, and procedural) for tive of the, samples, unless wisely selected and r which standard deviation values have been calculated careful 1y handled.

1 by statistical methods for random events. To do s '

TNe characteristics required of CRMs include:

- involves assumptions as to the randomness of these variances that may not be at all valid. The variances a. Sufficiently small and known uncertainties in may, in fact, be due to a combination of systematic the assigned values. (Normally, the uncertainties of errors that appear to be randomly distributed over the the CRMs will contribute only a small fraction of the long run but that are not at all random in their total uncertainty of the measurement.)

occurrence for a g,ven i analyst employing a given b. Predictability in the response produced in the combmation of standards,' tools; and instrumental measurement process. (Ideally, the measurement components. Thus, it is necessary to derive the un- process will respond to the reference materials in the certainty value of a measurement from methods that same way as to the item or material to be measured.

also involve a summation of the nonrandom (sys* If there is a difference in measurement response to tematic) uncertainties, not from the mathematics of the measured parameter arising from other random events alone. The valid determination of the measurement affecting factors, these effects must be uncertainty of a measurement relative to the NMS, known and quantifiable.)

and thus of the degree of traceability, is not a rigorous procedure but is the result of sound judg. c. Adequate stability with tespect to all' ment based on thorough knowledge and understand. measurement-affecting characteristics of the stand-ing of allfactors involved, ard. (This is r.ecessary to avoid systematic errors due to changes in such properties as density, concentra-

, Obviously, the sources of systematic error can be tion, shape, and distribution.)

reduced if the Working Reference Materials (WRMs)

. are included at least once in every series of related d. Availability in quantities adequate for the in-tended applications.

measurements by a given analyst and combination of

. tools,1 instruments, and conditions. The calibration It cannot be assumed that RMs will always remain and correlation factors so obtained cannot be applied wholly stable' as seen by the measurement system uncritically to successive measurements. It also fol- . employed, that working RMs will forever remain lows that the applicability of any given RM to a representative of the measured material for which

• l' series of measurements of process material should be - they were prepared or selected, or that the measured examined critically both periodically and with every material itself will remain unchanged in its measure-

- (,

5.58 3

__-,__ ._t

h ment characteristics. Therefore, it is essential that material or to evaluate (and give traceability to) these RMs, as well as the measurement instrumenta- non-NBS but substantially identical material from tion and procedures, be subject to a program of which matching WRMs are then prepared. This is continuing confirmation'of traceability. Many of the necessary because of both the wide diversity of proc-factors involved in such a program are discussed in ess materials encountered and the very small number Reference 1.' and variety of SNM SRMs available. These inter-mediate RMs may be used directly as WRMs, if

2. Mass and Volume Measurements appropriate, or may be reserved for less frequent use The national systems of mass and volume meas- in the calibration of suitable synthetic or process-unements are so well established that RMs meeting material WRMs of like characteristics, as well as for the above criteria arc readily available. Where neces- verifying instrumental response factors and other as-sary, the licensee can use the RMs to calibrate pects of the measurement system. However each WRMs that more closely match the characteristics level of subsidiary RMs adds another level of uncer-of the measured materialin terms of mass, shape, and tainty to the overall uncertainty of the SNM meas-density in the case of mass measurements or are more urement, j easily adapted to the calibration of volume-measurement equipment. SRMs can also be used to " spike" process sam-pies or WRMs to determine or verify the measurabil-Specific procedures for the use of mass and volume ity of incremental changes at the working SNM level.

