LA-UR-20-22603

Correlations Between Fission Fragment and Neutron Anisotropies in Neutron-Induced Fission

A. E. Lovell,1, ∗ P. Talou,1 I. Stetcu,1 and K. J. Kelly1 1Los Alamos National Laboratory, Los Alamos, NM 87545, USA (Dated: May 8, 2020) Several sources of angular anisotropy for fission fragments and prompt neutrons have been studied in neutron-induced fission reactions. These include kinematic recoils of the target from the incident neutron beam and the fragments from the emission of the prompt neutrons, preferential directions of the emission of the fission fragments with respect to the beam axis due to the population of particular transition states at the fission barrier, and forward-peaked angular distributions of pre- equilibrium neutrons which are emitted before the formation of a compound nucleus. In addition, there are several potential sources of angular anisotropies that are more difficult to disentangle: the angular distributions of prompt neutrons from fully accelerated fragments or from scission neutrons, and the emission of neutrons from fission fragments that are not fully accelerated. In this work, we study the effects of the first group of anisotropy sources, particularly exploring the correlations between the fission fragment anisotropy and the resulting neutron anisotropy. While kinematic effects were already accounted for in our Hauser-Feshbach Monte Carlo code, CGMF, anisotropic angular distributions for the fission fragments and pre-equilibrium neutrons resulting from neutron- induced fission on 233,234,235,238U, 239,241Pu, and 237Np have been introduced for the first time. The effects of these sources of anisotropy are examined over a range of incident neutron energies, from thermal to 20 MeV, and compared to experimental data from the Chi-Nu liquid scintillator array. The anisotropy of the fission fragments is reflected in the anisotropy of the prompt neutrons, especially as the outgoing energy of the prompt neutrons increases, allowing for an extraction of the fission fragment anisotropy to be made from a measurement of the neutrons.

I. INTRODUCTION emission of fission fragments and neutrons during the de- cay. There are several sources of this anisotropy: fission fragment emission with respect to the incident beam, Being able to accurately and reliably model the fis- neutron emission from the fission fragments (including sion process and the resulting observables is important whether neutrons are emitted from the fully accelerated for a variety of applications, including security, non- fragments), along with the emission of scission, multi- proliferation, energy, and fundamental science. While chance, and pre-equilibrium neutrons before fission. microscopic models for this process, beginning from fundamental nucleon-nucleon interactions, are available Although, there is no reason to expect that fission [1, 2], more often these models are constructed based fragments should be emitted anisotropically with respect on a combination of macroscopic and microscopic theo- to the incoming neutron beam from the compound nu- ries [3, 4]. Still, at the moment, these methods are lim- cleus, many experimental studies from the 1950’s through ited; they are computationally expensive, and the range today have shown that there can be a large degree of of observables that can be calculated generally ends at anisotropy - defined as the ratio of the cross section at the post-scission fission fragment yields. These results 0◦ to the cross section at 90◦. The enhancement with can then be used as input for codes that calculate the respect to the beam axis can be as much as 30% to 60%. prompt and delayed neutron and γ-ray observables. There have, in particular, been several measurements of the anisotropy of actinides, including 235U [12–16], 238U These codes (e.g, CGMF [5, 6], FREYA [7, 8], HF3D/BeOH [12, 15, 17–21], and 239Pu [15, 22, 23].

