LA-UR “- ---:

LA-UR--91-2461

DE91 016237

Los Alamo3 Nahonal Labofalofv IS opwmtad by lhe Unw~wiy olCmwnM for l~e United SIaIOI Oewrmentof Energy under comracl W.740S-ENG-3S

TITLE DEVELOPMENTOF NUCLEARMATERLALSACCOUNTINGFOR INTERNATIONALSAFEGUARDS:THE PAST - THE PRESENT- THE FUTJRE

AUTHOR(S) J. T. Markin, R. H. Augustson, G. W. Eccleston,and E. A. Hakkila

MASTERSUBMITTED TO, 32nd Annual Meeting of the Institute of

Nuclear Materiala Management, New Orleans,Louisiana, July 28-31, 1991

DISCLAIMER

..

..-.

,,

. .-

kos~[a~()~ LosAlamos,Los Alamos National Laboratory

,New Mexico 87545, ,. k ., .,

DEVELOPMENT OF NUCLEAR MATERIALS ACCOUNTING FORINTERNATIONAL SAFEGUARDS: THE PAST - PRESENT - FUTURE*

J. T. Martin, R. H. AugustsOn, G. W. F~cleston, and E. A. HakkilaLos Alamos National Laboratory

Los Alamos, New Mexico 87545, USA

ABSTRACT

Nuclear materials accountancy was introduced as aprimary safeguards measum in international safeguards fromthe inception of the EURATOM safeguards directorate in1959 and IAEA safeguards in 1961 with the issuance ofINFCIRC 26. As measurement technology evolved andsafeguarded facilities increased in both number and size,measurement methodology requirements increased asreflected in INFCIRC 66 (Rev 2.) in 1968 and later inINFCIRC 153 in 1972. Early measurements relied heavilyon chemical analysis, but in the 1960s it evolved more andmore toward nondestructive assay. Future nuclear materialsaccounhncy systems will increase in complexity, driven bylarger and more complex facilities; more stringent health,safety, and environmental considerations; and unattendedautomation in facility operations,

I. INTRODUCTION

Accounting for nuclear material under InternationalAtomic Energy Agency (IAEA) safeguards is specificallycalled for in the IAEA Statute, Article XII,

“send into the territory of the recipient State or Statesinspectors, designated by the Agency after ccmsulta-tion,,,, who shall have access at all times to all placesand da;a and to any person,,. as necessary to accountfor source and special fissionable materials,.. .“

Safeguards in IAEA member States arc applied undereither of two agreements, For States not signatory to theNon-Proliferation Treaty (NPT), safeguards are npplied

*This work supported by the US Dcptwtment of Encrby,Office of Safegwuds and Security,

under INFCIRC 66, originally published in 1965 andamended in 1968. During routine inspections, ‘Je inspectionactivities may include,

“49(b) Verification of the amount of safeguardednucleax material by physicid inspection, meas-urement, and sampling” and

“(c) Examination of principal nuclear facilities,including a check of their measuring instru-ments and operating characteristics.”

For States signatory to the NPT, safeguards areapplied under INFCIRC 153 as published in 1971 andamended in 1972. This document is more specific in m@r-ing measurements for the accounting of nuclear rnateriaLThe operator is required to provide a

“43(d) Description of the existing and proposed pmcedures at the facility for nuclear materialaccountancy and control, with special referenceto material balance areas established by theoperator, measurements of flow and proce-dures for physicalinventory taking.”

Furthermore, by paragrapil 46, the operator is to iden-tify material b~iance areas (MBAs) and key measurementpoints for accounting purposes, According to paragraph 55,“the system of measurements ,,. shall either conform to thelatest international standards or be equivalent in quality tosuch standards.”

The inspection activities may,

“74(b) Make independent measurements of all nuclearmaterial subject to safeguards under theAgreement,” or

“(e) Use other objective methods which have beendemonstrated to be technicxdly feasible.”

Parugruph 74(e) is ptirticultirly importunt becuuse itwlows the Agency to use newly developed technology suchIISneur-reul-time accounting (NR?A) or other techniques tobe described Ititer in this report,

The European Atomic Energy Community(EURATOM) was created in 1957 to promote the peacefuluse of nuclear energy in the member States. TheEURATOM safeguards system was created in 1959. In1973, EURATOM and the IAEA signed an agreement toimplement the NPT. Because EURATOM is the State sys-tem of accounting for nuclear material in nonweaponsnuclear facilities of member States of the Commission ofEuropean Communities, measurements are a fundamentalpart of their accounting system.