RMs for the calibration of measurement processes However, because of possible " threshold" or "zero

, and equipment are given in the corresponding ANSI error" effects and/or nonlinearity or irregularity of standards (Refs. 2 and 3). Factors likely to affect measurement response with concentration, this pro-uncertainty levels in inventory measurements of mass cess does not of itself establish traceability.

and volume are discussed in other regulatory guides (Refs. 4, 5, and 6), 3.2 Working Reference Materials

3. Chemical Assay and Isotopic Measurements WRMs that closely match the effective compo-Methods for chemical analysis and isotopic meas- sition of process material, or a series of such WRMs

'urement often are subject to systematic errors caused that encompass the full range of variation therein, by the presence of interfering impurities, gross dif. serve as the traceability link in most chemical analy-ferences in the concentrations of the measured com- sis and isotopic measurements. The WRMs derive ponent(s) or of measurement-affecting matrix mate- traceability through calibration relative to either rials, and other compositional factors. Traceability in SRMs or, more often, synthesized intermediate these measurements can be obtained only if such CRMs containing either SRMs or other material effects are recognized and either are eliminated by evaluated relative to the SRM (see Section B.3.1 of adjustment of the RM (or sample) composition or, in this guide),

some cases, are compensated for by secondary meas-urements of the measurement-affecting variable com- The characteristics required of a WRM are that it penent(s) and corresponding correction of the meas- be chem,cally i similar to the material to be measured ured SNM value. The latter procedure involves addi- (Including interfering substances), that it be suffi-tional sources of uncertainty and therefore should be c ently stable to have a useful lifetime, and that it employed only if it has a substantial economic or have sufficiently low uncertainty m its assigned value time advantage, if the interferences or biasing effects to meet the requirements of the measurement methods are small and limited in range, if the corrected and of the accountability limits of error.

method is reliable, and if the correction itself is WRMs can be prepared (a) from process mate-verifiable and is regularly verified. rials characteristic of the material to be measured or 3.1 National Standards - Uses and Limitations (b) by synthesis using known quantities of pure SNM. The former method offers the advantage that NBS SRMs generally are not recommended for the WRM will include all the properties that can use directly as WRMs, not only because of cost and affect the measurement such as impurities. SNM required quantities but also because of differences in concentration level, and chemical and physical form; composition (or isotopic ratios) compared to the it suffers from the disadvantage that the assigned process materials to be measured. NBS SRMs are value is determined by analyses subject to uncertain-more often used to prepare synthesized intermediate ties that must be ascertained. The latter method in-RMs of composition and form matching the process volves preparations using standard reference material (not usually economical imless small amounts are

• Regulatory guides under development on measurement con. used) or SCRMs (see Section B.3.1) with the appro-tr i Programs for SNM accounung and un considerations for determining the systemssic error and the random error of SNM Priate combination of other materials to simulate the accounting measurements will also discuss the factors involved material to be measured. The advantages of the latter in a program of continuing confirmation of traceability. method include more accurate knowledge of the SNM 5.58-4

1 content and better control of other variables such as laboratory agreement or differences, unless tracea- j the amount of impurities and the matrix composition, bility of one or more of the samples in a set has been  ;

The chief disadvantage is that the synthesized RM established as above.

may not possess all the subtle measurement-affecting 4 Nondestructive Assay characteristics of the process material. Moreover, the preparation of symhesized WRMs may be substan- Nor destructive assay (NDA) measurement  ;

tially more costly than the analysis of WRMs pre- methods are those that leave the measured material l pared from process material. Detailed procedures for unchanged (e.g., gamma emission methods) or with .

preparing plutonium and uranium WRMs are de- no significant change (e.g., neutron activation) rela- l scribed in NRC reports (Refs. 7 and 8). tive to its corresponding unmeasured state (Ref.1). l NDA offers the advantages that the same RM or the  ;

The primary concern in the use of a WRM to establish traceability in SNM measurements is the 5 me SamPl e can be measured repeatedly and yields J

valu ble data on system uncertainties not otherwise validity of the assigned value and its uncertainty. btained, that the measurement made does not con-Considerable care is necessary to ensure that the l sume process material, and that measurements can be j WRMs are prepared with a minimal increase in the made more frequently or in greater number, usually uncertainty of the assigned value above that of the SRM upon which the WRM value is based, if the at a lesser unit cost than destructive chemical methods. These advantages often yield better process ,

assigned value of a WRM is to be determined by and inventory control and enhanced statistical signifi-analysis, the use of more than one method of analysis cance in the measurement data. However, like is necessary to enhance confidence in the validity of chemical analytical methods, NDA methods have the assigned value. The methods should respond dif-many s urces of interferences that may affect the,r i ferently to impurities and to other compositional j accuracy and reliability.

