ML19220A222

For Comment Issue of Reg Guide 5.58, Considerations for Establishing Traceability of Special Nuclear Matl Accounting Measurements

ML19220A222
Person / Time
Issue date:	11/13/1978
From:	NRC OFFICE OF STANDARDS DEVELOPMENT
To:
Shared Package
ML19220A223	List: ML19220A222 ML20148P724 ML20153A971
References
REGGD-05.058, REGGD-5.058, NUDOCS 7811290024
Download: ML19220A222 (11)

Text

f%

U.S. NUCLEAR REGULATORY COMMISSION November 1978

%.' %.. *)

REG U LAT O RY G L D E OFFICE OF STANDARDS DEVELOPMENT REGULATORY GUIDE 5.58 CONSIDER ATIONS FOR ESTABLISHING TRACEABILITY OF SPECIAL NUCLEAR MATERIAL ACCOUNTING MEASUREMENTS A. INTRODUCTION assigned value' is know n relative to national stand-Part 70 Domestic 1.icensing of Special Nuclear N1aterial. ' of Title 10 of the Code of Federal Regu-This guide presents conditions and procedural ap-lations requires that for approsal to possess and use proaches acceptable to the NRC staf f for establishing more than one ef fectne kilogram of special nuclear and maintaining traceability of SNN1 control and material (SNMP the licensee must proside proper accounting measurements. No specific methods will physical security and an adequate material control be presented herein since the methodofogy to be used and accounting o stem Nection 70 51 N1aterial flal-for any gisen measurement must be tailored to the ance, ins entory, and Records Requirements, re-needs and peculiarities of the relesant process ma:e-quires licensees to calculate material unaccounted for rial. reference standards, instrumentation, and cir-61U F) and the limit of error of the N1UF salue cumstances Rationales and pertment analytical fac-(1.ENIU F) follow mg each phy sical insentory and to tors will be pre.sented for consideration as to their compare the I.EN1UF with prescribed standards. Sec-applicabihty to the measurement at hand.

tion 70.5S.

' Fundamental Nuclear N1aterial Con-

trols, requires licensees to maintain a program for B. DISCUSSION the continuing determination of sy stematic ani ran-L Hackground dom measurement errors and for maintaining control of such errors within prescribed limits. Section SNM measurements for control and accounting are 70.57, N1easurement Control Program f or Special performed on a great sariety of material types and

% clear Staterials Control and Accounting.

pro-concentrations, with a disersity of measurement pro-s ides criteria f or establishing and mamtaming an ceduas. by a large number of licensees at all the acceptable measurement and control sy uem.2 s ariou s industrial, rescarth and deselopment, and academic f acihties in s olwd A way of Imking all Implicit in the criteria stated ja s 70.57 is the these measurements and their uncertainties to the requirement of traceability of all SNN1 control and N N15 is nCLCssJry to achies e s aliJ os erall accounta-accounting measurements to the National N1easure-bihts. Io this end, all measurement ss stems must be ment Sy stem (NN15) by means of reference stand-compatible with the N N1$.. and all measurement ards I.ranabdm means the ability to relate indi-results must be traceable to the appropriate national siJual measurement results to national standards or (primary ) ref erence standard or Primary (,crtified nationally accepted measurement ss stems through an Ref erence N1aterials (PCRN1a l.o obtain this neces-unbroken chain of comparisons, and reforna stad sary compatibility for any gnen SN.N1 measurement ard means a material. des ite, or instrument w hose task, secondary (intermediate, w orking ) re f ere nce standards or Secondary Certified Reference N1aterials

' t er atinsns. see paragraphs m 4,w and m or io ( i R (S('RN1s) appropriate f or each SNN1 type and meas-h " 7" urement sy stem are nearly alway s required Iable 1 2ihe lateJ regulanons do no' apply to spesial n ut lear defines the sarious ty pes of reterence materials.

rnmerials ins olv ed n the opermon of a rus lear reas tor, in w ete s urses

$ce paragra ph s

' T he term ' T a!ue inc ludes in s trumetual re spo n se a nJ ot her J:spmal opera

.ns or e sealcJ 31 sliel,To 53bi.and 7t> 5% ai of In O R Part To pernnent tat ters USNRC REGULATORY GUIDE S Cu

" +"4 Wa ?-1 te seM 'a **

S*'. 'ew v ("+ Cem"DW W q USNce Roy aa?ory C r m %nn W e,N ga OC ?N6 Avt < <

am1 Amwor, Gort s sa we.j fa <1ew,* c d m.o

• a.a ur a to toe poNc Sern e arc h

+P% wl% AM e% 'JI *# 10 The NRC $f,tH () :rg e"Ma r' t.r q y er 6 r pr*g q,f the 8

Comr w s em w,res to.1eeeate w hn pas e,-1 tw me svf in eva u I"e p +s m e w;,, reef m y -taul4vses dhOrJ Speau.!! f [W f 4@f91% y( { wig ?g,ldTgagj p (gjprifg gy gg {pryden ggjark p [O n( {WI.' W~T S SMUWM (regis)*S are rWpf gg{,g[itgleg f er gen.4,qf n g g gru E rym WW Od' h d 5 b

O'inM 4 I

i 08+ des P W'Th he1 ts f%C*t reeNerdNT M*91*> dis arW1 srMtm)nS tjeef er.f fer,m ew 6a*

