dbrown-Future CIELO Activities€¦ · 56Fe(n,ɣ) IRSN 56Fe RRR evaluation appeared like attractive option • Higher energy, up to 2nd excited state threshold • Many more resonances

Future CIELO ActivitiesD. Brown, G. Nobre

National Nuclear Data Center, BNL

We are interested in fluctuations (esp. near closed shell nuclei like Fe) and improved direct reaction modeling (esp. for (in)elastic cross sections)

Unfinished business: CIELO Fe

Our CIELO Fe evaluation was very much driven by data, and respected previous excellent evaluations

56Fe

Froehner (1991) RR

RR

JEF-2 (1993)

JENDL-4.0 (2010)

JENDL-3.3 (2002)

RR

Perey (1988) ORNL/TM-11742 (1990)

COMMARA-2.0 (2011)

CIELO & ENDF/B-VIII.0

(2017)

FastRR High-En. Cov.

ENDF/B-VI.1 (1991) ORNL-9964 (1986)FastRR Cov.

ENDF/B-VI.8 (2001)

ENDF/B-VI.5 (1997)

ENDF/B-VII.0 (2004)

ENDF/B-VII.1 (2011)

Cov.


FastRR High-En.



High-En.

LA150 (1997)

JEFF-3.2 (2014)

JEFF-3.1.1 (2009)

JEFF-3.1 (2006)

JEFF-3.0 (2005)

EFF-3.1 (2001) FastRR Cov.

FastRR Cov.

FastRR Cov.

IRDFF-v.1.05 (2014)

Fast Cov.




EFF-3.0 (1998) FastRR Cov.

JEF-1 (1989)

FastRR

Allen (1976); Kaeppler (1983); Wisshak (1984); Corvi (1991)

Atlas (1981) RR

Corvi (1988); Gayther (1988); Sowerby (1988); Liou (1979)

Froehner (1977) RR

Thousands of datasets, we could not get through it all. Used history to guide us.

•Better Resonances • LRF=7 option for 56Fe • The low energy background (from 10 to

100 keV) in 56Fe capture • EGAF thermal capture cross section for 56Fe • Elastic angular distribution on 56Fe

• Fusion cross section between elastic and inelastic in the energy range from 4 to 8 MeV

•Minor Fe covariances, esp. in RRR • Other steel constituents (Cr, Ni)







LRF=7 resonances needed

Legacy RRR evaluations (Froehner’s and hence JENDL-4.0, Atlas and ours) use LRF=3 format.

LRF=3 format uses Reich-Moore approximation, but channels limited to capture, elastic and fission

First and second excited states show fluctuations: we need resonance treatment

Angular distributions can be computed from RRR data, if they are trustworthy

56Fe(n,ɣ)

IRSN 56Fe RRR evaluation appeared like attractive option

• Higher energy, up to 2nd excited state threshold • Many more resonances

56Fe(n,ɣ)

IRSN 56Fe RRR evaluation appeared like attractive option

• Higher energy, up to 2nd excited state threshold • Many more resonances

Gripes about IRSN evaluation: • Resonances shifted from Atlas, ours, & J4.0

(aggressive use of ToF correction) • Not using all available data, focusing only on

ORNL measurements • Poor reproduction of MT51

(resonance JΠ assignments?) • Poor reproduction of angular distributions

(resonance JΠ assignments?) • Missing capture resonances • Given time constraints were unable to resolve

ENDF Request 6268, 2017-Oct-31,15:44:32

Incident Energy (MeV)

Cro

ss S

ectio

n (b

arns

)

10-8 2.10-8 5.10-8

1

10

0.2

0.5

2

5

20 ENDF/B-VIII.b5: FE-56(N,G)FE-572017 Firestone1977 Shcherbakov1977 Shcherbakov2017 Firestone

Firestone’s thermal capture (EGAF)

Which is it? Why? is it a 15% correction, or something worse?

