lA-nnoUC-34C

banal: March 1988

Analysis of Tritium Productionin Concentric Spheres ofOralloy and6LiD Irradiated by 14-MeV Neutrons

L. Raymond Fawcett, Jr.*

Roger R. Roberts II** LA—11110Raymond E. Hunter

DE88 006308

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United StatesGovernment. Neither the United States Government nor any agency thereof, nor any of theiremployees, makes any warranty, express or implied, or assumes any legal liability or responsi-bility for the accuracy, completeness, or usefulness of any information, apparatus, product, orprocess disclosed, or represents that its use would not infringe privately owned rights. Refer-ence herein tu any specific commercial product, process, or service by trade name, trademark,manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recom-mendation, or favoring by the United States Government or any agency thereof. The viewsand opinions of authors expressed herein do not necessarily state or reflect those of theUnited States Government or any agency thereof.

'Collaborator at Los Alamo*. Director »/ Physics ami Prc-F.ngina.ring Programs,Longivood College, Farmville, VA 23901."Graduate Research Assistant at /.os Alamos. Dcfiartimtit of Physics,Miami University. Oxford, Ohio 45056.

A \ ^iTTrPlf'rriCSJ *-os Alamos National Laboratory * \.r-\AU(SLlU U U ^ S ) Los Alamos,New Mexico 87545 / \\J

DISTRIBUTION Of T ^ ^ l : : ' U Z T IS UNLiMITEP ^

CONTENTS

ABSTRACT 1

I. INTRODUCTION 1

II. EXPERIMENT 1A. Source Neutrons 3B. Enriched Uranium and 6LiD Assembly 3C. Tritium Production 3D. Radiochemical Detector Foils 6

III. ANALYSIS 7A. Tritium Production 7B. Radiochemical Detector Foils 9C. Perturbations 10D. Errors 10

IV. RESULTS 10A. Tritium Production 10B. Radiochemical Detector Foils 11

1. (n.2n) Activation 112. (n,f) Activation 113. (n,-y) Activation 11

V. CONCLUSIONS 11

ACKNOWLEDGMENTS 16

APPENDIX A. Typical Tritium Production in 7LiH Input File 20

APPENDIX B. Comparison of Tritium Production in 7Li Samples with Cross Sectionsfrom ENDF/B-V vs. ENDF/B-V with 235U Softened Fission Spectrumat High Energy 21

APPENDIX C. Comparison of Tritium Production in 7Li Samples for InducedBackground Originating from Ampule Quartz vs.Contaminated Helium 22

APPENDIX D. Comparison of Cross-Section Evaluations 23

REFERENCES 25

ANALYSIS OF TRITIUM PRODUCTION IN CONCENTRIC SPHERESOF ORALLOY AND 6LiD IRRADIATED BY 14-MeV NEUTRONS

by

L. Raymond Fawcett, Jr., Roger R. Roberts II, and Raymond E. Hunter

ABSTRACT

Tritium production and activation of radiochemical detector foils ina sphere of cLiD with an oralloy core irradiated by a central source of14-MeV neutrons have been calculated and compared with experimen-tal measurements. The experimental assembly consisted of an oralloysphere surrounded by three solid GLiD concentric shells with ampules ofGLiH and 7LiH and activation foils located in several positions through-out the assembly. The Los Alamos Monte Carlo Neutron Photon Trans-port Code (MCNP) was used to calculate neutron transport throughoutthe system, tritium production in the ampules, and foil activation. Theoverall experimentally observed-to-calculated ratios of tritium produc-tion were 0.996 ± 2.5% in 6Li ampules and 0.903 ± 5.2% in 7Li ampules.Observed-to-calculated ratios for foil activation are also presented.

I. INTRODUCTION

The objectives of this experiment were 1) to deter-mine the integral of the 9Li and 7Li tritium productioncross sections in a system composed of a 6LiD-reflectedenriched urnanium sphere with a centra) source of 14-MeV neutrons and 2) to measure the transport of 14-MeV neutrons through the reflected assembly.

The experimental system was composed of a 15.2o-cm-diam oralloy sphere (with channels to provide fora central 14-MeV neutron source) centered in a 60.0-cm-diam 6LiD sphere consisting of a series of nestedsolid hemispherical shells (Figs. 1 a. b. and c). Inte-gral determination of the neutron energy and flux asa function of distance from the center of the assemblywas done by radiochemical detector foils placed insidethe oralloy and between the 6LiD shells at several dis-tances from the source. Tritium production measure-ments were made with GLiH- and 'LiH- filled quart/,ampules placed in alcoves on and between the "LiD

shells. Comparison of reaction rates determined by ex-periment and calculation tests the ability to calculatethe integral over energy of the produc. of evaluatedcross sections and calculated fluxes.

Although the experiment was performed in Septem-ber 1977. measurements and calculation of the val-ues of tritium production and radiochemi< al activationhave just been completed.

II. EXPERIMENT

The following description of the experiment is takenpredominately from three sources: an unpublished re-port by those who performed the experiment,1 the ex-periment notebooks of A. Hemmendinger and C. E.Ragan. and a memorandum from •'. S. Gilmore.2

(a) (b)

(c)

Fig. 1. Bottom half of oralloy and LiD assemby of hemishells is shown in (a). The tritium target is inside the steel ball at thecenter. Oralloy core and innermost LiD shell completed are shown in (b). Top two outermost hemishells are missing. Almostcomplete experimental assembly is shown in (c). Top outermost hemishell is yet to be put in place.

A. Source Neutrons C. Tritium Production

Neutrons of II MeV were produced by 3(M)-k»>Vdeuterons that were incident on a tritium target atthe center of the experimental assembly. The assem-bly was exposed to 3.815 x 1015 11-McV source neu-trons. The neutron source flux distribution for thesame target setup used in this experiment has alreadybeen mapped over I n sr for an earlier experiment.3

That distribution was used in the analysis of this ex-periment. A map of the fiux distribution is in Fig. 3of Ref. 3.

B. Enriched Uranium and 6LiD Assembly

The enriched uranium consisted"1 of two solid hemi-spherical shells of total mass 32.900 kg and isotopiccomposition 93.18 at.'-,' 3 3 S l ' . 5.82 at.% 2 3 8 l ' . and 1.00at.l>f 2 3 4 l ' . The cLiD was composed3 of six solid hemi-spherical shells of average density 0.7425 g/cm3 andisotopic composition 95.6 n\.% GLi and 4.4 at.% 7Li.Both the uranium core and GLiD shells were machinedso they could be fitted together to form a solid sphere,except for a small cavity at the center and three chan-nels to house and access the tritium target. The di-mensions and masses of the hemishells are in Table I.(The 6Li specifications in Table I were extracted fromRef. 3. Part I. Table I.) Figs. J a. b. and c show thepartially completed assembly.

1. Samples. Tritium was produced in exper-imental samples through the reactions °Li(n,a)?Hand 7Li(n,n')2He. 3H. The samples consisted of(mart z ampules some containing cLiH, others 7LiHlocated in several positions throughout the °LiDsphere. Those at 30.0 cm from the center of the assem-bly had 0.032-in.-thick cadmium covers to absorb anyroom-return thermal neutrons. Information on ampulespecifications and positions is in Table II. AlthoughTable III of R.ef. 1 places the innermost ampules at 7.6cm from the center of the assembly, they were actuallyabout 8.3 cm from the center. The oralloy sphere hadan external radius of 7.62 cm. No alcoves were cut inthe oralloy to accommodate ampules. The alcoves inthe innermost 6LiD shell were cut deeper and the am-pules were taped below the surface of the 6LiD. Theradii of these inner ampules were 0.50 cm; therefore,the distance from the assembly center to the ampulecenter was 8.3 ± 0.1 cm. Angular orientation of theampules is shown in Fig. 2. The ampules were spher-ical with 1.0-inm-thick walls and had small stems onthem for sealing after being loaded in a helium atmo-sphere.* (See Fig. 1 of Ref. 3 for a drawing of an am-pule.) The ampules were not loaded with LiD becausethe tritium content in deuterium produced a dps back-ground larger in most cases than that expected fromtritium production in the ampules.

