Nuclear data for accelerator-driven systems

Nuclear data for accelerator-driven systems

Progress in Nuclear Energy, Vol. 38, No. I - 2 , pp. 179-219, 2001 © 2001 Published by Elsevier Science Ltd. All rights reserved Printed in Great Brit...

2MB Sizes 1 Downloads 57 Views

Progress in Nuclear Energy, Vol. 38, No. I - 2 , pp. 179-219, 2001 © 2001 Published by Elsevier Science Ltd. All rights reserved Printed in Great Britain 0149-1970/01/$ - see front matter

Pergamon www.elsevier.com/locate/pnucene

PII: S0149-1970(00)00102-5

N u c l e a r D a t a for A c c e l e r a t o r - D r i v e n S y s t e m s M.B. Chadwick, H.G. Hughes, R.C. Little, E.J. Pitcher, and P.G. Young University of California, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA

Abstract

We review recent evaluations of neutron and proton reaction cross sections up to 150 MeV in the LA150 Library, for use in computer code simulations of accelerator-driven systems. An overview is provided of the nuclear theory together with measured cross section data used in the evaluations. The possible use of bismuth activation foils for high-energy neutron spectrometry is also discussed. We describe recent developments to the MCNPX radiation transport code, which merges MCNP and LAHET in one code and uses the LA150 evaluated data. A number of benchmark comparisons against integral experiments are described, for thick-target neutron production, neutron transmission through macroscopic slabs, and neutron kerma coefficients. These benchmarks help validate the transport code and the evaluated data for use in ADS simulations of neutron production in a spallation target (n/p), radiation shielding, heating, and damage. A brief summary is also given of future data needs for subcritical transmuters an~d spallation targets, in accelerator transmutation of waste technologies. © 2001 Published by Elsevier Science Ltd. All rights reserved.

1

Introduction

There are research and technology development programs underway in many countries, including the United States, Europe, Asia, and Russia, to design high-powered accelerator-driven systems (ADS). These ADS technologies make use of moderated neutrons produced from protons of approximately 1 GeV incident energy on a spallation target. One area of interest is to use an accelerator-driven system to provide an external neutron source for a subcritieal reactor, or "transmuter", in order to transmute long-lived radioactive waste into shorter-lived products. Such accelerator transmutation of waste (ATW) systems may also address other goals, such as the production of energy, and the destruction of excess weapons-grade plutonium. ADS are also being studied in the U.S. as a technology to produce isotopes, including 179

180

M. B. Chadwick et al.

tritium - indeed, the Accelerator Production of Tritium (APT) project has been a major supporter of recent research and development in accelerators, nuclear data, computer simulation capabilities, and material studies. Research is also underway in many countries to develop new-generation spallation neutron sources for materials research. This review describes nuclear data needed for accurate simulations of the spallation process, including data needed for design simulations involving shielding, radiation heating and damage. Data needed for calculations of the neutronics, criticality, and transmutation rates within an ATW transmuter are also discussed. We describe developments to the M C N P X radiation transport simulation computer code (Hughes e t al. , 1999). Accurate code simulations facilitate the materials and geometry choices for the spallation target and the transmuter. Because of the high cost of performing integral experiments to test the spallation target and transmuter designs when subject to a high-current high-energy proton beam, it is essential that various designs are first tested and compared in benchmarked computer simulations. In Section 2 we describe the LA150 Library, which consists of evaluated reaction cross sections and emission spectra up to 150 MeV for incident neutrons and protons, for over 40 target isotopes important in spallation targets, structural materials, and shielding. The evaluations are based on nuclear model calculations using the.GNASH code (Young e t al. , 1998), which applies Hauser-Feshbach, preequilibrium, and direct reaction theories to compute the nonelastic cross sections, and elastic scattering distributions obtained from the optical model. Experimental data were used in the evaluations to guide and validate the theoretical predictions, and in some cases (such as the neutron total cross section, and the total nonelastic cross sections) the evaluations were based largely on measured data. These data have been documented in an extensive article (Chadwick e t al. , 1999b), and have been incorporated into the U.S. nuclear data library E N D F / B - V I , as Release 6, and are therefore available through the National Nuclear Data Center at the Brookhaven National Laboratory (Dunford, 1998). Section 2 also gives illustrative examples of the nuclear cross sections in the LA150 data library. Our aim is to emphasize the need for accurate nuclear reaction theories to provide nuclear data for ADS design. The M C N P X radiation transport code is describe in Section 3. It is based on a merger of the M C N P code (Briesmeister, 1997) and the L A H E T intranuclear cascade code (Prael & Lichtenstein, 1989) and is able to make use of the new LA150 evaluated data tables. We give a number of integral benchmark examples in Section 4 of how accurate nuclear data can have a significant impact on quantities of practical importance in ADS system, such as neutron production and neutron transmission, shielding requirements, and radiation heating and damage. In this paper the simulations used a developmental version of M C N P X 2.1.6. A number of comparisons are made between measurements and M C N P X simulations that use two different sets of calculations - those using the LA150 data below 150 MeV, and those using L A H E T physics modules above 20 MeV for neutrons, and for all energies for protons.

Nuclear data for accelerator-driven systems

181

As discussed in a recent review (Van Tuyle & Beller, 1999), ATW systems in the U.S. are being considered to complement a geological repository for spent fuel (Bodansky, 1996). By separating plutonium, minor actinides, and long lived technetium and iodine fission products from the spent fuel, and transmuting these materials using ATW, some important concerns are addressed: plutonium is no longer available for unauthorized nuclear weapon proliferation; the inventory of long-lived radioactive materials is largely reduced; energy can be extracted from the plutonium and minor actinides; and technetium and iodine, which may be transported away from the repository through ground water movement, can be destroyed. Future nuclear data improvements needed for ATW are discussed in See. 5. At present the LA150 library does not include evaluated data for actinide targets important in ATW, though we hope that such developments will be supported at Los Alamos in the future. Preliminary work on higher-energy actinide calculations and evaluations, notably for the uranium isotopes, has begun at a number of laboratories (Young et al. , 1990; Maslov et al. , 1998; Ignatyuk et al. , 2000, in press)

2

LA150 neutron and proton cross sections

A suite of evaluated reaction cross section files collectively known as the LA150 Library has been developed (Chadwick et al. , 1999b) in support of accelerator-driven systems design. These evaluations are in ENDF-6 format, and have recently been accepted into the U.S.-standard ENDF/B-VI Library as Release-6. For incident neutrons, they extend the previously-existing ENDF/B-VI information from 20 MeV up to 150 MeV. For incident protons, the files extend from 1-150 MeV. To date, evaluations have been completed for isotopes of the following structural, shielding, and target-blanket materials: H, Li, C, N, O, A1, Si, P, Ca, Fe, Ni, Cr, Cu, Nb, W, Hg, Pb, and Bi. These LA150 evaluated cross sections are designed for use in radiation transport codes, such as the MCNPX code described in Section 3. The primary motivation for using these evaluated data is the accuracy improvements that one can expect to obtain in the below-150 MeV energy region. In most previous transport simulations, intranuclear-cascade methods have been used for neutrons above 20 MeV and for protons at all energies, even though the semiclassical assumptions inherent within such models do not hold at lower energies. By developing evaluated cross section libraries up to 150 MeV, using state-of-the-art nuclear reaction models as well as experimental data, one can expect to have the most accurate possible representation of the nuclear cross sections. Section 4 provides some examples that illustrate the accuracy gains accompanying the use of these data.

2.1

Cross s e c t i o n s r e p r e s e n t e d

The LA150 evaluations represent the nuclear interaction cross sections in terms of inclusive emission (or "production") spectra of the ejectiles (both light particles,

182

M. B. C h a d w i c k et al.

gamma-rays, and heavy nuclear recoils). This means that all reaction channels corresponding to separate exclusive channels that lead to a given ejectile of interest are summed into one inclusive spectrum. Thus, for instance, the inclusive neutron production cross section in proton-induced reactions, written as (p, ccn), is composed of contributions from the exclusive channels (p, ln), (p, 2n), (p, np), etc. This approach was adopted because of the fact that, as the incident energy becomes higher, the number of possible exclusive channels increases dramatically and it becomes impractical to represent them separately in an ENDF file 1 Additionally, for most applications this approach is perfectly adequate. The only case where problems might arise from having collapsed the detailed exclusive cross sections into inclusive cross sections is if correlation information is needed between the ejectiles, such as in some simulations of detector response. In ENDF format parlance (McLane, 1997b), the above approach is referred to using the reaction designator MT=5. The total nonelastic cross section as a function of incident energy is tabulated in File-3, MT=5. File-6 then contains, for each ejectile type emitted via nonelastic processes, a yield (multiplicity) and an emission energy spectrum (both as a function of incident energy). Information to determine angular distributions is also provided (Kalbach, 1988). This procedure was first used at Los Alamos for a suite of data evaluations developed in the late 1980s (Young et al. , 1990), and is also recommended by the Nuclear Energy Agency Subgroup 13 on intermediate-energy nuclear data (Koning et al. , 1998a; Koning et al. , 1998b). The cross sections and angular distributions for elastic scattering are also provided, as a function of incident energy. For incident protons, the elastic scattering angular distribution is only specified above a cut-off angle (usually a few degrees) since the Coulomb force causes a divergence at zero-degrees. Compared to previous nuclear cross section evaluations at higher energies (Young et al. , 1990; Chadwick et al. , 1996; Chadwick & Young, 1996; Alsmiller et al. 1986; Pearlstein, 1993), the LA150 data are more comprehensive in that they include information on all possible ejectile types, including heavy recoils. This is important since it allows calculations of radiation heating and damage. Because of their high LET, the heavy recoiling nuclides play a particularly important role in radiation damage, single-event-upsets in microelectronics (Chadwick & Normand, 1999), and relative biological effectiveness (RBE) in radiation therapy. To make the ENDF data available for a transport code such as MCNPX, they must be processed by the N JOY code (MacFarlane & Muir, 1994). This code has undergone significant development in recent years to support the use of LA150 data in transport simulations of accelerator-driven systems. 1For example, at 150 MeV the followingexclusive channels represent a small subset of all reaction channels that may be open for neutron-induced reactions: (n, 6nc~), (n, 4n2t), (n, 5ndt), (n, 6npt), (n, 6n2d), (n, 7npd), (n, 8n2p).

