Regulatory Guide 3.35

Assumptions Used for Evaluating the Potential Radiological Consequences of Accidental Nuclear Criticality in a Plutonium Processing and Fuel Fabrication Plant

ML12220A062
Person / Time
Issue date:	07/31/1979
From:	Office of Nuclear Regulatory Research, NRC/OSD
To:
References
RG-3.035, Rev. 1
Download: ML12220A062 (20)

U.S. NUCLEAR REGULATORY COMMISSIONRevision 1July 1979:*REGULATORY GUIDEOFFICE OF STANDARDS DEVELOPMENTREGULATORY GUIDE 335ASSUMPTIONS USED FOR EVALUATING THE POTENTIAL RADIOLOGICALCONSEQUENCES OF ACCIDENTAL NUCLEAR CRITICALITY IN APLUTONIUM PROCESSING AND FUEL FABRICATION PLANT

A. INTRODUCTION

Section 70.22, "Contents of Applications," of10 CFR Part 70, "Domestic Licensing of SpecialNuclear Materials," requires, that each appli-cation for a license to possess and use specialnuclear material in a plutonium processing andfuel fabrication plant contain a description andsafety assessment of the design bases of theprincipal structures, systems, and componentsof the plant. Section 70.23(a)(3) states thatapplications will be approved if the Commissiondetermines that, among other factors, theapplicant's proposed equipment and facilitiesare adequate to protect health and minimizedanger to life and property, and Sec-tion 70.23(b) states that the Commission willapprove construction of the principal struc-tures, systems, and components of the plantwhen the Commission has determined that thedesign bases of the principal structures, sys-tems, and components and the qualityassurance program provide reasonableassurance of protection against theconsequences of potential accidents.In plutonium processing and fuel fabricationplants, a criticality accident is one of thepostulated accidents used to evaluate the ade-quacy of an applicant's proposed activities withrespect to the public health and safety. Thisguide describes methods used by the NRC staffin the analysis of such accidents. Thesemethods result from review and action on anumber of specific cases and, as such, reflectthe lates~t general NRC-approved approaches tothe problem. If an applicant desires to employnew information that may be developed in thefuture or to use an alternative method, NRC*Lines indicate substantive changes from previous issue.will review the proposal and approve its use, iffound acceptable.

B. DISCUSSION

In the process of reviewing applications forpermits and licenses authorizing the construc-tion or operation of plutonium processing andfuel fabrication plants, the NRC staff hasdeveloped a number of appropriately conser-vative assumptions that are used by the staffto evaluate an estimate of the radiologicalconsequences of various postulated accidents.These assumptions are based on previousaccident experience, engineering judgment,and on the analysis of applicable experimentalresults from safety research programs. Thisguide lists assumptions used by the staff toevaluate the magnitude and radiological conse-quences of a criticality accident in a plutoniumprocessing and fuel fabrication plant.A criticality accident is an accident resultingin the uncontrolled release of energy from anassemblage of fissile material. The cir-cumstances of a criticality accident are difficultto predict. However, the most seriouscriticality accident would be expected to occurwhen the reactivity (the extent of the deviationfrom criticality of a nuclear chain reactingmedium) could increase most rapidly andwithout control in the fissile accumulation ofthe largest credible mass. In plutonium pro-cessing and fuel fabrication plants where con-ditions that might lead to criticality arecarefully avoided because of the potential foradverse physical and radiological effects, suchan accident is extremely uncommon. However,experience with these and related facilities hasdemonstrated that criticality accidents mayoccur.USNRC REGULATORY GUIDES Comments should be sent to the Secretary of the Commission, U.S. NuclearRegulatory Commission, Washington, D.C. 20566, Attention: Docketing andRegulatory Guides are issued to describe and make available to the public Service Branch.methods acceptable to the NRC staff of implementing specific parts of theCommission's regulations, to delineate techniques used by the staff in evalu- The guides are issued in the following ten broad divisions:sting specific problems or postulated accidents, or to provide guidance toapplicants. Regulatory Guides are not substitutes for regulations, and com- 1. Power Reactors 6. Productsphiance with them is not required. Methods and solutions different from tdose 2. Research and Test Reactors 7. Transportationset out in the guides will be acceptable if they provide a basis for the findings 3. Fuels and Materials Facilities 8. Occupational Healthrequisite to the issuance or continuance of a permit or license by the 4. Environmental and Siting 9. Antitrust and Financial ReviewCommission. 5. Materials and Plant Protection 10. GeneralRequests for single copies of issued guides (which may be reproduced) or forComments and suggestions for improvements in thease guides are encouraged at placement on an automatic distribution list for single copies of future guidesall times, and guides will be revised, as appropriate, to accommodate comments in specific divsions should be made in writing to the U.S. Nuclear Regulatoryand to reflect new information or experience. This guide was revised as a result Commission, Washington, D.C. 20555, Attention: Director, Division ofof substantive comments received from the public and additional staff review. Technical Information and Document Contro In plutonium processing and fuel fabricationplants, such an accident might be initiated by(1) the inadvertent transfer or leakage of asolution of fissile material from a geometricallysafe containing vessel into an area or vesselnot so designed, (2) introduction of excessfissile material solution to a vessel, (3) intro-duction of excess fissile material to a solution,(4) overconcentration of a solution, (5) preci-pitation of fissile solids from a solution andtheir retention in a vessel, (6) introduction ofneutron moderators or reflectors (e.g.,entrance of water