*LiH samples were prepared at Y-12 plant. Oak Ridge, TN.

TABLE

Material

I. Specifications for Hemispherical

Diameter (mm)

Inside Outside

Shells

Mass (kg)

Oralloy

6LiD

6LiD

6LiD

44.4

154.3

254.0

402.0

152.3

252.0

400.0

600.0

16.41016.3390.150°

2.4652.330

9.1339.430

30.00029.200

"Screw and plugs.

TABLE II. LiH Sample Specifications

LiIsotope

SampleNo.

Ampule OutsideRadius (cm)

Distance FromSource (cm)

Sample Mass0

LiH (g)

411412414

0.90.90.9

29.9529.9529.95

0.83810 96810.9073

410 0.9 20.15 0.9489

344346347

0.90.90.9

12.5512.7512.75

0.89120.87110.8966

338 0.5 8.37 0.1188

7 348349

340341342

401407408

355339

0.90.9

0.90.90.9

0.90.90.9

0.50.5

29.9529.95

19.9519.9520.15

12.5512.7512.55

8.228.37

"Taken from F. D. Bender, Document Transmittal Form,June 27, 1977. Available from L. R. Fawcett, Jr.

1.05421.0200

0.94640.95021.0104

1.03980.92670.9454

0.13980.1210

"Ampule Data,"

After irradiation, the LiH ampules were assayed fortritium content by Teledyne Isotopes, Westwood, NJ07675. The amount of tritium produced in each samplewas measured at Teldiyne by proportional counter andby liquid scintillator. All results were then correctedfor radioactive decay of tritium to September 20, 1977.Finally, the Teledyne count rates were normalized tothermal-neutron tritium-production results3 based onthe ratio of the thermal capture cross section of 6Lito that of 197Au. The result was that Teledyne's netcounting rate was divided by 1.058 for large ampulesand by 1.103 for small ampules.

2. Background. Before normalization, both nat-ural and induced backgrounds were subtracted from

Teledyne's measured counting rates. The backgroundsin unirradiated LiH samples are shown in Table III.*The induced background was measured by irradiat-ing helium-filled ampules along with the LiH samplesat the time of the experiment. Induced backgroundhad a significant impact on the 7Li part of the experi-ment because it was greater than 50% of the measuredcounting rate for small ampules. The large induced

"The information in this table is similar to that in Tablo II ofRef. 1. with one major exception. Ampule 349. an irradiatedsample, was mistakenly included in Table II of Ref. 1 as anunirradiated sample. Ampule 349 is correctly excluded fromTable III of this work. Then-fore, the 7Li average background

" 1 'i?~ 'is corrected from 3.88 s 'K ' to 2.69 S '»""'

c u = -

0)

+x

180°

Fig. 2. Orientation of ampules with respect to 2 H + beam. In views A-A and B-B, top and bottom refer to the assembly at thetime of the experiment.

TABLE III. Tritium Backgrounds in LiH Samples(Unirradiated)

Isotope SampleNo. Background

"1)

6Li 417418

0.273 ±0.110.247 ±0.11

7Li 343413416

Av.

Av.

0.260 ± 0.07

2.93 ± 0.182.35 ±0.152.78 ±0.17

2.69 ± 0.10

background is particularly worrisome since in the 14-MeV experiment (Ref. 3) this same measurement wassmall enough to be neglected. The experimenters say"there must have been some contaminant in the HE(loading atmosphere) whose reaction product was car-ried through the chemical processing."' The inducedbackground data are in Table IV.

3. Absolute Tritium Production. The abso-lute tritium production for each ampule is presentedin Table V (Results .section). Absolute tritium produc-tion is calculated from the experimentally determinednormalized net counting rate by

where

N dNN(obs) - — = —: • — ,

to XM dt (1)

N(obs) is the number of tritons produced per lithiumatom in the sample by the total number ofsource neutrons; that is, the absolute tritiumproduction.

N is thv number of tritium atoms present pergram of LiH,

dndT

M

A

is the normalized— 1 \

K )•is 8.520 x 1022 6L7LiH g"1 .= 1.7781 x l O ' s

net

iHg

- l

counting rate

" ' and 7.506 x 1022

Ampules 401. 407. and 408 were all at the samedistance from the central neutron source. However,their experimental values of tritium production varyby a factor of 2. The experimenters state1 that thereis something wrong with the results from those threesamples.

D. Radiochemical Detector Foils

Six packets of radiochemical detector foils wereplaced in a plane through the center of the assem-bly at 90° to the incident deutcron beam. One packetwas on the exterior surface of the largest GLiD shell at30.0 cm from the center of the assembly; two packetswere between 6LiD shell interfaces at 20.0 and 12.6 cmfrom the center; one packet was at the oralloy/cLiDinterface at 7.5 cm from the center: and two packetswere within the oralloy core at 4.8 and 2.3 cm from thecenter.2 A 0.5-in.-diam radial channel in the lower oral-loy hemisphere directly below and at 60° to the coolingchannel had been cut to house foil packets. This chan-nel was completely backfilled with solid oralloy plugsafter the foils were inserted. The several foil local ionsare shown in Fig. 3. The foil nudides were 45Sc. r'8Ni.8 9 y Q 0 Z r i 1 0 9 T m 191 j , . 1 9 3 , r . 9 7 A , , 2 3 5 ( J < „ „ , 238 , , :

A complete description of the foils is in Ref. 2.

TABLE IV

Isotope"Plane

6Li

7Li

Induced Background

DistanceFrom

Source(cm)

30.020.012.68.3

30.020.012.6&3

in He-Filled

SampleNo.

423, 424404, 405160, 402

331

421, 422426

403, 427333

Ampules (Irradiated)

AmpuleSize

LargeLargeLargeSmall

LargeLargeLargeSmall

Induced6

Background(dps/sample

0.2334.38

17.9827.00

0.174.97

14.540.5

aAll 6LiH ampules and corresponding He rilled ampules were placed in one planein the experimental assembly; all 7LiH and corresponding He ampules were placedin another.6Taken from Ref. 1, Table III.

HI. ANALYSIS

The Los Alamos Monte Carlo Neutron PhotonTransport Code (MCNP)5 was used to calculate neu-tron transport in the enriched uranium core and the6LiD. MCNP also calculated tritium production in theampules and foil activation through

(2)

where

<j>x is the neutron fluence (in neutrons cm 2MeV ')at ampule or foil position i, and

(j>j is the reaction cross section for reaction j .The geometry of the oralloy hemispheres was mod-

eled according to the specifications contained on anengineering drawing6 of the core, and that of the 6LiDshells was modeled according to the information con-tained in Table I. Three radial channels were cut intothe oralloy and 6LiD shells to accommodate an oc mon-itor, the 2 H f beam, and cooling tubes. These channelswere included in the MCNP model.

The neutron multiplication of the system exceeded20. This made it difficult to keep MCNP running be-

cause some histories required more than 20 s to com-plete. To make calculations, some artiBcal cutoffs wererequired in the input files. Even then the problems ranvery slowly. Many runs required more than 20 h com-puter time.