2bNFhFd IMI bMJ Ws 7 YrFMM df

• C net %f in $p gw1M Well t$ M t.Ppf,9tW it they pegvulp & IMS s IDF f"# bf Vh' M g 3

'JPM W'r1 Ma

• 5 E F hr e%

@ Ch C u%4 NHe#

4 F "V'# OrmeN O~1 S'I#9 l 40' I 5 Di F # *

'I It*

' 85' }v tN f @ 10 T' W % drW' @ Of COF'f

• Dud T W Of 3 pearmat rJt N PD%e IW fbP (orNMgeg g b MdWW% anij FW Prc twfon d Ger +r

• 9 Comme"ts am? suwestw, for imprommeus n mese 9, des a,a erw ni r+.3,1 pmoests 6,w swa cop s of x,oe.1 emes. *N< b rwy t e +pr ao* mi' w tw BT JN I1f7 #,

df M) @,d#s W 41 IM9 rPVISMe1 @% dMWr TW.4'@

fu F Cq emf 7w date Com peg prierst on an gy 1; Mdf if dist, I uf h C 9%f 89 Unge M 4 ' +M

('t Ouf'PP -@ s F 3P5 rwr'?S arvj ta re% f syw it'f tsNtoo or e generv e n ewevo, comerver's or en q ci

• c dw woe,5 shMj be rea.M er, we f r y Ts. the U S V r< s ar Rat Ja' Je>

i If'f% Qudf E! '94 @rVp4j Wr ft 4 AthOt,I TWO rFJ@el ag*Tef,t3 rNgu.gqr p 1$,'ll bda C om r%,, eon, ys JghrM,,'r n C(

JUf M O Fppf M O Cm t. w h v %r01 ef

[)elrTK'Mstly UMP'ul O $V4%df gTQ f**@ fieenrjb it3ri RAry reyggy o T sei-Or.se g Jrd ymg N in 3rwj OE FumPPI bi CfD'I

Traceability is a property of the overall measure-ordinarily possible, and alternative means for ment, including all Certified Reference N1aterials achiesing traceability must be employed This neces-(CRN1s), instruments, procedures, measurement con-sary linkage of measurement results and their uncer-ditions, techniques, and calculations employed Each tainties to the NNIS can be achiesed by:

component of a measurement contributes to the un-certamty of the measurement result relatise to the

a. Periodic measurements by the licensee of SRN1s NN1S. The NN1S itself comprises a number of com-or PCRNis w hose assigned s alues and uncertainties ponents, including Standard Reference N1aterials base been certified by the National Bureau of Stand-(SRN1s) or PCRN1s, national laboramries, calibration ards (N BS). ~l hese measurements may include inter-f acdities, and standard-writing groups. If the NN1S is national reference materials whose assigned values viewed as an entity capable of making measurements hase been approsed and accepted by the NHS. This without error. traceability can be defined as the abil-option applies only if the materials to be measured ity to relate any measurement made by a local station base a substantially identical etteet upon the meas-(e g.

licensee) to the ' 'cocrec t ' s alue as measured urement process as do the reference materials (RN1s) by the N.NIS. It it were possible f or the NN1S to make or if the diff erence is relatis ely small and e.nily measurements on the same item or material as the correctable by means of the know n ettects of all local station, this relationship, and hence interfering parameters Also, of course. the meas-traceability, could be diiectly obtamel Since the urement of the RN1s must he performed in a manner NNIS is largely an intangible reference system, not a identical to that employed f or the SNN1 measurements functioning entity, such direct comparisons are not isee Section B1I of this guide)

Table 1 TYPES OF REFERENCE N1 ATERI ALS 101I)peand

.1bbr e t ia tion thfimtion Example s Reference N1ateria! I R N1)

A general term that is recom-Any or all of the materials listed mended as a substitute for that below.

which pres iously has been re-ferred to as a standard or standard material.

Certified Reference Ntaterial iCRN1)

A general term for any PCRN1 or Any PCRN1 or SCRN1 or these SCRN1 or these materials as a materials as a g rou p. See eu group.

amples below.

Primary Certified Reference A stable material characterized, standard Reference N1aterials of N1aterial ( PCR N1) certified, and distribuied by a the National Bureau of Standards national or international standards (NBS SRN1s) and StanJard N1 ate-body.

rials of the International Atomic Energy Agency (IAEA) bearing the IAEA cLssification S.

Secondary Lc#ied Ref erence An RN1 characterited against Reference N1aterials available N1aterial (SCR N1)

PCR N1s, usually by seseral lab-from New Brunswick Laboratory oratories. Unlike P C R N1 s,

(NHL) or from I AE A. T hose SCRN1s can be typical, somewhat from the latter bear the IAEA less stable materials.

classification, R.