Because of the importance of materials accountancyand its required measurements, the US Department ofEnergy has supported research and development for inter-national safeguards since the early 1970s. This programalso supports bilateral activities with other IAEA memberstates in developing and testing new instruments andmethods at nuclear facilities that may not be available in theus.

Since the US Nuclear Nonproliferation Act of 1978,the US Department of State has supported a program oftechnical assistance in safeguards (POTAS) for the IAEA.This program supports the training of inspectors and thedevelopment of specific instruments and methods for IAEAinspector use.

11. DEVELOPMENT OF MEASUREMENT ANDSYSTEMS METHODOLOGY

A, Destructive Analysis

Destructive analysis has provided the basic measure-ment methods for nuclear materials since the inception ofinternational safeguards, The IAEA opened a laboratory atSeibersdorf in 1961 for research in several scientific disci-plines associated with nuclear energy. The safeguardsanalytical laboratory (SAL) was csmblished at the same sitein 1966 to analyze uranium and plutonium samples. TheIAEA sponsored a panel of experts on the topic of AnalyticalChemistry of Nuciear Fueis in Juiy 19701 and a Sympcwmon Analytical Chemistry of the Nucietir Fuel Cycle inNovember 19712 to identify possible problem areas andrecommend accepted methods.

3

The Davies-Gray methods bec~me established as themethod of choice for uranium analysis, and with variation isstill used today. Gravimetry is used for pure samples thatcan be oxidized to U3@. Electrometric techniques such aspotentiometry, amperometry, and coulometry, developedprimarily at the US, UK, and French weapons laboratories,were adapted to the commercial fuel cycle.

Mass spectmmetry has become the method of choicefor isotopic analysis of both plutonium and uranium.4 It iscapable of high precision when properly used but is limitedby throughput. Because samples must be transported to theSAL, the resin bead technique was developed to minimizesample size and problems associated with transportingnuclear materials

Recent trends in safeguards destructive analysis aredirected toward automating procedures for speed, accuracy,and decreased radiation exposure,

Portable mass spectrometers are being studied to see ifthey would enable inspectors to perform analyses on-site.The advantage of on-site analysis is gained through somesacrifice in precision because of poorer mass resolution,

An alternative to mass spectrometry is beingdeveloped-the isotope dilution gamma-ray resin bead tech-nique.~ This enables inspectors to perform analyses on-siteusing the same resin beads as are used for mass spectrom-etry. The primary advantage is the simplicity of the equip-ment compared to mass spectrometry equipment. Precisionis somewhat poorer than can be obtained at SAL, but isbetter than the precision that cm be obtained with portnblemass spectrometers, Further development of the method istaking place at Los Alamos, Tokai, and ElJRATOM.

Further developments in wet chemical analyses will bedriven by health, safety, and waste minimization concerns,The UK is developing in-line electrometric procedures thatdo not require handling the sample,

B, Nondestructive Assay (N DA) ApplicationEvolu(ion

NI’)Ainstrumcnttition using neutrons and gamma ray:;allows the inspectors to nwke mensurernents in the field,

4

The earliest such instruments emphasized portability; theinspectors carried them along as part of their luggage. Thishad the advantage that the equipment was always under thecognizance of the Agency. The measurements tended to beused for attribute checks of items and were not required to behighly accurate. As time went on, there were requests forbetter accuracy and improved opcraticmal specifications,e.g., ruggdness and reliability. Because the Agency veri-fied a wide variety of nuclear material in many physical andchemical forms, a numb of instruments were developed.To support these instruments, the Agency developed mainte-nance, shipping, and most important, training organizations.Support programs played an important role in developing theinstruments and the support functions. This continues to bethe case up to the present time and should continue in theforeseeable future.

As nuclear fuel cycle facilities evolved, so uld the NDAquipment. The portable instruments, such as the portablernultichanncl analyzer, incorporated microprocessor andCMOS battery operated technology and became morecapable while weighing less. Some larger and heavierinstruments (HLNC, AWCC, and UFBR) were stillretrofitted to make quantitative measurements on buik mate-rials. Shipping these instruments bcctune more cumbersomeand they were left more frequently at a facility and storedunder Agency seal. In a growing number of bulk facilities,permanently installed NDA systems were considered in thesame way that surveillance cameras were installed at thefacilities.