A. Tritium Production

All tritium production calculations were three di-mensional, with each ampule modeled in the MCNPinput file at the position where it was located in theexperimental assembly. Track length per unit volumetallies were used. A typical input file is attached as Ap-pendix A. Although the spherical ampules had stems,they were treated as spheres without stems to simplifythe model. The ampule radii used in the calculations(0.50 cm for small and 0.90 cm for large ampules)matched the outside rather than inside diameter tomake an approximate allowance for stem volume. Cad-mium covers of 0.032 in. were modeled on all ampulesat 30.0 cm. The correct mass of LiH was modeled ineach ampule. The correct mass was particularly im-portant to account properly for the "flux trap" effectdiscussed on page 7 of Ref. 7.

Fig. 3. Orientation of foils. Foils were located in a vertical plane through the center of the assembly and perpendicular to the2 H + beam. The numbers 1 through 6 locate the foil packets.

The cross sections used for both neutron trans-port and tritium production were from the ENDF/B-Vevaluation.*

Since all ampules contained a mixture of 6LiH and7LiH, tritium production in an ampule had to be cal-culated as the sum of two integrals as follows:

t prod. 6Li = 0.9590 f <t>i{E)aii6 (E)dE + 0.0410

f <$>x (E)<rLl7 {E)dE (3)

*ENDF/B-V cioss sections were used for transport and tritiumproduction, except for 2H and Li. for which the 1982 LosAlamos Group T-2 evaluations were employed. The Group T-2evaluation for 7Li is the same as ENDF/B-V Revision 2. The2H evaluation is the same as ENDF/B-V except the library usedhere has been updated with correlated energy' angular distribu-tion for the (n.2n) reaction.

t prod. 7Li = 0.001 / <j>t (E)<j>Ll6 (E)dE + 0.999

r [E)oLl7 (E)dE . (4)

where

<£,(E) is t n e average neutron fluencc over the volumeof ampule i.

4>i,6 's the reaction cross section for 6Li(n,a). and4>UY is the reaction cross section for 7Li(n.n'a).

The observed (experimental) values of absolute tri-tium production were derived from the Net CountingRate Normalized column of Table III. Ref. 1, usingEq.(l). The values of the normalized net counting ratefor 7Li were adjusted for a natural background of 2.69± .095 s~'g~' from the original, incorrect 3.88 ±1.13

In Table IV one finds that although the observed-to-ealculated (O/O) ratios for tritium generation in 7Liare generally not very good, for ampules 335 and 339the ratios are totally unacceptable in that- the experi-mental values are about half as large as the calculatedvalues.

Since the 7Li tritium-production cross section hasa threshold of about 2.8 MeV. and ampules 335 and339 are directly exposed to the fission neutrons fromthe oralloy ball, it was thought that, possibly the high-energy end of the ENDF/B-V 235U fission spectrum istoo hard. To test the effect of a softer fission tail, R.C. Little and R. E. Seamon8 modified the ENDF/B-V 235U cross section file so that it contained a fissionspectrum whose high-energy tail at E' = 5.0 MeV wasreduced to 50% of that of ENDF/B-IV and at E' =12 MeV was reduced to 10% of that of ENDF/B-IV.This modified fission spectrum greatly improved theobserved-to-calculated ratios for ampules 335 and 339but also produced lower calculated values in the other7Li ampules. The results are presented in AppendixB.

As mentioned in Section II.C.2 on background, theexperimenters felt there must have been an contami-nant that could be activated in the helium atmospherein which the ampules were loaded. If ir is assumedthat the induced background count rate in Table IVis produced totally by a reaction product in contami-nated helium, the observed values of tritium produc-tion can be properly adjusted. First, the free volume ofa loaded ampule is calculated by subtracting the LiHvolume from the ampule volume. Assume the free vol-ume to be completely filled with helium (since loadingoccurred in a helium atmosphere). Then the fractionof volume occupied by helium will be equal to the frac-tion of the induced background measurement (in thecorresponding helium-filled ampule) that must be sub-tracted from the measured counting rate (less naturalbackground). For example, the induced backgroundfor 7Li at 8.3 cm in a helium-filled ampule was mea-sured to be 40.5 ^ p . In ampule 339 the volume ofhelium is 58.4%. Therefore, the induced backgroundin 339 is (40.5 ^p)(0.584) = 23.7 ^ .

The measured counting rate (less natural back-ground) was 540.6 dis/s — g. Therefore, the net count-ing rate is given by

(540.6 dis g) - 23.7dis s - l

rather than 205 dis H~ Jg l , as is the case when the in-duced background is assumed to originate in the glassof the ampule.

Based on this helium contamination assumption,the O/O ratios for ampules 335 and 339 are substan-tially improved without much detriment to the other7Li O/C ratios. The results are presented in AppendixC.

B. Radiochemical Detector Foils

The radiochemical activation calculations werethree dimensional (using point detectors) for foil posi-tions 1, 2, and 3 inside the oralloy, and one dimensional(using surface tallies) for foil positions 4, 5, and 6 inand on the 6LiD. Of course, one would prefer to use thethree-dimensional model for all foil positions. How-ever, the point detectors require prohibitive amountsof computer time to produce good statistics in a low-flux environment.

The one-dimensional calculation is thought to bejustified on the premise that outside the oialloy theneutron fluence approaches isotropy.

Point detectors, which were modeled in the inputfile at the experimental coordinates of the foils, tallythe neutron fluence at the detector location. A surfacetally produces an integrated fluence over an entire sur-face.

Evaluated dosimetiy/activation cross sections forsome foil nuclides were available from several sources.9

In such cases the choice of dosimetry cross section wasbased on the advice of the authors of Ref. 9. The crosssections used for each nuclide and reaction are identi-fied in Tables VII and VIII in the Results section. Thecross sections used for neutron transport were takenfrom the ENDF/B-V evaluation.

Because the observed-to-calculated ratios for the ra-diochemical detector foils are in some cases unaccept-able, the calculated values were redone using a 235Utransport cross section with modified fission spectrum,as described in Section III.A and Ref. 8. This modifi-cation was expected to decrease calculated activationin (n,2n) reactions which would produce observed-to-calculated ratios closer to unity. Although the mod-ified 235U transport cross section did in fact improveO/C ratios for (n,2n) reactions, it caused significantdeterioration in the ratios for (n,f) and (n,f) reactions.

0.121 g

= 345 rfi.s tf-

C, Perturbations

The effects of several undesirable physical phenom-ena are inextricably woven into the experimental ro-sults. Although two of these phenomena will be men-tioned, their influences on the experiment are thoughtto be small enough not to warrant further investiga-tion.

One of these phenomena is room return. The ex-periment was performed in a large steel and concreteroom. In the 6LiD core experiment (Ref. 7), whichtook place in the same room, the influence of room re-turn on radiochemical activation foils was investigatedin some detail. In all cases room return contributedless than 1% of induced foil activity. Since in this ex-periment the outer ampules had cadmium covers, it isexpected that tritium production caused by room re-turn was negligible. Another phenomenon during theexperiment was self- and cross-activation of the foils.Inside the oralloy core the foils were packaged on topof one another. Thus not only would there be self-activation of a foil from neutrons born within it, but,also cross-activation from neutrons born in the super-imposed foils penetrating the first foil. Self- and cross-activation were investigated in Ref. 7. The largestcombined effect found there was 1.9% of the total ac-tivation from all other neutrons. This was for the238U(n,7) reaction. For all other reactions the effectwas much smaller. Therefore, the calculated activa-tion values in this experiment are not corrected forself- and cross-activation.

D. E^rorst

Tallies calculated by the MCNP transport code areaccompanied by a statistical error of one fractionalstandard deviation of the mean. No estimates of cross-section uncertainties are included. The precisions ofthe experimental values for the foils were taken fromTable I of Ref. 2. The precisions of the experimen-tal values for tritium production are given in Ref. 1.These are the uncertainties assigned to the observedand calculated values found in Tables V, VII, and VIII(Results section). In addition, for tritium productionthe experimenters estimated a systematic error of lessthan 6%. Generally the errors quoted on observed-to-calculated ratios are for one fractional standard de-viation and consist of the square root of the sum ofthe squared experimental and calculated fractional er-rors. When uncertainties were derived by other for-mulations, the method is explained by footnotes tothe tables.