ENDF Request 6268, 2017-Oct-31,15:44:32


Cro

ss S

ectio

n (b

arns

)

10-8 2.10-8 5.10-8

1

10

0.2

0.5

2

5



Firestone’s thermal capture (EGAF)

(no, lower point is primary gamma cross section, so it’s 15%)

ENDF Request 6268, 2017-Oct-31,15:44:32


Cro

ss S

ectio

n (b

arns

)

10-8 2.10-8 5.10-8

1

10

0.2

0.5

2

5



Firestone’s thermal capturePHYSICAL REVIEW C 95, 014328 (2017)

Thermal neutron capture cross section for 56Fe(n,γ )

R. B. Firestone,1,2 T. Belgya,3 M. Krticka,4 F. Becvar,4 L. Szentmiklosi,3 and I. Tomandl51University of California, Department of Nuclear Engineering, Berkeley, California 94720, USA

2Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA3Centre for Energy Research, Hungarian Academy of Sciences, H-1525 Budapest, Hungary

4Charles University in Prague, Faculty of Mathematics and Physics, V. Holesovickach 2, CZ-180 00 Prague 8, Czech Republic5Nuclear Physics Institute, Czech Academy of Sciences, CZ-250 68 Rez, Czech Republic

(Received 24 October 2016; published 30 January 2017)

The 56Fe(n,γ ) thermal neutron capture cross section and the 57Fe level scheme populated by this reaction havebeen investigated in this work. Singles γ -ray spectra were measured with an isotopically enriched 56Fe targetusing the guided cold neutron beam at the Budapest Reactor, and γ γ -coincidence data were measured with anatural Fe target at the LWR-15 research reactor in Rez, Czech Republic. A detailed level scheme consisting of448 γ rays populating/depopulating 97 levels and the capture state in 57Fe has been constructed, and ≈99% ofthe total transition intensity has been placed. The transition probability of the 352-keV γ ray was determined tobe Pγ (352) = 11.90 ± 0.07 per 100 neutron captures. The 57Fe level scheme is substantially revised from earlierwork and ≈33 previously assigned levels could not be confirmed while a comparable number of new levels wereadded. The 57Fe γ -ray cross sections were internally calibrated with respect to 1H and 32S γ -ray cross sectionstandards using iron(III) acetylacetonate (C15H21FeO6) and iron pyrite (FeS2) targets. The thermal neutron crosssection for production of the 352-keV γ -ray cross section was determined to be σγ (352) = 0.2849 ± 0.015 b.The total 56Fe(n,γ ) thermal radiative neutron cross section is derived from the 352-keV γ -ray cross section andtransition probability as σ0 = 2.394 ± 0.019 b. A least-squares fit of the γ rays to the level scheme gives the 57Feneutron separation energy Sn = 7646.183 ± 0.018 keV.

DOI: 10.1103/PhysRevC.95.014328

I. INTRODUCTION

Precise thermal neutron capture γ -ray spectra were mea-sured for all elements with Z = 1–83, 90, and 92, except forHe and Pm, using neutron beams at the Budapest Reactor[1,2]. The γ -ray energies and cross sections were determinedand combined, together with additional information from theliterature, to generate the Evaluated Gamma-ray ActivationFile (EGAF) [3] and they were also published in the Handbookof Prompt Gamma Activation Analysis with Neutron Beams[4]. These data can be used to determine total radiative thermalneutron capture cross sections, σ0. When the level scheme iscomplete, σ0 equals both the sum of transition cross sections,γ -ray plus conversion electron, feeding the ground state (GS),#σγ+e(GS), and the sum of transition cross sections deexcitingthe capture state (CS), #σγ+e(CS). Thermal neutron capturedecay schemes are typically completely determined for low-Zelements where all of the transitions are observed.