The Agency is using a mix of NDA equipment:portable, facility dedicated, and permanently installed. Mostof these instruments are connected to a computer thatcollects, organizes, and analyze~ the measurement data.Developing the software for these computer controlledinsrrumcnts has become an important activity at the Agency,

hwalled NDA equipment brings with it the need toauthenticate the operation during an inspection, Authentica-tion is the process to assure that genuine information isobtained for safeguards purposes using equipment for whichthe IAEA ~-cks sufficient control or knowlcdg~. TheAgency ht~sdeveloped and is developing the technologyused in the authcntimtion process. Often the installedcquiprncnt is operuled continuously in an untmendcd mudc

5

and the authentication then needs to be maintained continu-ously. This continuous, unattended operation has savedman y person days for facility operators and inspectors. Italso provides radiation-based monitoring data that comple-ment video surveillance information.

The accuracy of the measurement data for some kmate-rials has improved to the point that the data are :outinelyused for partial defect tests and, in some cases, are used bythe Agency data-evaluation section to calculate nuclear bal-ances and their attendant MUFS. This represents a majoradvance for NDA in the verification process. It opens thepossibility for expanded use of advanced nuclear materialaccountability systems, e.g., NRTA, in internationalsafeguards.

C. Systems Approaches

In the early years of international safcguatds, account-ing was performed conventionally. In item facilities, itemtracking was used with the goal of detecting one missingitem. In bulk handling facilities, h4UF accounting wasapplied using ihe conventional MUF equation

M~TF=I.()+~1.)71 ,

where I = input,o = output,131= beginning inventory, andE1 = ending inventory.

Acquiring these data requires a plant clcanout during aphysical inventory, In the mid seventies it became clear thatfor future large bulk handling facilities (reprocessing, mixedoxide) it would be impossible to detect loss cf significantquantities of nuclear material in a timely mannwi

10 Near=Real-Time Accounting. In the midseventies, the US DOE began funding a series cf studies atLos Alamos to apply NRTA to mixed oxide (MOX) andreprocessing facilities for domestic safeguards,7~fl Thesestudies were extended to interrmtiomd safeguards applica-tions in the late sevcnties,~

6

In 1978 the IAEA convened an advisory group meetingon safeguards for large reprocessing plants. 10 This led tothe formation of the International Working Group on Repro-cessing Plant Safeguards (IWG RPS), which met for threeyearn to discuss various approaches to safeguarding the largeplants planned for the eighties and nineties. Among itsrecommendations, the group included,l 1

“Work needs to be continued on assessing the impactof NRTA on plant design and operating procedures,”and

“New procedures and techniques for physical inven-tory determination should be investigated. Specificallyprocedures which permit the accurate measurement ofinventory quantities with minimum process shutdownand cleanout activities should be investigated.”

In the early eighties the Japanese began R&D activitieson applying NRTA at the Tokai reprocessing plant. A seriesof reports was published demonstrating the feasibility ofNRTA, with the later studies performed in conjunction withthe IAEA.12

The UK also initiated experiments in NRTA for a smallfast breeder fuel reprocessing plant and demonstrated thattimeliness and sensitivity goals could be met. 13 As a resultof these experiments and work performed by British NuclearFuels Limited, the Thorp reprocessing plant will use NRTAas a fundamental safeguards measure

Paper studies on NRI’A were performedby the FederalRepublic of Germany for the Wackersdorf facility, butstudies were concluded when the plant was cancellcd,

2. Adjusted Running Book Inventory(ARBI). Studies on ARB1 were initiated by the USNuclear Regulatory Commission,; h 1988. ARB1 can beconsidered as a form of NRTA, differing in the way in-process inventory is determined. When similar statisticaltests are applied to NRTA and ARB1 data, comparabledetection sensitivities should be achieved.

3. Cumulative FIux,14 The cumulative flux tech-nique was developed by France and extensive studies havebeen performed at the ~processing phmt at La Hague, The

7

technique differs from NRTA and ARBI in that all of the in-process inventory and its uncertainty is estimated fromprocess operating data. The method will be used at the UP-3plant at La Hague.