to a higly under-moderatedsystem), (7) deformation of or failure tomaintain safe storage arrays, or (8) similaractions which can lead to increases in thereactivity of fissile systems. Some acceptablemeans for minimizing the likelihood of suchaccidents are described in RegulatoryGuide 3.4, "Nuclear Criticality Safety inOperations with Fissionable Materials OutsideReactors. "1I. CRITICALITY ACCIDENT EXPERIENCE IN RELATION TOTHE ESTIMATION OF THE MOST SEVERE ACCIDENTStratton (Ref. 1) has reviewed in detail34 occasions prior to 1966 when the power levelof a fissile system increased without control asa result of unplanned or unexpected changes inits reactivity. Although only six of theseincidents occurred in processing operations,and the remainder occurred mostly in facilitiesfor obtaining criticality data or in experimentalreactors, the information obtained and itscorrelation with the characteristics of eachsystem have been of considerable value for usein estimating the consequences of accidentalcriticality in process systems. The incidentsoccurred in aqueous solutions of uranium orplutonium (10), in metallic uranium orplutonium in air (9), in inhomogeneous water-moderated systems (9), and in miscellaneoussolid uranium systems (6). Five occurred inplutonium systems, including reactors andcriticality studies, of which three were insolutions.The estimated total number of fissions perincident ranged from 1E+152 to 1E+20 with amedian of about 2E+17. More recently, anotherincident in a plutonium processing facility atWindscale (U.K.) was described in which atotal yield of about 1E+15 fissions apparentlyoccurred (Ref. 2). In ten cases, thesupercriticality was halted by an automaticcontrol device. In the remainder, the shutdownwas effected as a consequence of the fissionenergy release which resulted in thermalexpansion, density reduction from theformation of very small bubbles, mixing of light'Copies may be obtained from the U.S. Nuclear RegulatoryCommission, Washington, D.C. 20555, Attention; Director,Division of Document Control.21E÷15 = 1 x 1015. This notational form will be used in thisguide.and dense layers, loss of water moderator byboiling, or expulsion of part of the mass.Generally, the criticality incidents werecharacterized by an initial burst or spike inthe curve of fission rate versus time followedby a rapid but incomplete decay of the fissionrate as the shutoff mechanism was initiated. Asmore than one shutdown mechanism may affectthe reactivity of the system and the effect of aparticular mechanism may be counteracted, theinitial burst was frequently succeeded by aplateau period of varying length. This plateauwas characterized by a lesser and decliningfission rate and finally by a further dropoff asshutdown was completed. The magnitude of theinitial burst was directly related to the rate ofincrease of reactivity and its magnitude abovethe just-critical value but was inversely relatedto the background neutron flux, which is muchgreater for plutonium than for uraniumsystems.Those systems consisting only of solidfissile, reflector, or moderator materialsexhibited little or no plateau period, whereassolution systems had well developed plateaus.For solution systems, the energy releaseduring the plateau period, because of its dura-tion, provided the major portion of total energyreleased. For purposes of the planning neces-sary to deal adequately with criticalityincidents in experimental and production-typenuclear facilities, Woodcock (Ref. 3) made useof these data to estimate possible fission yieldsfrom excursions in various types of systems.For example, spike yields of 1E+17 and 1E+18and total yields of 3E+18 and 3E+19 fissionswere suggested for criticality accidentsoccurring in solution systems of 100 gallons orless and more than 100 gallons, respectively.Little or no mechanical damage was predicted atthese levels.2. METHODS DEVELOPED FOR PREDICTING THE MAGNITUDEOF CRITICALITY ACCIDENTSThe nuclear excursion behavior of solu-tions of enriched uranium has been studiedextensively both theoretically and experi-mentally. A summary by Dunenfeld and Stitt(Ref. 4) of the kinetic experiments on waterboilers, using uranyl sulfate solutions,describes the development of a kinetic modelthat was confirmed by experiment. This modeldefines the effects of thermal expansion andradiolytic gas formation as power-limiting andshutdown mechanisms.The results of a series of criticality excur-sion experiments resulting from the introduc-tion of uranyl nitrate solutions to verticalcylindrical tanks at varying rates are sum-marized by Ldcorchd and Seale (Ref. 5). Thisreport confirms the applicability of the kineticsmodel for solutions, provides correlations ofpeak power with reactivity addition rate, notes-J3.35-2 the importance of a strong neutron source inlimiting peak power, and indicates the natureof the plateau following the peak.Many operations with fissile materials in aplutonium processing plant may be conductedwith aqueous (or organic solvent) solutions offissile materials. Consequently, well-foundedmethods for the prediction of total fissions andmaximum fission rate for accidents that mightoccur in solutions (in process or other vessels)by the addition of fissile materials should be ofconsiderable value in evaluating the effects ofpossible plutonium processing plant criticalityaccidents. From the results of excursionstudies and from accident data, Tuck (Ref. 6)has developed methods for estimating (1) themaximum number of fissions in a 5-secondinterval (the first spike), (2) the total numberof fissions, and (3) the maximum specific fis-sion rate in vertical cylindrical vessels, 28 to152 cm in diameter and separated by >30 cmfrom a bottom reflecting surface, resultingfrom the addition of up to 500 g/1l solutions ofPu-239 or U-235 to the vessel at rates of 0.7 to7.5 gal/min. Tuck also gives a method forestimating the power level from which thesteam-generated pressure may be calculatedand indicates that use of the formulas for tanks>152 cm in diameter is possible with a loss inaccuracy.Methods for estimating the number of fis-sions in the initial burst and the total numberof