arXiv:2005.03206v1 [nucl-th] 7 May 2020 [9], FIFRELIN [10], GEF [11]) employ a variety of physics models describing how the excitation and kinetic energies This anisotropy was postulated, by A. Bohr [24], to are shared between the light and heavy fission fragments, stem from the populated transition states at the fission and determine neutron and γ-ray observables on either an barrier of the parent nucleus, which would account for the event-by-event basis or as averages. Event-by-event cal- changing anisotropy as a function of incident neutron en- culations allow for the determination of all observables, ergy. At low incident energies, few states are populated, including correlations, as consistently as possible, from resulting in an anisotropic distribution of fission frag- scission to prompt particle emissions, conserving energy, ments depending on the quantum numbers of the states spin, and parity at every step of the decay. One effect that are populated. As the incident energy increases, that has been studied over the years, but is not always more states are available in the parent nucleus (and the fully included in these types of codes, is the anisotropic angular distribution becomes more isotropic) until the second-chance fission channel opens up, at which point the A nucleus is fissioning instead of the A + 1 system, lowering the number of available states, increasing the ∗Electronic address: [email protected] anisotropy again. Of course, the picture is more compli- 2 cated than this since the opening of the second-chance II. THEORY fission channel leads to a possibility of fission from either the A+1 or A compound nucleus. A recent measurement There are several physical processes that could lead 239 of the anisotropy of Pu confirmed that the changes in to an anisotropic distribution of the prompt neutrons anisotropy track the shape of the fission cross section [16], emitted during the fission process. The following sec- in agreement with many previous measurements that ob- tions describe these processes along with the Monte Carlo served the same trends, e.g. [15, 22, 23]. Hauser-Feshbach code CGMF, used here to calculate the decay of the fission fragments by prompt neutron and There have also been several studies to assess whether γ-ray emissions. or not the prompt neutrons are emitted isotropically from the fission fragments, along with whether the neutrons are emitted from the fully accelerated fragments, both A. Fission Fragment Anisotropy of which are commonly assumed in the fission fragment deexcitation codes mentioned above. Often, differences The first process we discuss is the fission fragment an- between the measured and calculated neutron angular gular anisotropy coming from the population of the tran- distributions are attributed to scission neutrons, those sition states [24]. The angular distribution of the fission neutrons that are emitted during the rupture of the neck fragments can be described with the help of the Wigner of the compound nucleus. However, this interpretation d-matrices [33], defined as does not take into account the assumptions in the model J that could change the resulting calculations (such as that dM,K (θ) = all prompt neutrons are emitted from the fully acceler- 1/2 X [(J + M)!(J − M)!(J + K)!(J − K)!] ated fragments). The percentage of neutrons that are (−)n (J − M − n)!(J + K − n)!(M − K + n)!n! not emitted from fully accelerated fission fragments is n somewhat of a controversy [25–29].  θ 2J+K−M−2n  θ 2n+M−K × cos sin , (1) 2 2 As the incident neutron energy increases, reaction channels beyond first-chance fission open. Besides the where J is the total angular momentum of the compound scission neutrons that may be emitted, other pre-fission nucleus, K is its projection on the fission axis, M is its neutrons can be emitted before the compound nucleus projection on the beam axis, and θ is the angle of the fissions. Most often, these pre-fission neutrons are emit- fission fragments with respect to the beam axis. The ted from the fully equilibrated compound nucleus. How- angular distribution of the fission fragments in the lab frame would be ever, for incident neutrons with high enough energy, & 12 MeV, a neutron can be emitted with or without the dσ 2J + 1 f (J, K, M) = |dJ (θ)|2, (2) compound nucleus forming. When the compound nu- dΩ 2 M,K cleus does not form, the emitted neutrons are considered to be pre-equilibrium. These pre-equilibrium neutrons if the fission only occurred through one of these (J, K, M) have a different energy spectrum from the pre-fission neu- states. Instead, each state must be weighted by its pop- trons coming from the compound system [30] and have ulation, Ps(J, K, M), and the full angular distribution is a forward-peaked angular distribution, more akin to the given by angular distribution of an inelastically scattered neutron dσf X 2J + 1 than the isotropic distributions of the pre-fission neutrons W (θ) ≡ = P (J, K, M) |dJ (θ)|2. dΩ s 2 M,K emitted from the compound system [31]. Pre-equilibrium s(J,K,M) neutrons have been definitively observed and their angu- (3) lar distributions measured for the first time for neutron- The weight of each state is often extracted from exper- 239 induced fission of Pu [32]. imental measurements, e.g. [16], but these weights are P R dσf connected to the fission cross section, Ps = dΩ dΩ. Recently, the fission fragment anisotropy and pre- The anisotropy of the fission fragments is defined as the equilibrium angular distributions have been implemented ratio of the angular distribution at 0◦ to the angular dis- into the Monte Carlo Hauser-Feshbach code CGMF [5, 6]. tribution at 90◦ relative to the beam axis, often written In this work, we describe the implementation and study as W (0◦)/W (90◦). of the correlations between the fission fragment and neu- tron directions. This paper is organized in the following manner: in Section II, we discuss the underlying physics B. Pre-equilibrium Neutron Angular Distribution of the angular distributions as well as the CGMF code and the implementation of these sources of anisotropy; in Sec- When the incident neutron energy is high enough, pre- tion III, we present our results and compared to experi- equilibrium neutrons can be emitted before the com- mental data; and finally, we conclude in Section IV. pound system equilibrates. Both the energy spectrum 3 and angular distributions of these neutrons are signif- D. CGMF icantly different from those of the neutrons emitted from the fully equilibrated compound nucleus. Pre- The fission fragment anisotropy and pre-equilibrium equilibrium angular distributions can be well described angular distributions have recently been implemented by the quantum mechanical theory of Feshbach, Ker- into the Monte Carlo Hauser-Feshbach code, CGMF [5, 6], man, and Koonin (FKK) [31]. Here, the angular dis- developed at Los Alamos National Laboratory (LANL). tributions of the emitted particles can be separated into As input, CGMF requires yields in mass, charge, to- multi-step direct and multi-step compound parts. The tal fission fragment kinetic energy, spin, and parity, pre-equilibrium neutrons are described by the multi-step Y (A, Z, T KE, J, π), for the fission reaction of interest, ei- direct part of the calculation, similar to inelastic scatter- ther spontaneous fission or neutron-induced fission for in- ing. cident energies from thermal to 20 MeV. The fission frag- Recently, pre-equilibrium angular distributions were ments are sampled from this initial distribution. The to- measured from 239Pu(n,f) for the first time [32]. These tal excitation energy, TXE, defined through the Q-value neutrons were shown to be first emitted around an in- as TXE = Q - TKE, is calculated and shared between the cident neutron energy of 12 MeV and were significantly two complementary fragments based on a ratio of tem- forward peaked. These forward-peaked angular distri- peratures, currently fit to reproduce the average neutron butions were well described by the FKK theory, along multiplicity as a function of mass and does not include a with Kalbach-Mahn systematics [34], which represents dependence on the incident neutron energy. Each frag- the shape of the angular distributions as a sum of Leg- ment decays through the emission of prompt neutrons endre polynomials. and γ rays according to the Hauser-Feshbach statistical theory [31]. At each step in the decay, the probability for the emission of a prompt neutron is determined from the transmission coefficient calculated from an optical po- tential, and the γ emission probability is calculated from the γ-ray strength function. Energy, spin, and parity are conserved during the full decay path. C. Other Sources of Anisotropy CGMF records the initial conditions of the fragments along with the energies and momenta of the fragments, neutrons, and γ rays for each simulated fission event. Another source of angular anisotropy for the fission This allows us to reconstruct average quantities, such as fragments and the emitted neutrons is from the recoil average neutron energy, photon energy, and number of of the compound system due to momentum conserva- particles, as well as distributions, such as energy spectra tion between the neutron beam and the fissile target. and multiplicity distributions. In addition, correlations Although the neutrons are very light compared to the between all observables can be constructed. Numerous heavy target, this effect is non-negligible. The recoil had examples of the prompt quantities that have been studied already been included in CGMF, and the resulting effect on with CGMF can be found in [35–37]. the fission fragment and neutron anisotropy will be pre- As mentioned previously, CGMF now takes into account sented in Sec. III. Similarly, there is also a small amount the angular distributions of fission fragments and of the of recoil to the fission fragments from the emission of the pre-equilibrium neutrons. As far as we are aware, this is prompt neutrons; this kinematic effect was also already the first time that these two effects have been included included in CGMF. into a fission decay code such as this. In the sections There are several other potential sources of anisotropy below, we describe how these angular distributions were that depend on how the neutrons are emitted from the fis- parametrized and included within CGMF. sion fragments. In this work, we assume that the prompt neutrons are emitted isotropically from the fission frag- ments in their center-of-mass frame. In addition, we as- 1. Fission Fragment Anisotropy sume that the neutrons are emitted from the fully ac- celerated fission fragments. Anisotropic angular distri- The fission fragment anisotropy, W (0◦)/W (90◦), was butions of the neutrons in the center-of-mass frame of introduced in CGMF for all neutron-induced reactions the fragments could either raise or lower the resulting available. For most of these nuclei, there are several ex- neutron anisotropy depending on the distribution, while perimental data sets available, shown in Fig. 1 for 235U, neutrons emitted before the fragments are fully acceler- 238U, and 239Pu as the colored stars. For each nucleus, ated would reduce the anisotropy of the neutrons with these data were fit using cubic splines, shown as the cor- respect to the fragments as the neutrons would receive respondingly colored dashed lines of Fig. 1. The knots a smaller kinematic boost from the fragments. Both of of the splines, black circles, were adjusted to reproduce these effects should be investigated and then taken into the features within the experimental data. Cubic splines account, however, that is beyond the scope of this current fits were also constructed for the anisotropy of 233,234U, work. 237Np and 241Pu. However, for the 241Pu target, data 4