IV. RESULTS

A. Tritium Production

Table V lists the tritium production for each ampuleand the corresponding observed-to-calculated ratios.Whereas specific tritium production, f(r), is reportedin Ref. 1, Table III, the absolute tritium production,N(obs), is reported here. Absolute tritium produc-tion is defined as the number of tritons produced perlithium atom in an ampule for the total number ofsource neutrons. Specific and absolute tritium pro-duction arc related by

(5)

where

f(r) is in units oftritons produced • mm?

Li atom • source neutronwhere

nr

is the number of source neutrons, andis the distance from the neutron source tothe center of an ampule in millimeters.

The calculated tritium production in 6Li ampulesmatches the experimentally measured values quitewell. Experimentally observed-to-calculated ratios liebetween 0.87 to 1.09, with several ratios very nearunity. Observed-to-calculated ratios are unity withinthe limits of the quoted uncertainties for six of eightampules.

The observed-to-calculated ratios of tritium pro-duction in several of the 7Li ampules are unaccept-able. For ampules in the 20.0-cm and 30.0-cm radialpositions, the results, although not good, are at leastmarginally acceptable. Only observed and calculatedvalues are presented at the 12.6-cm radial position.Here the large discrepancy between observed values(approximately 100% from smallest to largest) appearsto be due to some unknown experimental problem. Asmentioned earlier, the experimenters comment in theirreport1 that there is obviously something wrong withthe measured values from ampules 401, 407, and 408.It should be noted that in the tritium assay, the am-pules were destroyed, so the source of the discrepancycannot be determined. The results from ampules 335and 339 in the 8.3-cm position are totally unaccept-able. Here the experimentally observed values are onlyapproximately 50% of the calculated values.

The experimenters1 comment in Ref. 1 about in-duced activity in contaminated helium prompted cal-culation of the influence contaminated helium may

10

TABLE V. Tritium Production - Observed and Calculated0

N(Obs)c x 1013 N(Calc)d x 1013

Experiment CalculatedLi

Isotope

6

7

Sample

No.

411412414

410

344346347

338

348349

340341342

401407408

335339

Location

r (mm) w b

299.5299.5299.5

201.5

125.5127.5127.5

83.7

299.5299.5

199.5199.5201.5

125.5127.5125.5

82.283.7

(deg)

20-20135

-145

30- 3 0

-135

40

-20120

3515

-125

30-30135

-4055

/ Tritons Produced \

\ Li Atom /

8.495 ± 6%9.104 ± 6%9.375 ± 6%

96.16 ± 6%

453.0 ± 6%468.7 ± 6%451.1 ± 6%

1004 ± 6%

0.336 ± 18%0.328 ± 20%

1.618 ± 8%1.551 ± 9%1.618 ± 9%

9.613 ± 5%14.71 ± 5%7.410 ± 7%

14.21 ± 14%14.01 ±. 16%

/ Tritons Produced \1 I^ Li Atom J

7.763 ± 6%9.003 ± 9%8.820 ± 7%

97.20 ± 3%

481.7 ± 2%453.6 ± 2%467.0 ± 2%

1153 ± 2 %

0.378 ± 12%0.240 ± 13%

1.49 ± 7%1.84 ± 7%1.42 ± 8%

7.42 ± 3%7.09 ± 3%7.13 ± 3%

30.1 ± 3%26.7 ± 3%

Observed*

Calculated

1.094 ± 9%1.011 ± 10%1.063 ± 9%

0.989 ± 7%

0.940 ± 6%1.033 ± 6%0.966 ± 6%

0.872 ± 6%

0.889 ± 22%1.367 ± 24%

i

1.086 ± 11%0.843 ± 11%1.139 ± 12%

0.472 ± 14%0.525 ± 16%

°The quoted uncertainities on both observed and calculated values are for one fractional standard deviation.In addition, there is an estimated <6% systematic error in the observed values which is not included in theerrors quoted here.6Angles above the parting plane in the experimental configuration are positive; angles below the parting planeare negative.cObserved tritium production extracted from Ref. 1, Table III. (Tritons produced per Li atom in an ampulefrom 3.815 x 1015 source neutrons.)rfTritium production calculated in this work using ENDF/B-V cross section data for both transport and tritiumproduction. (Tritons produced per lithium atom in an ampule from 3.815 X 1015 source neutrons.)T h e quoted errors were determined from a square root of the sum of the squares combination of the observedand calculated uncertainties.

I I

h»wc had on the observed count rate - hence, tritiumproduction, Those results are presented in Appendix( \ Clearly, if the induced background (which was, forampules 335 and 339, greater than 50% of the mea-sured counting rate) did result from contaminated he-lium, the observed-to-calculatod ratios in ampules 335and 339 are dramatically improved without substantialnegative effect on the other 7Li ratios. Induced activ-ity by contaminated helium would have a negligible ef-fect on the observed tritium production in 6Li becausethe measured counting rate is hundreds of times largerthan the fraction of induced background that could beattributed to activation of contaminated helium.

The overall observed-to-calculated tritium produc-tion ratio was obtained by equal-weight averaging ofratios for each isotope. The average ratio for 6Li (in-cluding all ampules) is

,„„,, „,"/„ =0-996 ±2.5%*

The average ratio for 7Li (excluding ampules 401,407, and 408) is

{Obs/Calc),prod ,n 7£it

= 0.903 ± 5.2

The information in Table V is summarized in TableVI by presenting the average tritium production ateach radius. The observed- to-calculated ratios for 6Lirange between 0.871 and 1.05. For 7Li they lie between0.497 and 1.07. The O/C ratios are unity within thelimits of the quoted uncertainties in four of seven casesfor the two mid ides.

B. Radiochemical Detector Foils

Table VII contains the experimentally observed andcalculated values of radiochemical activation. TableVIII is more interesting because the experimental re-sults are compared with calculated values throughobserved-to-calculated ratios. It appears that, exceptfor six or seven commonly well-known dosimetry crosssections such as 197 Au(n,'y) and 235U(n,f), a number ofless well-known cross-section evaluations are in error.The dosimetry cross sections were taken from severalsources, as described in footnote c of Tables VII and

VIII. Approximately 2/3 of the observed-to-calculatedratios differ from unity by more than 10%, and a lewdiffer by a factor of 3. It is believed thai this informa-tion will be useful to those who participate in deter-mining recommended cross sections.

Figures 4a through 4q arc graphs of the ratios inTable VIII. Specific comments on comparisons of iiievarious evaluations from Ref. 7 and this p.iper aregiven in Appendix D.

1. (n,2n) Activation. Except for 89Y(n.2n).the calculated induced activity is greater than thatobserved. The lack of a consistent pattern of increasingor decreasing ratio magnitudes with increasing radiusdoes not. suggest a code transport difficulty but impliesuncertainty in the dosimetry cross sections.

2. (n,f) Activation. Both the 235U(n,f) andthe 238U(n,f) cross sections are well known. Yet at allradii for both reactions the calculated activity is an av-erage of about 12% greater than that observed. Againthe ratio magnitude shows no pattern of increasing ordecreasing with increasing distance from the source,implying reliable neutron transport calculations.

3. (n,7) Activation. Here only 89Y(n,7),193Ir(n,7). and 197Au(n,7) results are acceptable. Onewould also have expected the 238U(n,-y) ratios to bedose to unity because the 238U(n,i) capture cross sec-tion should be reasonably well known up to 2 MeVin any dosimetry evaluation. Such apparently is notthe case. In the 20.0- and 30.0-cm radius positions,experimentally determined activities exceeded calcu-lated values by almost two to one. The 45Sc(n.7) and1C9Tm(n,7) ratios are unacceptable. It is thought thatthis is because their (11,7) cross sections are not as gen-erally well known as are those of 89Y. 193Ir, and l97Au.