fissions, derived from the work reported byL6corchi and Seale (Ref. 5), have also beendeveloped by Olsen and others (Ref. 7). Thesewere evaluated by application to ten actualaccidents that have occurred in solutions andwere shown to give conservative estimates inall cases except one.Fission yields for criticality accidentsoccurring in solutions and some heterogeneoussystems, e.g., aqueous/fixed geometry, can beestimated with reasonable accuracy usingexisting methods. However, methods for esti-mating possible fission yield from .other typesof heterogeneous systems, e.g., aqueous/powder, are less reliable because of theuncertainties involved in predicting thereactivity rate. The uncertainty of geometryand moderation results in a broad range ofpossible yields.Woodcock (Ref. 3) estimated that in solidplutonium systems, solid uranium systems, andheterogeneous liquid/powder systems (fissilematerial not specified) total fission yields (sub-stantially occurring within the spike) of 1E+18,3E+19, and 3E+20, respectively, could bepredicted. Mechanical damage varied fromslight to extensive. Heterogeneous systemsconsisting of metals or solids in water wereestimated to achieve a possible magnitude of1E+19 following an initial burst of3E+18 fissions. The possibility of a burst of3E+22 fissions resulting in a serious explosioncould be conceived for large storage arrayswhere prompt criticality was exceeded, e.g.,by collapse of shelving. It is recognized that insuch arrays, where reactivity is more likely tobe increased by the successive additions ofsmall increments of materials, only a delayedcritical condition with maximum yields of 1E+19fissions is likely. These estimates could aid inthe analysis of situations in plant systems.However, they should not be taken as absolutevalues for criticality assumptions for thepurpose of this guide.For systems other than solution systems,the estimation of the peak fission rate and thetotal number of fissions accompanying an acci-dental nuclear criticality may be estimated withthe aid of information derived from accidentexperience and from the SPERT-l reactor tran-sient tests with light- and heavy-watermoderated uranium-alumium and U02-stainlesssteel clad fuels (Ref. 8). Oxide core tests inthe latter group provide some information onenergy release mechanisms that may beeffective, for example, in fabricated fuelelement storage in a mixed oxide fuel fabrica-tion plant. Review of unusal process struc-tures, systems, and components for thepossibility of. accidental criticality should alsoconsider recognized anomalous situations inwhich the possibility of accidental nuclear cri-ticality may be conceived (Ref. 9).The application of the double-contingencyprinciple3 to fissile material processing opera-tions has been successful in reducing theprobability of accidental criticality to a lowvalue. As a consequence, the scenariosrequired to arrive at accidental criticalityinvolve the assumption of multiple breakdownsin the nuclear criticality safety controls. It hastherefore been a practice to simply andconservatively as'sume an accidental criticalityof a magnitude equal to, or some multiple of,the historical maximum for all criticality acci-dents outside reactors without using anyscenario clearly defined by the specific opera-tions being evaluated. In the absence ofsufficient guidance, there has been wide vari-ation in the credibility of the postulatedmagnitude of the occurrence (particularly thesize of the initial burst), the amount of energyand radioactivity assumed to be released, andthe magnitude of the calculated consequences.It is the staff's judgment that the evalua-tion of the criticality accident should assumethe simultaneous breakdown of at least twoindependent controls throughout all elements ofthe operation. Each control should be such thatits circumvention is of very low probability.Experience has shown that the simultaneous3The double-contingency principle is defined in ANSI N16. 1-1975, "Nuclear Criticality Safety in Operations with FissionableMaterials Outside Reactors," which is endorsed by RegulatoryGuide 3.4.3.35-3 failure of two independent controls is veryunlikely if the controls are derived, applied,and maintained with a high level of qualityassurance. However, if controls highlydependent on human actions are involved, thisapproach will call for some variation in theassumed number of control failures. Thecriticality accidents so conceived should thenbe analyzed to determine the most severewithin the framework of assumed controlfailures, using realistic values of suchvariables as the fissile inventory, vessel sizes,and pump transfer rates.3. RADIOLOGICAL CONSEQUENCES OF ACCIDENTAL CRITI-CALITYPast practice has been to evaluate theradiological consequences to individuals ofpostulated accidental criticality in plutoniumprocessing and fuel fabrication plants in termsof a fraction of the guideline values in 10 CFRPart 100, "Reactor Site Criteria."The consequences of a criticality accidentmay be limited by containment, shielding,isolation distance, or evacuation of adjacentoccupied areas subsequent to detection of theaccident. If the impact of a criticality accidentis to be limited through evacuation of adjacentoccupied areas, there should be prior formalarrangements with individual occupants andlocal authorities sufficient to ensure that suchmovements can be effected in the time allowed.The equations provided for estimatingdoses from prompt gamma and neutron radiationwere developed using experimental andhistorical data. The report, "Promp Neutronand Gamma Doses from an AccidentalCriticality," explains this development.* Theseequations cannot be expected to be as accurateas detailed calculations based on actualaccident conditions. Comparisons withpublished information indicate they may not beconservative for smaller accidents .(e.g. , 1-2E+17 fissions). However, for accidents thatare likely to be assumed for safety assessmentpurposes, they appear to be sufficientlyconservative. These equations are included inthe guide to provide a simplified method forestimatinK prompt gamma and neutron radiationdoses from a potential criticality accident.