Working Reference N1aterial (WRN1)

An RN1 derised f rom CRN1s or Process stream materials and any characterized against CRN1s, used RN1 prepared according to this to monitor measurement methods, and related reports.

to calibrate and test methods and equipment, and to train and test pe rson nel 144 124

@1 5 58-2

b. Periodic mea su re me n t s of w ells haracteri/ed ch.inge or hint ot' change in the measurement c harac-prot eu Ilulle rlJ ls or sy nthesi/cd artit.icts t h.it base teristics of tile prtwess ni.iterial been show n to be substantiala stable arn! either w it is doubtf ui that the WR\\1s can eser be ex.nt homocencous or a bi hasing small s arubility of representations of the material under incasurcment in know n hmits. I he uncertainties t relatn e to the N\\1S) any gn en mstance. es en f or highly umtrolled proc-associated w nh the s alues assigned to suth protess ess materulw sut h as f ormed f uel pieces or umf orm In.iterl Ils tir urtlIaCts are obblilled b) direct tir Indirect ptiw dered ()\\lde, shtiw n lti be substantiall) unitt'Inl til comparisons aith PUR\\1s or NHS SRN1s both composition arul measurement attecting phy sical characteristics (e p.

denuty or shape for nonde-Perio.hc submission of samples f or comparatise uructne away (NIM) measurements >. Ilow eser. in e

rneasurement by a recogni/ed f acjhty lusing estab-most cases RN1s that yield measurement uncertainties li hed traceability

'r the measurement insol.ed.

w ithin the selected hmits f or the outerul in question einploy me ont or buh ut the abo,e procedines, and can be athiesed ()bsiously. the erron resultmg trorn tilstilsing onl) %lniples Util suliject tan chJhge in their Intsinatch of the R\\1 w ith the riieasured iiniterial w ill IN C.l s u r e d s a l li e N d il r i rl y ' 14

• I i g e t)r t r a it s i t be !argest in hCterogCiletius nutter such as w ete i 'Round robu s.unple exchanges between t'acihties materuls. but in the se cases the SN N1 tons entrations Can be tisef ul in dinlirining 'tr Jenying aunpatibility nornnilly u ill be low.irn! the alh m able limits t il of results. but such esch mees do not of themsels es uncertamty coirespondmply less strmgent.

constitute the e s t J blis h m e n t or maintenance of tr.neabihts i l he un piu t a n t truth being stressed here i s.

that na rr nu aum nu ni mmt in a om n!< r ni. or all as Yahd assignment ot.m uncertamt> s alue to any

/>c< In a s <m on/n n/na/ Jcn r "unalmn s;Mn I to cr f or measureinent result demands a thorouch kmmledce In "" a t ar vn of s mn n conc vt uhnh mas l>c of all the obsersed or auigned untert$ unties m the w/< /\\ t em"' I T he all-too-natural tendencs to treat measurenient sy stem nMuJmg

..n understanding of sunewne measurements as routme muq be ryor-iusl) a s tilded l'hy sical R \\1s. in particular. temt tii the nature i)I the villit es tit these uncei jaintles, n< it t

iust a statistical measure -t their e mtent e it is not be nnstakenly accepted as true and urn ary mg, but NutIlctCnt, f or e\\ainple. li) derls e a riutt-1:h an-square IliC) n ul) w ell be sublCcl til changes in ettetti e s alue Itir a successisql tit ithsers ed Lit assigned un.

s alue i rneasured respinise t. as well :ts unrepresenta-(Cl!Jlnties < UR\\1 mstrumental. and proc ed ural t for tne of the samples. unless wisely selec ted and whlch standarsi des ialit)U s alues has e been calculated caref ully hamiled.

by statistital methods f or random esents T o d o '"

Ihe t haracter%in s requned of CR\\1s mc!ude-In s tils es as'unlpfltins.ls tii the randiimnew iit these sarunces that may not be at all s ahd t he s ariante' Suttioentis small and know n untertainties m a.

Illa), In tJct, bt due o' a ' t Mlibtnallt'n !

G sICnullC the assigned s alues. ( Ni rinall). lhe uncerta!ntlCs ad i

errors that appear te he randomly dhtributed our the the ('R \\1s will contribute only a small flattion of the hing run bilt that arc ned aI a!! Iaf abMll In IbClr (gital uncertainty i)! the meastis enlent. )

occm rence f or a gnen analy u employing a gnen b PreJ!ctabihty m the response proJuted in the tombmation ol standards tools and mstrume, cal measurement process. (IJeally. the me.nurement components t hus, it b u te w ary to Jern e the un-process will reg and to the relerente materuls m the

~

certainty s alue of a meeure:,ent f rom methods that sone w ay a s to the item or mater ol to be me.isured.

also unohe a summation of the nonrandom i sy s-11 there is a (htlerent e in na eurement response to te nnit lC ) unceltJ!ntie' thli Irlinl Ibe n' IIbCIIIallCs (I!

lhe nlC a s u re d l1JI a ln el C r aIising IIt'In til b C I random es ents alone / hc t < ol <!< tct "naanon el the meeur em nt -allectmp t.h tor s. these ettects nmsi be mn,it:m: o/ a me a u n "a 7" n lato, to the Y t/A-know n.. iJ quantituhle i owl ihm, >1 th, I erer

<t tw n shn. is not a Adequate stability w ith respec t to all ri eorom f o o, n n th< n u/t of w noi n. te-t nu nt /Nist t/ ( > n t /..

.l( /l d rN ' H [' tla' id 'h / 3 %[( ' i td 'h f.