4. Batch Follow Up.15 Batch follow up orF BOMB was applied as a primary safeguards measure atthe ALICEM MOX facility at Hanau. Lnput batches for theprocess are planned so a measurable difference exists in theplutonium isotopic composition of successive batches. Thusfrom input and output measurements the inspector candetermine when a new batch is introduced, when the pre-vious batch is completed, and thus the inventory associatedwith each batch.

III. FUTURE DIRECTIONS

The number of facilities and the quantity of nuclearmaterials under IAEA safeguards continues to increase.More countries are opening their nuclear facilities to Agencynspections. New plants with increased capacity and signifi-

cant throughput are being constructed, Many of these facil-ities and their nuclear materials will be added to the IAEAsafeguards inspection list. In contrast, the zero growthbudget of the IAEA, emerging requirements to lower inspec-tor and facility personnel radiation doses, and automation offacilities will require that fewer Agency inspectors will beavailable per kilogram of nuclear material and their access toverify the material will become increasingly more difficultand restricted.

A. Verification Activities

Agency inspectors will continue to conduct t)hysicalinventory verifications (PiVs), interim inspections, andunplunned/ad hoc inspections at facilities under theirsurveillance. These activities will be more effective andefficient with improved, newer generation, portable NDAinstrumentation and computers that use sophisticated assayand inventory software. The current generation of NDAinstrumentation and electronics used by the [AEA, such asthe portable multichannel analyzer, will be significantlyreduced in size but have more capabilities, New detectors,such M Csl/photodiode, that are small in size and have lowpower requirements will improve the captibility of’inspectors

u

to carry their equipment and conduct ad hoc field measure-ments and attribute checks on materials in facilities.

Inspector training, instrument calibration, and meas-urement contro! will continue to be impoxtant and integralfunctions in the proper operation of the NDA measurementprocess. These functions will increase in importance as theND . instrumentation expands to large integrated systemsthat incorporate NRTA capabilities.

B. Automated and Bulk Proeesaing Facilities

Large bulk processing facilities that have highthroughput will require integrated NDA measurement sys-tems installed in-line providing continuous quantitativeinformation for NRTA of nuclear materials. In particular,automated facilities that iimit access to the process areaduring operations will require NDA instrumentation to beinstalled at the appropriate locations to provide unattendedcontinuous measurements at critical processing axeas in thefacility. For NRTA to be effective, the in-line NDA systemsinstalled thnmghout the facility, coupled with video surveil-lance, will need to be linked by secure local area networksthrough which they can continuously transmit their statusand information to the central NRTA system computer. Inthese systems, authentication and reliability will take onincreasing importance,

NRTA systems connected to continuous measurementinstrumentation will produce vast quantities of data andrequire large storage systems to reliably hold the informa-tion. The difficulty in storing and sorting this data willrequire that sophisticated data compression algorithms bedeveloped to reduce the collection of unneeded data. Thedevelopment and application of pattern recognition, artificialintelligence, and neural net software will be needed to helpinspectors review the data. In addition, the ability to com-bine data from a variety of measurement stations, to searchfor patterns in nuclear material movement, and to check theconsistency of nuclear material flows thrcugh various pointsin a process will provide increased safeguards effectiveness.This information will complement safeguards inspection dataobtained from the advanced nuclear material accountabilitysystems.

9

Continuing improvements in NDA systems, instru-mentation, and physics analysis techniques coupled to anintegrated NRTA safeguards system with sophisticatedartificial intelligence software will be required to meet thechallenges facing Agency inspections of large, automatedbulk pxucessing facilities.

C. New Safeguards Approaches

Although the IAEA in most instances has adequateresources for maintaining acceptable verification of safe-guarded materials, future increases in the size, complexity,and number of nuclear facilities comb~hmdwith limitations onincreases in inspection resources could degrade safeguardseffectiveness. Anticipating these possible conditions ofreduced effectiveness, recent studies have developed twoalternative IAEA procedures, the zone approach and ran-domized inspections, for more efficiently applying safe-guards xesources. Although these new approaches are, ingeneral, not now needed to attain current Agency safeguardsgoals, they represent potential means for addressing antici-pated shortfalls in inspection resources.