V. CONCLUSIONS

Experimentally observed and calculated values fortritium production in 6Li match quite well. As notedearlier, the observed-to-calculated ratios for tritiumproduction in 7Li were marginally acceptable only inthe two regions of deepest penetration and unaccept-able elsewhere. The results from the radiochemicalactivation part of this experiment will most profitablybe used as feedback for those who recommend dosiinc-try cross sections.

•This uncertainty dors not include iiny estimate of systematicerror.

It is believed that the cause of poor tritium produc-tion in 7Li is that most of the 7Li experimental valueshave significant problems associated with them. Thosewe know about are listed below.1. An irradiated ampule was included in a group

of unirradiated ampules from which natural back-ground was determined. (It is believed that thisproblem has been corrected by removal of the ir-radiated ampule from the natural background database.)

2. The irradiated ampules used to measure inducedbackground may have contained contaminated he-lium whose reaction product produced a count rategreater than that due to tritium produced in am-pules 335 and 339. There appears to be no way

at tin's point to determine what portion of inducedbackground came from helium and what portionwas due to activated impurities in ampule quartz.Without this knowledge the experimental value ofnet counting rate cannot be determined correctly.

3. The experimenters contend that there is obviouslysomething wrong with the measured results fromampules 401, 407, and 408 in that the net count rateper gram of 7Li from 407 was twice that of 408 andthese ampules were all 12.6 cm from the neutronsource, where the neutron fluences to which theywere expo ~d should have been comparable.

4. The experimental error associated with ampules 348and 349 is unacceptably large.

TABLE VI. Average Tritium Production i t Each Radius'

Li

Isotope

6

7

Distance from

Source (mm)

300200

• 12783

30020012783

N(Obs)1>.e x 1 0 1 3

Experiment N

/Tritons

V Li i

8.99196.16457.61004.

0.3321.59610.5814.11

Produced

\tom ,

±3%±6%±3%±6%

± 13%± 5 %

± 11%

N(Calc)'

Observed and Calculated"

:,d x 1013

Calculated x

\ | Tritons 1Produced \

' \ Li Atom /

8.52997.20467.41153.

0.3091.5837.21328.40

±±±±

±±±±

4%3%1%2%

9%4%2%2%

Observede

Calculated

1.0540.9890.9790.871

1.0741.008

0.497

±±±±

±±

±

5%7%3%6%

16%6%

11%

°The quoted uncertainties on both observed and calculated values are for one frac-tional standard deviation. In addition, there is a less than 6% systematic error inthe observed values.'Observed tritium production (tritons produced per lithium atom from 3.81 x 1015

source neutrons).T h e errors assigned to the observed and calculated averages were determined by:

1/2/ 1 \

S =

dTritium production calculated in this work using ENDF/B-V cross-section datafor both transport and tritium production. (Tritons produced per lithium atomfrom 3.18 x 1015 source neutrons.)eThe quoted errors were determined from a square root of the sum of the squarescombination of the observed and calculated uncertainties.

13

TABLE VII.

React ion

SBY(n.-, |

89Y(n.2n)

l97Aii(n.->)

197Au(n.2n)

238I!(n.7)

23al'(».2n)

2MV(n.f)

" rT!ii . f )

I69Tni(n.-,)

169Tin(ii.2n)

""ZrfiiAi)

•"•'Sclu.-,)

191lr(n.-») +193Ir(n.2n)l9IIr(u.2ii)

193Ir(iiO)

193Ir(ii.n')

58Ni(n.p)

58Ni(n.2n)

58Ni(n.d)

Experimental" and

Activation1'-'

Cross Section

39089.7 IY39089.71Y

79I97.56C

79197.56*'

92238.30Y

92238.30Y

92238.30Y

92235.30Y

69169.70Y

69169.70Y

4(MHK).2(iY

2I045.20Y

77191.70Y77193.71Y77191.70Y

77193.71Y

77193.71Y

28058.30Y

28O58.24Y

28058.24Y

Calculated Values (x

Notes

Experiment' 'Calculated'ExperimentCalculatedExperimentCalculatedExperimentCalculatedExperiment('alculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculatedExperimentCalculated

l()1:i) for

2.3

3S.647

32.63 ±398.4304.6 ±704.7688.5 ±10191165 ±

572.1621.9 ±455.5530.2 ±17941946 ±

Activation

o.ry;

1%

0.2%

3"{

2%

2%

of Detector Foils

4.8

29. 7028.31 ± l"'r75.(5958.59 ± 3"-f581.9013.4 ± 1"{193.1249.8 ± 9'5f470.6531.8 ± 3"f123.1125.2 ± ;>"/,10531256 ± 4"<

60876769 ± I'll

785.6968.5 ± \'\165.0203.8 ± 4%48.7457.91 ± 5'Y43.5347.03 ± \l%1192653.2 ± 4%160 2212.5 ± 4"{756.8769.8 ± \"i999.03236 ± 1%

348.9394.2 ± 2%1.5052.398 ± 7%47.2450.13 ±6'?f

(ENDK/B-V

Distance

1 .'•)

19.8319.02 ± I','22.1020.92 i 22".'455.1485.2 ± 3'V57.70(57.(50 ± 5'^320.0365.7 ± 4'"i41.2258.02 ± 17'"{424.2497.5 ± C'i'32263566 ± \"i

542.2915.1 ± 1'?.47.8092.22 ± :«'•{14.4718.23 ± 6'!f.{4.(5539.06 ± 3'i1029463.9 ± 29'-.'50.9665.06 ± C'V696.5672.6 ± l"'{454 ± 5'X1399 ± \"'<

138.1152.3 ± 3"?0.475 ± (i't0.689 ± a"AH.3714.43 ± 5l>>

Transport Sections)

from Source (cm)

r>(5

7 9957.960 ± 1";5.0044.024 ± 81*241.625.5.1 ± 3'V14.7417.13 ± &/,133.9163.2 ± 3'i10.9012.14 ± 1%67.4372.93 ± 20;886.01(K)3 ± \"i

309.8549.0 ± \%13.9516.84 ± 0.4'V3.4543.964 ± 5"'('16.3120.86 ± l'-'f539.8231.5 ± a1'-!12.5916.61 ± 0.4"';385.3398.9 ± T\60.10 ± 10",'185.7 ± \"i21.1722.94 ± 1%0.110 ± 6'Y0.159 ± 0.5%3.6603.800 ± 0.4",'

20.0

1.9141.948 ± W,1.0540.9039 ± 7"f64.2369.24 ± 1";3.8834.356 ± VI73.4938.75 ± T'\2.8503.311 ± TV,11.4712.31 ± 2'V190.0213.5 ± 1".'80.70145.1 ± \%3.9664.315 ± V",0.6700.708 ± 3'V3.9705.110 ± I",'153.661.00 ± I"?3.4204.219 ± V<109.7107.9 ± l"<7.7(H) ± 10",'28.61 ± l"-f3.7834.151 ± 1";0.02(K) ± 10'V0.0320 ± 1".'0.8100.882 ± l''i

30.0

0.16280.1628 ±0.18850.184(1 +5.1425.456 i0.8981.024 ±5.7473.(K53 ±0.5700.836 ±1.6251.786 ±16.4718.65 ±7.69011.49 ±0.6871.025 ±0.1200.181 ±0.3710.430 ±12.945.00!) ±0.7871.000 ±7.0888.311 ±

0.6160 713 ±0.005900.006130.1(500.191 ±

r;i";

:J'»'

l'-.'