Il CasurCnlent attectlrlg c harat tcrisIlcN til lhC qJm!'

in g of all ta< ton un o!t c,1 ard iIhn a ncccuary to asoid sy stematic errors due to changes m sus h properties e density, wnt entra-()bs itiusly. The viurtes tit ss s!CnnlIiC Cril'r t all bC IlOU -

AIC d I II reduced it the Workmg Ref'erent e N1aterials iWRN1s >

d. As ailability m quantities adeq ute f or the in-are included at least once m es cry series of related tend.il applicathMis me as u re me n t s bs a pnen analy st and combmation of tthils. Instruments, and Conditiolls t he calibration it cannot be anurned tlut R\\1s w ill alw as s rennon and Girrelatiidl t at itirs Mi t)hI. lined cannlit be applicJ w holly st ible e seen by 'lic riieasurcinent sy stetii uncritically to successn e measurements it aho tol-employ ed. that w or k mp h \\1s w ill I ires er rema:n hiw s that t h i. Jpplicabillt) sit an) gis en Rh1 f t) a representalis e tit the illeasuret! lilaterial liir w hlt h senes of meeurements of process nutetul should be they were prerated or selectcil, or that the measurcJ exannned cntically both periodh ally and with es ery rul}t ihwill remam unt hanged m its me e ure-

ment characteristics. Therefore, it is essential that nuterial or to es aluate (ana gis e traceability to) these RMs. as well as the measurement instrumenta-non-NBS but substantially identical material from tion and procedures, he subject to a program of which matching WRMs are ther, prepared This is continump contirmation ( ' traceability. Many of the necessary because of both the wide disenity of proc-factors insohed in such a program are discussed in en materials encountered and the sery small number Reference i f and variety of SNM SRMs asaibble. These inter mediate RMs mas be used directis as WRMs. if

2..\\ lass and Volume.\\leasurements appropriate, oc may be resersed for Iess frequent use The national sy stems of maw and solume meae in the cabbration of suitable sy nthetic or process-urenients are so well established that RMs meeting material WRMs of like characteristics, as well as for the abose criteria are readily available. Where neces-serifying instrumental response factors and other as-sary, the licensee can use the RMs to calibrate pects of the measurement sy stem. Ilo ^ eser, each WRMs that more closely match the characteristics lesel of subsidiary RMs adds another esel of unser-l of the measured material in terms of mass, shape, and tainty to the oserall uncertainty of the SNM meas-density in the case of mass measurements or are more u re men t.

easily adapted to the calibration of s olume-SRMs can also be used to " spike ~, proce4s sam-measurement equipnient.

ples or WRMs to deternune or scrit.y the measurabil-Specific procedures for the use af mass and solume ity of incremental changes at the working SNM lesel.

RMs for the cahbration of measurement processes liow es er, because of possible " threshold or 'tero and equipment are pnen in the correspondmg ANSI error' ef fects and'or nonlinearity or irregularity of standards (Refs 2 and 3). Factors hkely to affect mea su reme,t re sponse with concentration, this pro-uncertainty lesels m insentory measurements of mass ce:,s does not of itself establish traceability.

and s olume are discussed in other regulatory guides (Refs. 4. 5 and 6).

3.2 Working Reference \\laterials

3. Chemical Assay and Isotopic Measurements WRMs that closely match the ef fectise compo-Method s f or chenncal analy sis and isotopic meas-sition of process material, or a senes of such WRMs urement otten are subject to systematie errors caused that encompass the full range of variation therein, by the presence of interfering impurities, grow dif-serse as the traceability link in most themical analy-ferences in the concentrations of the measured com-sis and isotopic measurements The WRMs derive ponentN or of meeurement-attecting matrix mate-traceability through calibration relatise to eithe.

rials, and other compositional factors. Traceability in SRMs or, more otten, sy nthesized intermediate these measurements can be obtained only il such CRMs containing either SRMs or other material ettects are recognized and either are eliminated by esaluated relative to the SRM bee Section B.3.1 of adjustment of the RM lor sample) composition or, in this guide).

some cases, are compensated for by secondary meas-urements of the measurement-alfecting s ariable com.

The characteristics required of a WRM are that it be chemically similar to the material to be measured ponenu s) and corresponding correction of. the meau ured SNM value..I.he latter procedure insolses addi-(including interferine substances), that it be sut.l..i-tional sources of uncertainty and therefore should be ciently stable to hase a useful lifetime, and that it employed only if it has a substantial economic or hase suf ficiently low uncertainty in its assigned salue

~

time ads antage, if the interf erences or biasing effects to met tk requhenwnh of We meawenwnt nwtM 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 he prepared (a) from process mate-serifiable and is regularly serified.

rials characteristic of the material to be measured or qb) b) ss nthesis usine know n quantities of pure 3.1 Nat.mnal Standaids - Uses and Limitations SNM.'.Ihe former method of fers the advantage that Nh5 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 af fect 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 suf fers f rom the disads antage that the awigned procen materials to be measured. NHS SRMs are salue is determined by analyses subject to uncertain-more of ten used to prepare sy nthesized intermediate ties that must be ascertained. T he latter method in-RMs of composition and form matching the process volscs preparations using standard reference material (not usually economical unlew small amounts are

• Regul.uory guiJn unJer Jeselopmem on measuremem um-used) or SCRMs (see Section B.3.1) with the appro-trol programs for SN s1 an uuntmg aoJ on tonuder ahe ns for Uriate ombination of other materials to simulate the Jetermunne the u stemant error and the random error of S's N1 am untin[ measu'rement s will aho Jm uu the latfors n s eh eJ matertal to be measured. T he athantages of the latter m a progrhm of t'ononumg tonbrmaron of trateahihty method include more accurate knowledge of the SNM 144 126 s.ss"

content and better control of other sariables such as laboratory agreement or dif f erences, unless tracea-the amount of impurities and the matris composition.

bility of one or more of the samples in a set has been

~Ihe chief disadvantage is that the sy nthesi/ed RN1 established as abose.