C. REGULATORY POSITION

I. FOLLOWING ARE THE PLANT ASSESSMENT AND ASSUMP-TIONS RELATED TO ENERGY RELEASE FROM A CRITI-CALITY ACCIDENT AND THE MINIMUM CRITICALITYACCIDENT TO BE CONSIDERED:a. When defining the characteristics of anassumed criticality accident in order to assess*A copy of Charles A. Willis' report, "Prompt Neutron andGamma Doses, from an Accidental Criticality," is available forinspection at the NRC Public Document Room, 1717 H StreetNW., Washington, D.C.the adequacy of structures, systems, andcomponents provided for the prevention ormitigation of the consequences of accidents,the applicant should evaluate crediblecriticality accidents in all those elements of theplant provided for the storage, handling, orprocessing of fissile materials or into whichfissile materials in significant amounts could beintroduced. To determine the circumstances ofthe criticality accidents, controls judgedequivalent to at least two highly reliable,independent criticality controls should beassumed to be circumvented. The magnitude ofthe possible accidents should then be assessed,on an individual case basis, to estimate theextent and nature of possible effects and toprovide source terms for dose calculations. Themost severe accident should then be selectedfor the assessment of the adequacy of theplant. In order to determine the source termsfor release of plutonium, the powder mixtureshould be the maximum weight percent pluto-nium to uranium compound to be used in amixed oxide fuel fabrication plant.Calculation of the radioactivity of fis-sion products may be accomplished by computercode RIBD (Ref. 10). An equivalent calculationmay be substituted, if justified on anindividual case basis.b. If the results of the preceding evalu-ation indicate that no possible criticalityaccident exceeds in severity the criticalityaccident postulated in this section, then theconditions of the following example may beassumed for the purpose of assessing theadequacy of the facility. A less conservativeset of conditions may be used if they are shownto be applicable by the specific analysesconducted in accordance with paragraph C.l.aabove.An excursion that produces an initialburst of 1E+18 fissions in 0.5 seconds followedsuccessively at 10-minute intervals by47 bursts of 1.9E+17 fissions for a total of1E+19 fissions in 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> is assumed to occur.The excursion is assumed to be terminated byevaporation of 100 liters of the solution.2. ASSUMPTIONS RELATED TO THE RELEASE OF RADIO-ACTIVE MATERIAL ARE AS FOLLOWS:4a. It should be assumed that all of thenoble gas fission products and 25% of the iodineradionuclides are released directly to aventilated room whose construction is typical ofthe plant's Class I structures. If the accidentis assumed to occur in a solution, it should alsobe assumed that an aerosol, which is generatedfrom the evaporation of solution during theexcursion, is released directly to the roomatmosphere. The aerosol should be assumed to4Certain assumptions for release of radioactive material, doseconversion, and atmospheric diffusion reflect the staff'sposition indicated in Regulatory Guide 1.3 (Ref. 20).3.35-4 comprise 0.05% of the salt content of thesolution that is evaporated. The room volumeand ventilation rate and retention time shouldbe considered on an individual case basis.b. The effects of radiological decay duringtransit within the plant and in the plantexhaust system should be taken into account onan individual case basis.c. The reduction in the amount of radio-active material available for release to theenvironment through the plant stack as aresult of the normal operation of filtrationsystems in the plant exhaust systems may betaken into account, but the amount of reduc-tion in the concentration of radioactive mate-rials should be evaluated on an individual casebasis.d. Table 1 lists the radioactivity of sig-nificant nuclides released, but it does notinclude the iodine depletion allowance.* 3. ACCEPTABLE ASSUMPTIONS FOR DOSE AND DOSE CON-VERSION ARE AS FOLLOWS:a. The applicant should show that the con-sequences of the prompt gamma and neutrondose are sufficiently mitigated to allowoccupancy of areas necessary to maintain theplant in a safe condition following the accident.The applicant should estimate the promptgamma and neutron doses that could bereceived at the closest site boundary andnearest residence. The following semi-empiricalequations may be used for these calculations.Because detailed evaluations will be dependenton the site and plant design, different methodsmay be substituted on an individual case basis.Potential total dose attenuation due to shieldingand dose exposures should be evaluated on anindividual case basis.(I) Prompt5 Gamma DoseD = 2.IE-20 Nd-2 e-3.4dwwherefirst foot, and a factor of 5.5 for each addi-tional foot.