4'y,

2";

2'V

i";

r-;I";

l'V

2'7

4",'

l";

1'V

i'-r± 33± I1'

1",'

"Experimental values taken from Ref. 1.'The number to the left of the decimal is the MCNI' uuclide identification number (atomic number followed by mass number). The number to the right of the decimalis the neutron cross section set identifier.^Activation cross sections identified by .niiY are de.«riti."J in Ref. 9. Those (.'escribed by 56C are ENDF/B-V cross .sections updated fiv Los Al;:m»s group T-2evaluations. Those described by .70Y and .71Y are Los Alamos group T-2 evaluations.^Experimentally observed foil activation (activations produced per foil nucleus by 3.815 x 1015 source neutrons).'Statistical uncertainties in the calculated values are for one standard deviation.'Unless otherwise stated, uncertainties in precision of experimental values are 3%. (See Ref. 1.)

TABLE VIII. Ratio of Observed to Calculated" Values for Activation of Detector Foils (ENDF/B-V Transport Cross Sections)

Activation6' Distance from Source (cm)

Reaction

89Y(n,7)89Y(n,2n)197Au(n,-y)197Au(n,2n)238U(n.7)238U(n.2n)238U(n,f)235U(n,f)169Tm(n,i)I69Trn(n,2n)9OZr(n,2n)45Sc(n,-y)191Ir(n/y)+193Ir(n,2n)191Ir(n,2n)193Ir(n,7)193Ir(n,n/)58Ni(n,p)58Ni(n,2n)58Ni(n,np)

Cross Section

39089.30Y39089.30Y79197.56C79197.56C92238.30Y92238.30Y92238.30Y92235.30Y69169.70Y69169.70Y40090.26Y21045.26Y77191.70Y77193.71Y77191.70Y77193.71Y77193.71Y28058.24Y28058.24Y2805S.30Y

2

1.0921.3071.0240.8740.9200.8590.922

.3

±±±±±±±

6%3%3%4%3%4%

4

1.0481.2920.9480.7700.8840.9800.8300.8990.8110.8090.8410.9251.824

0 ",o40.9830.3080.8800.6270.942

.8

±±±±±±±±±±±±±

±±±±±±

3%5%3%9%4%6%5%3%3%5%6%4%5%

5%3%3%4%7%7%

7.5

1.040 ±0.946 ±0.937 ±0.853 ±0.870 ±0.710 ±0.850 ±0.900 ±0.592 ±0.518 ±0.794 ±0.887 ±2.218 ±

0.780 ±1.035 ±0.320 ±1.102 ±0.690 ±0.990 ±

3%22%4%6%5%18%7%3%3%33%6%4%29%

6%3%5%4%8%6%

12

1.0041.2440.9540.8600.8200.8900.9320.8800.5640.8280.8710.7802.330

0.7581.0350.3560.9220.7330.963

.6

±±±±±±±±±±±±±

±±±±±±

3%9%5%6%4%3%3%3%3%3%6%3%7%

3%4%10%3%6%3%

20.0

0.9821.1660.9270.8911.8900.8600.9200.8900.5500.9190.8390.7702.518

0.8111.0160.2690.9110.6250.910

±±±±±±±±±±±±±

±±±±±±

3%7%3%3%5%3%3%3%3%3%5%3%3%

3%3%10%3%10%3%

30.0

0.999 ±1.021 ±0.942 ±0.870 ±1.870 ±0.682 ±0.910 ±0.880 ±0.660 ±0.670 ±0.660 ±0.862 ±2.583 ±

0.787 ±0.925 ±

0.864 ±0.962 ±0.830 ±

3%10%4%3%5%4%4%3%3%3%3%3%5%

3%3%

3%33%3%

"The neutron transport cross sections used to obtain calculated values are those of ENDF/B-V, as described in the footnote inSection III.A.6The number to the left of the decimal is the MCNP nuclide identification number (atomic number followed by mass number).The number to the right of the decimal i- the neutron cross section set identifier.cActivation cross sections identified by .nnY are described in Ref. 9. Those described by .56C are ENDF/B-V cross sectu r<supdated by Los Alamos group T-2 evaluations. Those described by .70Y and .71Y are Los Alamos group T-2 evaluations.dThe quoted errors were determined from a square root of the sum of the squares combination of the observed and calculatedfractional uncertainties. (No estimates of cross section uncertainties are included.)

The induced background for ampules 340, 341, and342 was less than 20% of the measured counting rate.Also, the experimental errors were 8-9%. Thus it waspossible to obtain acceptable ratios for these three am-pules. In Ref. 7. our reanalysis of the eLiD core ex-periment, we demonstrated a match between observedand calculated values of tritium production in 7Li towithin 6% of unity at all radial experiments! positions.If there is any lingering concern over our ability to han-dle the 7Li(n,t) reaction, the 7Li part of the oralloycore experiment needs to be redone.

Except for the 7Li experimental problems men-tioned above, tritium production from eLi and 7Li hasbeen wrung out. This should provide stronger confi-dence in our ability to handle these processes than wehave had previously.

A C K N O W L E D G M E N T S

The authors are grateful to Robert Schrandt.Robert Seaman. Robert Little, and Charles Ragan IIIfor their assistance.

16

LA

TtD

rao

L5

U>E

/I33O

1 4

1 ?

1 0

0 B

06

t 5

89Y|n.y ) Activation

39089.30Y

• *" f ... j..

-

20 50 100 200

RADIUS (mm)

1 0

»9Y(n,2n) Activation

39089.71 Y

4 ' '

. ' • if

. . . . .f.

20 50 100 200

RADIUS (mm)

(b l

} 4

UJV-< * ?

I ' *Zloo 1 0

OB

SE

RV

ED

^

06

'9?Au(n,y) Activation

79197.56c

. i , • •

-

I

20 50 100 200

RADIUS (mm)

i o

197Au(n,2n) Activation

79197.56c

20 50 100 200

RADIUS (mm)

300

(c) (d)

=3O

1 2

"8U(n,y) Activation

92238.30Y

^ I

20 50 100 200

RADIUS (mm)

1.0 -

enmo

23«U(n,2n) Activation

92238.30 Y

; i

20 50 100 200

RADIUS (mm)

(f)

Fig. 4. Ratio of observed-to-calculatod activation as a function of foil distance from the center of the assembly.

17

OBSERVED/CALCULATEDOBSERVED/CALCULATED OBSERVED/CALCULATED

3of

OBSERVED/CALCULATED OBSERVED/CALCULATEDOBSERVED/CALCULATED

>3" <=

OBSERVED/CALCULATED OBSERVED/CALCULATED

OBSERVED/CALCULATED

1 1 1

t-a-t

' 1

roatoenCOre

-

Z

^cti

5"

OBSERVED/CALCULATEDOBSERVED/CALCULATED

S" - M i .