may not possess all the subtle measurement-affectine 4 Nondestructne Assay charurenstics of the process material. N1orcoser. the preparation ot 9 nthesi/cJ WRNis may be substan-Nendestructis e assay < N D A, measurement tially more costly than the analp:s of W RN1s pre-methods are those that lease the measured material pared f rom procco material Detailed procedures f or unchanged te.g.

ganuna emiwion methods) or with preparing plutonium and uranium W RN1s are de-no significant change te p.

neutron actis ation) rela-scribed in NRC repora IRefs. 7 and S).

th e to its corresponthng unmeasured state t Reti I L NDA otters the adsantages th it the same RN1 or the

.l he primary concern in the use of a W RN1 to same sample can be measured repeatedly and yields establish traceability in SNN! measurements is the saluable data on 9 stem uncertainties not otherwise salidity of the assigned s alue arxl its u ncertainty.

obtained, that the measurement made does not con-C,onsiderable care is accessary to ensure that the sume process material, and that measurements can be W.R N1 s are prepared with a nunimal increase in the made niore f requently or in greater number, usually uncertaints of the assiened s alue abos e that of the at a lever unit cost than destructise chemical SRN1 upon which the W.RN1 salue is based ll the methodt.t hese ads antages of ten yield better proceu assiened s,due of a W.RN1 is to be determined bs

.tna!') sis, lhe use of Inilre th,in one meihiiJ iit analy sis and insentory control and enhanced statistical signit~i-cance ir, the measurement data. Ilow es er, like is necewars to enhante confidence in the s alidits ot mc t hids, NIM methods base

~

~

chemical a! al)tiCa!

I the assigned salue. lhe methods should respond ds inany soortes of interf erences that may attect their terentl) Itl illlpliIities and to other compositional aCcuracs afh} rCliabilits.

s ariat h ms. It the W.RN1 has been 9 ntliesi/ed trinn standard ref erence material or trom intermediate reti in nearly all NIM methods,' the iniecrity and erCnce inateria!. the conipositi ni and SNN1 content traceability of the measureriients depend

<>n the s a-can be serified by stibsequent analy ses-hdity i.t the R N1s by w hich the N!M 9 stem is calibrated. C ilibrations generally are based on t he composition of a W.RN1 tan t hange w ith W R N1s that are or are intended to be w ell-tilne, e p.

changCs in tisidJthm state, cr)stallinC char,tC:crl/Cd And represental.s e ill the process in IIC-torin, g )dratitm. I'r adsorption. T hese t hances, n o, a

rial tir items tti he incasured. W.hile the inatchine of.

thCir CIIeCts tin incasureWen! JiC InIn!ml/Cd b) ap-R N1s to procen items, and consequent s alid tracea-propriaIe packJuine anJ proper storage conJitions.

biht). is not ditticult to JChies e f or homogCnCous dilditional assur'ance is atta!ned by distrihtiting pre-materiah of substantially constant compodtion te e incasurcJ amounts of the material inti) indis idual alloy s i hn mg tised si/e and shape te p.

m at hin'ed packets at the tune of preparation. and these packets can be appropriarch sized so that the entire packet is Inew a su e

con ons are not okund W.

nbhl DN1 InCasurements N]Jn) tit the filaterid!s Jnd used f or a smgle ca'libration or test. B en among such It cIn s enstillntered are Utinh4tind)peneous. WinCtIn-subsamples there may be s ariabihty in SN NI centent, tornnng in di tribution, site, or shape, and hichh and this sarlabilil) must be (M en into Jccount m

' ~

determining the uncert.unty of the auigned s alue.

s ariable in ty pC til InJICrlal and compositiiin in i>rder to ensure tras CJbilit) ol the n eeurement results lo 3,3 Standard I ahoratories and Sample Inter-the NNIS. s ariations in the phy sical charat teristics change and composmon of process items and in their ef f ects Ljuin the respamse of the NI) A incasurenant sy stern

_I.rJs Ja!'tht) oI' t hemiCal e sa) and is,'IoplC anJI-nlust be es JlualCd anJ carettills ccmsiilered in the y sis salues aho may be obtainable through compara~

M n of WM N wmmem y tls e J lal)'es of identical samples under parallel con-g

,g y g j,3, Jithmt A compara t n e-mee u remen t procram may tJke ellher t'r bilth itI tu ti f orms

%.R N!s usually i a are prepared f rom procew ma-teriah that hac been characterized by meeurement a Pe r n h c submissun or procen samples for methoJs w hose uncertainties hae been aw ertained analy sis by a recogm/ed facihty haine demonstrated relaine to the NNIS li e.

are traceable) or ib> are trasCabkllt) In the desired incasurement.

artif acts sy nthesi/cJ f r(irn well-characterized niate-rials to repht ate the pms eu material ^ llow es er, b Interf acility interthang and meeurement of calibrathin i t the NIM iitethtid by mea.is ist s tic h w ell-t h.n as ten /ed and re presentatn e ma:enals with salues awgne i by a tasihty hn mp demonstrated traccJbilits in the measulenient

' usenoc e rm e oi u si on k m, <he ou n a Round r6,hin propr.Uns m w hh h Igresentath e N

P

**"MCe samples are analy /eJ by a number ut laboratenes do

-ru a,,, ohc

.f e i

r u usi. nee not cuahhsh traseability but can only iniht ate inter.

sc w 11 4 t a N

m

R Nis does not automatically establish continuing 4.2 Characteriration by a Second Niethm!

traceabihty of all process item measurement results It tne process items or materials being measured