Iron is an important structural and shielding material innuclear reactors and other nuclear installations that has animportant impact on thermal neutron flux distribution ina reactor pressure vessel [5]. Despite its importance, the56Fe(n,γ ) total radiative thermal neutron cross section, σ0,is only known to an accuracy of ≈5% based on only twoearly measurements from over 40 years ago. In this work the56Fe(n,γ ) reaction has been studied with a thermal equivalentneutron beam impinging on an enriched 56Fe target. Thecorresponding 57Fe γ -ray decay scheme has been nearlycompletely determined with only minor corrections necessaryto account for the weak, missing, or unplaced γ -ray intensity.The new γ -ray data have been internally calibrated withthermal cross section γ -ray standards to determine a new value

of the total radiative thermal neutron cross section accurate to≈0.8%.

The 56Fe(n,γ ) reaction was previously studied by Venninket al. [6], who placed 191 γ rays that populated/depopulated62 levels in 57Fe. Levels and γ rays were assigned by Venninket al. on the basis of γ -ray energy sums but without the aid ofγ γ coincidence data. That procedure can be unreliable due to ahigh probability of chance energy sums matching known levelenergies resulting from the complexity of the (n,γ ) spectrum.In this work we have also exploited γ γ coincidence spectra,originally measured for studying two-step γ cascades [7] usinga natural Fe target, to confirm the placement of more than 70%of the transitions observed in γ -ray singles measurements, addnew transitions, and divide the intensities of γ rays that couldbe multiply placed in the decay scheme.

II. EXPERIMENT

The singles 56Fe(n,γ ) neutron capture γ -ray spectrum wasmeasured in the guided cold neutron beam at the 10-MWBudapest Reactor [1]. Neutrons entered the evacuated targetholder and continued to the beam stop at the rear wallof the guide hall. The target station, where both primaryand secondary γ rays can be measured in low backgroundconditions, is located 30 m from the reactor. The thermal-equivalent neutron flux at the target was 1.2 × 108 cm−2 s−1

during this experiment.Prompt γ rays from the target were measured with an n-type

high-purity, 27% efficient, germanium (HPGe) detector withclosed-end coaxial geometry located 23.5 cm from the target.The detector is Compton-suppressed by a bismuth germanate

2469-9985/2017/95(1)/014328(11) 014328-1 ©2017 American Physical Society

10-4

10-3

10-2

10-1

100

101

102

10-3 10-2 10-1

Cro

ss S

ect

ion

(b

)

Incident Neutron Energy (MeV)

56Fe(n,γ)H.Pomerance, 1952

B.J.Allen, 1976

Huang Zheng-De, 1980

O.A.Shcherbakov, 1977

O.A.Shcherbakov, 1977

ENDF/B-VII.1

ENDF/B-VIII.0

A background was added to 56Fe capture,we want to get rid of it

0

5

10

15

20

25

100 101 102 103 104 105 106

Un

cert

ain

ty (

%)

Incident neutron energy (eV)

0

2

4

6

Cro

ss s

ect

ion

(m

b) Background

ENDF/B-VIII.0

10-2

10-1

100

0 20 40 60 80 100 120 140 160

Angle (degree)

Einc = 1.30 MeV

S.G.Malmskog, 1972

10-2

10-1

100

dσ

/dΩ

(b

/sra

d)

Einc = 2.00 MeV

W.E.Kinney, 1976

ENDF/B-VII.1

ENDF/B-VIII.0

EMPIRE

10-2

10-1

100

56Fe (n,n)

Einc = 3.00 MeV

A.P.D.Ramirez, 2017

J.C.Hopkins, 1964

I.A.Korzh, 1977

A.Smith, 1980

Evaluation of Iron . . . NUCLEAR DATA SHEETS M. Herman et al.

M. Herman et al, ND2016, Bruges, Sept. 11-16, 2016

<CIELO-219>

CIELO-219JEFF-3.2

Inelastic Dupont norm. to Negret

Total CIELO-219

Elastic CIELO-219

<Elastic CIELO-219>

<Total

CIELO-21

9>

CIELO transitions to Negret

nat-Fe: (n,tot), (n,el) (n,inel)

2.0 2.1 2.2 2.3 2.4 2.5 Incident Neutron Energy (MeV)

Cro

ss S

ectio

n (b

arns

)0.

