Regulatory Guide 5.21
| ML13064A082 | |
| Person / Time | |
|---|---|
| Issue date: | 04/30/1974 |
| From: | US Atomic Energy Commission (AEC) |
| To: | |
| References | |
| RG-5.021 | |
| Download: ML13064A082 (7) | |
April 1974U.S. ATOMIC ENERGY COMMISSIONREGULATORY GU I D EDIRECTORATE OF REGULATORY STANDARDSREGULATORY GUIDE 5.21NONDESTRUCTIVE URANIUM-235 ENRICHMENT ASSAYBY GAMMA-RAY SPECTROMETRY
A. INTRODUCTION
Section 70.51, "Material Balance, Inventory, andRecords Requirements," of 10 CFR Part 70, "SpecialNuclear Material," requires, in part, that licenseesauthorized to possess at any one time more than oneeffective kilogram of special nuclear material (SNM)determine the material unaccounted for (MUF) and itsassociated limit of error (LEMUF) for each element andthe fissile isotope for uranium contained in material inprocess. Such a determination is to be based onmeasurements of the quantity of the element and of thefissile isotope folr uranium.The majority of measurement techniques used inSNM accountability are specific to either the element orthe isotope but not to both. A combination oftechniques is therefore required to determine the MUFand LEMUF by element and by fissile isotope for'uranium. Passive gamma-ray spectrometry is anondestructive ýmethod for measuring the enricdment, orrelative concentration, of the fihuile isotope U-235- inuranium. As such, this technique is used in conjunctionwith an assay for the element uranium in order todetermine the amount of U-235.This guide details conditions for an acceptableU-235 enrichment measurement using gamma-rayspectrometry, and prescribes procedures for operation,calibration, error analysis, and measurement control..DISCUSSIONThe alpha decay of U-235 to Th-231 is accompaniedby the emission of a prominent gamma ray at 185.7 keV(4.3 x 104 of these 185.7-keV gamma rays are emittedper second per gram of U.235). The relatively lowenergy and consequent low penetrating power of thesegamma rays implies that most of those emitted withinthe interior of the material are absorbed within thematerial itself. These thick ' materials therefore exhibita 185.7-keV gamma ray activity which approximates theactivity characteristic of an infinite medium: i.e., theactivity does not depend on the size or dimensions ofthe .material. Under these conditions, the 185.7-keVactivity is directly proportional to the U-235enrichment. A measurement of this 185.7-keV activitywith a suitable detector forms the basis for anenrichment measurement technique.The thickness of the material with respect to themean free path of the 185.7-keV gamma ray is theprimary characteristic which determines the applicabilityof passive gamma-ray spectrometry for the measurementof isotope enrichment. The enrichment technique isapplicable only if the material is thick. However, inaddition to the thickness of the material, otherconditions must be satisfied before the gamma-rayenrichment technique can be accurately applied. Anapproximate analytical expression for the detected185.7-keV activity is given below. This expression hasbeen separated into several individual terms in order toaid in identifying those parameters which may interferewith the measurement. Although approximate, 'thisrelationship can be used to estimate the magnitude ofinterfering effects in order to establish limits on therange of applicability and to determine the associateduncertainties introduced into the measurement. Thisrelationship is:'Thick" and -thin" am used throughout this guide torefer to distances in relation to the mean free path of the 185.7keV gammn ray in the material under consideration. The meanfree path is the I/e-folding distance of the gamma-ray flux or, inother terms,'.the average distance a gamma ray traverses beforeinteracting.USAEC REGULATORY GUIDES Copies of pub" Id Sui*es my be obtained by request Indicating the divisiondesired to the US. Atomic Energy Commission., WhIngon, DZC. 20546,Regulatory Guides ae issued to describe and make maildable to the public Attention: Director of Regulatory Steondaerd. Comments snd suggetions formnthods acceptable to the AEC Regulatory staff of Implementing speciffic pats of Imlwovaetlr;s In theme guides en and id'ould be sent to tdw Secretary'the Commission's regulations, to -dwli usd by V.w staff in of the Commission. US. Atomic Energy Commission. Washington. D.C. 2M346.evaluating spiedfii problems or postuletiodLa:ements, or to provide guidene to Attention: Chief. Public PrFt rnis Staff.appllmnst. Rogulatory Guides ar, not substitutes fo reguletlohss and c pllaneswith them is not required. Methods and tolutios dlast frosm diw a ull mt in Th guides ae issued I. tht following ten brood divisic.:the guides will be acoosoehie' If they provide a both for the fInidis tothe isuance or continuance of a permit or lianso by the Commission. 