2. Pre-equilibrium Neutron Angular Distributions Spline Knots ) 2.0 235 0 U(n,f) The pre-equilibrium neutron angular distributions 9 ( 1.8 238U(n,f) were fit to theoretical calculations from the FKK model W

/ 239 ) Pu(n,f) [31, 38]. These calculations were performed on a two- 1.6 ∗ 0 dimensional grid of incident, E , and excitation, E , ( inc

W energies and were individually fitted assuming Kalbach

y 1.4

p systematics [39], o r t

o 1.2 dθ

s ∗ ∗ ∗ i (Einc,E ) = a(Einc,E )sinh(cosθ)+b(Einc,E ). (5) n dΩ A 1.0 Here, both a and b were taken to be polynomial in Einc 0 5 10 15 20 and E∗. This parameterization allows an angular dis- Incident Neutron Energy (MeV) tribution to be calculated for any incident neutron en- ergy up to 20 MeV. The resulting distributions are then FIG. 1: (color online) Fission fragment anisotropy as a func- normalized to form a probability distribution function tion of incident neutron energy. Stars with error bars show and are sampled when a pre-equilibrium neutron is emit- the data, black circles are the knots that anchor the cubic ted. These distributions were calculated for 239Pu but splines, given by the dashed lines of the corresponding color, are taken to be the same for all targets. For the ac- 235 238 239 U(n,f) in blue, U(n,f) in red, and Pu(n,f) in green. tinides considered in CGMF, these angular distributions These represent some of the fits now included in CGMF. calculated with the FKK model are almost unchanged with the addition or removal of a few nucleons. The emitted neutrons are then sampled from a cumu- were only available for incident neutron energies below 8 lative distribution function, CDF. The CDF was con- MeV, and because these data have the same shape and structed for 50 values of cos(θ) ∈ [−1, 1], and points in 239 magnitude as those for Pu, the anisotropy as a func- between were interpolated linearly. Here, 50 values of 241 tion of incident neutron energy for Pu was taken to cos(θ) are enough for converged results, and this grid and 239 be the same as that of Pu, where the data extend up interpolation does not introduce any uncertainty com- to 20 MeV. For implementation into CGMF, these fits are pared to a finer grid. The angular distributions of the tabulated in steps of 0.1 MeV incident neutron energy. fission fragments were sampled in the same way. This energy is then matched with the incident neutron energy for each CGMF calculation. The uncertainty intro- duced from this rounding is less than the accuracy of the 3. Extracting the Anisotropy resulting anisotropy determined from a fit to the Monte Carlo results. Because CGMF calculates the fission events in a Monte The angle, θ, with respect to the incident neutron Carlo fashion, for each run of CGMF, we end up with a beam axis of the light fission fragment is then sampled list of properties for each of the fission fragments and from the distribution associated neutrons and γ rays. To extract the neutron anisotropy, we bin the neutron angles and fit the resulting  W (0◦)  P (cos(θ)) ∝ 1 + − 1 cos2(θ), (4) distribution to Legendre polynomials, W (90◦)

W (cosθ|a) = a0P0(cosθ) + a1P1(cosθ) + a2P2(cosθ), (6) in the center-of-mass frame of the compound system. ◦ ◦ When W (0 )/W (90 ) = 1, cos(θ) is sampled isotropi- where P0, P1, and P2 are the first three Legendre poly- cally. In all cases, the angle around the beam axis, φ, is nomials, and a = [a0, a1, a2] are the associated weights sampled isotropically. The direction of the heavy frag- of each polynomial. From each fit, we get best-fit values, ∗ ∗ ∗ ment is determined from the back-to-back momentum a0, a1, and a2, along with a covariance matrix between conservation of the fragments in the center-of-mass. the three fit parameters. To quantify the uncertainty This method, however, does not rigorously take into coming from the polynomial fit, we draw samples from account the opening of multi-chance fission channels. a multivariate Gaussian distribution centered at the best Ideally, when a fissioning nucleus emits one or more pre- fit with the associated covariance matrix. The anisotropy ∗ ∗ fission neutrons, the anisotropy of the Ac − ν system - of the neutrons is then defined as W (1|a )/W (0|a ), and where ν is the number of pre-fission neutrons emitted - 95% confidence bands are constructed from the samples should be considered in Eq. (4), instead of the anisotropy from the covariance matrix. This procedure is used to of the Ac system. From the measurement of the fission mitigate the lack of statistics when looking at specific fragment anisotropy, it is impossible to disentangle the neutron energies. The same procedure is followed to ex- joint effect of the anisotropy of each fissioning system tract the anisotropy of the fission fragments from the without further modeling. CGMF calculations. 5