(2) Prompt Neutron DoseDn = 7E-20 Nd"2 e-5.2dwhereDn = neutron dose (rem)N = number of fissionsd = distance from source (kin)For concrete, the dose should bereduced by a factor of 2.3 for the first 8inches, 4.6 for the first foot, and a factor of20 for each additional foot.b. No correction should be made for deple-tion from the effluent plume of radioactiveiodine due to deposition on the ground or forthe radiological decay of iodine in transit.c. For the first 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, the breathingrate of a person off site should be assumed tobe 3.47E-4 mS/sec. From 8 to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> follow-ing the accident, the breathing rate should beassumed to be 1.75E-4 m3/sec. These valueswere developed from the average daily breath-ing rate (2E + 7 cm3/day) assumed in thereport of ICRP Committee 11-1959 (Ref. 12).d. External whole body doses should becalculated using "Infinite Cloud" assumptions,i.e., the dimensions of the cloud are assumedto be large compared to the distance that thegamma rays and beta particles travel. "Such acloud would be considered an infinite cloud fora receptor at the center because any additional(gamma and] beta emitting material beyond thecloud dimensions would not alter the flux of[gamma rays and] beta particles to thereceptor." [See Meteorology and AtomicEnergy--1968 (Ref. 13), Section 7.4.1.1;editorial additions made so that gamma and betaemitting material could be considered.] Underthese conditions, the rate of energy absorptionper unit volume is equal to the rate of energyreleased per unit volume. For an infiniteuniform cloud containing X curies of betaradioactivity per cubic meter, the beta doserate in air at the cloud center isD- = 0.457EPXThe surface body dose rate from beta emittersin the infinite cloud can be approximated asbeing one-half this amount (i.e., pDoo = 0.23EX).For gamma emitting material, the dose rate inair at the cloud center isDo, = o.5o07E XYD = gamma dose (rein)¥N = number of fissionsd = distance from source (kin)Data presented in The Effects of NuclearWeapons (Ref. 11, p. 384) may be used todevelop dose reduction factors. For concrete,the dose should be reduced by a factor of 2.5for the first 8 inches, a factor of 5.0 for theSyost of the gamma radiation is emitted in the actual fissionprocess. Some gamma radiation is produced in various second-ary nuclear processes, including decay of fission products. Forthe purposes of this guide, "prompt" gamma doses should beevaluated including the effects of decay of significant fissionproducts during the first minute of the excursion. Forconditions cited in the example, the equation given includesthese considerations.3.35-5 From a semi-infinite cloud, the gamma dose ratein air is= o.25EYxwhereID-= beta dose rate from an infinite cloud(rad/sec)Do, = gamma dose rate from an infinite¥ cloud (rad/sec)E = average beta energy per disintegration(MeV/dis)EY = average gamma energy per disintegration¥ (MeV/dis)X = concentration of beta or gamma emittingisotope in the cloud (Ci/m3)e. The following specific assumptions areacceptable with respect to the radioactive clouddose calculations:(1) The dose at any distance from theplant should be calculated based on the maxi-mum concentration time integral (in the courseof the accident) in the plume at that distance,taking into account specific meteorological,topographical, and other characteristics thatmay affect the maximum plume concentration.These site-related characteristics should beevaluated on an individual case basis. In thecase of beta radiation, the receptor is assumedto be exposed to an infinite cloud at themaximum ground level concentration at thatdistance from the plant. In the case of gammaradiation, the receptor is assumed to beexposed to only one-half the cloud owing to thepresence of the ground. The maximum cloudconcentration should always be assumed to beat ground level.(2) The appropriate average beta andgamma energies emitted per disintegration maybe derived from the Table of Isotopes (Ref. 14)or other appropriate sources, e.g. , Ref. 23.(3) The whole body dose should beconsidered as the dose from gamma radiation ata depth of 5 cm and the genetic dose at adepth of 1 cm. The skin dose should be thesum of the surface, gamma dose and the betadose at a depth of 7 mg/cm2.The beta skindose may be estimated by applying an energy-dependent attenuation factor (Dd/DB) to thesurface dose according to a method developedby Loevinger, Japha, and Brownell (Ref. 15).(See Figure 1.)f. The "critical organ" dose from the in-haled radioactive materials should be estimated.The "critical organ" is that organ that receivesthe highest radiation dose after the isotope isabsorbed into the body. For the purpose ofthis guide, the following assumptions should bemade:(1) The radionuclide dose conversionfactors are as recommended by the report ofCommittee 11, ICRP (Ref. 12) or other appro-priate source.(2) The effective half-life for the nu-clide is as recommended in ICRP Publication 6(Ref. 16) or other appropriate source.(3) The plutonium and other actinidenuclide clearance half time, or fraction of nu-clide clearing the organ, is as recommended bythe ICRP task group on lung dynamics(Ref. 17). A computer code, DACRIN(Ref. 18), is available for this model. Taskgroup lung model (TGLM) clearance parametersare presented in Table 2; the model is shownschematically in Figure 2.g. The potential dose exposure for all sig-nificant nuclides should be estimated for thepopulation distribution on a site-related basis.4. ACCEPTABLE ASSUMPTIONS FOR ATMOSPHERIC DIFFU-SION ARE AS FOLLOWS:a. Elevated releases should be consideredto be at a height equal to not more than theactual stack heigh Certain site-dependentconditions may exist, such as surroundingelevated topography or nearby structures, thatwill have the effect of reducing the actualstack height. The degree of stack