1. Power Reactors 6. Products2. Research end Teot Reactors 7. Transportation3. Fuels end Materials Fac1lit1is 8. Occupational HealthPublished quides will be revised perlodicallv, as appropriate. to accommodate 4. Environmental and Shing 9. Antitrust Reviewcommenn and to reflect new information or exparien. 5. Maerials and Plant Protectloo 10. General Effective source of 185.7.-ceVpmrm rays men by the detectorC E (a/tu) A [I + e (fa/4ir) e-PcIcdenrichmftnt detecor containerefecien/c absorptionPhysical are material geometricalconstants defined by composition efficiencycollimator(1)wherePu,pi,pcAuAi, AC = detected 185.7-keV activityE = enrichment of the uranium (. -1)= density of the uranium (u), matrix material (i), and container wall(c), respectively, in (g/cm3)c = mass attenuation coefficient for 185.7-keV gamma rays in uranium(u), matrix material (i), and container wall (c) in units of (cm2/g)a = specific 185.7-keV gamma ray activity of U-235= 4.3 x 104 gamma rays/sec-ge = net absolute detector full energy peak efficiency for detecting185.7-keV gamma rays (< 1)E2 = solid angle subtended by the detector (11 < 2w)A = cross-sectional area of material defined by the detector collimatord = container wall thicknessA derivation of this expression, as well as othernecessary background information relevant to this guide,may be found in the literature. 2 As evident in Eq. 1,the activity (C) is proportional to the enrichment (E)'but is affected by several other characteristics as well.Material Thicknm EffectsIn order for Eq. 1 to be applicable, it is necessarythat the material be sufficiently thick to produce strongattenuation of 185.7-keV gamma rays. To determinewhether this criterion is met, it is useful to compare theactual thickness of the material with a characteristiclength xo, where xo is defined as that thickness ofmaterial which produces 99.5% of the measured185.7-keV activity, i.e.,Calculated values of xc, the critical distance, for .several common materials are givn in Table 1.TABLE 13Material Density Critical Material(g/cm3) Distance CompositionxO lcm) 'TermPi tai.1 + 2:-i Pu MuU (metal) 18.7 0.20 1.000UF6 4.7 1.08 1.040U02 10.9 0.37 1,012U308 7.3 0.56 1.015Uranyl Nitrate 2.8 2.30 1.095Values of the mass attenuation coefficient, A, may befound in J. H. Hubbell, "Photon Cross Sections, AtteniationCoefficients, and Energy Absorption Coefficents From 10 keVto 100 GeV," NSRDS-NBS 29, 1969.X0 I n(.005) = 5.29 Xwhere(2)(3)IA = u.u + 7- plip2 L. A. Kull, "Guldejiws for Gamm&-gray SpectroscopyMeasuremente of U-235 Enrichment," BNL-50414, July 1973.5.21-2 Note: Other nondestructive, techniques are capable ofdetecting SNM distributed within. a container. Theenrichment technique, however, is inherently a surfacemeasurement. Therefore, the "sample" observed-i.e., thesurface, must be representative of all the material in thecontainer. In this respect the enrichment mesurement ismore analogous to chemical analysis than other NDAtechniques.Material Composition EffebIf the gamma-ray measurement is to be dependentonly on the enrichment, the term related to -thecomposition of the matrix should be approximatelyequal to one, i.e.,detector. The fractional change in the measured activityAC/C due to a small change Ad in the container wallthickness can be expressed as follows:AC-- .= -ZcPcAd(5$)+ pi' L C. I;li -P A(4)Calculted values. of this quantity for commonmaterials are given in Table 1. The deviation of thenumbers in Table I from unity indicate that a bias can'be introduced by ignoring the difference in materialcomposition.Inhomogeneities in matrix material composition,uranium density, and uranium enridunent within themeasured volume of the maierial (as chariterized bythe depth xo and the collimated area A) can producechanges in the measured 185.7-keV activity and-affectv-the accuracy of an enrichment calculated on the bais ofthat activity. There is a small to negligble effect on themeasurement accuracy due to variations in the contentof low-atomic-number (Z<30) matrix materials. Careshould be exercised, however, in applyin this techniqueto materials having.high-atomro-number matria" (Z>50)or materials having uranium concentuations 1. thanapproximately 