2

0.5

1.0

2.0

5.0

1

0

Consistent usage of exp. data - (i) total from Berthold (corrected for minor isotopes), (ii) inelastic - Dupont normalized to Negret up to 2.4 MeV, Negret above, (iii) elastic = total - inelastic. Resolution broadened elastic agrees with Kinney data.

CIELO

Total CIELOEla. CIELO

<CIELO>

2.0 2.1 2.2 2.3 2.4 2.5Incident Neutron Energy (MeV)

Cro

ss S

ectio

n (b

)

0.5

1.

0

2.0

5.0

>

>

FIG. 8. (Color online) Example of fluctuating structure intotal, elastic and inelastic on natFe illustrating consistent con-struction of the recommended CIELO (ENDF/B-VIII.0) crosssections. Lines indicated with <> represent energy averages.

the remaining levels.Fig. 9 compares total inelastic cross sections in the

CIELO evaluation with experimental data by Negret andNelson [141]. Both experiments derived total inelasticcross sections from the intensity of the 847 keV -linethrough which the first excited state in 56Fe decays tothe ground state. The Nelson results were correctedby our Chinese collaborators for the contribution of the57Fe(n, 2n) reaction and monitor cross section updatedin 2004. For more recent Negret measurement carried onthe enriched sample these corrections were not needed.Both experiments agree perfectly up to around 6 MeVwhile at higher energies Negret data tend to be lower.Authors consider, however, their data accurate only upto 4.5 MeV where there is no model-dependent contri-bution from the continuum. Above that limit Negretdata represent the lower limit. Our new evaluation re-produces very well Negret data up to about 7 MeV andthen follows results of Nelson. In the plateau of the in-elastic cross section no adjustment of the model param-eters was needed. Above 12 MeV EMPIRE calculations,adjusted to reproduce 56Fe(n, p) reaction, were slightlylower than measurements and other libraries. To removethis discrepancy, without upsetting agreement for otherchannels, we increased -strength function in 56Fe whichmakes more likely for ’s to win competition with neu-trons just above the threshold of the 56Fe(n, 2n) reaction.To this end, E1 -strength in 56Fe was switched from theGDR default to Weisskopf single-particle estimates withthe scaling factor set to 0.1 instead of the default 0.01.

54

Fe,

57

Fe,

58

Fe(n,inelastic) - For the three mi-nor isotopes results of EMPIRE calculations for wereadopted.

Limited experimental data, available only for54Fe(n,inelastic), indicate possibility of a fluctuat-ing structure similar to the one observed in the case

0.5

1.0

1.5

2.0

2 4 6 8 10 12 14 16 18 20

Cro

ss S

ect

ion

(b

)Incident Neutron Energy (MeV)

56Fe(n,inel)

Nelson*, 2004Negret, 2013ENDF/B-VII.1ENDF/B-VIII.0

FIG. 9. (Color online) Evaluated 56Fe(n, n0) neutron inelasticcross section compared with data retrieved from EXFOR andwith previous evaluation.

0.2

0.4

0.6

0.8

1

1.2

5 10 15 20 25 30

54Fe (n,inel)

Cro

ss S

ect

ion

(b

)


P.T.Guenther, 1986M.B.Fjodorov, 1973

ENDF/B-VII.1ENDF/B-VIII.0

FIG. 10. (Color online) Evaluated 54Fe(n, n0) neutron inelas-tic cross section compared with data retreived from EXFORand with previous evaluation.

of 56Fe (see Fig. 10). Such a pattern is expected in allisotopes of iron, at least in the even ones, as the nuclearstructure of these nuclei is similar. The data, however,are not sucient to unambiguously establish shape ofthe fluctuating cross sections, thus we resort to smoothmodel calculations. Fig. 10 shows that calculations arein fair agreement with the experimental data and thatthe new evaluation agrees with ENDF/B-VII.1 up to 6MeV and then it is consistently higher.

57Fe is a special case since the inelastic threshold lies in-side the resonance region requiring the Reich-Moore for-malism and the ENDF-6 option LRF=7. The details werealready described in Sec. II C. The cross sections for scat-tering to the first excited state in 57Fe are displayed inFig. 5.