heightreduction should be evaluated on an individualcase basis.Also, special meteorological and geo-graphical conditions may exist that can con-tribute to greater ground level concentrationsin the immediate neighborhood of a stack. Forexample, fumigation should always be assumedto occur; however, the length of time that afumigation condition .exists is stronglydependent on geographical and seasonal factorsand should be evaluated on a case-by-casebasis.' (See Figure 3 for elevated releasesunder fumigation conditions.)b. For plants with stacks, the atmosphericdiffusion model should be as follows:Scredit for an elevated release should be given only if thepoint of release is (1) more than two and one-half times theheight of any structure close enough to affect the dispersion ofthe plume or (2) located far enough from any structure thatcould have an effect on the dispersion of the plume. For thoseplants without stacks, the atmospheric diffusion factorsassuming ground level releases, as shown in RegulatoryPosition 4.c, should be used.7For sites located more than 2 miles from large bodies ofwater, such as oceans or one of the Great Lakes, a fumigationcondition should be assumed to exist at the time of the accidentand continue one-half hour. For sites located less than 2 milesfrom large bodies of water, a fumigation condition should beassumed to exist at the time of the accident and continue for4 hours.I3.35-6 (1) The basic equation for atmosphericdiffusion from an elevated release isexp(-he2/2Cz2 )X/Q =iiua ayzwherex = the short-term average centerline valueof the ground level concentration (Ci/m3)Q = rate of material release (Ci/sec)u = windspeed (m/sec)a = the horizontal standard deviation of theY plume (m). [See Ref. 19, Figure V-l,p. 48.]a = the vertical standard deviation of thez plume (m). [See Ref. .19, Figure V-2,p. 48.]h = effective height of release (m)8e(2) For time periods of greater than 8hours, the plume from an elevated releaseshould be assumed to meander and spreaduniformly over a 22.50 sector.9 The resultantequation is8 to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />suming various stackheights] windspeed 1 m/ sec;uniform direction.See Figure 5 for Envelope of-Pasquill diffusion categories;windspeed 1 m/sec; variabledirection within a 22.50sector.x/Q =2.032 exp(-h e2/2 z2)e UXy uxzwherex = distance from the release point (m);other variables are as given in b(l).(3) The atmospheric diffusion model'0for an elevated release as a function of thedistance from the plant is based on the infor-mation in the following table.c. If no onsite meteorological data areavailable for facilities exhausted wihout stacks,or with stacks that do not meet the elevatedrelease criteria, the atmospheric diffusionmodel should be as follows:(1) The 0-to-8 hour ground level re-lease concentrations may be reduced by afactor ranging from one to a maximum of three(see Figure 6) for additional dispersionproduced by the turbulent wake of a majorbuilding in calculating nearby potential expo-sures. The volumetric building wake correctionfactor, as defined in Section 3.3.5.2 ofMeteorology and Atomic Energy--1968(Ref. 13), should be used in the 0-to-8 hourperiod only; it is used with a shape factor ofone-half and the minimum cross-sectional areaof a major building only.(2) The basic equation for atmosphericdiffusion from a ground level point source isx/Q= 1nuraayzwhereX = the short-term average centerline valueof the ground level concentration(Ci/m3)Q = rate of material release (Ci/sec)u = windspeed (m/sec)a = the horizontal standard deviation of they plume (m) [see Ref. 19, Figure V-i,p. 48]a = the vertical standard deviation of thez plume (m) [see Ref. 19, Figure V-2,p. 481(3) For time periods of greater than 8hours, the plume should be assumed tomeander and spread uniformly over a 22.50sector.9 The resultant equation is2.032x/Q= a uxzwhereX = distance from point of release to thereceptor; other variables are as givenin c(2).Time FollowingAccidentAtmospheric Conditions0 to 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> See Figure 4 for Envelope ofPasquill diffusion categories[based on Figure A7,Meteorology and AtomicEnergy--1968 (Ref. 13), as-8h =h -h , where h is the height of the release above plantgrads, SanA ht is tde maximum terrain height, above plantgrade, between the point of release and the point at which thecalculation is made, he should not be allowed to exceed hs.gThe sector may be assumed to shift after 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> if localmeteorological data are available to justify a wind directionchange. This should be considered on an individual case basis.'l°n some cases, site-dependent parameters such as meteor-ology, topography, and local geography may dictate the use ofa more restrictive model to ensure a conservative estimate ofpotential offsite exposures. In such cases, appropriate site-related meteorology should be developed on an indivdual casebasis.3.35-7 (4) The atmospheric diffusion model'0for ground level releases is based on the infor-mation in the following table.Time FollowingAccident0 to 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />s8 to 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />sAtmospheric ConditionsPasquill Type F, windspeed1 m/sec, uniform directionPasquill Type F, windspeed1 m/sec, variable directionwithin a 22.50 sector.