75%. Inhomogeneities in uraium densitywill also produce small to negligible effects on theaccuracy if the matrix isu of low-atomic-numberelements. Sifjkuw inacraeieas cn. a.Ni, howem,when the urnium enrichment itself ce. be expected tovary throughout the sample.The above , gonclusions about the effects ofinhomogeneities are based on the assumption that thethickness of the material exceeds the critical distance,xo, and that the inhomogeneities exist within this depth.In the case of extremely inhomogeneous materiah muchas scrap, the condition of sufficient depth may notalways be fulfllled,-or inhomogeneitiesmay exist beyondthe depth xo; i.e., the "sample" is not representative.Therefore, this technique is not applicable to suchinhomogeneous materials.Container Wafl EffectsVariations in the thickness of the container walls-can significantly affect the activity measured by theCalculated values of AC/C, corresponding to achange in container thickness Ad of 0.0025 cm, forcommon container materials, are given in Table 2.TABLE 2Material Density(g/cm3l CSteel 7.8 -.003Aluminum 2.7 -.0009Polyethylene 0.95 -.0004Therefore, the container wall thickness should beknown, e.g., by measuring an adequate number of thecontainers before loading. In some cases an unknowncontainer wall thickness can be measured using anultrasonic technique and a simple correction applied tothe data to account for attenuation of the 185.7-keVgamma rays (see eq. 5). Commercial equipment isavailable to measure wall thicknesses ranging from about0.025 to 5.0 cm to relative accuracies of approximately1.0% to 0.1%, respectively.Area and Geometrical EfficiencyThe area of the material viewed by the detector andthe geometrical efficiency are variables which may beadjusted, within limits, to optimize a system. It isimportant to be aware that once these variables arefixed, changes in these parameters will affect the resultsof the measurement.It is also important to note that the placement ofthe material within the container will affect the detectedactivity. The 'material should fill the volume of thecontainer to a certain depth, leaving no void spacesbetween the material and the container wall.Net Deteetw BffidncyThallium-activated sodium iodide, NaI(T1),scintillationw detectors and lithium-drifted germanium,Ge(LI), solid-state detectors have been used to performthese measurements. The detection systems are generallyconventional gamma-ray spectrometry systems presentlycommercially available in modular or single-unitconstruction.5.21,3 The following factors influence detector selectionand the control required for accurate results.1. Backgrounda. Compton Background. This background ispredominately produced by'the 765-keV and ICOl-keVgamma rays of Pa-234m, a daughter of U-238. Since, inmost cases, the Compton background behaves smoothlyin the vicinity of the 185.7-keV peak, it can be readilysubtracted, leaving only the net counts in the 185.7-keVfull-energy peak.b. Overlapping Peaks. The observable peak fromcertain gamma rays may overlap that of the 185.7-keVpeak due to the finite energy resolution of the detector;i.e., the difference in energies may be less than twice theFWHM. This problem is common in enrichmentmeasurements of recently separated uranium from areprocessing plant. The peak from a strong 208-keVgamma ray from U-237 (half-life of 6.75 days)- canoverlap the 185.7-keV peak when an Nal detector isused. Analytical separation of the two unresolved peaks,i.e., peak stripping, may be applied. An alternativesolution is to use a Ge(Li) detector so that both peaksare clearly resolved.The U-237 activity ;present in reprocesseduranium will depend on the amount of Pu-241 presentbefore reprocessing and also on the time elapsed sinceseparation.c. Ambient Background. The third source ofbackground originates from natural sources and fromother uranium-bearing materials located in the vicinityof the measuring apparatus. This last source can beparticularly bothersome since it can vary with timewithin wide limits depending on plot operatingconditions.2. Count-Rate LoAmes. Calculation of the detectorcount rates for purposes of making dead time estimatesrequires that one calculate the total count rate, not onlythat due to U-235. Total count rate estimates forlow-enrichment material must therefore take intoaccount the relatively important background fromU-238 gamma rays. If other radioactive materials arepresent within the sample, their contributions to thetotal count rate must also be considered.Count-rate