~

obtamed by that method..l.he ettech of small saria-are subject to non-SN N1 s ariations that a t t'e c t the tions in the materials being assay ed may lead to biased results es en w hen the W.R \\1 anJ the materul bNhl illCJstirCinellt, it ma) bC poulb!C to einploy onC or more additional methods of analy sis to measure Llrkler Jswl) w ere tibt ained t rilin niinll'1Jll) the sinie these sariathins and thus to c haracteri/c priwess ma-process material It theref ore may be neceswiry either terials in tCIrns of slit h analy sis re s til t s 11 the see-M to estabhsh traseabihty of process item incasure-ondary analy ses alui are by an NDA methoJ. they nient results by comparing the NI) A measurement results with those obtaineil by means of a reliable

- olten be pCr!Ormed routmCl) With the SN \\1 inas rneasurements. In nuny cases. the results of sec-alternJIls e IlleasuIeme!): s) stellt ol' know n traceabil-onJary analy ses may be used to deris e simple cor-It), e.g.

b) to!JI disvilutisin ath! tht niical analy sis rections to the SN.N1 measurement resulw (,o r rec t io n

( see hectitirl l$ 4 1) tir (b ) t(i establish.iJeqtiate also may be obtained and traceability prescrs ed bs sample chara< teri/ation to permit the selection of a the iudicious modification of R\\1s so as to mcorpo-si mll.id s t haracterizeJ W R\\1 f or method cahbranon rate the wime s ariable f actors. i.e.

so ilut they can bec Section IL4_3L prottuce the winle relatis e ellet ts in the SN \\1 and non-SN \\1 measurements as do the process ' a ri-4.1 Traceability Assay by a Semnd Niethm!

ablem Altenutn ely it may be aJs antageous to prepare Any NI)A method w ould be of httle practical W R N1s that sp.m the normal range of.ariabihty ot use il es ery measurement alui required a confirma-the measurement-at tecting non-SN N1 parameteri s t tor) analy sts. klii% es er, in t asCs in W hish there are a iathi alui the SN \\1-o uis ept range, it appr(1pr iate i iluniber til ilents tir rnatCIla! sJnlplCs Lil' Cst ab!!shed IhCsC s t a n d a rd s C.i n thCH bC Charat teri/ed sin the practic.il to establish basis of their thin-SN N1 measurement results or of similar c h.irac teris tit s.

it is traiCabl!it) lor a series til Ok asurCinents b) nie ins 61l umle lunJtiom s) et bS Nl and non-SN \\l measurCment it an apprii-results and tan be assig ried a cis t r e s pi iid i n g t r.h C.ible Net tinJ riiethtid es altiatii ns i

i i

priate propiirlia m i'l ranJi ml) selec ted sJ!nples il the

' T har.h teristic fig u r e.

It this procedure can be car-i then ti und to r!ed ilul % lth aJequate sciisitis ity and speciticity rei-tiirtClatlim bCt % Cen the tu ti nlet!hkis is i

be consistent, traceahl!lt) is establish 'd ti'r all Nl) \\

athe to the mterf ering tat tors. and w ithm atteptahlJ nicJsuremt m on that lot of SN\\1 and on other hiehly hnuts of u m. e r t.u n t y. the process material can be slinilar nialertal routinC!) tharacteri/ed m like manner and tht appro-priate WR N1 selected on the basis o! suc h characteri-l'ilr ihiminall) li n i t t ir l!l pititess t ir priid ds t h in material iiI W hish nitiltiple subsamples can be ob-t.uned f rom a gross sample, the umf ornuty s an be

5. Continuing ~lraceability Assurance dedut ed titHO thC distribl! thin of the NI);\\ incasuIC-Initi il iir ( A casli nal denuinstralliin thJt a labilratiir) i nient data For thus c haracterized material. trascabil-has made measurer ants compatthle with the N\\1N is ity can be establisheJ f or all subsamples that ap-not sut t h ient to support a clann ol' traccahility.

prosinlate the incan' lrtun the separate t rat eabl;

\\leasurernent prii esses are by tlicir nature sly na'ilic t

second Inethi d analy sis ist a ist the subsan:ples l hey are s tilrierahle ti) sniali changes in t!.e skill and i

()ther like subsamples tan then bC sCICGed as Ir.h e ~

t are With w hich thes.ne pert urmeJ 1)eterioration in able W N\\ls W htise anlenCd s aluCs arc re!u cd Io thC the rehability of the!r me.tsurement FCsh!!s tJn bC separJte[) Jnal)/eJ subsainples thrin;gh their re-causeil by t a i s ha ges in per uinliel per tiirniante. (b) spet tis c NI) \\ !!icasureniCnt !Csult s deteI tiratis m in tir !!!c dC s C itipule n t lit detCcts lil 5

R N1 s, inst:umentanon. or other des tces, oric) sana-l'i)I subsan!plc populJtitins C shlbitiilg a rang.e o!

t hin in the e ns tr(in nlerl'al (.tindilitm s under Whh h the NI)A salues. especially where a dest ruc tis e pe r t on n e d. I ht ' t hmques dis measurements are second nlethiid anais sis ix used, the

.t w innin e

~

cuned in presedine set ti ms enstir cability sinly nicthi)d til saniplJ selet tiiin fila) be en pl())ed lii this il t 'le s are useJ within a ciintir 3 pri$g r a ni ol methoJ. pairs of subumples are mah hed by their measurement t ontrol N(IA inJasurenlent s alues, and the Wals hes are ain-tirmeJ by NI)\\ reruns (ine member of each p.ur