1. Zone Approach. In the zone approach, MBAsin a State’s fuel cycle that contain materials of similar safe-guards significance, i.e., material categories [e.g., low-enriched umniurn (LEU), spent fuel, or direct-use material],are combined into a single zone for purposes of clofiing amaterials balance on this larger accounting area. Althoughthe State’s System of Accounting and Control would con-tinue to report on the current MBA basis, for purposes ofsafeguards conclusions, the Agency would velify the zonebalance. Resources are saved by this approach because thezone balance eliminates the need to verify intra-zone flows ofmaterial. Instead, only flows crossing the zone boundaryand simultaneous inventories of material within the zonewou!d be verified.

These resource savings should be weighed against theloss of verified information about material flows within thezone, which implies an inability to verify the facility mate-rials balances. Instead, a positive statement abcnu the zonebalance would be tk e basis for a positive statement abouteach MBA. However, in the event of an anomaly in theTone information, localizing the armmaly to a single MBA

10

within the zone and resolving it could require more resourcesthan for anomaly resolution under the current facilityapproach.

Among the zones that could be defined according tomaterial category are (1) an LEU zone with input transfer atthe receipts area of an enrichment plant, iunoutput transferwhere fresh fuel enters the reactor core, and material inven-tories in enrichment facilities, fuel fabrication facilities, andfish fuel stores at reactors; ~q)a spent fuel zone with inputtransfer from the rezctor core to the reactor spent fuel pond,output transfer from the spent fuel pond at a reprocessingplant to the dissolution tank in the separations area, andinventories of sl t iuel in the reactor core, and spent fuelponds at the reactor and the reprocessing plant or at interimstorage facilities and (3) a plutonium zone with input trans-fer at the dissolution tank of a reprocessing plant, outputtransfer where fresh fuel is moved into a reactor core, andinventories in chemical separation and storage areas ofreprocessing plants, conversion plants, fuel fabricationplants, and fresh fuel storage at reactors.

Advances in technology for unattended measurementsat modem automated facilities may reduce the utility of thezone approach. Because materials in automated facilities aredifficult to access, in some instances, the IAEA relies on in-Iine NDA equipment to carry out flow measurements in anunattended mode. These accounting measures are comple-mented b! advanced containment/surveillance methods thatallow the IAEA to authenticate the integrity of the unattendedmeasurements. Savings in inspection effort in the zoneapproach by eliminating intra-zone flow verification can berealized by the automated verification of flows without theneed to eliminate these verification measumments.

“f’heAgency has practical experience in applying thezone approach to the natural uranium fuel cycle u:wd by theCANDU reactors in Canada where there are nearly simulta-neous PIVS. There is also some limited experience with thezone approach for the LEU fuel cycle in the Republic ofKorea.

2. Randomized Inspections. The principle ofrandom sampling of a population and extrapolation of acharacteristic of the sample to the entire population is cur-rently applied by the Agency in verifying the integrity of

11

items in a material stratum. Recent studies suggest tht thissame principle be applied to random sampling of facilities forinspection. In this instance, ii positive safeguards conclu-sion for the random] y selected facilities would be the basisfor a positive statement about all facilities in the population.In eid~ercase, for items in a stmtum or individual facilities ina group of facilities, the conclusions drawn horn the sampleare equally valid fi-oma statistical viewpoint.

3* Population of Inspection Opportunities.The first stage of a randomized inspection strategy consistsof random selection horn a population of inspection oppor-tuni~es. These opportunities could include PIV, interim, orflow verification inspections at a single facility or at multiplefacilities. Example populations include all PIV inspectionsin a State’s fuel cycle facilities, all interim inspections at aState’s reactors, or all flow verification inspmtions at asingle facility. Thus, a population of inspectionopportunities can include both spatial and temporal elemer.ts;for example, random selection of PIVS across a State’s fuelcycle could be simultaneous in time but spread acrossmultiple facilities, whereas randomized flow verification at asingle facility involves just one physical location withopportunities spread over time.