corrections can be made by determiningthe dead time or by making measurements for known4 FWHM- full width of the spectrum peak at half itsmaximum height.live-time s intervals. The pile-up or overlap of electronicpulses is a problem which also results in a loss of countsin the full-energy peak for Ge(Li) systems. A pulser maybe used to monitor and correct for these losses.Radiation which provides, no useful -information can beselectively attenuated by filters; e.g., a one-millimeter-thick cadmium filter will reduce x-ray interference,eliminating this source of count-rate losses.3. Instability in Detector Electronics. The gain of aphotomultiplier tube is sensitive to changes intemperature, count rate, and magnetic field. Provisioncan be made for gain checks and/or gain stabilization forenrichment measurement applications. Various gainstabilizers that automatically adjust the system gain tokeep a reference peak centered between two presetenergy limits are available.
C. REGULATORY POSITION
Passive gamma-ray spectrometry constitutes anacceptable means for nondestructively determiningU-235 enrichment, if the following conditions aresatisfied:Range of Application1. All material to be assayed under a certaincalibration should be of similar chemical form, physicalform, homogeneity, and impurity level.2. The critical distance of the material should bedetermined.. Only those items of the material havingdimensions greater than -this critical distance should- beassayed by this technique.3. The material should be homogeneous in all respectson a mnacroscopic 6 scale.- The material should behomogeneous'with respect to uranium enrichment' on amicroscopic -wscale.4. The containers should all be of similar size,geometry, and physical and chemical composition.System RequirementsI. Nal('I) scintillation detectors having a resolution ofFWHM < 16% at the 185.7-keV peak of' U-235 ares"Live time" means that portion of the measurementperiod during which the instrument can record detected events.Dead time refers to that portion of the measurement periodduring which the instrument is busy processing data alreadyrecehed anldcannot accept new data. in order to compare6fferent data for which dead times are appreciable, one mustcompare counts measured for equal live-time periods.(actual measurement period) -(dead time) = live ,time6 Macroscopic refers to distances greater than the criticaldistance; miuoscopic to distances les than the critical distance.5.21-4 generally adequate for measuring the enrichment ofuranium containing more than the natural (0.71%)abundance of U-235. Crystals With a thickness of ~- 1.25cm are recommended for optimum efficiency. If other-1- radionuclides Which emit significant quantities of gammaradiation in an energy region E = 185.7 keV +/- 2 FWHMat 185.7 keV are present:a. A higher-resolution detector. e.g., Ge(Li),should be used, orb. A peak stripping procedure should be used tosubtract the interference. In this case, data should beprovided to. show the range of concentration of -theinterfering radionuclide, and the accuracy and precisionof the stripping technique over this range.2. The detection system gain should be stabilized bymonitoring a known reference peak.3. The system should measure live time or provide ameans of determining the count-rate losses based on thetotal counting rate.4. Design of the system should allow reproduciblepositioning of the detector or item being assayed..5. The system should be capable of determining thegamma-ray activity in at least two energy regions toallow background subtraction. One region shouldencompass 185.7 keV, and the other region should beabove this but not overlapping. The threshold and widthof the regions should be adjustable.6. The ýsystem should have provisions for filteringlow-energy radiation which could interfere with the185.7-keV or background regions.Data ReductionI. if the total counting rate is determined primarily bythe 185.7-keV gamma ray, the counting rate should berestricted (absorbers, decreased geometrical efficiency)below those rates requiring correction. The systemsensitivity will be reduced by these measures and, if nolonger adequate,' separate calibrations should be made intwo or more enrichment regions.Ifrthe total counting rate is determined primarily byevents other than those due to 185.7-keV gamma rays,counting rate corrections should be made.2. To determine the location and width of the185.7-keV peak region and the background region(s),the energy spectrum from each