C. REGULATORY POSITION es aluated by the tr.h eable second-method ana!> sis lhe nicasurement t iin t r i >l pri graili used by tht i

the tither IOCn ber iit that pair is then assigned the Ik ensee sn $illJ insluJe pros isions tii ensure that in-s Jllie dJICr nlined liir its twin and may serse thereat-dis Idual niedsule ent iesults

e trat cahle rii the ter as a trateable W R\\1 tor the meeurement of that national IllCJ.ulenlCnt s) stCin ( N \\lh j N \\ls used Io process matenal by that NI)A meihod.

cstabb sh tr.h eabihty of incasurement results to the

+su to n whose unt eriam N \\lS should has e anlened s alues

.wh i : + wheu 2casm a m s mun,

IIC* JN

• • U IC I\\ C I

UWI I I'

tb c.i o i e

!lscrs i ars m ir eI to d: :th.

o mh u u t..r w R s t s tomhtion. the hcensee sh.iuld maintam a tontinuing

~

the WRN1 should not be used. Typical statistical and program f or cahbrating each measurement process, using RNis that meet the criteria in the following analytical procedures acceptable to the NRC statt f or paragraphs.

preparing WRNis are found in References 7 and S.

1. Reference Niaterials Storage and packaging of WRN1s should folhm rm ures pn to mininun any danges hkeh I.I The National flureau of Standards to attect the salidity of the awigned salues. When-Desices, instruments, and matenals calibrated or eser practical, the WRN1 should be disided into small approsed by the NHS are acceptable RNts* tor cali-measured quantities at the time of preparation..n. I the bratn.g eitt:er methods or WRNh. Iloweser. it is scry quantities uould be of appropriate si/e so that tach important that the licensee be able to demonstrate that entire unit is used f or a single cabbration or cahbra-the RNb are stable under the conditions f or which tion L,t.

they are used, that their s ahdity has not beca com

/

Lid i bA RW NDA & Id promise (J. and tha they meet the accuracy reamre be prepared f rom w ell-charat teri/ed materials w hose ments of the mienJed applications.

SNN1 contents base been measured by methods the I.2 Secondary Celtilled 14ference and Nrk.

hase been calibrated with CRNh or f rom synthetic materiah of know n SNN1 content. 't he NDA RNh ing Reference N1ateria;s should closely resemble in al! key characteristics the 1 ower-order SCRNh or WP.N1s that bas e been items to be measured bs the ssstem Smce proceu produced by the licensee or by a conunercial supplier destructise measurements ordinIrih cannot be made are acceptable prosided their uncertainties relatise to on NDA RNb in order to serify makeup, as required PCRN1s are know n.

for WRN1s f or chemical assay and isotopic analy ses, A statemt at of uncertainn should be asuened to RNb should be prtpared in sets of at least t h ree,

each RN1 based on an es aluation et the uncer'tainties using procedures that guard against errors common to all membet of the set. lhe consntency of the NI>A of the cahbcatmo process 'lhe statement should con.

tain both the standard desiation and the estimated sy stem respons; to all the RNh in the set prosides a bounds of the systematic errors associated with the basis f or judging the s ahdity or the set of RNh. If one or more of the RNh in the set dif f ers significantly assigned ndue.

from the espected response, no RNb f rom that set

/.2. / R.th for Chemic al and /wmpw A nah m.

should be used Statistical tests f or this companson WRNb used for calibrating chenucal away and c n be found in References 7 and S.

notopic measmements may be prepared hum stan b The dcsien and tabrication of the RNb should ard reference matenah (SRNh> supphed by NHS or take into account the measurement process parame-f rom other w ell-tharactenzed matenah as ailable t" ters af fecting the response of the sy stem (Ret 1 ).

the industry Suth W RNh should be prepared under lu h conditions that ensure high rehabdity and should be

a. SNM content, packaged and stored in a way that cluninates any b Isotopic content.

potential for degradation of the WRN1.

c Mains matenal.

The awigned salues of WRNb prepared f rom

d. Density,

procen ma:eriah should be deternuned by analysis-

e. Container material and danensions using two ditterent methods wheneser possible A

t. Self-absorpt on ef fects. and i

sutticient numbcr of analyses should be done by both

g. Absorption a.:d moderation ef fects, methods to :diow a reliable estimate of the compo-Studies shouM be carned out b suWeient detail nents of random s ariation that af fect the measure.

to dentity the proceu item characteristics and the ment. If two methods are not asailable, as mas be the s anation s of the characteristi.s that can cause sys-case for isotopic anah sis it is recommendc[1 that a teniatic error. The results of the studies should be senfication analysn be obtained from another lab-used to estabhsh reasonabfe bounds for the systematic oraton.

erro s.

If WR. b are prepared f rom NHS SRMs or other N

PCRNh. they should be analy zed to scrity that the NDA s> stems w hose uncertamties relatne to the makeup value is correct, i.e.

that no mistakes bas e NMS cannot be satisfactorily es:ablished dirr Jy been made m their preparation For this scritication, through the calibration procew should be testeu oy at least tne samples should be analy zed, using the comparatise analy sis. This test should be done by most reliable methou asadable. Should the anahtical periodically analy/ing randomly seiected process results differ significantly f rom the makeup Eatue, itenn with the NDA sy stem in question and by another method with know n uncertainty. ~l he s criti-cation analysis can be done on samples obtained af ter n ducnon of the entire item to a homogeneous f orm

" International R sts and iclerent e material sm h as I M A R' t, are miaaed d ' erred b NHs in some cast, serification analy sn by small-sample 144 129 5.58-7

NDA or by other NDA methods may be acceptable if Table 2 the uncertainties of the serification method are known relatise to the NMS.

RECERTIFICATION OR REPl.ACEMENT INTERVAES FOR CRMs l'est Objccts and Detices Sta timum l'eriods

2. Measurement Assurance Mass 5 yr Length 5 sr The traceability of each measurement process to Volumetric Prosers 5 hr the NMS should be maintained by a conunuing Thermometers and program of measurement assurance. This program should include planneJ periodic serifications of the Thermocouples 3 sr Calorimetric Standards 2 kr assigned salues of all RMs used f or calibrations.

~

Certified Reference Stateriah Plutonium Metal 2.1 Verification of Calibrations (af ter unpacking) 3 mo U aO- (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 interlaboratory stronels dependent on system stability and should be mhange programs when such programs are relesant deteriidned for each case by using historical perform _

to the ty po of measurements perfonned in his ance experience. Current performance of the meas-laboratory. The data obtained through this participa-urement sy stem based on measurement control pro-non and other comparatise measurement data auch cram data mas sienal the need for more f requent as shipper-receiser ditterences and insentory seriin erifications. Aho,' the ef fects of chances in process cation analy ses) should be used to substantiate the parameters such as composition of maierial or mate-unanainty uatements of b measurementt rial f'ow s should be evaluated when they occur to When sienificant desiations in the results of the deternune the need for new calibrations.

comparatise Incasurements occur. indicating lack of consistency in measurements. the licensee shoulJ WRMs that are subject to detenoration should be conduct an ins estication. The insestication should recerufied or replaced on a predetermined sthedule.

identify the cause of the inconsistency and, if the T he frequency of recertification or replacement cause is within his organization, the litensee shoulJ should be based on performance history. It the initiate correctise actions to remose the inconsis-integrity of an RM is in doubt. it must be disarded tency. The insestication may insohe a reesaluation or recabbrated.

of the measurement process and the CRMs to locate sources of bias or sy stematic error or a rees aluation of the n:easurement errors to determine if the stated 2.2 Recertification or Replacement of CRMs

3. Records Objec t s, in s t ru me n t s. or materiah calibrated by The licensee should rer.un all records reles ant to NPS or other authoritalise laboratories and used as the uncertainty of each measurement process for 5 CRMs by the h ce n see should be monitored by y e ars.

~l he records should include doc u ment s or intercomparisoPs with other CRMs to establish their certificates of CRMs. the meeurement and statistical coatinued s aliday. In any cee, the salues should be Jata used for assigmng s a;ues to WRMs, and the reueter,; 'ned periodic dly accordmg to lable 2.

calibranon procedures used in preparing the WRMs

-f

\\hh

\\

REFERENCES

1. Regulatory Guide 5.11.

Nondestructise Assay cration s-Sy stem s for N1easuring the N1 ass of of Special Nuclear.\\1aterul Contained in Scrap I.iquids ' ( 1975 ).

and Waste ' (1973)

7. G. C. Sw anson, S. F. \\1arsh, J. ii. Rein, G. l..

2. ANSI Standard N 15. lS.

'Ntass Calibration

l..ietien, R. h../cipler, and G. R. W.aterbun.

.I.echniques for Nuclear N1aterial (,ontrol,

,P sion of Workine Calibration and Test American National Standards Institute, 1430 N1aterials-Plutonium Nitrate Solution. NRC llroadway, New York, New York i 1975)~

report NUREG-01 IS (1977).

3. ANSI Standard N15.19 Volume Calibration Techniques for Nuclear N1aterial Control.

8. S. S. Yamamura, F. W. Spraktes, J. N1. Italdwin American National Standards Institute, 1430 R.

l..

Iland, R.

P.

1. ash, and J.

P. Clark, llroadway, New York, New York (1975).

' Preparation of Working Calibration and Test

.\\1aterials: Urans t Nitrate Solution, ' NRC report

4. Reeulatorv Guide 5.25,

' Design Considerations NUREG-025311977L for.\\linimizing Residual lloldup of Special Nu-clear N1aterial in Equipment f or Wet Process 9 ANSI Standard N15.20, GuiJe to Calibratine Ope rat ion s " i 1974 L Nondestructise Assay Sy stems.

American Na-tional Standards Institute,1430 llroadw ay, New

5. Reculaton Guide 5.42, ' Design Considerations York, New York (1975 L for.\\1inhui/ine Residual llolJup of Special Nu.

clear N1ateriai in Equipment for Dry Process

10. Regulatory G,u.de 5.53, " Qualification, Calibra-i Ope ration s ' ( 1975 ).

tion, and Error Estimation N1ethods f or Nondes-6.

Reculatory Guide 5 48,

' Design Consid-tructis e Assay " ( 1975).

\\h4 \\b\\

-7 5.58-9

UNITE D ST A TE S NUCLEAR HEGULATOHY COMM(5SION PO5 7 A t,E ANO F E E 5 P AID W A S HIN G T O*4. O C.205$5

~

v s Nuctr AH wruvun rony C OMM15% ON o"lg' h""AT15 2o t

(15 M ML h

94 02 g