Nuclear data for accelerator-driven systems

2.2

Nuclear

theory

and modeling

183

methods

The bulk of the data in the LA150 Library have been generated from nuclear model calculations using the GNASH code (Young e t al. , 1998), though these calculations have been extensively benchmarked and validated through comparisons with measurements (see below). The nuclear models used are based on theoretical approaches that are appropriate for the energies in the few-MeV to 150 MeV range: the Hauser-Feshbach compound nucleus theory; preequilibrium calculations based on the Feshbach-Kerman-Koonin theory (Feshbach e t al. , 1980) or the exciton model; direct reactions calculated from the optical model using collective excitation form factors; and elastic scattering fi'om the optical model. The GNASH code was demonstrated to be one of the most accurate codes available for model calculations below 150 MeV in a Nuclear Energy Agency-code intercomparison (Blann e t al. , 1994). The GNASH code models the sequential emission of particles and gamma-rays using the aforementioned nuclear reaction mechanisms, until a final residual nucleus attains its ground, or isomeric, state by gamma-ray or particle emission. A recent development has been a capability to compute the recoiling'energy of the product residual nucleus (Chadwick e t al. , 1999b). This is done by the GSCAN code (Chadwick & Young, 1999) which uses a GNASH output, converts it into ENDF-6 format, and additionally performs the kinematical calculations needed to determine the energy-dependent spectrum of the heavy recoils. The optical model is used for predictions of the total, reaction, and elastic scattering cross sections, making use of nucleon potentials at higher energies developed by Madland (Madland, 1988), Chiba, and Koning. It is particularly useful for accurately representing the angular distributions in elastic scattering, allowing more accurate neutron transport simulations. (Many previous intranuclear cascade transport codes instead represent elastic scattering using a black-disc diffraction formula or use even simpler approaches, which poorly approximate reality below a few hundred MeV.) The optical model also provides transmission coefficients for use in the Hauser-Feshbach compound nucleus theory, and, when using a Distorted-Wave Born Approximation or coupled-channel formalism, it is used to predict the cross sections for inelastic scattering to low-lying excited states in the target nucleus. For neutrons and protons incident on light nuclei such as hydrogen and deuterium, R-matrix scattering theory and phase-shift analyses are applied, and extensive use is made of experimental data. For hydrogen, G.M. Hale produced a new ENDF neutron evaluation that extends up to 150 MeV as part of the LA150 Library. Like the previous ENDF/B-VI version, the evaluation uses R-matrix parameterizations below approximately 26 MeV. At higher energies, though, this evaluation is smoothly merged onto Arndt's VL-40 phase-shift analysis, as recommended by the Nuclear Energy Agency and IAEA's standards group (Condd, 1992). Hydrogen is particularly important to model accurately due to its abundance in water, plastics, and human tissue. Its light mass results in a high fraction of a neutron's kinetic energy being transferred to the recoiling hydrogen nucleus in n-p elastic scattering, making it a

184

M. B. Chadwick et al.

7.0

n+Pb total cross section Evaluation ¢..-

+ Finlay (1993) Lisowski (1980) Larson (1980) o Complied data, from the NNDC

6.0

._o ffl

m 5.0

e0 ~ 4.0

F--

3.0

L_ 2O

40 60 80 100 120 Incident neutron energy (MeV)

140

Figure 1: Evaluated neutron total cross section for lead compared with measurements compiled at Brookhaven (Finlay et al. , 1993; McLane, 1997a). crucial contributor to neutron energy deposition in hydrogenous materials. Furthermore, the n-p scattering cross section at 180-degrees plays a key role as a "standard" cross section for neutron fluence determinations in experimental neutron cross section work.

A detailed description of comparisons between the LA150 data and measurerments has been given (Chadwick et al. , 1999b). Below, we present a selection of examples that illustrate these cross section data, emphasizing reactions on important spallation target and structural materials.

2.3

Total and elastic s c a t t e r i n g cross s e c t i o n s

An illustrative example of the neutron total cross section is shown in Fig. 1 for lead. The measured data shown in the figure were obtained from an experimental program at the Los Alamos Weapons Neutron Research facility, which obtained high-precision neutron total cross sections from 6 - 600 MeV for a wide range of target nuclei (Finlay et al. , 1993; Dietrich et al. , 1997). The calculated elastic scattering cross section in the LA150 Library for neutrons on lead is shown in Fig. 2 for energies ranging from 20 to 136 MeV. A coupled channel optical potential was used in these calculations. The diffraction patterns predicted by the optical model are in good agreement with the measurements, though at 65 MeV there is an indication that the integrated elastic scattering cross section may be underpredicted. A recent paper (Ibaraki et al. , 1999) presented new measurements at the TIARA facility of neutron elastic scattering in the 40 - 90 MeV region on carbon,

Nuclear data for accelerator-driven systems

185

silicon, iron, zirconium and lead, and made comparisons with the LA150 data. Again, agreement was generally good, though there was an indication of an underprediction of the elastic scattering cross section in this region.

107

IO'

n+Pb elastic scattering

10s 10'

t,,)

:

~ DeVito (1980), xO.001

lo'

vE 102 ~ (.-

o

Calculation (1983) FFinla~((19831:x01101

• Finlay

~

.

~

)

'

xO-O011

120.0

150.0

10~

o 1o0 o

10-2 10"3 10-4 i0 -s

0.0

30.0

60.0 c.m.

90.0

angle (deg)

180.0

106

lo'

n+Pb, elastic scattering ---"~_ ~ Calculation ,o' ~ ~. Hjort (1990) "Q ~, ~ ~ ~ Bratenahl (1950) l°3 ~ . ~ " ~h, 6 5 M e V Salmon(1960) E ~ ' ~ % _ ~ ~ Van (1956) .~0 102

°\, 10'

k,f'~

(-~ 10° ,o,

~

~

MeV

\ 96 MeV

.

136 MeV \

~°-~

(xo.ool). 0.0

v-~ 30.0

~_ ~

.

-~,. . 60.0 c.m. angle (deg)

, 90.0

120.0

Figure 2: Evaluated neutron elastic scattering cross section for 2°8pb compared with measurements (McLane, 1997a).

186

M. B. Chadwick et al. 104

~- 103

!

26 MeV2°gBi(~,xn)

[%

--

!

Marcinkowski (1983) GNASH calc.

i

Equilibrium ~k~ emission

..O

!

E

Pree:uilib[i:nm

102

Direct reactions

101 0

5

10 15 20 Neutron emission energy

25

Figure 3: Measured secondary angle-integrated neutron spectrum for 26 MeV neutrons on bismuth compared with GNASH calculations used in the LA150 Library. The peaks in calculations at the highest energies, due to direct reaction processes, would be in better agreement with the measurements if they were spread using the experimental detector resolution. 2.4

Neutron

multiplication

i n (n,

xn) r e a c t i o n s

Neutron multiplication reactions play an important role in spallation targets, and when the nucleon energies fall below 150 MeV much of the neutron production comes fl'om (n, xn) rather than (p, xn) reactions. This is simply because the probability that protons below 150 MeV stop before having a nuclear reaction is significant. Because of this, we have placed a large emphasis on accurately modeling the (n, xn) reactions. An example is shown in Fig. 3 for the 26 MeV 2°gBi(n, xn) angle-integrated secondary neutron spectrum, where measurements exist (Marcinkowski et al. , 1983). The figure shows the components of the calculated spectrum due to compound nucleus emission, preequilibrium emission, and direct reactions. The direct reactions were calculated using Distorted-Wave Born Approximation theory. Collective excitations of excited vibrational states were treated using the weak-coupling model, with a h9/2 proton coupled to a 2°8pb core. This allows extensive detbrmation-length information from direct reactions on 2°sPb to be used in the bismuth calculations. In this particular example, the neutron yield at 26 MeV is calculated to be 2.54; i.e. for every neutron inducing a reaction, 2.54 neutrons are emitted on average. This quantity is only predicted accurately if the calculated neutron emission spectrum is accurately modeled. The good agreement seen between calculation and experiment in Fig. 3 suggests that neutron multiplication is accurately treated in the LA150 Library.

Nuclear data for accelerator-driven systems

187

10~ 10' 103

-

},...

>

10 2

-

113 MeV 2°aPb(p,xn) GNASH calculation o Exp., Meier (1989)

=oo S°Ix2°°°)

10 ~

oo

10 0

•.~ I0-~ LU 13

1 0 -2

%~ 10-3 1 0 -4

10 -5

0

20

40

60

80

10(

120

Neutron emission energy (MeV)

Figure 4: Evaluated 2°SPb(p, xn) double-differential neutron emission spectra compared with experimental data (Meier e t al. , 1989) at 113 MeV incident energy.

2.5

Proton-induced neutron production

A number of experiments have measured the proton-induced neutron production spectra from thin targets of lead and bismuth, in the < 150 MeV region. Below, we compare our calculated emission spectra in the LA150 Library with measurements from Los Alamos at 113 MeV (Meier e t al. , 1989), and from the University of Maryland at 90 MeV (Kalend et al. , 1983; Wu e t al. , 1979). When a proton at these energies impinges on a heavy target material such as lead or bismuth, one or more high-energy preequilibrium proton or neutron ejectiles are likely to be emitted in the early stages of the reaction. Following this preequilibrium phase, an excited compound nucleus is produced from which sequential equilibrium decay occurs until the available excitation energy is exhausted. These evaporation particles are almost exclusively neutrons with energies in the few-MeV range, since the large Coulomb barrier strongly suppresses charged-particle emission. Figure 4 shows the calculated 113 MeV Pb(p, xn) neutron energy spectra at various angles, compared with data (Meier e t al. , 1989). Agreement is good except at the backward angle where the calculations underpredict the experiment in the preequilibrium region. At low energies the large evaporation peak from sequentially-emitted compound nucleus neutrons is evident. Above about 10 MeV, the spectra have a hard preequilibrium tail that extends up to the incident energy minus the Q-value. These preequilibrium neutrons are seen to be strongly forward-peaked, whereas the evaporation neutrons are approximately isotropic. For 90 MeV protons on bismuth, Fig. 5 shows comparisons between the angle-

188

M. B. Chadwick et al. 10 3

MeV p+2°°B, - ......

> (D

10 2

E ~ o/ 1~

"

(p,xn) GNASH calculation (p,xp) GNASH calculation • 2°gBi(p,xn) data, Kalend (1983) o 2°gBi(p,xp) data, Wu (1979)

L" - ~ - ' ° ' - - - ~° ~T---..O. . / . . . . . .". . . .

o -~. o /i/i

10 ~

fO ID if} ffJ

if i

10 o

o

i

l t

i

10-1

i t

~ . 0.0

, 20.0

._L

~_ 40.0

.J 60.0

,

, 80.0

Emission energy (MeV)

I 100.0

Figure 5: Evaluated 2°9Bi(p, xn) and (p, x p ) cross sections compared with angleintegrated data at 90 MeV incident energy (Kalend et al. , 1983; Wu et al. , 1979). integrated neutron and proton spectra, compared with measurements (Kalend et al. , 1983; Wu et al. , 1979). Good agreement is seen between theory and measurement. The suppression of low-energy compound nucleus proton emission, relative to neutron emission, is evident. Figs. 4, 5 also suggest that the calculated total neutron production cross section, obtained by integrating the calculated spectra over all emission angles and energies, is expected to be fairly accurate. This is important in transport simulations, where the total neutron production from a beam of stopping protons incident on a spallation target is a key design quantity that drives the overall neutron economy. If the calculations had, for instance, over-predicted the high-energy preequilibrium neutron spectra, then energy-balance would dictate that the compound-nucleus low energy spectrum would be largely underpredieted, and therefore the total neutron production would be underpredicted. These kinds of considerations demonstrate the importance of accurately modeling the thin-target neutron production spectra.

2.6

B i s m u t h for high-energy n e u t r o n s p e c t r o m e t r y

If high-powered accelerator-driven facilities are built, it will be useful to have a diagnostic capability to infer information about the neutron spectrum at various locations. This will be important for two primary reasons: (1) much of the ADS design will be based on computational simulation tools, and such measurements can be used to validate the code's predictions of the neutron energy fluence; (2) For radiation protection it will be important to understand the neutron fluence spectrum - for ex-

Nuclear data for accelerator-driven systems

189

ample, the spectrum of neutrons that penetrate shielding materials. Experimental nuclear physics methods that are commonly employed in laboratory conditions, such as neutron time-of-flight, cannot be used in such conditions. However, activation foils of various materials can be used to infer the shape of a neutron spectrum. Radiochemical activation techniques for neutron spectrometry work in the following way. When neutrons interact in the target material, nuclear reactions such as (n, 7), (n, n') to a metastable state, (n, 2n), (n, p) etc. may lead to a product nucleus that is radioactive. Each of the excitation functions for these reactions has a different threshold, shape, and peak. For instance, the capture cross section is large at neutron energies below an MeV, whereas the (n,2n) excitation function may peak at about 10-15 MeV. Therefore, by irradiating an activation foil and subsequently measuring the amounts ofvarious radioactive nuclides produced, one can infer information about the neutron spectrum shape. To make use of this method, the radioactive product nuclei must have half-lives suitable for measurement after an irradiation, and the cross sections for production of the nuclei must be known. Clearly, the inference of a neutron spectrum shape from such an activation measurement is an inverse problem, which generally does not have a unique solution due to cross section uncertainties. But even with these uncertainties the approach provides valuable information. For neutrons with energies below 20 MeV, many radiochemical detectors have been used in nuclear technologies for inferring fission and fusion neutron spectra. However, high-energy neutron spectrometry presents a new difficulty. The problem is that, until a recent measurement program was completed in Japan, there were no materials for which the higher energy activation cross sections were known with precision. This is because few experimental facilities exist to produce quasi-monoenergetic neutron sources over a large range of energies, to measure the excitation functions, and ideally one would not want to rely exclusively on theory predictions. Recently, though, measurements have been made for 2°gBi(n,xn), x=3-12, for neutrons with energies up to 150 MeV (Kim et al. , 1998). The measured excitations are shown in Fig.6 compared with GNASH model calculations from the LA150 evaluation. These measurements have been tremendously important for developing a high-energy neutron spectrometry capability, and it is satisfying to see such a good agreement with the theoretical predictions. It is evident that the various excitation functions peak at increasing energies as the neutron emission multiplicity increases, providing an ideal opportunity for neutron spectrometry up to 150 MeV.

190

M. B. C h a d w i c k et al.

10'

104

10'

~Bi(n,3n)

,

~

2~Bi(n,5n)

Bi(n,4n)

3c 103

"~ 103 oc o

o

o 102 ~3

8 10= Kim (1998) GNASH calc



- 10'

50 1O0 Incident neutron energy (MeV)

0

I __ --

150

104

10'

103

~

o 102

~3

10 ~

80 100 Incident neutron energy (MeV)

150

104

50 100 Incident neutron energy (MeV)

10 ~

150

'1

• Kim (1998) - GNASH calc.

10~

g

159

2°~Bi(n,8n)

•1 ~'E 10a i

8

5

50 1O0 Incident neutron energy (MeV)

15(3

• Kim (1998) - -

GNASH calc

1(32

10 ~

50 1O0 Incident neutron energy (MeV)

150

104

20913i(n,10n)

• Kim (1998) - GNASH calc

__

50 1O0 ~ncident neutron energy (MeV)

10'

103

.

10~

102

~'e~Bi(n,gn)

3g

O

2°9Bi(n,7n)

• Kim (1998) - GNASH t a l c

• Kim (1998) GNAS H c alc

101

10'

~Bi(n,6n)

_. ~

- -

2°gBi(n, 11 n)

• Kim (1998) GNASH

calc.

~ 1(33

', ~o ~

- -

• Kim (1998) GNASH calc.

102

10'



,

50 1O0 Incident neutron energy (MeV)

150

10 ~

50 100 Incident neutron energy {MeV)

150

50 100 Incident neutron energy (MeV)

150

Figure 6: Measured excitation functions (Kim et al. , 1998) for 2°9Bi(n, z n ) , x=3-11, for neutrons with energies up to 150 MeV, compared with GNASH calculations that are used in the LA150 Library.

Nuclear data for accelerator-driven systems

3

MCNPX

191

code development

The associated development of an enhanced particle transport code for high-energy applications has also been the focus of a major effort at Los Alamos. Primarily supported by the APT program, the MCNPX Monte Carlo code is intended to combine the capabilities of several major transport codes into a unified, coherent whole, and to extend both the physics models and the available nuclear data to support a consistent prediction of the coupled particle cascade. MCNPX incorporates the essential features of the MCNP code (Briesmeister, 1997), all of the basic LAHET (Prael & Lichtenstein, 1989) nuclear physics modules, improvements to multiple-scattering models for charged particles, the recently developed Cascade-Exciton Model (Mashnik e t al. , 1998; Mashnik & Sierk, 1999), a new photonuclear-reaction capability (White e t al. , 1999), several new tallying (Snow, 1998) and variance reduction features useful for practical applications, and consistent use of the LA150 libraries for incident neutrons and protons. A recent review (Hughes e t al. , 1999) of the status of MCNPX provides references for most of the physics models and many of the new features of the code. MCNPX has evolved into a significant computational tool for accelerator-related applications, and has recently been released to the Radiation Safety Information Computational Center (RSICC) at Oak Ridge National Laboratory, from which the code is generally available to the transport community.

4 4.1

B e n c h m a r k s o f L A 1 5 0 d a t a in M C N P X n / p in spallation targets &: nuclear reaction physics

The total number of neutrons per proton (n/p) produced in a spallation target is a crucial quantity in ADS systems, since it governs the overall economy of the system. In this subsection we present an example of how neutron inelastic scattering physics, computed using theories of Distorted-Wave Born Approximation excitation of nuclear collective states, can have an impact on the calculated n/p. Our aim is to show that detailed modeling of the nuclear reactions is important, and that nuclear reaction physics matters! In the course of spallation target design, it has been useful to study the impact of different physics transport and cross section assumptions by using a simple idealized problem. This artificial model consists of a 1 GeV beam of protons incident on an infinite tungsten spallation target, and computing how many thermalized neutrons are finally produced per incident proton. When MCNPX was used using LAHET-physics (implementing the Bertini intranuclear cascade model together with multistage preequilibrium) the answer was n/p=34.33 with a l-sigma confidence interval of 0.14%. However, when the neutron and proton LA150 tungsten data are used within MCNPX to model all neutron interactions below 150 MeV, this result decreases by 3.6~0 to n/p=33.17. (We note that most of this decrease is due to the use of n e u t r o n LA150 data tables, since when only neutron LA150 data were used and protons below 150

192

M. B. Chadwick et al.

MeV were modeled with LAHET physics a similar value of 33.11 was obtained.) At first sight, this decrease might seem confusing because below 150 MeV, the overall neutron nonelastic cross section (reaction cross section) is higher in the LA150 Library than calculated using LAI-IET physics since LA150 uses a more accurate optical model for tungsten (Prael & Chadwick, 1997). However, this fact is irrelevant in our infinite-target idealized problem since the total cross section determines the mean-free-path between collisions, and in an infinite medium a nuclear collision will eventually occur no matter what the magnitude of the reaction cross section. Instead, what is most important in determining the total neutron production, n/p, is the neutron multiplication that occurs in (n, z n ) reactions. The neutron yield, or multiplicity, in a collision is given by the total neutron production cross section divided by the nonelastic cross section. It is this quantity that governs neutron multiplication, and calculated values from LA150 and from LAHETphysics are shown in Fig. 7 (upper graph). The lower values for LA150 compared to L A H E T is the reason that the infinite spallation target n / p drops from 34.33 to 33.11. Therefore, we have to understand why the calculated data in LA150 have lower neutron multiplicities for tungsten. This can be understood by considering the spectrum of secondary neutrons calculated in the two approaches. At 26 MeV incident energy, W(n,xn) neutron inelastic scattering data have been measured (Marcinkowski et al. , 1989), and so it is at this energy that we make the comparison, see Fig. 7 (lower graph). The LA150 evaluated results are seen to be considerably higher than the LAHET-physics results at the higher emission energies, and are in better agreement with the measurements. This harder spectrum implies that the low-energy emission cross section from evaporation particles must be lower for LA150, and the overall neutron multiplicity lower, because of energy balance. The reason that the LA150 Library has a harder spectrum at the highest outgoing energies is that it includes a calculation of direct reaction processes where the neutron inelastically scatters off the nucleus, exciting giant-resonance and rotational collective states (Mareinkowski et al. , 1996; Chadwick et al. , 1999b). Finally, we note that a similar result is obtained for an idealized problem of an infinite lead target bombarded by 1 GeV protons. When LA150 neutron and proton data are used in MCNPX, the calculated value of n / p is 32.59 a value that is 2.8% lower than the L A H E T result of 33.516. Again, this is due to lower multiplicities for (n, zr~) neutron production in our GNASH calculations compared to LAHET. The implication of these reductions in n / p from an applications perspective is that the proton beam current delivered to the target must be increased by an amount inversely proportional to the drop in neutron production yield (n/p) in order to obtain the desired neutron production rate from the spallation target. Alternatively, the proton beam energy could be increased.

Nuclear data for accelerator-driven systems

n+~W neutronyield (multiplicity) ~

6 .....

/

LAHET

J

193

"'" ..""

._o

o- 5 -1

¢-

04 "'1

z

k

20

~

40

a

i

i

140

26 MeV W(n,xn) angleintegratedspectrum

103

',~

\ \ \ ~

> •

10 2

t

60 80 100 120 Incident neutron energy (MeV)

UI. g "~

101

- ---

100

.

0

.

\

G N A S H calc (LA150) • Marcinkowski (1989) LAHET, semiclassical INC calc. .

.

,

.

.

.

.

,

.

.

.

.

i

.

.

.

.

i

5 10 15 20 Neutron emission energy (MeV)

.

.

.

.

i

25

Figure 7: Upper graph: Neutron emission multiplicity for neutrons incident on tungsten, calculated by GNASH and by LAHET; Lower graph Angle-integrated spectrum of neutrons following 26 MeV neutrons bombardment on tungsten (Marcinkowski et al. , 1989), compared with GNASH and LAHET calculations. The LAHET calculations underpredict the data over the whole emission energy range because the calculated reaction cross section is too low (Prael & Chadwick, 1997) by approx. 25%, but even when the LAHET results are scaled up accordingly, they still underpredict the highest energy emission due to omission of collective effects.

194

4.2

M. B. Chadwick et al.

Proton-induced

thick-target

neutron

production

It is important for nuclear simulation codes to adequately predict secondary neutron production from protons as they traverse target, blanket, and shielding materials. Secondary neutron production is central to spallation neutron target design, beam-stop design, and shielding considerations. Thick-target neutron production measurements provide an ideal way to test the microscopic nuclear data, in this case the LA150 proton data library, and test the proton transport algorithms used within MCNPX. Such measurements are usually made for targets that are sufficiently thick to stop the incident protons, though the targets are generally "thin" as far as the secondary neutrons are concerned. Thus, the probability that a secondary neutron produced has a subsequent nuclear interaction within the target is small. The protons slow down from their maximum energy, and for the maximum energy and all lower energies they have a probability of undergoing nuclear collisions leading to neutron production. Thick-target measurements therefore represent an integral over all incident energies up to the maximum of the differential "thin-target" cross sections, weighted by the stopping power function. In this work we focus on the following proton-induced thick-target neutron production measurements that test the LA150 data: 30 and 52 MeV data (Nakamura e t al. , 1982) for lead, iron, copper, and carbon; 68 MeV data (Meigo e t al. , 1997) for carbon and aluminum, and 113 MeV data (Meier e t al. , 1989) for carbon, aluminum, and iron. The use of LAHET-physics in MCNPX for simulating the Nakamura data was described by Sweezy at the SARE-4 conference (Sweezy e t al. , 1998). Before comparing MCNPX simulations of thick-target neutron production (using LA15O data and using LAHET-physics) with measurements, we first show a result (Fig. 8) that validates the accuracy of computational implementation in the MCNPX calculations. In this figure, MCNPX results are shown for 30 MeV protons on iron, compared with measurements (Nakamura e t al. , 1982), using the same angular-bins as used in earlier simulations (Sweezy e t a l . , 1998). We also show a similar calculation of the neutron spectra based solely on a deterministic integration of the LA150 data using a stopping-power weighting (dashed-dot line). The excellent agreement between the two calculations indicates that MCNPX is simulating the nuclear collisions and the proton stopping process correctly. The only discrepancy - at 135 degrees - is due to the very large back-angle bin (120-180 °) used to simulate the 135 ° detector. MCNPX calculations, using LA150 data and using LAHET-physics, for thicktarget neutron production spectra from 30 MeV protons are shown in Fig. 9, compared with measurements (Nakamura e t al. , 1982) for lead, iron, and copper. The MCNPX results using the LA150 data, are shown as solid lines, whereas MCNPX results using LAHET physics are shown as dashed lines. It is evident that considerable improvements are obtained in modeling neutron production using the library data. This is to be expected since the (LAHET physics) intranuclear cascade model cannot be expected to work well at low energies where semiclassical physics begins to fail. The only significant discrepancy between the MCNPX calculations and the measurements

Nuclear data for accelerator-driven systems

102 . 10 t

100 10-t

30MeV thick-targe(Fe(p,xn)'

" ~"'~ . ~ ,'~ ®

""

0 .3

g

1 0 -4

"-...

®

- . ,, ~

1

t

"'"®

®

_

=

I

o Nakamura (1983) -MCNPX with LA150 data ----- '~Fe LA150: stopping integration

~"

195

- 3o°(xl,o).

~,'~

[ ,arge~MCNPX-"'L~_'2. ® _ - - ~-,,,~,= ,, , ] F angularbin ~'-,~-~ ® 135~ - ~ ,,!~ 1 0 -6 r (12-0-1500) ~ "~X~i,i 0.0 5.0 10.0 15.0 20.0 25.0 Neutron energy (MeV) 1 0 -s

Z

Figure 8: Thick target neutron production spectra for 30 MeV protons incident on Fe (Nakamura e t al. , 1982) compared with calculations. The solid lines show MCNPX using the LA150 data, the dashed lines show a deterministic calculation that involves an integration of the LA150 data with a stopping-power weighting. is at the 135 degrees back angle, and as discussed above and shown in Fig. 8, much of this failing is an artifact related to the large angular bin used in the MCNPX simulation. Comparisons between calculation (C) and experiment (E) for integral neutron production per proton are shown in Table 1. The average deviation from unity of C/E is 8% for MCNPX, but 27% for LAHET. Similar information is presented in Table 2, for neutrons with energies above 3 MeV (corresponding to contributions above the detector threshold in the measurements). For neutrons above the 3-MeV detector threshold, MCNPX is seen to underpredict the measurements by about 23%, though the underprediction obtained by using LAHET physics is larger.

196

M. B. Chadwick et al.

Target Iron Copper Lead

MCNPX with LA150 data Calc./Exp. n/p 1.09 + 0.21 0.86 + 0.17 0.98 + 0.20

LAHET physics Calc./Exp. n/p 0.86 4- 0.17 0.74 + 0.13 0.58 4- 0.12

Table 1: MCNPX calculations using LA150 data, and LAHET physics, for total n/p for 30 MeV protons, relative to measured values (Nakamura et al. , 1982). We estimated experimental uncertainties of 20% since the experimental values involved an extrapolation to emission energies below the detector thresholds (3 MeV).

Target Iron Copper Lead

MCNPX with LA150 data Calc./Exp. n/p (En > 3 MeV) 0.77 0.77 0.78

LAHET CMc./Exp. n/p (E, > 3 MeV) 0.35 0.38 0.58

Table 2: MCNPX calculations using LA150 data, and LAHET physics, for total n/p for 30 MeV protons, relative to measured values (Nakamura et al. , 1982). Only neutrons with energies above 3 MeV, the detector threshold, are included.

Nuclear data for accelerator-driven systems 10; 10 ~ '~

! [

197

30 MeV thick-target Fe(p,xn) ~ ~~ •

--

O Nakamura (1983) MCNPX with LA150 dala

100

~ 1 0 -~ '~ 10-~

I ~~ ~

~~

~ ~_~'~

30° xlO 3

x 10-~ ~-" 10~4

£

Z

I0 -s 10-~

00

5.0

10.0

150

20.0

Neutron energy (MeV)

25O

102 30 MeV thick-target Cu(p,xn) 10 ~

~

O

~

~

\\

@

NackN~ rwaiI ~ 983~. . . . . . LAHET physics

10o ~'t0 ~ '~10 2 x 10-~ 1 0 -4

Z

10 ~ 10 6

50

10 O

15.0

200

Neutron energy (MeV)

25.0

102 ~.~. 30 MeV thick-target Pb(p,xn) 10 ~



\ \ ~

\\

'~,~. " =~ ~ ®

(3

~

0 Nakamura (1983) - MCNPX with LA15O data - LAHETphysics

~ \

® •



• O°(x~O ')

10 ~ 1 0 .2

=;lO i to



i

10~ oo

--¢L ~ 50

100

135°

\ 150

Neutron energy (MeV)

200

25.0

Figure 9: Thick target neutron production spectra for 30 MeV protons incident on Fe, Cu, and Pb (Nakamura e t al. , 1982) compared with calculations. The solid lines show MCNPX using the LA150 data, the dashed lines show MCNPX using LAHET physics.

198

M. B. Chadwick et al. 10 2

52 MeV p+C thick target n prod. 101 c o

>

• Nakamura (1983)

10 ° 10 -1 10 -2

iT 10 -3 0

1 0 -4 0 Z

10 -5 ,

,

'----.. .\,

10 -6 0

10

20

30

40

Neutron energy (MeV)

Figure 10: Thick target neutron production spectra for 52 MeV protons incident on C (Nakamura et al. , 1982) compared with calculations (an integration of the LA150 data with a stopping-power weighting) We do not show a result for protons on carbon at 30 MeV, since at present the LA150 Library does not include the 13C isotope. However, the role of laC becomes relatively unimportant by 52 MeV, and so in Fig. 10 we show our calculated thicktarget neutron production compared with data. Unlike the results shown in Fig. 9, that were based on an MCNPX simulation using the LA150 data, here the calculation was performed by directly integrating these data in the lab-frame with a stopping power weighting. At 68 MeV, proton-induced thick-target neutron production has been recently measured (Meigo et al. , 1997). These data are compared with our MCNPX calculations in Figs. 11, 12. For 68 MeV protons on carbon (Fig. 11), the MCNPX calculations with LA150 data and with LAHET-physics are seen to be of comparable quality, whereas for aluminum (Fig. 12) the MCNPX with LA150 simulation appears to more accurately describe the high-energy neutrons. The measured data at low neutron emission energies are described well by both calculations. Figures 13, 14, 15 show experimental thick-target neutron production spectra from 113 MeV protons on carbon, aluminum, and iron, (Meier et al. , 1989) compared with two sets of calculations: MCNPX using the LA150 proton data; and MCNPX using LAHET physics. Results are shown for carbon, aluminum, and iron. The angular bins for the detector locations were simulated as follows: 7.5 degrees - 7 to 8 degrees; 30 degrees - 29 to 31 degrees; 60 degrees 58 to 62 degrees; 150 degrees - 148 to 152 degrees. For the two lighter targets, Figs. 13, 14 show that low-energy (<2 MeV) yields

Nuclear data for accelerator-driven systems

199

predicted by LA150 are in better agreement with the measured data than those obtained using LAHET physics. For iron (Fig. 15), both calculations agree well with measured low-energy yields. For all targets, and for most angles, LA150 tends to slightly overpredict yields in the 3- to 10-MeV range, whereas LAHET shows generally good agreement with measurements in this energy range. At high energy (>20 MeV) and at forward angles, LA150 generally underpredicts yields (except for iron) whilst LAHET physics tends to overpredict yields. Overall, the quality of the MCNPX simulations using LA150 data, and using LAHET physics, are about the same.

100

~

e

V

p+C thick-targetn prod.

•e q~'d"•'o~ "e"e"q~' -. "•,~,

lP ~

~0.. 10-2 •

.

10_3 ~- 10-4

"...

"~, 10 s 10-6 Z 10-7 -81

10 2

4

6

8 10

30

50 ' 7'0

Neutron energy(MeV)

Figure 11: Thick target neutron production spectra for 68 MeV protons incident on carbon (Meigo et al., 1997) compared with calculations• The solid lines show MCNPX using the LA150 data, the dashed lines show MCNPX using LAHET physics.

200

M. B. Chadwick et al.

10° I~..... ~

>

'

. . . . . .

~ . . . . . . 68-MeV p+AI thick-target n prod.

:i-

"."~_'". ~

lo 3 r ~ " - ' ~ - ~

"O

o

£

: 7g::oOOO

10"

......

'L

• 6 0 ° x 10

z

-"'..,"~

• 150 ~

10.7

"'.." a ,

...... ~

~L

"'-...~ "'".

AAtridEnT

MCNPXwith LA150

10s 2

ll0

;, ' ~ ' ~

3'0

'

'

., , ~0 50

Neutron energy (MeV)

Figure 12: Thick target neutron production spectra for 68 MeV protons incident on aluminum (Meigo e t al. , 1997) compared with calculations. The solid lines show MCNPX using the LA150 data, the dashed lines show MCNPX using LAHET physics.

101

'

'

' ' 1

'

'

.

.

.

.

.

.

I

'

'

'

'

'

'

' ' 1

t 13-MeVp+C thick-targetn prod. 100 "E" O

10-1

{3.

>

&l

10-2 10 3

X3

&

10-4

c-

10 .5 oD

Z 1 0 -e

10-7

1

10 Neutron energy (MeV)

100

Figure 13: Thick target neutron production spectra for 113 MeV protons incident on carbon (Meier e t al. , 1989) compared with calculations. The solid lines show MCNPX using the LA150 data, the dashed lines show MCNPX using LAHET physics.

Nuclear data for accelerator-driven systems 101

' , ',i ...... :.:-. ........

.

.

.

.

.

.

.

.

I

'

201

. . . . . . .

I

1 1 3 - M e V p+Al thick-target n prod.

10°

,?, "5

10 -1

> 10-2 ~G)

~,

10-3

10 .4

e--

e -5 lO.5 z

1

10 Neutron e n e r g y (MeV)

100

Figure 14: Thick target neutron production spectra for 113 MeV protons incident on aluminum (Meier et al. , 1989) compared with calculations. The solid lines show MCNPX using the LA150 data, the dashed lines show MCNPX using LAHET physics.

I0 ° ¢O

.o

101

Q..

~ -

rod. "

.-

> (D

lO--,

¢v

10-3

"o

10 .4 to



10-5

30 ° , , 1 0 0

--e~._

'H

,6OoxlO

z

• 150° 10 .6 10"

....

.7%--

M C N P X with L A H E T

~Xw!th,LA1,50,,, 1

q.. .

i "

...~:F'Aq

10 Neutron e n e r g y (MeV)

00

Figure 15: Thick target neutron production spectra for 113 MeV protons incident on iron (Meier et al. , 1989) compared with calculations. The solid lines show MCNPX using the LA150 data, the dashed lines show MCNPX using LAHET physics.

202

4.3

M. B. Chadwick et al.

Neutron

transmission

and

shielding

design

Since the incident proton beam in ADS systems has an energy as high as 1 GeV, secondary neutrons and protons are produced via nuclear collisions in the target, with a range of energies extending up to approximately a GeV. Some of these particles will leave the spallation target, and require adequate shielding. Radiation transport codes can predict the neutron and proton transmission through shielding materials, and therefore provide a useful tool for ADS design. Neutron transmission experiments provide a useful test of MCNPX and the LA150 neutron-induced cross sections, particularly the elastic scattering differential cross sections and the nonelastic (n,xn) cross sections. Japanese measurements (Nakashima e t al. , 1996) have been particularly useful for assessing the capabilities of a simulation code for shielding studies. A quasimonoenergetic neutron source was generated by bombarding a Li target with 68 MeV protons, using the Li(p,n) reaction. These neutrons subsequently impinged upon macroscopic slabs of iron and concrete, of various thicknesses. The transmitted neutrons were then measured at locations both on and off the beam's axis. We demonstrated (Chadwick e t al. , 1999b) that, compared to LAHET simulations, significant gains can be obtained when modeling neutron transmission through iron by use of the LA150 data in MCNPX. At the recent American Nuclear Society Topical Meeting on Nuclear Applications of Accelerator Technology, a large range of results for the use of MCNPX in calculating neutron transmission through bulk shielding materials were presented (Sweezy e t al. , 1999). Unlike the experiment described above that made use of quasimonoenergetic neutrons from the Li(p, n) reaction, Sweezy e t al. focussed on modeling broad-spectrum neutron sources created by 52 MeV protons on a graphite target (Uwamino e t al. , 1982), and 65 MeV protons on a copper target (Shin e t al. , 1991). The neutrons produced through these reactions were then modeled as they passed through shields of iron, lead, concrete, and graphite. In these MCNPX simulations, the neutron source reaction was simulated using the LAHET models in MCNPX (rather than using the LA150 proton-induced libraries, which were not available in the version used at the time). The subsequent neutron interactions in the shielding were based upon the LA150 neutron data. Figure 16 shows the calculated neutron transmission through concrete and lead shields, compared with data, for neutrons generated by the 65 MeV protons on copper, taken from Sweezy e t aI.'s paper (Sweezy e t al. , 1999). There is seen to be good agreement between the calculations and the measurements. However, for the 100 cm concrete shield, the shape obtained in the experiment is probably an artifact of the data reduction procedure. Similar information on neutron transmission through graphite and water shields is shown in Figure 17, for neutrons generated by 52 MeV protons on graphite. Again, the agreement between the MCNPX simulation and measured data is seen to be fairly good, though the simulation appears to overpredict the data at the highest energies for transmission through water.

203

Nuclear data for accelerator-driven systems

I

I

Measured

o

]

MCNPX

--

-

o%

I

I

I

Measured MCNPX

10-9

OlU

~o

"T

~ 10-11

--

~ 10 "1°

.... o~

-

-

lOcm °°°°

oo

ooo °

ao

a

o

°a o

Q

o 0-11

~ 10-12

t~ 10-13 *5

r-

100 cm 00000%

-

o

o ~ ° %

z 0

o

I

I

I

20

40

60

10 -14

-

- 1 0 12

Z0

10-13 0

Neutron Energy (MeV)

I

I

I

20

40

60

Neutron Energy (MeV)

Figure 16: Neutron transmission through concrete (left graph) and lead (right graph) shields, compared with data, for neutrons generated by 65 MeV protons on copper (Shin et al. , 1991). The MCNPX calculations (Sweezy et al. , 1999) use the LA150 neutron data.

I

i

I

Measurement

,,-

10-7

~

'tO

MCNPX

I

Measurement

I

MCNPX

10 .9

• -

I

-

=

t =21.4cm

o..

E o

I

I •

10 .8

~•~,_

"T >

10 -10

t = 4 2 . 8 cm

O~

10 .9 X :D

X

t-

r"

O

"5

10 "1

9

10.10

z

z 10 -11

0

10

20

30

Neutron Energy (MeV)

40

10.12

I , , tt t, 0

10

20

30

40

Neutron Energy (MeV)

Figure 17: Neutron transmission through graphite (left graph) and water (right graph) shields, compared with data, for neutrons generated by 52 MeV protons on graphite (Uwamino et al. , 1982). The MCNPX calculations (Sweezy et al. , 1999) use the LA150 neutron data.

204

4.4

M. B. Chadwick et al.

N e u t r o n heating &: kerma

The LA150 cross section data were developed with an emphasis on providing an accurate description of energy deposition, for radiation heating calculations. One of the reasons for this was their use in medical applications for calculations of absorbed dose in fast neutron and proton radiotherapy (Chadwick e t al. , 1999a; ICRU Report 63, 2000). Radiation heating is also important in accelerator-driven systems. As protons traverse matter they can deposit their energy either through electromagnetic interactions with electrons ("dE/dx"), or via nuclear collisions in which the incident energy is transfered to secondary nuclear particles, including heavy recoils. Neutrons, on the other hand, interact only through the nuclear force and deposit their energy via a two step process: first, a nuclear interaction occurs; and second, the emitted chargedparticle nuclear fragments deposit their energy via electromagnetic interactions as they slow down. Emitted neutrons and gamma-rays usually have longer mean-free paths and therefore do not lead to energy deposition near the local nuclear collision site. The concept of kerma (kinetic energy released in matter) is particularly useful for neutron energy deposition and dosimetry considerations, when the incident energy is not too high. Here, "too high" is defined by requiring that the range of the secondary charged-particles is sufficiently small that the energy they deposit is local, i.e. in the vicinity of the collision site. In this case, the energy deposition (or "absorbed dose", to use the medical physics terminology) can be obtained approximately be convoluting the kerma coefficient with the neutron fluence. Since kerma coefficients can be experimentally determined, as well as calculated from the evaluated neutron cross section data, comparisons between measured and calculated kerma coefficients provide a valuable test of the LA150 cross section data (ICRU Report 63, 2000). The total kerma coefficient k~ (kerma per unit fluence ~) is calculated from cross sections by summing the partial kerma coefficients for each type of secondary charged particle including recoils, using: k~(Einc) = N ~i(Einc) ui _prod/~ (~inc},

(1)

where k~(Einc) is the kerma coefficient of ejectile type i at incident energy Einc, O.prod ~ i [~'-~inc) is the inclusive production cross section of ejectile i expressed in barns, and ~i(Einc) is the average energy of ejectile i expressed in MeV. The factor N = 9.64853 x 10-15/MA, where MA is the atomic mass of the target in units of u, converts the partial kerma coefficient from units of MeVb to S.I. units of f G y m 2 [femto (f) = 10-15; Gray (Gy) = Jkg-1]. Of course, with a transport code such as MCNPX, one does not need to compute energy deposition by making the kerma approximation - the code will automatically simulate the nuclear collisions and the trajectories of the light secondary charged particles as they slow down and stop, transferring their kinetic energy to heat. Nevertheless, kerma remains a valuable concept. If the kerma coefficients derived from the microscopic cross sectiondata are in agreement with measurements, then one can

Nuclear data for accelerator-driven systems

205

have confidence in the energy deposition predictions by the transport code. Furthermore, by convoluting the kerma coefficient and the neutron fluence - a "poor man's" estimate of energy deposition - one can easily validate the transport code's prediction of energy deposition. As an illustrative example of the calculated kerma coefficients compared with measurements, Figs. 18, 19 show the kerma coefficient for A1 and Fe as a function of neutron energy, for energies up to 150 MeV. The kerma coefficient increases over this range because of the increasing energy brought in by the neutron projectile. Agreement between calculation and experiment is seen to be good. For additional information, we point the reader to other papers that summarize neutron energy deposition and kerma from the LA150 data (Chadwick et al. , 1999a; ICRU Report 63, 2000). In particular, the new International Commission on Radiation Units and Measurements report (ICRU Report 63, 2000) provides a comprehensive summary of nuclear data for neutron and proton radiotherapy and for radiation protection. 5

....

I ....

~ ....

I ....

I ....

I ....

i ....

I ....

I'

''l

....

E U

=2 i

al

~'''-

o

Ikmh~w

id

aL

=

Oe.~a

~

aL

o

S e h ~ ' c n m ed e~, 1 9 e 7

-

m

Evdu=Uon

........

C4I~I

.t

~.

leo

0 0

10

20

30 40 50 60 Neufron energy, E /

70 MeV

80

90

100

Figure 18: Calculated A1 kerma coefficient from LA150 data, as a function of incident neutron energy, compared with measurements (Chadwick et al. , 1999a). 4

....

i ....

[] "-3



i ....

i ....

~'w

o

N~wt~mer

• e

Wuu and MIk~k~a.. Harlrncmnlggl



~ewa

........

J ....

,t d. w78

i ....

i ....

i'

S6Fe

~

C~llwd

el

al. M

=c2

:_~ /

o 1

E 0 0

10

20

30 40 50 60 Neutron energy, E /

70 MeV

I

I

80

90

100

Figure 19: Calculated Fe kerma coefficient from LA150 data, as a function of incident neutron energy, compared with measurements (Chadwick et al. , 1999a).

206

4.5

M. B. C h a d w i c k et al.

Radiation damage

The N JOY Nuclear Data Processing System has, since 1979, been capable of calculating the neutron damage energy cross section. A description of the theory of damage energy and the computation of damage energy are described in the N JOY Manual, Chapter 4 (MacFarlane & Muir, 1994). This cross section is used to calculate lattice defect production rates in materials, which are an important measure of radiation damage. The LA150 Library is the first Los Alamos release of a continuous-energy MCNPX library to include the damage energy cross section. For accelerator applications, this cross section is a welcomed addition as radiation damage effects are particularly severe in accelerator-driven neutron sources. At energies above nuclear data evaluations, the LAHET code (Prael & Lichtenstein, 1989) may be used to generate the damage energy cross section. The LAHET code relies on physics models of the intranuclear cascade in calculating the damage energy cross section, and again these are known to be less reliable at energies below approximately 150 MeV. The community that uses damage energy cross section data has noted a substantial discrepancy between evaluated damage energy cross sections and those generated by LAHET at the traditional transition energy of 20 MeV (Wechsler e t al. , 1985; Wechsler e t al. , 1997). The extension of this transition energy to 150 MeV brought about by the advent of the LA150 Library reduces the observed discrepancy. In figure 20 we plot the damage energy cross section from the LA150 Library over the energy range 400 keV to 150 MeV for four isotopes: 12C, 27A1, SaFe, and 2°SPb. Also plotted is the damage energy cross section as calculated using LAHET version 2.8 over the energy range 16 MeV to 3.1 GeV for the elements corresponding to the isotopes (Pitcher, 1997). For the representative isotopes considered here, the agreement between the evaluated data and the LAHET-generated data is generally better at 150 MeV than at 20 MeV (Barnett e t al. , 1999). This is consistent with expectations since LAHET should be less reliable at the lower energy. Plots of the damage energy cross section for all the isotopes in the LA150 library have been published in a laboratory report (Chadwick e t al. , 1999c). They are also available over the internet at http://t2.1anl.gov/cgi-bin/nuclides/endind by viewing the "ENDF/B-VI ACE Postscript Plots" produced by MacFarlane's N JOY code for a given ENDF/B-VI Release 6 isotope.

Nuclear data for accelerator-driven systems

207

A1;o~ 27AI

012 o lo i

! .... I N [ 006'

J

27AI LAHET ve~ion 28

~ JL~

"...... ...... . . . . .

0041 oo21

OOO

.... '

I

. . . . . . . .

'

10

. . . . . . . .

'

100

........

'

1000

'

'

000

~

1

10

energy (MeV)

0

3

5

1

energy

~

100 (MeV)

1000

30

e l e m e n t a i Pb LANET version 2 8 " ~

S

": :';

2o

/

t

LA~l~e~traslioFne28

/

N 15 /

g g

F 1

r'

/

o5

lO

1()0 etlergy ( M e V )

Figure 20: The from the LA150 S6Fe, and 2°SPb. LAHET version

1Dco

~Pb

1

.

lo energy

100 (MeV)

10oo

neutron damage energy cross section (MacFarlane & Muir, 1994) Library over the energy range 400 keV to 150 MeV for 12C, 27A1, Also plotted is the damage energy cross section as calculated using 2.8 over the energy range 16 MeV to 3.1 GeV.

208

5

M. B. Chadwick et al.

N u c l e a r d a t a n e e d e d for A T W

Major technology advances are needed for the accelerator transmutation of waste (ATW), also known as "Hybrid reactors" (Nifenecker et al. , 1999). Reliable highcurrent proton accelerators that can operate with minimal breaks in the beam's operation are needed. This is important for avoiding temperature fluctuations in the transmuter target/blanket, and for providing a constant electrical energy source to the grid. Chemical separation capabilities are needed that allow the separation of uranium from reactor spent-fuel, leaving the minor actinides (transuranic waste) for subsequent transmutation by fission in a fast neutron spectrum, and technetium and iodine fission products for transmutation in a moderated neutron spectrum within the transmuter. Technology developments needed for the spallation target and the transmuter design are also extensive. This will require extensions to the LA150 cross section library, and is discussed in more detail below. The Department of Energy recently sponsored a "Road Map" exercise to establish the science and technology research needed to develop an ADS capability, and the resulting publication (Barrett, 1999; Venneri et al. , 1999) concluded that there were no obvious show-stoppers to prevent the eventual deployment of such systems, although significant funding will be needed over a period of many years. Likewise, ADS research has begun in Europe, Japan, Russia (Gudowski, 1999). There are certain design differences between target/blanket ADS transmuters in the U.S. compared to those under consideration in Europe and Japan (Gudowski, 1999): The U.S. design includes a large component of plutonium from dismantled weapons, unlike the Japanese design; The European designs are considering the use of a thorium fuel cycle. Our focus here is the ADS designs that are currently being considered in the U.S., i.e. a high-current proton beam of energy approximately 1 GeV; a lead-bismuth spallation target and coolant; and a subcritical system that operates at fast neutron energies. We note, though, that alternative schemes are being discussed, such as spallation targets made of tungsten, tantalum, mercury, or depleted uranium; coolants based on liquid sodium or helium, as well as systems that operate at thermal energies (Bowman, 1999).

5.1

Subcritical t r a n s m u t e r and spallation target

The main attraction of accelerator-driven-systems for transmutation and energyproduction is the ability to operate a reactor at a subcritical level, and therefore, in principle, to have a safer system, over which one has more flexibility and control. However, from a nuclear physics perspective, ADS introduce a number of uncertainties compared to conventional reactor technologies that must be better understood, and these can be distinguished into issues related to the transmuter, and to the spallation target, addressed below. For the transmuter, the fundamental nuclear data for minor actinides are in many cases poorly understood. This affects our ability to predict the criticality, k~ff, as

Nuclear data for accelerator-driven systems

209

well as the change in criticality with time (burn-up swing) as the minor actinides and fission products are transmuted. This is important because the overall economy of an ADS system increases as criticality is approached, yet one needs to have confidence in the predicted criticality values, and uncertainties in these values, in order to operate near critical. Additionally, data uncertainties affect our ability to predict the rates of transmutation of the various radioactive species. Since most designs have, at present, focussed on transmuters that operate at fast neutron energies, the most important energy range is below 20 MeV, and primarily energies in the hundreds-of-keV to few-MeV region for the minor actinides, and lower energies for the fission products. More accurate data are desirable for the minor actinides, for the fission products, and for the coolants such as lead and bismuth (Palmiotti e t al. , 1999). Data at higher energies are needed in certain cases, such as for estimations of radiation damage, and shielding requirements. For the spallation target, accurate nuclear data are needed in order to predict the number of spallation neutrons produced per incident proton, n/p, which affects the overall economy. Data are also needed to better understand the radionuclides produced through spallation and fission nuclear reactions, for studies of activation, and materials properties. A lead-bismuth spallation target is currently being studied at many laboratories, and so it is data on these nuclei that are particularly important, for energies up to approximately a GeV. Other materials, such as tungsten, have also been discussed for the spallation target. Lead-bismuth has been used as a coolant in submarine reactors designed at Obninsk, Russia, and the response of this coolant to neutron radiation has been well studied. However, the effects of a high-energy proton beam on a lead-bismuth target has not been extensively studied, and such studies are a major objective of future ADS research. 5.2

Nuclear

data

measurements

Future nuclear data cross section measurements will be needed to support ADS system design. A detailed review of the experimental work needed is beyond the scope of the present article, though here we provide a brief overview. New measurements can address two types of needs: (1) cross section data for nuclear reactions that cannot be well-predicted by nuclear theory and model calculations; (2) cross section data that can be used to test, validate, and improve, nuclear model predictions that form the basis of transport code ADS simulations. The first category includes cross section data for minor actinides and fission products, where, in many cases, few previous measurements exist. At the energies present in a subcritical transmuter, the most important reaction channels open are fission, inelastic and elastic scattering, and radiative capture. Calculations of criticality depend sensitively on neutron fission cross sections and fission neutron multiplicities. Because of theoretical uncertainties in fission barrier heights and fission nuclear level densities, it is very difficult to a-priori predict a fission cross section from theory, and measurements are needed. Likewise, even though there are calculational capabilities,

210

M. B. Chadwick et al.

and phenomenological systematics, for predicting fission neutron multiplicities and spectra, new measurements may be needed to measure these data to the required accuracy. Other nuclear data that cannot always be well predicted by theory are radiative capture cross sections and inelastic scattering cross sections. A document was prepared by researchers at Livermore and Los Alamos summarizing nuclear data measurements that may be needed to support an ADS program (Slaughter et al. , 1998). The focus was on measurements that can be made at the Los Alamos Neutron Science Center (LANSCE) and included: • Nuclear fission cross sections for minor actinides from thermal to 100 MeV. Radiative capture cross sections for minor actinides and fission products, from thermal to 500 keV. Radiative capture cross sections are needed to better understand, and predict, transmutation rates. A capture measurement on bismuth is also important to resolve discrepancies in previous measurements, since neutron capture followed by beta decay produces 21°Po in the spallation target - a major radiological hazard. Measurements of spallation/fission products in thin and thick targets of lead and bismuth, for GeV-energy proton beams. Spallation and fission reactions lead to the production of radioactive nuclides, radioactive gases and alpha-emitting nuclides, in the Pb-Bi eutectic spallation target. • Measurements of fission product Z,A, distributions for minor actinides and for subactinide fission in Pb and Bi. For many minor actinides the fission prompt neutron multiplicity, F, is not well known• This quantity is important since it directly affects the transmuter's criticality. New measurements would be valuable, though it is recognized that experimental capabilities to achieve this would have to be re-started and would be costly. The second category, i.e. measurements to validate nuclear model predictions, includes data at intermediate energies (tens to hundreds of MeV). Such measurements may include (p, x n ) , (p, x p ) , (n, x n ) , (n, x p ) etc. cross sections, energy spectra, and angular distributions, and can be used as benchmarks for nuclear model predictions that implement compound nucleus, preequilibrium, and intranuclear cascade reaction mechanisms. Significant progress has already been made in this area by measurement programs in the U.S. (Meier et al. , 1989; Meier et al. , 1992; Haight et al. , 1997; Scobel et al. , 1990), Japan (Nauchi et aI. , 1999; Meigo et al. , 1997; Nakamura et al. ,1982), Louvain-la-Neuve (Benck et a l . , 1998), Saturne (Ledoux, 1999). Production cross sections of various radionuelides in subactinide spallation and fission can also be used for testing intranuclear cascade code theoretical predictions, and work has been undertaken in this area (Michel & Nagel, 1997; Titarenko et al. , 1998).

Nuclear data for accelerator-driven systems

6

211

Summary

The LA150 nuclear data described in this review are available, in ENDF-6 format, from the U.S. National Nuclear data Center (Dunford, 1998) as Release 6 of the ENDF/B-VI Library. They can be also obtained from the Group T-2 Nuclear Information Service W W W site: h t t p : / / t 2 . 1 a n l . g o v / d a t a . They are documented in a very extensive laboratory report (Chadwick et al. , 1999c) which is also available over the WWW at http://t2.1anl.gov/publieations/la150/la150.html, and a journal article (Chadwick et al. , 1999b). The MCNPX code has been released and is available from the Radiation Safety Information Computational Center (RSICC).

7

Acknowledgments

We would like to thank Dr. Laurie Waters in the Accelerator Production of Tritium Technical Project Office for many useful discussions and for supporting the work described here. Dr. Robert MacFarlane played an important role in this project by developing the N JOY code to process LA150 data, and by processing the library for use by MCNPX. We also benefited from numerous helpful discussions with Drs. David Madland, Gerald Hale, Richard Prael, William Wilson, Garry Russell, Stepan Mashnik, Arjan Koning, Satoshi Chiba, Tokio Fukahori, Gregory Van Tuyle, Prancesco Venneri, and Philip Finck. We would like to acknowledge useful discussions on ATW nuclear data measurement needs with Drs. Robert Haight, Jerry Wilhelmy, Stephen Wender, Dennis Slaughter, and Frank Dietrich. We are grateful to Nolan Hertel and Jeremy Sweezy for permission to reproduce their neutron transmission MCNPX calculations. We thank Drs. Nakamura and Meigo for kindly providing us with their thick-target stopping data.

212

M. B. Chadwick et al.

References Alsmiller, R. G., Barnes, J. M., & Drischler, J. D. 1986. Neutron-Photon Multigroup Cross Sections for Neutron Energies up to 400 MeV (Revision 1). Tech. rapt. ORNL/TM-9801. Oak Ridge National Laboratory, Oak Ridge, TN. Barnett, M. H., Wechsler, M. S., Dudziak, D. J., & Corzine, R. K. 1999. Calculations of radiation damage at SNS. Pages 555-559 @ VanTuyle, G. (ed), Proceedings of the ANS on Nuclear Applications of Accelerator Technology. (La Grange Park, IL: American Nuclear Society), Long Beach, California, November 14-18, 1999. Barrett, L. et al. 1999. A Roadmap for developing Accelerator Transmutation of Waste (ATW) technology. Tech. rapt. DOE/RW-0519. Report to Congress. Benck, S., Slypen, I., Meulders, J. P., Corcalciuc, V., Chadwick, M. B., Young, P. G., & Koning, A. J. 1998. Light charged-particle production in neutron induced reactions on aluminum at E~=62.7 MeV. Phys. Rev., C58, 1558-1568. Blann, M., Gruppelaar, H., Nagel, P., & Rodens, J. 1994. Paris, France: Organization for Economic Cooperation and Development (OECD) Nuclear Energy Agency. Bodansky, D. 1996. Nuclear Energy: Principles, Practices, and Prospects. American Institute of Physics, NY: AIP Press. Bowman, C. D. 1999. Long-term lower-cost nuclear energy without weapons or reprocessing using accelerator-driven systems. Pages ~53-~62 of." Van Tuyle, G. (ed), Proc. of the International Topical Meeting on Nuclear Applications of Accelerator Technology. (La Grange Park, IL: American Nuclear Society), Long Beach, California, November 14-18, 1999. Briesmeister, J. F. 1997. MCNP-A general Monte Carlo n-particle transport code, version ~, B. Tech. rapt. LA-12625-M (1997). Los Alamos National Laboratory, Los Alamos, NM. Chadwick, M. B., & Normand, E. 1999. Use of New ENDF/B-VI Proton and Neutron Cross Sections for Single Event Upset Calculations. IEEE Trans. Nucl. Sci., 46, 1386-1394. Chadwick, M. B., & Young, P. G. 1996. Calculation and Evaluation of Cross Sections and Karma Factors for Neutrons up to 100 MeV on 160 and 14N. Nucl. Sci. Eng., 123, 1-16. Chadwick, M. B., & Young, P. G. 1999. The GSCAN code: From GNASH reaction calculations to ENDF nuclear data files. Pages 188-197 of." Oblozinsky, P., & Gandini, A. (ads), Proc. of the IAEA Workshop on Nuclear Reaction Data and Nuclear Reactors. (Singapore: World Scientific Publishing, Ltd.), Trieste, Italy, February 23 - March 27, 1999.

Nuclear data for accelerator-driven systems

213

Chadwick, M. B., Blann, M., Cox, L. J., Young, P. G., & Meigooni, A. S. 1996. Calculation and Evaluation of Cross Sections and Kerma Factors for Neutrons up to 100 MeV on Carbon. Nucl. Sci. Eng., 123, 17-37. Chadwick, M. B., Barschall, H. H., Caswell, R. S., DeLuca, P. M., Hale, G. M., Jones, D. T. L., MacFarlane, R. E., Meulders, J. P., Schuhmacher, H., Schrewe, U. J., Wambersie, A., & Young, P. G. 1999a. A consistent set of neutron kerma coefficients from thermal to 150 MeV for biologically important materials, ivied. Phys., 26, 974-991. Chadwick, M. B., Young, P. G., Chiba, S., Frankle, S., Hale, G. M., Hughes, H. G., Koning, A. J., Little, R. C., MacFarlane, R. E., Prael, R. E., & Waters, L. S. 1999b. Cross Section Evaluations to 150 MeV for Accelerator-Driven Systems and Implementation in MCNPX. Nucl. Sci. Eng., 131,293-328. Chadwick, M. B., Young, P. G., MacFarlane, R. E., Moller, P., Hale, G. M., Little, R. C., Koning, A. J., & Chiba, S. 1999c. LA150 documentation of cross sections, heating, and damage: Part A (incident neutrons) and Part B (Incident protons). Tech. rept. LA-UR-99-1222 (1999). Los Alamos National Laboratory, Los Alamos, NM (http://t2.1anl.gov/publications/la150/la150.html). Cond6, H. 1992. In: Cond6, H. (ed), Nuclear Data Standards for Nuclear Measurements. Paris, France: Organization for Economic Cooperation and Development (OECD) Nuclear Energy Agency. Dietrich, F. S., Abfalterer, W. P., Haight, R. C., Morgan, G. L., Bateman, F. B., & Finlay, R. W. 1997. Recent measurements of neutron total cross sections on a wide range of targets from 5 to 600 MeV at LANSCE/WNR. Pages 402-406 of: Reffo, G. (ed), Proc. of the International Conference on Nuclear Data for Science and Technology. (Bologna, Italy: Societa Italiana di Fisica), Trieste, Italy, May 18-24, 1997. Dunford, C. L. 1 9 9 8 . Internet connection to the National Nuclear Data Center (NNDC). Brookhaven National Laboratory: WWW address: http://www.nnde.bnl.gov/. Feshbach, H., Kerman, A., & Koonin, S. 1980. The Statistical Theory of Multi-Step Compound and Direct Reactions. Ann. Phys. (N. Y.), 125, 429-476. Finlay, R. W., Abfalterer, W. P., Fink, G., Montei, E., Adami, T., Lisowski, P. W., Morgan, G. L., & Haight, R. C. 1993. Neutron Total Cross Sections at Intermediate Energies. Phys. Rev. C, 47, 237-247. Gudowski, W. 1999. Accelerator-driven transmutation projects. The importance of nuclear physics research for waste transmutation. Nucl. Phys., A654, C436C457.

214

M. B. C h a d w i c k et al.

Haight, R. C., Bateman, F. B., Sterbenz, S. M., Grimes, S. M., Wasson, O. A., MaierKomor, P., & Vonach, H. 1997. An Update on (n,charged particle) Research at WNR. Fusion Eng. Des., 37(1), 73-77. Hughes, H. G., Adams, K. J., Chadwick, M. B., Comly, J. C., Egdorf, H. W., Frankle, S. C., Hendricks, J. S., Little, R. C., MacFarlane, R. E., Mashnki, S. G., Prael, R. E., Sierk, A. J., Waters, L. S., White, M. C., Young, P. G., Gallmeier, F. X., & Snow, E. C. 1999. MCNPX for Neutron-Proton Transport. Page 939 of." Aragones, J. M. (ed), Proc. of the Mathematics and Computation, Reactor Physics and Environmental Analysis in Nuclear Applications. (Spain: Senda Editorial, S. A., Madrid), Madrid, Spain, September 27-30, 1999. Ibaraki, M., Baba, M., Takako, M., Hirasawa, M., Nauchi, Y., Nakashima, H., Meigo, S., Iwamoto, O., & Tanaka, S. 1999. Experimental method for neutron elastic scattering cross section measurement in 40 - 90 MeV region at TIARA. Nucl. Instr. Meth. A, (submitted). ICRU Report 63. 2000. Nuclear Data for Neutron and Proton Radiotherapy and for Radiation Protection. Bethesda, MD: International Commission on Radiation Units and Measurements. Ignatyuk, A. V., Lunev, V. P., Shubin, Y. N., Gai, E. V., Titarenko, N. N., Ventura, A., & Gudowski, W. 2000, in press. Neutron cross section evaluations for 238U up to 150 MeV. Nucl. Sci. Eng. Kalbach, C. 1988. Systematics of Continuum Angular Distributions: Extensions to Higher Energies. Phys. Rev. C, 37, 2350-2370. Kalend, A. M., Anderson, B. D., Baldwin, A. R., Madey, R., Watson, J. W., Chang, C. C., Holmgren, H. D., Koontz, R. W., Wu, J. R., & Machner, H. 1983. Energy and Angular Distributions of Neutrons from 90 MeV Proton and 140 MeV AlphaParticle Bombardment of Nuclei. Phys. Rev. C, 28, 105-118. Kim, E., Nakamura, T., Konno, A., Uwamino, Y., Nakanishi, N., Imamura, M., Nakao, N., Shibata, S., & Tanaka, S. 1998. Measurements of neutron spallation cross sections of 12C and 2°9Bi in the 20- to 150-MeV Energy Range. Nucl. Sci. Eng., 129,209-223. Koning, A. J., Fukahori, T., & Hasegawa, A. 1998a. Volume 13: Intermediate Energy Data. Tech. rept. NEA/WPEC-13. Nuclear Energy Agency, Paris. Koning, A. J., Bersillon, O., & Delaroche, J.-P. 1998b. Nuclear Data for AcceleratorDriven Systems: Nuclear Models, Experiments, and Data Libraries. Nucl. Inst. Meth. A, 414, 49-67. Ledoux, X. et al. 1999. Spallation neutron production by 0.8, 1.2, and 1.6 GeV protons on Pb targets. Phys. Rev. Lett., 82, 4412 4415.

Nuclear data for accelerator-driven systems

215

MacFarlane, R. E., & Muir, D. W. 1994. The NJOYNuclear Data Processing System, Version 91. Tech. rept. LA-12740-M (1994). Los Alamos National Laboratory, Los Alamos, NM. Madland, D. G. 1988. Recent Results in the Development of a Global Medium-Energy Nucleon-Nucleus Optical Model Potential. Pages 103-116 of: Strohmaier, B. (ed), Proc. of a Specialists Meeting on Preequilibrium Reactions. (Paris, France: Organization for Economic Cooperation and Development (OECD) Nuclear Energy Agency), Semmering, Austria, 1988. Marcinkowski, A., Finlay, R. W., Randers-Pehrson, C., Brient, C. E., O'Donnell, J. E., & Stankiewicz, K. 1983. Neutron Emission Spectra and Angular Distributions at 25.7 MeV Neutron Bombarding Energies. Nucl. Phys. A, 402,220-234. Marcinkowski, A., Finlay, R. W., Rapaport, J., Hodgson, P. E., & Chadwick, M. 13. 1989. Nucl. Phys. A, 501, 1. Marcinkowski, A., Demetriou, P., & Hodgson, P. E. 1996. Preequilibrium and collective reactions in neutron inelastic-scattering by the rotational nuclei 184W and ls6W. J. Phys. G, 22, 1219. Mashnik, S. G., & Sierk, A. J. 1999. Improved cascade-exciton model of nuclear reactions. Pages 29-51 of: Gabriel, T. A. (ed), Proe. of the Fourth Workshop On Simulating Accelerator Radiation Environments (SARE4). (Oak Ridge, Tennessee: Oak Ridge National Laboratory), Noxville, Tennessee, Sept 14-16, 1998. Mashnik, S. G., Sierk, A. J., Bersillon, O., L: Gabriel, T. A. 1998. Cascade-Exciton Model Detailed Analysis of Proton Spallation at Energies From 10 MeV to 5 GeV. Nucl. Instr. Meth., A414, 68-72. Maslov, V. M., Hasegawa, A., & Fukahori, T. 1998. Neutron-induced fission cross sections of uraniums up to 40 MeV; Proceedings of the third specialists' meeting on high energy nuclear data. In: Proc. of the specialists' meeting on high energy nuclear data. Tokai, Japan, 30-31 March 1998. McLane, V. 1997a. CSISRS experimental nuclear data file, National Nuclear Data Center (NNDC). Brookhaven National Laboratory: WWW address: http://www.nndc.bnl.gov/. McLane, V. 1997b. ENDF-102 Data Formats and Procedures for the Evaluated Nuclear Data File ENDF-6. Tech. rept. BNL-NCS-44945, Rev. 2/97. Brookhaven National Laboratory, National Nuclear Data Center, Upton, NY. Meier, M. M., Clark, D. A., Goulding, C. A., McClelland, J./3., Morgan, G. L., Moss, C. E., & Amian, W. t3. 1989. Differential neutron production cross sections and neutron yields from stopping-length targets for 113-MeV protons. Nucl. Sci., 102. 310-321.

216

M. B. Chadwick et al.

Meier, M. M., Amian, W. B., Goulding, C. A., Morgan, G. L., & Moss, C. E. 1992. Differential Neutron-Production Cross-Sections for 256-MeV Protons. Nucl. Sci. Eng., 110, 289 298. Meigo, S., Takada, H., Nakashima, H., Sasa, T., Tanaka, S. Shin, K., & Ono, S. 1997. Measurements of neutron spectra from stopping-length targets irradiated by 68MeV protons, 100 MeV alpha and 220 MeV carbon particles. Pages ~i3-~15 @ Reffo, G. (ed), Proc. of the International Conference on Nuclear Data for Science and Technology. (Bologna, Italy: Societa Italiana di Fisica), Trieste, Italy, May 18-24, 1997. Michel, R., & Nagel, P. 1997. Pages 1-353 of." International Codes and Model Intercomparison for Intermediate Energy Activation Yields, NSC/DOC(97)-1. Issyles-Moulineaux, France, May 30-June 1, 1994, Organization for Economic Cooperation and Development (OECD) Nuclear Energy Agency, Paris, France. Nakamura, T., Fujii, M., & Shin, K. 1982. Neutron production from thick targets of carbon, iron, copper and lead by 30- and 52-MeV protons. Nucl. Sci. Eng., 83, 444-458. Nakashima, H., Nakao, N., Tanaka, S., Nakamura, T., Shin, K., Tanaka, S., Takada, H., Meigo, S., Nakane, Y., Sakamoto, Y., & Baba, M. 1996. Transmission Through Shields of Quasi-Monoenergetic Neutrons Generated by 43-MeV and 68 MeV Protons. 2. Iron Shielding Experiment and Analysis for Investigating Calculational Method and Cross Section Data. Nucl. Sci. Eng., 124, 243-257. Nauchi, Y., Baba, M., Sanami, T., Ibaraki, M., Iwasaki, T., Hirakawa, N., Tanaka, S., Meigo, S., Nakashima, H., Takada, H., & Nakamura, T. 1999. Measurement of (n,xp) and (n,xd) Double Differential Cross Sections of Carbon and Aluminum for 65 and 76 MeV Neutrons. J. Nucl. Sci. Tech., 36, 143-151. Nifenecker, H., David, S., Loiseaux, J. M., & Giorni, A. 1999. Hybrid nuclear reactors. Prog. in Part. and Nucl. Phys., 43, 683-827. Palmiotti, G., Finck, P. J., Gomes, I., Micklich, B., & Salvatores, M. 1999. Uncertainty assessment for accelerator-driven systems. In: Proc. of the ANS GLOBAL '99 International Conference on future nuclear systems. (La Grange Park, IL: American Nuclear Society), Jackson Hole, WY, August 29-September 3. Pearlstein, S. 1993. Nuclear Data for Neutron and Proton Interactions with 12C in the Energy Range 0-10 GeV. Health Phys., 65, 185-189. Pitcher, E. J. et al. 1997. The Effect of the New Nucleon-Nucleus Elastic Scattering Data in LAHET Version 2.8 on Neutron Displacement Cross Section Calculations. In: Wechsler, M. S., Mansur, L. K., Snead, C. L., & Sommer, W. F. (eds), Proceedings of the Symposium on Materials for Spallation Neutron Sources. Orlando, Florida, February 10-12, 1997.

Nuclear data for accelerator-driven systems

217

Prael, R. E., & Chadwick, M. B. 1997. Application of evaluated nuclear data in the LAHET code. Pages 2416 2426 of." Reffo, G. (ed), Proc. of the International Conference on Nuclear Data for Science and Technology. (Bologna, Italy: Societa Italiana di Fisica), Trieste, Italy, May 18-24, 1997. Prael, R. E., & Lichtenstein, H. 1989. Users guide to LCS: The L A H E T code system. Tech. rept. LA-UR-89-3014. Los Alamos National Laboratory, Los Alamos, NM. Scobel, W., Trabandt, M., Blann, M., Pohl, B. A., Remington, B. R., Byrd, R. C., Foster, C. C., Bonetti, R., Chiesa, C., & Grimes, S. M. 1990. Preequilibrium (p,n) Reactions as a Probe for the Effective Nucleon-Nucleon Interaction in Multistep Direct Processes. Phys. Rev. C, 41, 2010-2020. Shin, K., I., Yoshiaki., Miyahara, K., Uwamino, Y., Sakai, H., & Numata, S. 1991. Transmission of Intermediate-Energy Neutrons and Associated Gamma Rays Through Iron, Lead, Graphite, and Concrete Shields. Nucl. Sci. Eng., 109, 380-390. Slaughter, D., Haight, R. C., Chadwick, M. B, Dietrich, F. S., & Wilhelmy, J. et al. 1998. A 5-year plan for development of nuclear data and simulation tools for accelerator transmutation of waste (ATW). Tech. rept. (unpublished). Lawrence Livermore and Los Alamos National Laboratories. Snow, E. C. 1998. Mesh Tallies and Radiography Images for MCNPX. Page 113 of: Gabriel, T. A. (ed), Proc. of the Fourth Workshop On Simulating Accelerator Radiation Environments (SARE4). (Oak Ridge, Tennessee: Oak Ridge National Laboratory), Noxville, Tennessee, Sept 14-16, 1998. Sweezy, J. E., Hertel, N., Rahnema, F., Waters, L. S., Hughes, H. G., & Prael, R. E. 1998. Comparison of MCNPX with measurements of angular-dependent neutron production from protons on thick targets. Pages 209 220 of." Gabriel, T. A. (ed), Proc. of the Fourth Workshop On Simulating Accelerator Radiation Environments (SARE4). (Oak Ridge, Tennessee: Oak Ridge National Laboratory), Noxville, Tennessee, Sept 14-16, 1998. Sweezy, J. E., Hertel, N., Waters, L. S., & Hughes, H. G. 1999. MCNPX benchmark calculations: Transmission of neutrons below 65 MeV through bulk-shielding materials. Pages 391-398 of." G.Van Tuyle (ed), Proc. of the International Topical Meeting on Nuclear Applications of Accelerator Technology. (La Grange Park, IL: American Nuclear Society), Long Beach, California, November 14-18, 1999. Titarenko, Y. E., Shedov, O. V., Igumnov, M. M., Mashnik, S. G., Karpikhin, E. I., Kazaritsky, V. D., Batyaev, V. F., Koldobsky, A. B., Zhivun, V. M., Sosnin, A. N., Prael, R. E., Chadwick, M. B., Gabriel, T. A., & Blann, M. 1998. Experimental and computer-simulation study of the radionuclides produced in thin

218

M. B. Chadwick et al.

2°9Bi targets by 130 MeV and 1.5 GeV proton-induced reactions. Nucl. Intsr. Meth. A, 414, 73-995. Uwamino, Y., Nakamura, T., & Shin, K. 1982. Penetration Through Shielding Materials of Secondary Neutrons and Photons Generated by 52-MeV Protons. Nucl. Sci. Eng., 80, 360-389. Van Tuyle, G. L., & Beller, D. E. 1999. Accelerator Transmutation of Waste Technology and Implementation Scenarios. Pages 337-346 of." G.Van Tuyle (ed), Proc. of the International Topical Meeting on Nuclear Applications of Accelerator Technology. (La Grange Park, IL: American Nuclear Society), Long Beach, California, November 14-18, 1999. Venneri, F., Crawford, D., Khalil, H., Li, N., Allen, T., Louthan, M., Woloshun, K., Brunsvold, A., Ehrman, C., Wade, D., Cappiello, M., Chadwick, M. B., Lisowski, P., Hayes, S., Herceg, J., & MacDonald, P. 1999. Roadmap for the development of accelerator transmutation of waste: Target and blanket system report of the target and blanket technical working group. Tech. rept. LA-UR-99-3022 (1999). Los Alamos National Laboratory, Los Alamos, NM. Wechsler, M. S., Davidson, D. R., Greenwood, L. R., & Sommer, W. F. 1985. Calculation of displacement and helium production in the LAMPF irradiation facility. Pages 1189-1198 @ Garner, F.A., & Perrin, J.S. (eds), Effects of Radiation on Materials: Twelfth International Symposium, STP 870. (American Society of Testing and Materials), Philedelphia, 1985. Wechsler, M. S., Lin, C., Ferguson, P. D., Mansur, L. K., & Sommer, W. F. 1997. Calculations of Radiation Damage in Target, Container, and Window Materials for Spallation Neutron Sources. Pages 968 974 of." CondO, H. (ed), Proc. of Second International Conference on Accelerator-Driven Transmutation Technologies and Applications. (Stockholm, Sweden: Gotab), Kalmar, Sweden, June 3-7, 1996. White, M. C., Little, R. C., & Chadwick, M. B. 1999. Photonuclear physics in MCNP(X). Pages 515-519 of: VanTuyle, G. (ed), Proceedings of the ANS on Nuclear Applications of Accelerator Technology. (La Grange Park, IL: American Nuclear Society), Long Beach, California, November 14-18, 1999. Wu, J. R., Chang, C. C., & Holmgren, H. D. 1979. Charged-Particle Spectra: 90 MeV Protons on A1-27, Ni-58, Zr-90, and Bi-209. Phys. Rev. C, 19, 698-713. Young, P. G., Arthur, E. D., Bozoian, M., England, T. R., Hale, G. M., Labauve, R. J., Little, R. C., MacFarlane, R. E., Madland, D. G., Perry, R. T., & Wilson, W. B. 1990. Transport data libraries for incident proton and neutron energies to 100 MeV. Tech. rept. LA-11753-MS. Los Alamos National Laboratory, Los Alamos. NM.

Nuclear data f o r accelerator-driven systems

219

Young, P. G., Arthur, E. D., & Chadwick, M. B. 1998. Comprehensive nuclear model calculations: Theory and use of the GNASH code. Pages 227-~0~ of." Gandini, A., & Reffo, G. (eds), Proc. of the I A E A Workshop on Nuclear Reaction Data and Nuclear Reactors - Physics, Design, and Safety. (Singapore: World Scientific Publishing, Ltd.), Trieste, Italy, April 15 - May 17, 1996.