calibration standard (seeCalibration, next section) should be determined and theposition of the 185.7-keV peak and neighboring peaksnoted. The threshold and width of each energy regionshould then be selected to avoid including anyneighboring peaks, and to optimize the system stabilityand the signal-to-background ratio.3. The net response attributed to 185.7-keV gammarays should be the accumulated counts in the peakregion minus a multiple of the counts accumulated in anearby background region(s). A single upper backgroundregion may be monitored or both a region above thepeak region and one below may be monitored.If only an upper background region is monitored, thenet response, R, should be given byR = G-bBwhere G and B are the gross counts in the peak regionand the background region, respectively, and b is themultiple of the background to be subtracted. This netresponse, R, should then be proportional to theenrichment, E, given byE = C, R = C, (G-bB)where C, is a calibration constant to be determined (seeCalibration, next section). The gross counts, G and B,should be measured for all the standards. The quantitiesG/E should then be plotted as a function of thequantities B/E and the slope of a straight line throughthe data determined. This slope is b, the multiple of theupper background region to be subtracted, i.e..G/E = b(B/E) + I/CIThe data from all the standards should be used indetermining this slope.If both an upper and a lower background aremonitored, the counts in each of these regions should beused to determine a straight line fit to the background.Using this straight line approximation, the area ornumber of counts under this line in the peak regionshould be subtracted from the gross counts, G. to obtainthe net response. An adequate technique based on thisprinciple is described in the literature.Calibration s1. Calibration standards should be obtained by:a. Selecting items from the production material. Agroup of the items selected should, after determinationG. Gunderson, 1. Cohen, M. Zucker, "Proceedings: 13thAnnual Meeting, Institute of Nuclear Materials Management,"Boston, Mass. (1972) p. 221." None of the calibration techniques or data reductionprocedures exclude the use of automated direct-readout systemsfor operation. The procedures described in this guide should beused for adjustment and calibration of direct-readoutinstruments.5.21-5 of the gamma-ray response, be measured by anindependent, more accurate technique traceable to, orcalibrated with, NBS standard reference material, e.g.,mass spectrometry. The other items should be retainedas working standards.b. Fabricating standards which represent thematerial to be assyed in chemical form, physical form,homogeneity, and impurity level. TheU-235 enrichmentof the material used in the fabrication of the standardsshould be determined by a technique traceable to, orcalibrated with, NBS standard reference material, e.g.,mass spectrometry.2. The containers for the standards should have ageometry, dimensions, and composition whichapproximate the mean of these parameters in thecontainers to be assayed.3. The values of enrichment for the calibrationstandards should span the range of values encountered innormal operation. No less than three separate standardsshould be used.4. Each standard should be measured at a number ofdifferent locations, e.g., for a cylinder, at differentheights and rotations about the axis. The mean of thesevalues should be used as the response for thatenrichment. The dispersion in these values should beused as an initial estimate of the error due to materialand container inhomogeneity.5. The data from the standards, i.e., the net responseattributed to 185.7-keV gamma rays and the knownuranium enrichment, should be used to determine theconstants in a calibration function by a weightedleast-squares technique.Operations1. The detection system and counting onometry(collimator and container-to-detector distance) shouldbe identical to those used in calibration.2. The data reduction technique and count-rate losscorrections, if included, should be identical to thoseused in calibration.3. Data from all measurements should be recorded inan appropriate log book.4. At least two working standards, should be measuredduring each eight-hour operating shift. The measuredresponse should beý compared to the expected response(value used in calibration) to determine if the differenceexceeds three times the expected standard deviation. Ifthis threshold is exceeded, repeat measurements shouldbe made