D. IMPLEMENTATION

The purpose of this section is to provideinformation to applicants and licensees regard-ing the staff's plans for using this regulatoryguide.Except in those cases in which the applicantproposes an alternative method for complyingwith specified portions of the Commission'sregulations, the method described herein willbe used in the evaluation of submittals forspecial nuclear material license applicationsdocketed after December 1, 1977.If an applicant wishes to use this regulatory*guide in developing submittals for applicationsdocketed on or before December 1, 1977, thepertinent portions of the application will beevaluated on the basis of this guide.(5) Figures 7A and 7B give the groundlevel release atmospheric diffusion factorsbased on the parameters given in c(4).I3.35-8 REFERENCES1. W. R. Stratton, "Review of CriticalityIncidents," LA-3611, Los Alamos ScientificLaboratory (Jan. 1967).2. T. G. Hughes, "Criticality Incident atWindscale," Nuclear Engineering Inter-national, Vol. 17, No. 191, pp. 95-7(Feb. 1972).3. E. R. Woodcock, "Potential Magnitude ofCriticality Accidents," AHSP(RP) R-14,United Kingdom Atomic Energy Authority.4. M. S. Dunenfeld, R. K. Stitt, "SummaryReview of the Kinetics Experiments onWater Boilers." NAA-SR-7087, AtomicInternational (Feb. 1973).5. P. Lgcorch6, R. L. Seale, "A Review of theExperiments Performed to Determine theRadiological Consequences of a CriticalityAccident, " Y-CDC-12, Union Carbide Corp.(Nov. 1973).6. G. Tuck, "Simplified Methods of Estimatingthe Results of Accidental Solution Excur-sions," Nucl. Technol., Vol. 23, p. 177(1974).7. A. R. Olsen, R. L. Hooper, V. 0. Uotinen,C. L. Brown, "Empirical Model to EstimateEnergy Release from Accidental Criticality,"ANS Trans., Vol. 19, pp. 189-91 (1974).8. W. E. Nyer, G. 0. Bright, R. J. McWhorter,"Reactor Excursion Behavior," InternationalConference on the Peaceful Uses of AtomicEnergy, paper 283, Geneva (1966).9. E. D. Clayton, "Anomalies of Criticality,"Nucl. Technoj., Vol. 23, No. 14 (1974).10. R. 0. Gumprecht, "Mathematical Basis ofComputer Code RIBD," DUN-4136, DouglasUnited Nuclear, Inc. (June 1968).11. The Effects of Nuclear Weapons, RevisedEdition, Samuel Glasstone, Editor, U.S.Dept. of Defense (Feb. 1964).12. "Permissible Dose for Internal Radiation,"Publication 2, Report of Committee II,International Committee on RadiologicalProtection (ICRP), Pergamon Press (1959).13. Meteorology and Atomic Energy-- 1968,D. H. Slade, Editor, U.S. Atomic EnergyCommission (July 1968).14. C. M. Lederer, J. M. Hollander, I. Perl-man, Table of Isotopes, 6th Edition,Lawrence Radiation Laboratory, Univ. ofCalifornia, Berkeley, California (1967).15. Radiation Dosimetry, G. J. Hine and G. L.Brownell, Editors, Academic Press, NewYork (1956).16. Recommendations of ICRP, Publication 6,Pergamon Press (1962).17. "The Metabolism of Compounds of Plutoniumand Other Actinides," a report preparedby a Task Group of Committee II, ICRP,Pergamon Press (May 1972).18. J. R. Houston, D. L. Strenge, and E. C.Watson, "DACRIN--A Computer Programfor Calculating Ocean Dose from Acute orChronic Radionuclide Inhalation," BNWL-B-389(UC-4), Battelle Memorial Institute,Pacific Northwest Laboratories, Richland,Washington, (Dec. 1974).19. F. A. Gifford, Jr., "Use of RoutineMeteorological Observations for EstimatingAtmospheric Dispersion," Nuclear Safety,Vol. 2, No. 4, p. 48 (June 1961).20. Regulatory Guide 1.3, "Assumptions Usedfor Evaluating the Radiological Consequencesof a Loss of Coolant Accident for BoilingWater Reactors," U. S. Nuclear RegulatoryCommission, Washington, D. C.21. J. M. Selby, et al., "Considerations in theAssessment of the Consequences of Efflu-ents from Mixed Oxide Fuel FabricationPlants," BNWL-1697, Rev. 1 (UC-41),Pacific Northwest Laboratories, Richland,Washington (June 1975)..22. "Compilations of Fission Product Yields,"NEDO-12154-1, M. E. Meek and B. F.Rider, General Electric Vallecitos NuclearCenter, TIC, P.O. Box 62, Oak Ridge,Tennessee 37830 (January 1974).23. "Nuclear Decay Data for RadionuclidesOccurring in Routine Releases from NuclearFuel Cycle Facilities," ORNL/ NUREG/TM-102, D.C. Kocher, Oak Ridge NationalLaboratory, Oak Ridge, Tennessee 37380(August 1977).3.35-9 TABLE 1RADIOACTIVITY (Ci) AND AVERAGE BETA AND GAMMA ENERGIES (MeV/dis)OF IMPORTANT NUCLIDES RELEASED FROM CRITICALITY ACCIDENT IN THIS GUIDENuclide Half-life b aKr-83mKr-85mKr-85Kr-87Kr-88Kr-89Xe-131mXe-133mXe-133Xe- 135mXe- 135Xe- 137Xe- 1381-1311-1321-1331-134,1-1351.84.510.776.32.83.2hhymhm0-0.5 Hr.1. 