variations. If the WRM has been synthesized from l standard reference material or from intermediate ref- In nearly all NDA methods,5 the integrity and I erence material, the composition and .SNM content traceability of the measurements depend on the va-can be verified by subsequent analyses. lidity of the RMs by which the NDA system is ]

The composition of a WRM can change with c librated. Calibrations generally are based on time, e.g., chang:s in oxidation state, crystalline WRMs that are or are intended to be well-form, hydration, or adsorption. These changes and Characterized and representative of the process mate-

. , l rial r items to be measured. While the matching of their effects on measurement are minimized by ap-propriate pacisaging and proper storage conditions. RMs to process items, and consequent valid tracea- l bility, is not difficult to achieve for homogeneous Additional assurance is attained by distributing pre- l measured amounts of the material into individual matenals of substantially constant composition (e.g.,

packets at the time of preparation, and these packets all ys) having fixed size and shape (e.g., machmed l can be appropriately sized so that the entire packet is Pi eces), such ideal conditions are not obtamed for j used for a single calibration or test. Even among such m st SNM measurements. Many of the materials and subsamples there may be variability in SNM content, items engountered are nonhomogeneous, noncon- l and this variability must be taken into account in f nmng in distribution, size, or shape, and highly determining the uncertainty of the assigned value. variable i,n type of material and composition. In order to ensure traceability of the measurement results to 3.3 Standard Laboratories and Sample Inter- the NMS, variations in the physical characteristics ,

change and composition of process items and in their effects l up n the response of the NDA measurement system Traceability of chemical assay and isotopic anal-must be evaluated and carefully considered m the ysis values also may be obtainable through compara-selection or design of WRMs and measurement pro-tive analyses of identical samples under parallel con-cedures (Refs. 9 and 10).

ditions. A comparative-measurement program may take either or both of two forms: }VRMs usually (a) are prepared from process ma-terials that have been characterized by measurement

a. Periodic submission of process samples for methods whose uncertainties have been ascertained analysis by a recognized facility having demonstrated relative to the NMS (i.e., are traceable) or (b) are traceability in the desired measurement. artifacts synthesized from well-characterized mate-rials t replicate the process material.* However,

b. Interfacility interchange and measurement of well-characterized and representative materials with ca ibration of the NDA method by means of suen values assigned by a facility having demonstrated traceability in the measurement,

, Absolute calorimetry of SNM of known chemical and Round-robin programs in which representative i5otopic comPosinon 18 an exception-samph:s are analyzed by a number of laboratories do *The advantages stated for similarly derived WRMs (see l not establish traccability but can only indicate inter- section a.33 also apply here.

l v.

1 S.58 5

'J

RMs does not automatically establish continuing 4.2 Characterization by a Second Method traceability of all process item measurement results obtained by that method. The effects of small varia- If the process items or materials being measured tions in the materials being assayed may lead t are subject to non-SNM variations that affect the biased results even when the WRM and the material SNM measurement, it may be possible to employ one under assay were obtained from nominally the sam or more additional methods of analysis to measure proce(s material, it therefore may be necessary either these variations and thus to characterize process ma-(a) to establish traceability of process item measure- terials in terms of such analysis results. If the sec-ment results by companng the NDA measuremen' ondary analyses also are by an NDA method, they results with those obtained by means of a reliable .nay often be performed routinely with the SNM g g gg alternative measurement system of known traceabil-ity, e.g., ondary analyses may be used to derive simple cor-(see Sect,by total dissolution ion B.4.1) and chemical or (b) to establish adequate analysis rections to the SNM measurement results. Correction sample characterization to permit the selection of a also may be obtained and traceability preserved by similarly characterized WRM for method calibration the j,udicious modification of RMs so as to incorpo-(see Section B.4.2). rate the same variable factors, i.e., so that they can produce the same relative effects in the SNM and non-SNM measurements as do the process vari-4.1 Traceability Assay by a Second Method able(s).