III. RESULTS AND DISCUSSION 1.30 Fission fragments

) (a) 1.25 We now present results for the case of the neutron- 0 239Pu(n,f) 239 9 induced fission on Pu for incident energies up to 20 ( W

MeV. Because the sources of anisotropy can become en- / 1.20 ) tangled and more difficult to understand when we con- R 0 sider them together, we first examine, in sequence, the ( 1.15 R+FF

W R+FF+PE

compounding effects due to the recoil of the target, the y p 1.10 o r

anisotropic distribution of the fission fragments, and then t o s the angular sampling of the pre-equilibrium neutrons. i

n 1.05

Figure 2(a) shows the compounding effect on the fission A fragment angular distribution when each of these three 1.00 effects is added. The red dotted line shows the effect due 0 5 10 15 20 to the recoil of the target from momentum conservation Incident Neutron Energy of the interaction with the beam, which can be up to 5% at 20 MeV. To this, the fission fragment anisotropy due Prompt neutrons to the transition states is added (black solid line). Fi- ) 1.08 (b) (no pre-fission) nally, the angular sampling of the pre-fission neutrons 0 239

9 Pu(n,f) is included (blue dashed line). As expected, there is (

W 1.06 very little change in the fragment angular distribution / ) when the pre-equilibrium neutrons are sampled from an 0 anisotropic distribution compared to the isotropic one. ( 1.04 W

The angular anisotropy of the emitted neutrons is y p o r

lower compared to the fission fragment anisotropy. The t

o 1.02 s same additions as in Fig. 2(a) are shown in panel (b) i R n but for the angular distributions of the neutrons emitted A R+FF from the fission fragments, without pre-fission neutrons 1.00 R+FF+PE included. Because the neutrons are emitted isotropically 0 5 10 15 20 in the center-of-mass frame of each fission fragment, the Incident Neutron Energy values for the anisotropy are reduced. Although the pre- equilibrium neutrons amount to less than 20% of the 1.10 (c) Prompt neutrons (all) pre-fission neutrons, there is a noticeable boost to the ) 239 angular distribution at forward angles when these events 0 1.08 Pu(n,f) 9 are included. The forward-peaked angular distribution ( W translates to an increase in the anisotropy, as seen in the / ) 1.06 difference between the black solid and blue dashed lines 0 in Fig. 2(c). ( W 1.04 We can define a lower bound on the outgoing neutron y p o r energy, the neutron threshold energy, Ethres. As Ethres t o

s 1.02 is raised, the magnitudes of the neutron anisotropies are i R n enhanced. In Fig. 3, we show the neutron anisotropy A R+FF R+FF+PE as a function of Ethres for several incident neutron ener- 1.00 gies, indicated by the different colored symbols. Panel 0 5 10 15 20 (a) shows the neutron anisotropy when pre-fission neu- Incident Neutron Energy trons are removed from the analysis, and the neutron anisotropy when they are included is shown in panel FIG. 2: (color online) Anisotropy for (a) fission fragments, (b). In both panels, the hashed areas give the uncer- (b) prompt neutrons without pre-fission neutrons, and (c) all tainties coming from the fit to Legendre polynomials, as prompt neutrons as a function of incident neutron energy for described in Sec. II D 3. When pre-fission neutrons are different sources of angular anisotropy. In each subplot, the not considered, as in Fig. 3(a), we see that the neutron red dotted line shows the effect due to the kinematic recoil anisotropy trends toward the fission fragment anisotropy (labeled R), the black solid line shows the joint effect of both as Ethres increases. Although this picture is complicated the kinematic recoil and the fission fragment angular distribu- by the addition of pre-fission neutrons, there appear to be tion (R+FF), and the blue dashed line depicts the anisotropy strong correlations between the neutron anisotropy and when the pre-equilibrium angular sampling is added to the first two effects (R+FF+PE), for 239Pu(n,f). Note the differ- Ethres that could be used to extract information about the fission fragment anisotropy. ence in the y-scale across the three panels, especially for (a) compared to (b) and (c). As shown in Fig. 2, the emission of prompt neutrons dilutes the signature of the anisotropy coming from the 6

1.8 Fission Fragments Einc = 0 MeV 239Pu(n,f) (a)

) Prompt neutrons (no pre-fission) ) Einc = 5 MeV 1.3 Prompt neutrons (all)

Einc = 10 MeV 0

0 1.6

9 Fragment data 9 Einc = 15 MeV ( (

Einc = 20 MeV W W / / ) ) 1.4 1.2 0 0 ( ( W

W

y

y 1.2 p p

o 1.1 o r r t t o o s s i i

1.0 n 239 n Pu(n,f) A A Neutrons from only fission fragments 1.0 Ethres = 0 MeV 0.8 0 2 4 6 8 10 0 5 10 15 20 Neutron Threshold Energy (MeV) Einc (MeV)

1.8 Einc = 0 MeV 239Pu(n,f) (b) FIG. 4: (color online) Neutron anisotropy as a function of ) Einc = 5 MeV incident energy without (blue) and with (red) pre-fission neu- Einc = 10 MeV 0 1.6 trons included, compared to the experimental data (grey) that