5E+19.9E01. 2E-46. OE+13'. 2E+11.8E+30.5-8 Hr. Total11.9 d2.0 d5.2 d15.6 m9.1 h3.8 m14.2 m8.0 d2.3 h20.8 h52.6 m6.6 h1.4E-23.1E-13.8E04.6E+25. 7E+ 16.9E+31. 5E+31. 5E01.7E+22.2E+l6.OE+26.3E+l9.5E+16. 1E+I7.2E-43.7E+22.0E+21. 1E+48.6E-21. 9E02.3E+12.8E+33.5E+24.2E+49.5E+39.5E01.OE+31.4E+23.7E+33.9E+21. 1E+27. IE+18. 1E-44.3E+22.3E+21. 3E+41.OE-12.2E02. 7E+I3.3E+34. IE+24.9E+41. IE+41. 1E+I1.2E+31.6E+24.3E+34.5E+25.9E-42.7E-55.8E-51.8E-24.3E-72.41E-5CY2.6E-31.6E-12.2E-37.8E-12. OEO1.6E02.OE-24. 1E-24.6E-24.3E-12.5E-11.6E-11. lEO3.8E-12.2E06. IE-12.6E01.5E0C02.5E-12.5E-11. 3E03.5E-11. 3E01.4E-11.9E-11.1E-19.OE-23.7E-11. 8EO6.2E-11.9E-15.OE-14. 1E-16.1E-13.7E-1Pu-238 dPu-239Pu-240Pu-241Pu-242Am-241aTotal curies, except for Pu and Am, are based on cumulative yield for fission energy spectrum using data in Ref. 22. Theassumption of cumulative yield is very conservative, e.g., it does not consider appropriate decay schemes. Calculationsregarding individual nuclide yields and decay schemes may be considered on an individual case basis. Data in this table doesnot include the iodine reduction factor allowed in Section C.2.a of this guide.by = yearh = hourd = daym = minutescHalf-lives and average energies derived from data in Ref. 23.dTotal radioactivity assumes the isotopic mix to be the equilibrium mix for recycled plutonium and 1 mg of Pu 02 released(Ref. 21).3.35-10 I.TABLE 2VALUES OF THE CLEARANCE PARAMETERS FOR THE TASK GROUP LUNG MODELaCOMPARTMENTNPabTB cdCLASS Dbcd fd0.01 0.50.01 0.50.01 0.950.2 0.05CLASS WcTd0.010.40.010.2CLASS yCdkfdk0.1 0.010.90.50.50.40.010.2k0.010.990.010.990.050.40.40.150.9P efghL i0.5n.a.en.a.0.50.50.8n.a.n.a.0.21.0501.05050500.15 5000.40.41.05000.05 5001.0 10000aSee Figure 2 for the task group lung model (TGLM) schematic diagram.bData for soluble plutonium is included. To maintain dose conversion conservatism, this class should only be con-sidered if justified on an individual case basis.Cclass D = readily soluble compounds where removal time is measured in days.Class W = compounds with limited solubility where removal time is measured In weeks.Class Y = insoluble compounds where removal time is measured in years.dTk is the biological removal half time in days; fk is the fraction of original deposit leaving the organ via pathwayindicated on the schematic model shown in Figure 2. Data are based on a mass median aerodynamic diameter of1 micron and were developed by Battelle Memorial Institute, Pacific Northwest Laboratories, and presented in aninterim report by E. C. Watson, J. R. Houston, and D. L. Strenge, April 1974.en.a. means not applicable.3.35-11 WI1.00.0 g/cm' ' 00-4-410 A fL0.05I._ : A...../ -10-2S I 0.0. 01210-3 LS0.FIGUR I0.1 1 3. 10.a. .n.u. Bea E ery eRAIOO IET DOS TO SUFC DOEAI!UCTO EAEERYSETAfo ni it I Pln oreo InIntThcesadfoAlwdSptr...1.De0e2oped frmCosdrain Prsne inRfrnc 5 hatr1: : : FIGURE 1: :: : : 3:.3 5-12; #

LM, CLMF LSCHEMATIC DIAGRAM DEVELOPED FROM ICRP TASK GROUP LUNG MODEL (Refil17FIGURE 2I.3.35-13 10-2i ..... -..-:-.' ......... .ELEVATED RELEASE

• ATMOSPHERIC DISPERSION FACTORS i.FOR FUMIGATION CONDITIONS T---ATMOSPHERIC CONDITIONS-PASQUILL TYPE FWINDSPEED 1 METERISEC-7:i44i..t ..-ra10 -6Distance from Release Point (meters)FIGURE 3 (Ref. 20)3.35-14 10-3 I I I I I I I I II IIELEVATEDATMOSPHERIC DIFI0-8 HOUR REU.0r-E0h =125 meters*h = 150 meters10-610-7 1 1 1 1 idl I I L -I I sall102 103 104Distance from Release Point (meters)FIGURE 4 (Ref. 20)3.35-15 ELEVATED RELEASE-... ATMOSPHERIC DIFFUSION FACTORS .-.... 8--24 HOUR RELEASE TIME-,L.- .-.-r".-,-- --f- -V --7 -- .. ., , .. ...----,- ---II I i I INýDistance from Release Point (meters)FIGURE 5 (Ref. 20)3.35-16

-w32.50.5A = 500 meters20.52 0.5A = 1000 meters20.5A = 1500 meters2, 0.5A = 2000 meters2.1.5CoC0.50-102 103Distance from Structure (meters)FIGURE 6 (Ref. 20)104 Ii i i i.i i i -I i i10-10-: : *-.GROUND LEVEL RELEASE-ATMOSPHERIC DIFFUSION FACTORS FORVARIOUS TIMES FOLLOWING ACCIDENTw+- howun..-- + +]* I ............_ I ." --4 --L -... ...... z,- -- --l f ..-24 houri, i.4-"-- I5 .--102103Disunce from Structure (mutre)FIGURE 7A (Ref. 20)105K3.35-18 fo'Distman from 8tructure (metesr)FIGURE 7B (Oef. 20)3.35-19 UNITED .STATESNUCLEAR REGULATORY COMMISSIONWASHINGTON, D. C. 20555OFFICIAL BUSINESSPENALTY FOR PRIVATE USE, $300POSTAGE AND FEES PAIDWEIED STATES NUCLEAREGIRA TORY COMMISSION0