to verify that the response is significantlydifferent and that the system should be recalibrated.5. All containers should be agitated, or the materialmixed in some manner, if possible, prior to counting.One container from every ten should be measured at twodifferent locations. Other items may be measured atonly one position. (If containers am scanned to obtainan average -enrichment, the degree of inhomogeneityshould still be measured by this method.)The difference between the measurements atdifferent locations should be used to indicate a lack ofthe expected homogeneity. If the two responses differby more than three times the expected standarddeviation (which should include the effects of the usualor expected inhomogeneity), repeat measurementsshould be made to verify that an abnormalinhomogeneity exists. If the threshold is exceeded, thecontainer should be rejected and investigated todetermine the cause of the abnormal inhomogeneity. 96. In the event that all containers are not filled to auniform height, the container should be viewed at aposition such that material fills the entire volume viewedby the detector. The procedure for determining the fillof the container should be recorded' e.g., by visualinspection at the time of filling and recording on thecontainer tag.7. The container wall thickness should be measured.The wall thickness and location of the measurementshould be indicated, if individual wall thicknessmeasurements are made, and the gamma-raymeasurement made at this location. If the containers arenominally identical, an adequate sampling of thesecontainers should be representative. The mean of themeasurements on these samples constitutes anacceptable measured value of the wall thickness whichmay be applied to all containers of this type or category.8. The energy spectrum from a process item selected atrandom should be used to determine the existence ofunexpected interfering radiations and the approximatemagnitude of the interference. The frequency of this testshould be determined by the following guidelines:a. At leat one item in any new batch of material.b. At ieast one item if any chanps in the materialprocesing occur.c. At least one item per material balance period.If an interference appears, either a higher-resolutiondetector must be acquired or an adequate peak strippingroutine applied. In both cases additional standards whichinclude the interfering radiations should be selected andthe system recalibrated.The difference nmy also be due to a large variation in wallthickness.Il5.21-6 9. No item should be assayed if the mesured responseexceeds that of the highest enrichment standard by morethan tvice the standard deviation in the reponse fromthis standard.Error AnysisI. A least4quares technique should be used todetermine the uncertainty in the calibration constants.2. The measurement.to-measurement error should bedetermined by periodically observing the net responsefrom the standards and repeating measurements onselected process items. Each repeat measurement shouldbe made at a different location on the container surface,at different times of the day, and under differingambient conditions.' "The standard deviation should bedetermined and any systematic trends corrected for.' The statistical error due to counting (Includingbackipound) and the erron due to inhomopamsity, ambientconditions, etc. will be include in this measurement-to-measurement error.3. The item-to-item error due to the uncertainty inwall thickness should be determined. The uncertainty inthe wall thickness may be the standard deviation aboutthe mean computed from measurements on randomlyselected samples, or it may be the uncertainty in thethickness measurement of individual containers. Thisuncertainty in wall thickness should be multiplied by theeffect of a unit variation in wall thickness on themeasured 185.7-keV response to determine thiscomponent uncertainty.4. Item-to-item errors other that those measured, e.g.,wall thickness, should be determined by periodically (seeguidelines in paragraph 8. of the Operation Section)selecting an item and determining the enrichment by anindependent technique traceable to, or calibrated with,NBS standard reference material. A recommendedapproach is to adequately sample and determine theU-235 enrichment by calibrated mass spectrometry. Inaddition to estimating the limit of error from thesecomparative measurements, the data should be added tothe data used in the original calibration and newcalibration constants determined.5.21-7