Alternatively, it may be advantageous to prepare Any NDA method would be of little practical WRMs that span the normal range of variability of use if every measurement also required a confirma- the measurement affecting non-SNM parameter (s) tory analysis. However, in cases in which there are a (and also the SNM-concept range, if appropriate).

number of items or material samples of established These standards can then be characterized on the similar characteristics, it is practical to establish basis of their non-SNM measurement results or of traceability for a series of measurements by means of some function (s) of SNM and non SNM measurement traceable second method evaluations of an appro- results and can be assigned a corresponding priate proportion of randomly selected samples. If the " characteristic figure." If this procedure can be car-correlation between the two methods is then found to ried out with adequate sensitivity and specificity rel-be consistent, traceability is established for all NDA ative to the interfering factors, and within acceptable measurements on that lot of SNM and on other highly limits of uncertainty, the process material can be similar material. routinely characterized in like manner and the appro-priate WRM selected on the basis of such characteri-For nominally uniform process or production zation.

material of which multiple subsamples can be ob.

tair,ed from a gross sample, the uniformity can be 5. Continuing Traceability Assurance deduced from the distribution of the NDA measure- Initial or occasional demonstration that a laboratory ,

ment data. For thus charactenzed material, traceabil- has made measurements compatible with the NMS is l ity can be established for all subs.imples that ap- not sufficient to support a claim of traceability.

proximate the mean7 from the separate traceable Measurement processes are by their nature dynamic.

second method analysis of a few of the subsamples. They are vulnerable to small changes in the skill and j Other like subsamples can then be selected as trace- care with which they are performed. Deterioration in  ;

able WRMs whose assigned values are related to the the reliability of their measurement results can be '

separately analyzed subsamples through their re- caused by (a) changes in personnel performance, (b) spective NDA measurement rcsults. deterioration in or the development of defects in For subsample populations exhibiting a range of RMs, instrumentation, or other devices, or (c) varia-NDA values, especially where a destructive tion m the environmental conditions under which the second-method analysis is used, the " twinning" measurements are performed. The techniques dis-method of sample selection may be employed. In this cussed m preceding sections ensure traceability only method, pairs of subsamples are matched by their if they are used within a continuing program of NDA measurement values, and the matches are con- measurement control.

firmed by NDA reruns. One member of each pair is evaluated by the traceable second method analysis: C. HEGULATORY POSITION the other member of that patr is then assigned the The measurement control program used by the value determined for its twin and may serve thereaf- licensee should include provisions to ensure th'at in-ter as a traceable WRM for the measurement of that dividual measurement results are traceable to the process material by that NDA method. national measurement system (NMS). RMs used to establish traceability of measurement results to the 75ubsamples whose measured values markedly deviate froni NMS should have assigned values whose uncertain-the mean ti.e., "Ayers") are not used for second method ties are known relatn e to the NMS. To meet this analysis or for WRMs. ,;ondition, the licensee should maintain a continuing 3.58-6

program for calibrating each measurement process, the WRM should not be used. Typical statistical and using RMs that meet the criteria in the following analytical procedures acceptable to the NRC staff for

~

paragraphs. preparing WRMs are found in References 7 and 8.

1. Reference Materials Storage and packaging of WRMs should follow pr cedures designed to minimize any changes likely 1.1 The National Bureau of Standards to affect the validity of the assigned values. When-Devices, instruments, and materials calibrated or ever practical, the WRM should be divided into small approved by the NBS are acceptable RMss for cali- measured quantities it the time of preparation, and the brating either methods or WRMs. However, it is very quantities shoulo 'oe of appropriate size so that each important that the licensee be able to demonstrate that entire unit is used for a single calibration or calibra-the RMs are stable under the conditions for which tion test.