2 Pto -*

< ? .•<

OS

1

APPENDIX A

IMC UP

IMP

VERSION BB2B 07/14/11

• 0II7CH.I

9450CP2ON? I) O7/M/II6 II 34 31

I-I-3-4-5-67-8-9-

10-11-12-13-14-It i-16-17-in19-20-21 -22-33-34-35-26-37-38-39-30-3 1 -32-33-34-33-36-37-3839-40-4 I -434344-4546-47-48-49-50-51-52-53-54-55-56-57-58-59-fiO6 1 -53-03-0465-66-67-68-69-70-7 1 -73-73-74-75-76-77-78-79-80-81-82-8384-65-86-87-88-89-90-91-92-93-91 •95-96-97-98-99-

100-101-10} •

IPI1IUH13345G7

89I O1 11?1.114ir>i n17

2 1353 94 84 95 ?530 4575961C364*jr>

or.67335339341)

3413423481 4 4

4014 0 /4 0 1

1

234

567

S910I 1131314

15I S

2223353 T4 84 91 1 53 3 93 4 034 1312J 4 H3 4 94 O 14 U 7

4 0 1

I NMODE

sue

000

0110404040O0000

00

311 •1 -44444441 -9999999

999

SOCX

cz

PRO0UC1IUN IN L

18 631518.6315

-0.7300

•0.7396

-0 7512

8 643-8.64218 6315ID 631518 6315-0 7300-O 7300•0.7396- O . 7 5 1 2

•0 73960.7513

• 0 7 5 1 2

18.6315-0.267• 0 2 3 1• 0 . 3 1 00 . 3 1 2

• 0 . 3 3 1

-0 346• 0 334• 0 . 3 4 1-0 3040 310

2.22250.85000 5842

1 CX 0 .8500SOSOCX

soczso

4.58607.616191 .270

i : 550.940

20.051 CX 1.350SOCXC ZC!

30 CO2. 1901 250? 830

1 CX 2.420PXP iZsssssssssssss

0 . 5-0 5

•5 575- 4 . 3 5 2

- 2 5 . 5 113.S7-5 703- 4 . 3 4 a

- 14 81• IT 46

10.47• 2 5 . 5 1

13.57-9.85

- 10.018.04

1 IOR 2 1 0 00.2

FILES 14COF 4FM4f 3 4M l

M2•54MSM7HBM9NP1C1ME

1 - 3348

1 4HB9 14BBSRC1-2 I I 2

666a6

1 08a

to105

1266

4 84 9

111

6668a

10105

-335• 3 3 9• 3 4 0• 34 1- 3 4 2-348•349-401• 4 C 7-40S

•5 1626.701

10.2425 94- 5 2B 16 053

11.445 16

1 6 . 5 110.242 5 . 9 4

6 286 . 3 88 . 8 7

1 1 0

. 13 0

- 1 5-1349 340 34 1 343

0 3S13E-8 7348

305349 340 34 1 343

92235 50 0 001840OO0 51 1 01003.55 0 53006.50 1 03007.65 1.01001.50 0 5IO01.50 0 .55O0O0400

TOTNUERGNCUTN

0 20.01.O(S 0 OOI - 8

PROHP 25000 25000F C N 0 19R 1 1 0 I3R 1 1

W l l

23

-4• 56

•7• a• a

• 1 0

- 10• 12- 12- 10

12• 1 2

6

4 8

353 9

3 4 8349

• 5•6• e•i• 8

1012101212• 6

21

1 1 .• 6

2268 .

-4t 1 .• 6

44 .

• 3 .

1 1

0 0

1 04 0 1

4 O 1

• r)y

50

• 023

• 835• 1

3 4 0- 1 1340• 13- 14- 15- 16

-74 9

3353 3 91212

-1•3- 4

- B• 9

- 1 1• 13• 14- 15

- Ifi• 7

f.OO9030 9336 0 90279 1148 0

0 93 359677 5

4 4

uiHf C:M

o00p1

0

utl00001)

0

1

0 520

J 04 0 7

4 0 7

922 3«.!iO

3006

3O063007

- 4

1 1

. 5 0

. 5 0

.35

1 0

•,

3223

-2234

223 923

34 1-2234 t

222323

-23J2

- 22• 2 3

2 2- 2 2• 3 3

22-22•33

2332

-22

77

9 B 118 1559• }

999999

6R 0 0

14.13

0 10 0408 335

408 335O.U5B2

0

00

O

47B2

4 7955

0

KNNI1S

234

7

401

34223

3 4 2

2 3

- 3 3

- 2 3

• 2 3

1 3R

. 5 tO5

1 3 03 3 9

339

HQDtLtOI

407 408

401 407 4

348 349

2 2R B B 1 1

1 4 . 1 3 0

15 5 20.O

92234 51 0 0100

3007

3O07

55 0.0218

55 0 0205

I I 13 14

•1RI O O 0 45 9O 135 9O 0 90 45 9U 45

20

APPENDIX B. Comparison of Tritium Production0 in 7Li Samples with Cross Sections from ENDF/B-V vs. ENDF/B-V with235U Softened Fission Spectrum at High Energy

Sample

No.

Location

r (mm) u>6(deg)

N(Obs)r x 1013

Experiment

(Tritons Produced7 Li Atom

N(Calc) x 1013

ENDF/B-VTritons Produced \

k 7Li Atom J

N(Calc) x 1013

ENDF/B-V(Modified235!!

Spectrum**)/ \.' Tritons Produced \\ 7L; Atom J

ENDF/B-V(Modified

ENDF/B-V Spectrum)Observed Observed

Calculated Calculated

348349

340341342

401407408

335339

299.5299.5

199.5199.5201.5

125.5127.5125.5

82.283.7

-20120

3515

-125

30-30135

-4055

0.336 ± 18%0.328 ± 20%

1.618 ± 8%1.551 ± 9%1.618 ± 9%

9.613 ± 7%14.71 ± 5%7.41 ± 7%

14.2114.01

14%16%

0.378 ± 12%0.240 ± 13%

1.49 ± 7%1.84 ± 7%1.42 ± 8%

7.42 ± 3%7.09 ± 3%7.13 ± 3%

30.1 ± 3%26.7 ± 3%

0.299 ± 13%0.197 ± 14%

1.17 ± 8%1.38 ± 8%1.06 ± 9%

5.01 ± 4%4.70 ± 4%4.78 ± 4%

17.1 ± 4%15.9 ± 4%

0.899 ± 22% 1.12 ± 22%1.367 ± 24% 1.66 ± 24%

1.086 ± 11% 1.38 ± 11%0.843 ± 1 1 % 1.12 ± 12%1.139 ± 1 2 % 1.53 ± 1 3 %

0.472 ± 14% 0.831 ± 15%0.525 ± 16% 0.881 ± 16%

"The quoted uncertainties on both observed and calculated values are for one fractional standard deviation. In addition, there isan estimated <6% systematic error in the observed values which is not included in the errors quoted here.6Angles above the parting plane in the experimental configuration are positive; angles below the parting plane are negative.cObserved tritium production values were generated from Ref. 1, Table III. (Tritons produced per 7Li atom in an ampule from3.815 x 1015 source neutrons.)dR. C. Little and R. E. Seamon modified the ENDF/B-V 235U cross section file so that it contained a fission spectrum whosehigh energy tail at E' = 5.0 MeV was reduced to 50% of that of ENDF/B-IV and at E' = 12 MeV was reduced to 10% of that ofENDF/B-IV. (See Ref. 8.)

APPENDIX C. Comparison of Tritium Production"Contaminated Helium

Sample

No.

348349

340341342

401407408

335339

Location

r (mm)

299.5299.5

199.5199.5201.5

125.5127.5125.5

82.283.7

w*(deg)

-20120

3515

-125

30-30135

-4055

N(Obs)c x 1013

. ExperimentJ Tritons Produced \

V 7Li Atom J0.336 ± 18%0.328 ± 20%

1.618 ± 8%1.551 ± 9%1.618 ± 9%

9.613 ± 5%14.71 ± 5%7.41 ± 7%

14.21 ± 14%14.01 ± 16%

in 7Li Samples for Induced Background

N(Obs) x 1013

(He AssumedContaminated)

/Tritons Produced

V 7Li Atom )0.343 ± 18%0.335 ± 20%

1.81 ± 8%1.74 ± 9 %1.81 ± 9 %

10.22 ± 7%15.27 ± 5%7.98 ± 7%

23.65 ± 14%23.42 ± 16%

Originating from Ampule Quartz vs.