9 Einc = 15 MeV ( were used to construct the spline fit for CGMF and resulting fis- Einc = 20 MeV W

/ sion fragment anisotropy (black) when no neutron threshold ) 1.4 energy is considered. The dashed lines show the associated 0 ( 95% confidence bands of the Legendre polynomial fit. The W

y 1.2 difference between the CGMF calculation for the fission frag- p

o ment anisotropy and the experimental data is discussed in r t

o the text. s i 1.0 n A All neutrons 0.8 emitted, around E = 12 MeV, the difference between 0 2 4 6 8 10 inc Neutron Threshold Energy (MeV) the two neutron anisotropies becomes more visible, as some of the pre-fission neutrons are now emitted prefer- entially in the direction of the incident neutron beam. FIG. 3: (color online) Neutron anisotropy as a function of In most cases, the most energetic neutrons are those threshold energy (filled circles) compared to the fission frag- ment anisotropy (dashed lines in the corresponding colors) (a) that are emitted along the direction of the fission frag- only neutrons emitted from the fission fragments are consid- ments, as they gain the strongest kinematic boost from ered and (b) pre-fission neutrons are included in addition. the motion of the fragments. Therefore, as can be seen in the differences between panels (a), (b), and (c) in Fig. 5, as the energy threshold for the neutrons increases, the fission fragments, as the neutrons are emitted isotrop- neutron anisotropy is more aligned with the anisotropy of ically in the center-of-mass frame of the fission frag- the fission fragments. When Ethres = 7 MeV, Fig. 5(b), ments. However, as seen in Fig. 3(a), as the outgo- the neutron anisotropy and associated uncertainties from ing neutron energy increases, the neutron anisotropy ap- the fit to the Legendre polynomials covers the spread proaches the fission fragment anisotropy. Therefore, if in the experimental data, with 62% of the data falling neutrons with a high enough energy are measured, the within the confidence bands for the neutron anisotropy fission fragment anisotropy could be extracted from the when experimental uncertainty is taken into account. We neutron anisotropy. Of course, pre-fission neutrons com- do see that the fission fragment anisotropy extracted from plicate this picture, as in Fig. 3(b), since they are emitted CGMF, black solid lines in Fig. 5, is somewhat higher than isotropically in the rest frame of the compound nucleus. the experimental anisotropy, due to the kinematic boost A hint of this difference when the pre-fission neutrons are from the incident neutron beam. The correction for the included is seen in Fig. 4 which shows the anisotropy of neutrons from the recoil of the target is less than 4%. the fission fragments in black, prompt neutrons from the Raising the neutron threshold energy pushes the effect fragments in blue, and all neutrons in red, compared to of the pre-fission neutrons to higher incident energies. the experimental data for the fission fragments, light grey For Ethres = 5 MeV, as in Fig. 5(a), pre-fission neu- circles. Around Einc = 5 MeV, where the second chance trons only begin to have an effect when the incident neu- fission channel opens, very slight differences can be seen tron energy is above 10 MeV. As the threshold energy between the calculations of the neutron anisotropy with is increased, the pre-fission neutrons have an increas- and without pre-fission neutrons, but these differences ingly smaller effect on the overall neutron anisotropy; are primarily within the uncertainty bands from each the difference in the anisotropy is only seen for higher calculation. When the pre-equilibrium neutrons are first incident neutron energies. The pre-fission neutron en- 7

10 1.36 Fission Fragments (a)

) Prompt neutrons (no pre-fission) 1.3 Prompt neutrons (all) 1.28 0 8

9 Fragment data ( 1.20 W / ) 1.2 6 1.12 0 ( W 1.04 y 4 p

o 1.1 r t

o 0.96 s Ratio of Anisotropies i 239 n Pu(n,f) Threshold Energy (MeV) 2 A E = 5 MeV 0.88 1.0 thres 239Pu(n,f) 0 0.80 0 5 10 15 20 0 5 10 15 20 Einc (MeV) Incident Energy (MeV)

Fission Fragments (b) FIG. 6: (color online) Ratio between the neutron anisotropy

) 1.5 Prompt neutrons (no pre-fission) Prompt neutrons (all) (all prompt neutrons) and the fission fragment anisotropy as a 0 function of the incident neutron energy and neutron threshold 9 Fragment data

( 1.4 energy for 239Pu(n,f). W / ) 1.3 0 ( W

y 1.2 ergy spectrum is typically peaked around 2 MeV and p o

r drops off at higher outgoing neutron energies. However, t o

s 1.1 around E ∼ 9 − 10 MeV, the pre-fission neutrons be- i 239 inc

n Pu(n,f)

A come more numerous than the neutrons emitted from 1.0 Ethres = 7 MeV the fission fragment and dominate the anisotropy when only higher neutron energies are considered. As the neu- 0 5 10 15 20 E (MeV) tron threshold energy continues to increase, the emission inc of high-energy pre-fission neutrons would not leave the residual compound with enough energy to fission. Thus, Fission Fragments (c) the neutrons with the highest energies are emitted from ) 1.6 Prompt neutrons (no pre-fission) Prompt neutrons (all) the fission fragments, in the same direction as the frag- 0

9 Fragment data ments. The trade-off here is that fewer neutrons are emit- (

W ted with these highest energies, which is clear from the / ) 1.4 growing uncertainty bands between each of these energy 0 ( thresholds. W

y We can then calculate a correlation factor between the p 1.2 o r

t neutron anisotropy and the fission fragment anisotropy, o s

i defined as the ratio of the neutron anisotropy (including n 239

A Pu(n,f) 1.0 pre-fission neutrons) to the fission fragment anisotropy Ethres = 9 MeV at every incident and threshold neutron energy. In Fig. 0 5 10 15 20 6, this ratio is plotted as a function of incident neutron Einc (MeV) energy and threshold neutron energy. Thus, knowing the incident and outgoing neutron energies and measuring the neutron anisotropy, the fission fragment anisotropy FIG. 5: (color online) Neutron anisotropy as a function of could be extracted with this correlation factor. It is incident energy without (blue) and with (red) pre-fission neu- worth noting that the correlation factor rises above one trons included compared to the experimental data (grey) that at the highest incident energies and neutron threshold was used to construct the spline fit for CGMF and resulting fis- sion fragment anisotropy (black). The dashed lines show the energies due to the inclusion of the pre-fission neutrons, associated 95% confidence bands of the Legendre polynomial consistent with what is shown in Fig. 5 in red. fit. In (a) Ethres = 5 MeV, (b) Ethres = 7 MeV, and (c) We also performed these studies for neutron-induced Ethres = 9 MeV. fission on 233,234,235,238U, 237Np, and 241Pu, all avail- able in CGMF. In all cases, we found very similar results to what was shown here for 239Pu(n,f). The most no- table difference was for even isotopes of uranium, which were the only even-even targets studied here. Differences in the spin of the target and the available transition 8 states can lead to large differences in the fission frag- ment anisotropy between even-even and even-odd target 238 1.3 U(n,f) nuclei [15, 40]. An increase of anisotropy is seen for all targets at the opening of the higher multi-chance fission channels, due to the population of few transition states 1.2 in residual fissioning nucleus with less excitation energy. ) 1.1 E