13

High resolution Cierjacks data not used, data from Ramirez et al. came out after evaluation finished

Must improve angular distributions






0.5 0.7 2 3 5 71 100.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8ENDF/B-VII.1

ENDF/B-VIII.0

C/E

Leaking Gamma Energy, MeV

10-5

10-4

10-3

10-2

10-1

direct γ from 252Cf(s.f.)γγ from54Fe(n,γ)

γ from56Fe(n,γ)

ENDF/B-VIII.0

ENDF/B-VII.1

Fe sphere (R/r=30/1cm) +252Cf(s.f.)

γ−le

akag

e, 1

/MeV

/n-S

ourc

e

0.03 0.3 30.01 0.1 1 10 200.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

σtot(natFe): VIII.0 - VII.1

ENDF/B-VII.1 ENDF/B-VIII.0

C/E

n-L

eaka

ge

Neutron Energy, MeV

10-5

10-3

10-1

101

ENDF/B-VII.1 ENDF/B-VIII.0

σtot(natFe)

ENDF/B-VIII.0ENDF/B-VII.1

Fe-sphere (R/r=30/1cm) +252Cf

n-Le

akag

e, 1

/MeV

/n-S

ourc

e

252Cf TFNS

0.2

1

10

50

-400-2000200

σtot(natFe

): VI

II.0-

VII.1

, mb

σtot(natFe

), b

252Cf(sf) source in Fe sphere

0.5

1.0

1.5

2.0

2 4 6 8 10 12 14 16 18 20

Cro

ss S

ect

ion (

b)


56Fe(n,inel)

Nelson*, 2004Negret, 2013ENDF/B-VII.1ENDF/B-VIII.0

We followed Nelson, maybe we should have followed Negret or split difference

18

56Fe(n,n’ɣ) powerful test of many things

19

56Fe(n,n’ɣ) powerful test of many things

• CC, DWBA • Level schemes (gammas & energies) • Competition with other channels • LD far above discrete levels






Correlation for 56Fe(n,γ)

10-3 10-2 10-1 100 101 102

Energy [MeV]

10-3

10-2

10-1

100

101

102

En

erg

y [M

eV

]

-100

-50

0

50

100

Must adapt RRR covariance code for Reich-Moore approximation

Our workaround for 56Fe only worked because Atlas==JENDL-4.0






103

104

105

106

107

E (eV)

10-4

10-3

10-2

10-1

100

101

σ(n

,γ)(E

) (b

)

ENDF/B-VIII.0L.Coquard, et al. (2006)

Allen et al. (1976)

grouped EMPIRE

grouped proposed

54Fe(n,γ)

Background

10-4

10-2

100

102

10-8 10-6 10-4 10-2 100 102

54Fe (n,γ)

Cro

ss S

ect

ion (

b)


EXFORENDF/B-VIII.0ENDF/B-VII.1

Questions about 54Fe capture background determination

Need more capture data above 500 keV

Fe window

Cr, Fe Sensitivities for HMI-001

Green=Cr(n,el), Pink=Cr(n,g)

Fe window

Green=Ni(n,el), Red=Ni(n,g)

Ni, Fe Sensitivities for HMI-001

Fe window

Blue=Al, Cyan=Ni, Red=Cr

Elastic Sensitivities for HMI-001

suspicious disagreement…

53Cr(n,ɣ)

Anomaly in Cr(n,el) SAD

50Cr

54Cr53Cr

52Cr

50Cr

54Cr53Cr

52Cr

• All isotopes have same distribution, taken from ENDF/B-V natCr

• Natural SAD built from isotopic SADs • If SAD is smooth, is OK to replace

isotopic with natural SADs

THIS IS NOT THE CASE

Anomaly in Cr(n,el) SAD

Documents

dbrown-Future CIELO Activities€¦ · 56Fe(n,ɣ) IRSN 56Fe RRR evaluation appeared like attractive option • Higher energy, up to 2nd excited state threshold • Many more resonances