N(Calc)d x 1013

Calculated11 Tritons Produced | Observed6

V 7Li Atom0.378 ± 12%0.240 ± 13%

1.49 ± 7%1.84 ± 7%1.42 ± 8%

7.42 ± 3%7.09 ± 3%7.13 ± 3%

30.1 ± 3%26.7 ± 3%

/ Calculated

0.889 ± 22%1.367 ± 24%

1.086 ± 11%0.843 ± 11%1.139 ± 12%

0.472 ± 14%0.525 ± 16%

(He AssumedContaminated)

Observed

Calculated

0.907 ± 22%1.396 ± 24%

1.215 ± 11%0.946 ± 11%1.275 ± 12%

0.786 ± 14%0.877 ± 16%

"The quoted uncertainties on both observed and calculated values are for one fractional standard deviation. In addition, there isan estimated <6% systematic error in the observed values which is not included in the errors quoted here.6 Angles above the parting plane in the experimental configuration are positive; angles below the parting plane are negative.cObserved tritium production extracted from Ref. 1, Table III. (Tritons produced per 7Li atom in an ampule from 3.815 x 1015

source neutrons.)dTritium production calculated using ENDF/B-V - updated where available - cross sections for both transport and tritiumproduction. (Tritons produced per 7Li atom in an ampule from 3.815 x 1015 source neutrons.)eThe quoted errors were determined from a square root of the sum of the squares combination of the observed and calculateduncertainties.

Appendix DComparison of Cross-Section Evaluations

Comparing the foil activation results from Ref. 7and this report elicits some remarks. It should benoted that in Ref. 7, the principal source of cross-section evaluations for activation reactions was theBarr-Hendricks library based on integral data from ac-tivation analysis by INC Division at Los Alamos. Withfew exceptions, agreement between observed and cal-culated values is good. Only for n,2n results on ^ Z rand l91Ir, and n, 7 on 160Tm and 45Sc (a single pointonly), do the O/C ratios deviate from unity by as muchas 20%.

In examining the 11,2n results in this paper, we seethat except for 89Y all O/C ratios show calculatedvalues that are too high by 15-40%. Uniquely, for 89Ythe calculated values are too low by some 20%.

The n,2n evaluations investigated in the calcula-tions in this paper were obtained from ENDF/B-V.Los Alamos Group T-2, and the Lawrence LivermoreNational Laboratory (LLNL) ACTL library. Thoseevaluation!: appear to have been based on microscopic(monoenergetic) cross-section measurements, whichcan be subject to background problems. Failure toremove all of the experimental background, especiallyduring the rise above the threshold, would result incross sections that are too large in magnitude; all ofthe evaluated cross sections (except for 89Y) appear,from our results, too large.

A comparison of the evaluated n,2n cross section for238U for the Barr-Hendricks and ACTL evaluations isshown in Fig. D-l. Note thp.t the ACTL curve risesmore rapidly, to a higher peak, than does the Barr-Hendricks curve. Review of extant microscopic datafrom the compilation of Garber and Kinsey10 showsthat the ACTL curve indeed follows thobe measure-ments. Because of the shape of the neutron spectrum,the difference between the two evaluations between 7and 10 MeV results in a measurable difference in O/Cratios in these experiments.

The n,7 results from Ref. 7, using Barr-Hendricksevaluations, are generally quite good. Again, the T-2 evaluation of 169Tm and ENDF/B-V for 45Sc givescalculated values that are 20-40% too high. Compari-son of the evaluated cross section for 45Sc is shown inFig. D-2.

It should be noted that in Ref. 7 the 238U n,7 evalu-ation was modified to fit the measurements of Poenitz,

Fawcctt, and Smith,11 which lowered the evaluatedcurve at high energy. Although the O/C ratios forthat reaction in this paper are a bit low at inner foillocations, results for the outer two locations are verylarge. This appears to be a glitch, as the cross sec-tion varies smoothly with neutron energy; the neutronspectrum does not change with foil location sufficientlyto produce the observed results.

Of all neutron cross sections, 235U and 238U fissionshould be weli known. ENDF/B-V evaluations gavevery good results in Ref. 7; ACTL evaluations gavean O/C ratio some 10% low in this paper for bothnuclides at all positions.

Finally, 191Ir(n,7) + 193Ir(n,2n) was well repre-sented by the Barr-Hendricks evaluations in Ref. 7.The T-2 evaluations gave O/C ratios off by a factorof 3 in this paper. The 193Ir(n,n') results in Ref. 7were in good agreement, with calculations using theBarr-Hendricks evaluation (with admittedly large un-certainties placed on the evaluated curve). The T-2evaluation gave O/C ratios off by a factor of 3 in thispaper.

In summary, evaluations of activation cross sectionsbased on integral radiochemical data give good results.Evaluations that do not include such data are generallyooor.

23

ZAID - 92238 30YFrom LLLD0S2

«« « • IOO no itg ita in »• «i

Fig. D-l . Comparison of the evaluated n,2n cross sectionfor 2 3 8U for the Barr-Hendricks and ACTL evaluations.Note that the ACTL curve rises more rapidly, to a higherpeak, than does the Barr-Hendricks curve.

ZAID = 2I045.26YFrom 532D0S2

Ntulnn Entrgy (UtV)Id

45CFig. D-2. Comparison of the evaluated cross section for Sc

24

REFERENCES

1. A, Henimendinger, C. E. Ragan, E. R. Shunk, .1.M, Anaya, and G. P. Estes, "Tritium Productionin Concentric Spheres of 235U and eLiD Irradiatedby 14-MeV Neutrons," unpublished report, revisedMay 14, 1979. Available from C. E. Ragan, LosAlamos National Laboratory.

2. J. S. Gilmore, "Report of INC-11 Results on Blue-beard II Sphere Experiment,'1 Los Alamos Na-tional Laboratory memorandum INC 11-84-308(August 1, 1984).

3. A. Hemmendinger, C. E. Ragan, E. R. Shunk, A. N.Ellis, J. M. Anaya, and Jon M. Wallace, "TritiumProduction in a Sphere of 6LiD Irradiated by 14-MeV Neutrons," Los Alamos Scientific Laboratoryreport LA-7310 (October 1978). An abbreviatedform was published in N.S.E. 70, 274-280, (1979).

4. C. E. Ragan III, G. F. Auchampaugh, A. Hem-mendinger, and M. G. Gilbert, "Neutron Spec-trum for a Uranium-235 Sphere Bombarded by 14-MeV Neutrons," Nucl. Sci. Eng. 61, 33 (1976).

5. Los Alamos Monte Carlo Group, "MCNP - A Gen-eral Monte Carlo Code for Neutron and Pho-ton Transport," Los Alamos National Labora-tory report LA-7396-M, Rev., Versions 2B and 3A,(April 1981).

6. Los Alamos National Laboratory, "Core," DrawingNo. 32Y-20935. Available from L. R. Fawcett, Jr.,Los Alamos National Laboratory.

7. L. R. Fawcett, Jr., "Reanalysis of Tritium Produc-tion in a Sphere of 6LiD Irradiated by 14-MeV Neu-trons," Los Alamos National Laboratory report LA-10506-MS, (August 1985).

8. R. C. Little and R. E. Seamon, "235U and 7Li CrossSections and Fission Neutron Spectra," Los AlamosNational Laboratory memorandum X-6:RES-86-329 (August 7, 1986).

9. R. C. Little and R. E. Seamon, "Dosimetry/Acti-vation Cross Sections for MCNP," Los Alamos Na-tional Laboratory memorandum, (March 13, 1984).Available from R. C. Little or R. E. Seamon, LosAlamos National Laboratory.

10. D. I. Garbcr and R. R. Kinsey, "Neutron Cross Sec-tions: Vol. II, Curves," BNL-325, 3rd Ed., Jan-uary 1976.

11. W. P. Poenitz, L. R. Fawcett, Jr., and D. L. Smith,"Measurements of the 238U(n,7) Cross Section atThermal and Fast Neutron Energies," Nucl. Sci.Eng. 78, 239; 1981.

25