The density of states in the residual nucleus depends on ( the odd/even character of the nucleus. For these iso- topes, the values of the anisotropy of the fission fragments 1.0 could reach nearly to 2, significantly larger than any of the values shown here for 239Pu(n,f). When the fragment 0.9 CGMF anisotropy was this large, the neutron anisotropy did not Chi-Nu reach those values, regardless of Ethres (within the limits 0.8 0 5 10 15 20 of the statistics of our calculations). When the fragment Incident Neutron Energy (MeV) anisotropy is high, increasing the energy threshold for the neutrons has only a small effect on the resulting neutron anisotropy. For this case, the neutrons seem to already FIG. 7: Comparison between CGMF calculations, red circles, be boosted in the direction of the fission fragments. In- and preliminary Chi-Nu data, black stars, for ρ(Eα) as given 238 creasing the neutron threshold energy would then have in Eq. (7) for U(n,f). a smaller effect on the resulting neutron anisotropy, as fewer neutrons are being excluded based on these cuts. Further studies on this feature should be performed. the number of fissions, the neutron detector efficiencies, Recently, several measurements of the angles and en- and the neutron scattering corrections, leaving primar- ergies of prompt neutrons from the fission of major ac- ily statistical uncertainties. The ratio resulting from Eq. tinides have been made with the Chi-Nu detector array (7) is intrinsically a shape quantity as a function of in- at LANL. The Chi-Nu experimental setup consists of cident neutron energy that contains a reflection of the two detector arrays used in separate measurements: a fission fragment anisotropy. Because this shape is, by 22-detector Li-glass array to measure outgoing neutrons default, scaled to have ρ(Eα) = 1 at the lowest incident with energy from 0.01 – 1.59 MeV and a 54-detector liq- energy (Einc = 1.54 MeV), and there is no measurement uid scintillator array to measure outgoing neutrons from at thermal energies where the ratio should be one, this approximately 0.89 – 10 MeV. The spectra from these shape was scaled to reproduce the magnitude of the CGMF detectors are combined to form prompt fission neutron calculations from 6 − 14 MeV incident neutron energy. spectrum (PFNS) measurements from 10 keV to 10 MeV The same ratio can be calculated from the CGMF his- for incident neutron energies from 1 – 20 MeV. More tories and compared to the measurement from Chi-Nu. details can be found in [41–43]. Prompt neutrons from 238 This comparison is shown in Fig. 7. For the best trade- neutron-induced fission on U were measured with the off between statistics and high incident neutron energies Chi-Nu liquid scintillator array to study the PFNS up to where the neutron anisotropy most closely tracks the fis- 20 MeV incident neutron energy. The Chi-Nu liquid scin- sion fragment anisotropy, the sum in Eq. (7) is taken tillator array covers nine outgoing neutron angles from ◦ from Ei = 5 − 10 MeV. The data from Chi-Nu are shown 30 – 150 in the lab frame, allowing for a comparison be- as the black filled stars, compared to the CGMF calculation tween the CGMF calculations of the neutron anisotropies of Eq. (7) as the red filled circles. The CGMF calculations to those measured by Chi-Nu. are scaled such that ρ(Eα) = 1 for thermal neutrons, and In order to exclude many sources of systematic error Chi-Nu was scaled to match the relative magnitude of the from the experiment, the neutron anisotropies from Chi- ◦ CGMF calculations to be able to compare the shapes of the Nu are presented as a ratio of experimental data at 30 two ratios. and 90◦, normalized to the lowest incident neutron en- ergy range, The shapes of the distributions from Chi-Nu and CGMF are mostly consistent within the experimental uncertain- P ◦ ◦ C(Ei,Eα, 30 )/C(Ei,Eµ, 30 ) ties. The experimental uncertainties are large compared i to those of the calculation, which are only statistical and ρ(Eα) = P ◦ ◦ . (7) C(Ei,Eα, 90 )/C(Ei,Eµ, 90 ) do not account for model defects. In the calculated and i measured ratios, there is a rise up to Einc ∼ 2 MeV, Ei are the outgoing neutron energies, Eα and Eµ are in- followed by a flattening before the second-chance fission cident neutron energies, where Eµ is the same for each of channel opens around incident energies of 6 MeV. Both the ratio calculations and is taken to be the lowest mea- ratios flatten off again, and then there is another increase sured incident neutron energy, and C(Ei,Eα, θ) is the in ρ(Eα) where the third-chance fission channel opens, number of foreground counts in the detector at angle θ although there could be a disagreement between the in- for Ei and Eα. This formulation allows many sources of cident neutron energy where this channel opens experi- uncertainty to cancel or cancel approximately, including mentally and in CGMF. The Chi-Nu experiment was not 9 optimized to extract the neutron anisotropy, and a dif- higher incident neutron energies if the neutron threshold ferent set-up could be put in place to more accurately energy is increased, but statistics become significantly measure this quantity. The CGMF calculations could then worse. A correlation factor, such as the one calculated be used to convert the experimental values of ρ(Eα) to here with CGMF, could be used to extract the fission frag- fission fragment anisotropies and compare with data mea- ment anisotropy from the neutron anisotropy or vise- sured directly. versa depending on the incident and outgoing threshold neutron energies, even when pre-fission neutrons are in- cluded in the analysis. Here, we performed a preliminary IV. CONCLUSIONS comparison with 238U(n,f) data measured at Chi-Nu, us- ing a ratio method which showed rather good agreement In summary, angular distribution sampling for both between the shapes of the CGMF calculations and experi- the fission fragment anisotropy and the pre-equilibrium mental measurement. neutrons were recently introduced into the Monte Carlo There are still other effects that we have not taken into fission code CGMF. Experimental values for the fission account in this study that could also contribute to neu- fragment anisotropy were fitted using cubic splines while tron anisotropy, including the angular distribution of the the pre-equilibrium neutron angular distributions were neutrons in the center-of-mass frame of the fission frag- calculated with the semi-direct model of Feshbach, Ker- ments, the emission of neutrons from fission fragments man, and Koonin and fitted as a function of neutron that are not fully accelerated, and the emission of scis- incident energy and compound excitation energy using sion neutrons. These effects should be studied theoreti- Kalbach systematics. In itself, the inclusion of these ef- cally and experimentally and included in the model. fects is important to interpret experimental results from Acknowledgments detector setups that measure neutron angles and ener- gies, such as the Chi-Nu array. We also found that high- energy neutrons track the fission fragment anisotropy rel- The authors would like to thank Matt Devlin and atively well, and this congruence improves as the energy John O’Donnell for their feedback on the comparison threshold of the neutrons is increased. This was studied to the Chi-Nu data and Toshihiko Kawano for his pre- for neutron-induced fission reactions on 233,234,235,238U, equilibrium neutron calculations and helpful discussions 237Np, and 239,241Pu. about those calculations. This work was performed Excluding pre-fission neutrons, for a threshold energy under the auspice of the U.S. Department of Energy of 7 MeV for the outgoing neutrons, the uncertainty by Los Alamos National Laboratory under Contract of the theory coming from the Legendre polynomial fit 89233218CNA000001 and was supported by the Office to the CGMF calculations is similar to the spread in the of Defense Nuclear Nonproliferation Research & Devel- available data. When the pre-fission neutrons are in- opment (DNN R&D), National Nuclear Security Admin- cluded, the neutron anisotropy does not change until af- istration, U.S. Department of Energy. We gratefully ac- ter the opening of third-chance fission. This difference knowledge the support of the U.S. Department of Energy in the neutron anisotropy when not including and in- through the LANL/LDRD Program and the Center for cluding pre-fission neutrons is only seen at increasingly Non Linear Studies.