4. Inspection Timing. Choice of the time to carryout a randc ‘-J selected inspection can be based on a pre-planned scht~ . from which opportunities are selectedrandomly or cm an unannounced basis at times that areunknown to the inspected facilities, Currently, planning forinspections of facilities is based on a six-month schedule oftimes when PIV, interim, and other inspections will becarried out, This schedule is negotiated with States toaccommodate a facilities operation Sdiduk. This coordina-tion is essential in those instances such as a core change in areactor that presents the only opportunity for making aninventory of all the material at the reactor,

A schedule could be the basis of a randomized inspec-tion strategy in which opportunities are randomly selectedfrom the scheduled dates for inspection, Under this regime,the facility opemtor would know the dates of potentinlinspections but not those to be actually implemented. Thisprocedure hits the ttdvitnt;ige of ttccommoditting the opertt-tor’s need to mttke prcpmttions such M uccess to fncilityiUt!iiS, inventory listings, or retrievitl of mittcriul,

12

Alternatively, the time of the inspection could bechosen randomly with no notice to the operator. These so-called short notice random inspections (SNRI) have beensuccessfully employed for flow verifications at a fuel fabri-cation facility and for inspections to detect highly enricheduranium production in centrifuge enrichment plants. How-ever, where the inspection activities require substantialpreparations by the operator, the SNRI strategy may not befeasible.

5. Method of Randomly Selecting Inspec-tions to be Implemented. Within the framework ofeither an announced schedule of inspection opportunities oran unannounced schedule of possible inspection times (forexample, short notice random inspections), there should be amechanism for selecting some fraction of these opportunitiesfor implementation. Beginning with a specified fiction a ofthe inspection opportumties to be carried out, two methodsarc proposed for this sel=tion: (1) at the beginning of aninspection period, randomly select the required fraction afrom the inspection opportunities for implementation or (2)as the time for each inspection oppmmmity arrives, randomlydecide to implement the inspection with probability p. Thelatter method results in a fluctuation of the number ofinspections actually carried out in each inspection period.

6. Operational Consideration Related to Ran-domization. The principal operational change under aregime of randomized inspections is an increased need forconfidentiality in the inspection planning process. Becauseknowledge by a facility operator that a planned inspectionwould not be carried out invalidates its deterrent effect, con-fidentiality of the planned inspections is essential to ensurethe validity of safeguards conclusions based on randomiza-tion. The operational implementation of confidentiality ininspection planning conflicts with the need for an inspectorto arrange visas, trrvel, ship equipment, and so forth beforean inspection, Presence or absence of these activities wotdddisclose the intent with respect to the inspection. Alterntt-tives for maintaining the deterent element for inspections notcarried out are to complete till aspects of planning M if theinspection were to be done or to keep the absence of plan-ning cortfidentiul,

At some time before the dtite of a plnnned inspectionthtit is not to be implernentcd, it would be ncccssary to

inform the facility operator of that fact. To ensure that dis-closure of the inspection plan dots not invalidate the safe-guards conclusions based on random sampling, it is essentialthat the facility operator commit to a physical inventory list-ing describing the status of materials before the notificationof intent by the inspcctoratc. In practice, this could beaccomplished by telexing such a list a few days before theplanned inspection date or using a “madbox” at the facilitythat would automatically date the declaration and preventsubsequent alteration.

Where inspections am randomly selected from the six-month schedule of some fraction a of the planned inspec-tions that arc to be actually impleme?tcd and that fraction isnot confidential, completion of the fraction a before the endof the six-month period discloses the absence of inspectionsfor the remainder of that period. This deficiency is rcmcdicdby keeping the &action a confidential or by independentlydeciding with probability p whether to carry out a plannedinspection. Although the latter tactic would cause the totalfraction of inspections carried out to fluctuate around a, itavoids prematurely disclosing inspection plans.

Frequently, facilities to be inspected arc divikd intogeographically close clusters with facilities in a clusterinspected during that inspection tour. This procedure mightrestrict the benefits of randomized inspections at facilitieswithin the same cluster because the biggest resourceexpcndi:ure, inspector travel to the cluster, would not besaved, For example, if there am several facilities in a clusterand a randomized inspection plfin sclcctcd only one fncilityfor inspection, traveling to the .~luster for onc inspectionwould reduce the efficiency gained by the clusteringprinciple.