they are used, that their validity has not been com- 1.2.2 Nondestructive Assay. RMs for NDA should promised, ano that they meet the accuracy require- be prepared from well-characterized materials whose ments of the mtended applications. SNM contents have been measured by methods that 1.2 Secondary Certified Reference and Work. have been calibrated with CRMs or from synthetic ing Reference Materials materials of known SNM content. The NDA RMs should closely resemble in all key characteristics the Lower-order SCRMs or WRMs that have been process items to be measured by the system. Since produced by the licensee or by a commercial supph,er destructive measurements ordinarily cannot be made are acceptable provided their uncertainties relative t on NDA RMs in order to verify makeup, as required PCRMs are known, for WRMs for chemical assay and isotopic analyses, l A statement of uncertainty should be assigned to RMs should be prepared in sets of at least three, I each RM based on an evaluation of the uncertainties using procedures that guard against errors common to of the calibration process. The statement should con. all members of the set. The consistency of the NDA tain both the standard deviation and the estimated system response to all the RMs in the set provides a l bounds of the systematic errors associated with the basis for judging the validity of the set of RMs. If l assigned value. one or more of the RMs in the set differs significantly l from the expected resporsse, no RMs from that set I

/.2.1 RMs for Chemical and Isotopic Analyses, should be used. Statistical tests for this comparison l WRMs used for calibrating chemical assay and can be found in References 7 and 8. l isotopic measurements may be prepared from stand- The design and fabrication of the RMs should f ard reference materials (SRMs) supplied by NBS or

, take into account the measurement process parame-from other well-characterized materials available t ters affecting the response of the system (Ref.1),

the mdustry. Such WRMs should be prepared under including:

conditions that ensure high reliability and should be packaged and stored in a way that eliminates any a. SNM content, potential for degradation of the WRM. b. Isotopic content,

c. Matrix material, The assigned values of WRMs prepared from d. Density, process materials should be determined by analysis, e. Container naterial and dimensions, using two different methods whenever possible. A r, self. absorption effects, and sufficient number of analyses should be done by both g. Absorption and moderation effects, methods to allow a reliable estimate of the compo-nents of random variation that affect the measure. Studies should be carried out in sufficient detail ment, if two methods are act available, as may be the to identify the process item characteristics and the case for isotopic analysis, it is recommended that a variations of the characteristics that can cause sys-verification analysis be obtained from another lab. tematic error. The results of the studies should be oratory, used to establish reasonable bounds.for the systematic errors.

If WRMs are prepared from NBS SRMs or other PCRMs, they shoeld be analyzed to verify that the NDA systems whose uncertainties relative to the makeup value is correct, i.e., that no mistakes have NMS cannot be satisfactorily established directly been made in their preparation. For this veritication, through the calibration process should be tested by at least five samples should be analyzed, using the comparative analysis. This test should be done by most reliuble method availaNe. Jhould the analytical Periodically analyzing randomly selected process results differ sigrtificandy from the makeup value, items with the NDA system in question and by another method with known uncertainty. The verifi-cation analysis can be done on samples obtained after

" International RMs and reference mater:al such as IAEA RMs reduction of the entire item to a homogeneous form.

are included, if accepted by NBS. In some cases, varification analysis b) small-sample 5.58-7

- _. ~

l NDA or by other NDA methods may be acceptable if' Table 2 l the uncertainties of the verification method are known relative to the NMS. RECERTIFICATION OR REPLACEMENT INTERVALS FOR CRMs Test Objects and Devices Ma.timum Periods

2. Measurement Assurance Mass 5 yr

,- Length 5 yr The traceability of each measurement process t

, , Volumetric Provers 5 yr

. the NMS . should be maintained by a contmumg Thermometers and program of measurement assurance. This program Thermocouples 3 yr i

. should include planned periodic verifications of the assigned values of all RMs used for calibrations. Calorimetric Standards 2 yr l

' I Certified Reference Materials \

Plutonium Metal j

2.1 Verification of Calibrations (after unpacking) 3 mo U sO s (after unpacking) 1 yr A formal program fixing the frequency at which 2.3 Interlaboratory Exchange Programs calibrations and calibration checks are performed . .

should be established. The required frequencies are The licensee should participate in mterlaboratory strongly dependent on system stability and should be exchange programs w s such programs are relevant i determined for each case by using historical perform- t the types of rr ,urements parformed m his I ance experience. Current performanc., of the meas- laboratory. The data obta.ined through this participa-urement system based on measurement control pro- ti n and other comparative measurement data (such gram data may signal the need for more frequent as shipper-receiver differences and inventory verifi-verifications. Also, the effects of changes in process cation analyses) should be used to substantiate the j

, parameters such as composition of material or mate- uncertainty statements of his meas >.rements. I rial flows should be evaluated when they occur to When significant deviations in the results of the determine the need for new calibrations, ,

comparative measurements occur, indicating lack of I consistency in measurements, the licensee should l WRMs that are subject to deterioration should be conduct an investigation. The investigation should recertified or replaced on a predetermined schedule, identify the cause of the inconsistency and, if the The frequency of recertification or replacement cause is within his organization, the licensee should should be based on pe.d mance history. If. the initiate corrective actions to remove the inconsis-integrity of an RM is in dc it must be discarded tency. The investigation may involve a reevaluation or recalibrated. of the measurement process and the CRMs to locate sources of bias or systematic error or a reevaluation of the measurement errors to determine if the stated uncertainties are correct.

2.2 Recertification or Replacement of CRMs

3. Records

• Objects, instruments, or materials calibrated by The licensee should retain all records relevant to NBS or other authoritative laboratories and used as the uncertainty of each measurement process for 5 CRMs by the licensee should be monitored by years. The records should include documents or intercomparisons with other CRM, to establish their certificates of CRMs, the measurement and statistical continued validity. In any case, tl'e values should be data used for assigning values to WRMs, and the redetermined periodically accordirg to Table 2. calibration procedures used in preparing the WRMs.

5.58-8

l REFERENCES

^ 1. Regulatory Guide' 5.11, " Nondestructive Assay erations-Systems for Measuring the Mass of of Special Nuclear Material Contained in Scrap Liquids" (1975).

)

and Waste" (1973).

7. G. C. Swanson, S. F. Marsh, J. E. Rein, G. L.

2. ANSI Standard N15.18, " Mass Calibratio9 Techniques for Nuclear Material Control,,

Tietjen, R. K. Zeigler, and G. R. Waterbury,

" Preparation of Working Calibration and Test

)

American National Standards Institute,1430 Materials-Plutonium Nitrate Solution," NRC '

Broadway, New York, New York (1975), report NUREG.0ll8 (1977)

3. ANSI Standard N15.19, " Volume Calibration l

8. S. S. Yamamura, F. W. Spraktes, J. M. Baldwin Techniques for Nuclear Material Control," R. L. Hand, R. P. Lash, and J. P. Clark, l

I American National Standards Institute,1430 Broadway, New York, New York (1975). " Preparation of Working Calibration and Test l Materials: Uranyl Nitrate Solution," NRC report i I

4. Regulatory Guide 5.25, " Design Considerations NUREG-0253 (1977).

for Minimizing Residual Holdup of Special Nu-clear Material in Equipment for Wet Process 9. ANSI Standard N15.20, " Guide to Calibrating  !

Operations" (1974). Nondestructive Assay Systems," American Na-tional Standards Institute,1430 Broadway, New

5. Regulatory Guide 5.42, " Design Considerations York, New York (1975).

for Minimizing Residual Holdup of Special Nu.

clear Material in Equipment for Dry Process - 10. Regulatory Guide 5.53, " Qualification, Calibra.

Operations (1975). tion, and Error Estimation Methods for Nondes. )

6. Regulatory Guide 5.48, " Design Consid- tructive Assay" (1975).

1 h/

i l

l 1

h l

v l l

5.58-9 l

4 UNITE D ST ATES f' 3' NUCLEAR REGULATORY COMMIS$10N Post AGE AND F EEs paso

• WASHINGTON D. C. 20555 v.s. M UC LE A R R E GU L A TO R Y OFFICI AL UUSINESS COMMisslON -

PE N ALTY FOR PRIV AYE USE $300 yug k

l l

l l

l 1

l l

l