[1] N. Schunck, D. Duke, H. Carr, and A. Knoll, Phys. Rev. 4655, URL http://www.sciencedirect.com/science/ C 90, 054305 (2014), URL https://link.aps.org/doi/ article/pii/S0010465515000466. 10.1103/PhysRevC.90.054305. [8] J. Verbeke, J. Randrup, and R. Vogt, Computer [2] A. Bulgac, P. Magierski, K. J. Roche, and I. Stetcu, Phys. Physics Communications 222, 263 (2018), ISSN 0010- Rev. Lett. 116, 122504 (2016), URL https://link.aps. 4655, URL http://www.sciencedirect.com/science/ org/doi/10.1103/PhysRevLett.116.122504. article/pii/S001046551730293X. [3] J. Randrup and P. M¨oller, Phys. Rev. Lett. 106, [9] S. Okumura, T. Kawano, P. Talou, P. Jaffke, and 132503 (2011), URL https://link.aps.org/doi/10. S. Chiba, J. Nucl. Sci. Tech. 55, 1009 (2018), 1802.01248. 1103/PhysRevLett.106.132503. [10] O. Litaize, O. Serot, and L. Berge, The European Phys- [4] P. M¨ollerand T. Ichikawa, The European Physical Jour- ical Journal A 51, 177 (2015), ISSN 1434-601X, URL nal A 51, 173 (2015). https://doi.org/10.1140/epja/i2015-15177-9. [5] P. Talou, T. Kawano, I. Stetcu, P. Jaffke, M. E. Rising, [11] K.-H. Schmidt, B. Jurado, C. Amouroux, and and A. E. Lovell, Computer Physics Communication (in C. Schmitt, Nuclear Data Sheets 131, 107 (2016), progress). ISSN 0090-3752, special Issue on Nuclear Reaction [6] B. Becker, P. Talou, T. Kawano, Y. Danon, and I. Stetcu, Data, URL http://www.sciencedirect.com/science/ Phys. Rev. C 87, 014617 (2013), URL https://link. article/pii/S0090375215000745. aps.org/doi/10.1103/PhysRevC.87.014617. [12] A. S. Vorobyev, A. M. Gagarski, O. A. Shcherbakov, [7] J. Verbeke, J. Randrup, and R. Vogt, Computer L. A. Vaishnene, and A. L. Barabanov, JETP Letters Physics Communications 191, 178 (2015), ISSN 0010- 102, 203 (2015), ISSN 1090-6487, URL https://doi. 10