7, Surveillance. Among the inspection activitiescarried out by the Agency, surveillance is most sensitive tointerference by randomization. Indeed, for facilities such asLWRS where an inspection was not carried out, the failure toretrieve and cvakmte the surveillance record would result innot attaining the timeliness god us current’j dcftncd in theSIR critcriti. Further, because the criteria require a rcvcrifi-ctition of the inventory when surveillance is not successfulfor tiny three-month period, there would bc winspection resource requirement thtit wouldsuvings gained by mndomizing.

additionalrcducc the

4

Sunwillance could accommodate ~anuomized inspec-tions through technology advances that would cKtend thecurrent three-month maximum internal for unattended opera-tion of the surveillance device. Methods for achieving alonger surveillance period are using multiple closed-circuittelevision (Cm) units tind automatically sequencing theinitiation of recording by each unit; increasing the storagecapacity of the CCT’V devices by using optical disk storagemedia; introducing fkont-end processing of suneillance datamd selectively recoding only those scenes in which motionhas occumd; randomly increasing the interval betweenecorded images to extend the recording life of the medium,or developing units that can continue to record over previ-ously recorded images whc~~the ‘ape has been completelyused. All of these technological optit.ms allow the inspectorto randomly eliminate some inspection while maintainingcontinued smweillance until the next inspection.

REFERENCES

10

2,

3.

4.

5. .

AnalyticalChemist~ of Nuclear Fuels-Proceedingsof a Panel, Vienna (international Atomic EnergyAgency, Vienna. 1972),

AnatjticalMethodsin the Nuclear Fuel Cycle—Pro-ceedings of a Symposium, Vienna, 29 November-3 December 197/ (International Atomic EnergyAgency, Vienna, 1972).

W, Davies and W, Gray, “A Rapid and Specific Titri-metric Method for the Prt sc Determination ofUranium Using Iron (11) S phate as Reluctant,”Takznta11, 1203-1211 (1%4).

See for example, R. G. Gutmacher and C. C, Thomas,“Chemictil Methods of Analysis,” in Handbook ofNuclear Safeguards Measurement Methods, D, R,Rngers, Ed, (US Nuclear Regulatory Commission,Washington, DC, 1983), pp. 153-407,

H, Shimojima et al,, “Resin Bead Ntmpling andAnalytical Technique, “ in TASTEX, Tokat’AdwncedSq$quard$ TechnologyExercise (Intemwiontd AtomicEnergy Agency, Vienna, I!M2), pp. 163-176,

Is

7.

8.

9.

10.

110

12.

13.

14!

T. K, Li, J. L. Parker, Y. Kuno, S. Sate, and T.Akiyama, “A New Technique for Determining Pluto-nium Concentration and Isotopic Composition in Dis-solver Solutions at Reprocessing Plants,” Los AlamosNational Laboratory repotiLA-UR-91 -1165 (1991).

J. P. Shipley et al., “Coorclinated Safeguards forMaterials Management in a Mixed-Oxide Fuel Facil-ity,” Los Alamos Scientific Laboratory report LA-6536(February 1977).

E. A. Hakkila et al., “Coordinated Safeguards forMaterials Management in a Fuel Reprocessing Plant,”Los Alamos Scientific Laboratory report LA-6881(September 1977).

E. A. Hakkila et al., “Materials Management in anInternationally Safeguarded Fuels ReprocessingPlant,” Los Alamos Scientific Laboratory report LA-8042, VO1.I (April 1980).

“Advisory Group Meeting on Safeguaxhg of Repro-cessing Plants— Secretariat Working Paper,” Inter-national Atomic Energy Agency report AG- 188 (May1978).

“lntemational Working Group on Reprocessing PlantSafeguards-Overview Report to the Director Generalof the IAEA,” International Atomic Energy Agencyreport (September 1981).

M, Miura et al,, “Field Testing of Near-Real-TimeMaterials Accounting at the PNC-Tokai ReprocessingPlant,” International Atomic Energy Agency reporiSTR-202 (January 1987),

T, L. Jones and D. Gordon, “Near-Real-Time NuclearMaterials Accountancy at Dounrwiy During Reproces-singCampaign PR 7/8,” USAEA report SRDP-R. 132(November 1986),

M, Delunge, “The Cumulative Flux VerificationApproach,” Proceedings oj the Third InternutfonulCoqfermce on FacilityOperutitJn.f--S#egurdf lnter-fucc, (Americun Nucletir Society, La Gr;mge Purk,Illinois, 19W) pp. 222-229.

16

15. E. Haas and W. Hagenbcrg, “A New SafeguardsConcept at ALKEM MOX Fuel Fabrication Plant,”Proceedings of the Third International!Conference onFacility Operations-Sqfeguar& Interface, (AmericanNuclcas Society, Lz Grw?ge Park, Illinois, 1988) pp.51-56.

17