org/10.1134/S0021364015160134. //www.sciencedirect.com/science/article/pii/ [13] S. Ahmad, M. M. Islam, A. H. Khan, M. Khaliquz- 0029558263907855. zaman, M. Husain, and M. A. Rahman, Nu- [29] N. Carjan and M. Rizea, Phys. Rev. C 99, 034613 (2019), clear Science and Engineering 71, 208 (1979), URL https://link.aps.org/doi/10.1103/PhysRevC. https://doi.org/10.13182/NSE79-A20413, URL 99.034613. https://doi.org/10.13182/NSE79-A20413. [30] E. Gadioli and P. E. Hodgson, Pre-Equilibrium Nuclear [14] J. E. Brolley, W. C. Dickinson, and R. L. Henkel, Phys. Reactions (Oxford Scientific Publications, 1992). Rev. 99, 159 (1955), URL https://link.aps.org/doi/ [31] H. Feshbach, A. Kerman, and S. Koonin, An- 10.1103/PhysRev.99.159. nals of Physics 125, 429 (1980), ISSN 0003- [15] J. E. Simmons and R. L. Henkel, Phys. Rev. 120, 4916, URL http://www.sciencedirect.com/science/ 198 (1960), URL https://link.aps.org/doi/10.1103/ article/pii/0003491680901402. PhysRev.120.198. [32] K. J. Kelly, T. Kawano, J. M. O’Donnell, J. A. Gomez, [16] V. Geppert-Kleinrath, F. Tovesson, J. S. Barrett, M. Devlin, D. Neudecker, P. Talou, A. E. Lovell, M. C. N. S. Bowden, J. Bundgaard, R. J. Casperson, D. A. White, R. C. Haight, et al., Phys. Rev. Lett. 122, Cebra, T. Classen, M. Cunningham, D. L. Duke, 072503 (2019), URL https://link.aps.org/doi/10. et al. (NIFFTE Collaboration 2), Phys. Rev. C 99, 1103/PhysRevLett.122.072503. 064619 (2019), URL https://link.aps.org/doi/10. [33] C. Wagemans, The Nuclear Fission Process (CRC Press, 1103/PhysRevC.99.064619. 1991). [17] R. L. Henkel and J. E. Brolley, Phys. Rev. 103, [34] C. Kalbach and F. M. Mann, Phys. Rev. C 23, 1292 (1956), URL https://link.aps.org/doi/10. 112 (1981), URL https://link.aps.org/doi/10.1103/ 1103/PhysRev.103.1292. PhysRevC.23.112. [18] J. E. Brolley and W. C. Dickinson, Phys. Rev. 94, [35] P. Talou, R. Vogt, J. Randrup, M. E. Rising, S. A. 640 (1954), URL https://link.aps.org/doi/10.1103/ Pozzi, J. Verbeke, M. T. Andrews, S. D. Clarke, P. Jaf- PhysRev.94.640. fke, M. Jandel, et al., The European Physical Journal A [19] E. Birgersson, A. Oberstedt, S. Oberstedt, and F.-J. 54, 9 (2018), ISSN 1434-601X, URL https://doi.org/ Hambsch, Nuclear Physics A 817, 1 (2009), ISSN 0375- 10.1140/epja/i2018-12455-0. 9474, URL http://www.sciencedirect.com/science/ [36] I. Stetcu, P. Talou, T. Kawano, and M. Jandel, Phys. article/pii/S0375947408007975. Rev. C 90, 024617 (2014), URL https://link.aps.org/ [20] I. Ryzhov, M. Onegin, G. Tutin, J. Blomgren, doi/10.1103/PhysRevC.90.024617. N. Olsson, A. Prokofiev, and P.-U. Renberg, Nu- [37] M. J. Marcath, R. C. Haight, R. Vogt, M. De- clear Physics A 760, 19 (2005), ISSN 0375- vlin, P. Talou, I. Stetcu, J. Randrup, P. F. Schus- 9474, URL http://www.sciencedirect.com/science/ ter, S. D. Clarke, and S. A. Pozzi, Phys. Rev. C article/pii/S0375947405009103. 97, 044622 (2018), URL https://link.aps.org/doi/ [21] F. Vivs, F.-J. Hambsch, H. Bax, and S. Oberst- 10.1103/PhysRevC.97.044622. edt, Nuclear Physics A 662, 63 (2000), ISSN 0375- [38] T. Kawano, private communication (2018). 9474, URL http://www.sciencedirect.com/science/ [39] C. Kalbach, Phys. Rev. C 37, 2350 (1988), URL https: article/pii/S0375947499004133. //link.aps.org/doi/10.1103/PhysRevC.37.2350. [22] R. B. Leachman and L. Blumberg, Phys. Rev. 137, [40] J. M. Mueller, M. W. Ahmed, B. Davis, J. M. Hall, S. S. B814 (1965), URL https://link.aps.org/doi/10. Henshaw, M. S. Johnson, H. J. Karwowski, D. Markoff, 1103/PhysRev.137.B814. L. S. Myers, B. A. Perdue, et al., Phys. Rev. C 85, [23] L. Blumberg and R. B. Leachman, Phys. Rev. 116, 014605 (2012), URL https://link.aps.org/doi/10. 102 (1959), URL https://link.aps.org/doi/10.1103/ 1103/PhysRevC.85.014605. PhysRev.116.102. [41] M. Devlin, J. A. Gomez, K. J. Kelly, R. C. Haight, J. M. [24] A. Bohr, in International Conference on the Peaceful O’Donnell, T. N. Taddeucci, H. Y. Lee, S. M. Mosby, Uses of Atomic Energy, edited by N. Y. United Nations B. A. Perdue, N. Fotiades, et al., Nuclear Data Sheets (Geneva, 1956), vol. 2, p. 151. 148, 322 (2018), ISSN 0090-3752, special Issue on Nu- [25] H. R. Bowman, S. G. Thompson, J. C. D. Milton, and clear Reaction Data, URL http://www.sciencedirect. W. J. Swiatecki, Phys. Rev. 126, 2120 (1962), URL com/science/article/pii/S0090375218300279. https://link.aps.org/doi/10.1103/PhysRev.126. [42] K. J. Kelly, M. Devlin, J. A. Gomez, J. M. O´Donnell, 2120. T. N. Taddeucci, R. C. Haight, H. Y. Lee, S. M. [26] H. R. Bowman, J. C. D. Milton, S. G. Thompson, and Mosby, N. Fotiades, D. Neudecker, et al., EPJ Web W. J. Swiatecki, Phys. Rev. 129, 2133 (1963), URL Conf. 193, 03003 (2018), URL https://doi.org/10. https://link.aps.org/doi/10.1103/PhysRev.129. 1051/epjconf/201819303003. 2133. [43] K. J. Kelly, J. A. Gomez, J. M. O’Donnell, M. Devlin, [27] S. S. Kapoor, R. Ramanna, and P. N. R. Rao, Phys. Rev. R. C. Haight, T. N. Taddeucci, S. M. Mosby, H. Y. 131, 283 (1963), URL https://link.aps.org/doi/10. Lee, D. Neudecker, M. C. White, et al., Tech. Rep., Los 1103/PhysRev.131.283. Alamos Nat. Lab. LA-UR-18-28140 (2018). [28] K. Skarsvg and K. Bergheim, Nuclear Physics 45, 72 (1963), ISSN 0029-5582, URL http: