Available online at www.sciencedirect.com
Nuclear Data Sheets 118 (2014) 230–232 www.elsevier.com/locate/nds
Angular Momentum Distribution of Fission Fragments I. Stetcu,∗ P. Talou, T. Kawano, and M. Jandel Los Alamos National Laboratory, Los Alamos, NM 87545, USA Latest generation fission experiments provide an excellent testing ground for theoretical models. In this contribution we compare the measurements obtained with the DANCE calorimeter at LANSCE with our full-scale simulation of the primary fragment de-excitation, using the recently developed CGMF code, based on a Monte-Carlo implementation of the Hauser-Feshbach theoretical model. We compute the isomeric ratios as a function of the initial angular momentum of the fission fragments. Comparison with the available experimental data allows us to determine the initial spin distribution. Finally, we study the sensitivity to the discrete spectra input. I.
INTRODUCTION
For a detailed description of the properties of prompt fission neutrons and gamma rays, current approaches depend on the availability of accurate experimental information and a handful of theoretical assumptions [1, 2]. During the fission process, mostly short-lived neutronrich nuclei are formed, for which the experimental data are scarce. In addition to nuclear structure information, theoretical models which describe in detail the deexcitation of the fission fragments rely on a number of other non-observables like the optical potential used to model the neutron emission, the densities of states, etc. The effect of those uncertainties is difficult to quantify, especially if one only considers average quantities like the average number of neutrons/gammas per fission event. Hence, it has become increasingly necessary to consider other observables and correlations, which can provide excellent diagnosis tools. In this paper, we investigate the role of the average angular momentum of the initial fission fragments to the description of the average γ-multiplicity distribution measured recently at DANCE, and the production isomeric ratio for select nuclides. The initial spin distribution of the fission fragments plays an essential role in our approach, as it controls the competition between neutron and photon emissions. For example, too low initial spins usually produce too many neutrons and too few gamma rays. However, direct measurements of the initial spins are not possible, and the only information available is extracted from other fission observables like isomeric ratios [3, 4], gamma-ray de-excitation feeding patterns of the ground-state bands [5], and angular anisotropy of prompt-fission gamma-rays [6]. Such procedures rely on statistical models [7] but are
∗
Corresponding author:
[email protected]
http://dx.doi.org/10.1016/j.nds.2014.04.044 0090-3752/2014 Published by Elsevier B.V.
model dependent. Hence, we assume an initial distribution of the spins of the fragments and compare our results against measured isomeric ratios. We find that the isomeric ratios are sensitive not only to to the initial spin distribution, but also to the accurate knowledge of highspin states in the discrete spectra of select nuclides. II.
FRAMEWORK AND RESULTS
The main tool at our disposal is the recently developed CGMF code [2] for fission modeling. We make the assumption that all the prompt neutrons and γ-rays are emitted from the fully accelerated fragments. The fragments are generated using experimental information about the yields of the primary fission fragments, in mass, charge and total kinetic energy (TKE), Y (A, Z, T KE), produced in a particular fission reaction. Theoretical calculations of Y (A, Z, T KE) are not yet available and such a detailed quantity can only be obtained by combining several experiments. The release of the internal excitation energy of each fragment by neutron and gamma evaporation is modeled in a Monte-Carlo HauserFeshbach statistical approach [8]. In order to model the de-excitation, the initial conditions in which the fragments are formed play an important role. In the HauserFeshbach approach, the probability to emit neutrons or γ rays depend on the transmission probabilities and the nuclear level densities of the final state. For neutrons, the transmission coefficients are obtained using a global optical potential like Koning-Delaroche [9], while γ-rays are extracted from the γ-strength function using the Kopecky-Uhl formalism [10]. For transitions between discrete levels, we use the experimental data available in the RIPL3 database [11, 12]. In the following, we concentrate on the neutron induced fission of 235 U, but the framework is general, and can be applied with minimal changes, to other fission processes
Angular Momentum Distribution . . .
NUCLEAR DATA SHEETS
0.8
10-2
DANCE α=1.0
10-3
α=1.3 α=2.0 0
2
4
6
(b)
0.7 0.6 0.5
135
0.4
Isomeric Ratio
Probability
10-1
I. Stetcu et al.
8 10 12 14 16 18 20 M’γ
FIG. 1. The prompt-γ multiplicity distribution in the thermal neutron induced fission of 235 U. The simulation results for different input parameters αI are further processed using GEANT4 to account for the detector efficiency and geometry and then directly compared against the experimental data (squares).
2
(Z,A,T ))
,
0.9 0.8
83
Se
RIPL-3 RIPL-3 2012 exp
0.6 0.5 0.4
(a) 0.5
1.0
1.5
αI
2.0
2.5
3.0
FIG. 2. The isomeric ratio for selected isotopes as a function of the parameter αI . We calculate the yield ratio of the high spin state to the total yield for the same isotope. Both the standard [11] and 2012 updated [12] RIPL-3 discrete levels have been used in the calculations.
Before discussing the isomeric ratios, defined here as the ratio of the yield of the highest-spin state (either the isomer itself or the ground state) to the total production of the isotope, we consider the case of a global αI . For 235 U, we found that using αI = 1.3 represents the best overall agreement with DANCE results, even though in Fig. 1 our results for the multiplicity distribution underestimate the newly measured low-multiplicity data [13]. While αI = 1 gives better agreement up to multiplicity 10, a higher value is necessary to describe the energy spectrum [13], and αI ≈ 2 is needed to reproduce ν¯ with good accuracy. For the isomeric ratio it is not necessary to perform a full simulation of the de-excitation of all possible fission fragments. Thus, for a specific isomeric ratio, we generate only those fragments that could contribute to this observable keeping only those events that produce the isotope of interest after the neutron emission. We interrupt the gamma cascade if the lifetime of the decaying state is greater than 0.8 s, value smaller than the lifetimes of the isomers considered here. We have performed calculations for a large number of isotopes for which experimental data on isomeric ratios exist, and the results are summarized in Table I. The allowed intervals for the parameter αI are extracted so that our predicted isomeric ratios are in agreement with the experimental data, taken from the compilation in Ref. [14]. In Fig. 2 we plot the isomeric ratio as a function of the parameter αI for selected isotopes, the shadowed region being allowed by the experiment. As illustrated in the table, the αI depends heavily on the isotope investigated, and, because of model dependence in all calculations, our extracted root mean square values of angular momenta, Jrms , generally differ from previous estimates. No single parameter αI consistent will all data can be
(1)
with B defined in terms of the fragment temperature T : B 2 (Z, A, T ) = αI I0 (A, Z)T /2 ,
0.2
0.7
(γ-induced or spontaneous). Once the primary fragments with a given T KE are sampled from Y (A, Z, T KE), their total excitation energy (T XE) can be calculated from the energy balance of the reaction. While no direct measurement of individual excitation energies is available, experimental information can be indirectly obtained from the average neutron multiplicity ν¯(A), which is strongly influenced by the amount of excitation energy available to each fragment. In our current approach, the sharing between the two fragments is governed by the ratio of the initial fragment temperatures RT = T0l /T0h . In current calculations, RT is mass-dependent, and was inferred by fitting the ratio ν¯l /¯ νh as a function of the fragment mass in a Weisskopf formalism [1]. Another important quantity for which no direct measurement exists is the initial spin distribution of the two fragments. Unlike RT , which influences mostly the neutrons and has little effect for γ-rays, the spin distribution has a large impact on the prompt-γ multiplicities, with less effect for prompt neutrons. In the current implementation, we assume a Gaussian probability distribution P (J) ∝ (2J + 1)e−J(J+1)/(2B
Xe
0.3
(2)
where I0 (A, Z) is the ground-state moment of inertia of the fragment with mass A and atomic mass Z, and αI is an adjustable parameter that we can fit to reproduce select observables. In our previous publication [2], we had fixed a global value of αI to reproduce correctly the average number of neutrons per fission event, ν¯. In general, we found that, to obtain a reasonable value for ν¯, it is necessary to increase significantly the average angular momentum, which in turn increases the competition between the neutron and γ emissions [2]. This translates into average values of angular momentum somewhat larger than those existent in the literature [2]. However, because the angular momentum extraction is model dependent, the focus should be a description of a directly measured quantity like isomeric ratios for select isotopes. 231
Angular Momentum Distribution . . .
NUCLEAR DATA SHEETS
I. Stetcu et al.
state, rather than the high-spin state. This is, indeed, what happens in the case presented in Fig. 2(b), where in the 2012 updated RIPL-3 database [12], two states at 2.3565 and 2.3875 MeV, respectively, both with spin 19/2 have been identified for 135 Xe. The existence of these two states changes the discrete transition path followed during the de-excitation in the case of high initial spin fragments, producing indeed the expected behavior. Less uncertainty impacts calculations close to stability. Thus, for neutron capture on stable nuclei, the initial conditions are well known and the process of de-excitation can be more reliably described. In this case, our predicted isomeric ratios are in general within 10 − 50% compared with experiment. Assuming a similar increased uncertainty for the fission isomeric ratios in Table I, we obtain better consistency for the αI parameter, although in this case the constraints for αI become too relaxed to be useful.
TABLE I. αI range extracted from the isomeric ratios measured after the prompt neutron and gamma de-excitation of the isotopes produced in the thermal neutron induced fission of 235 U. We also present our results for Jrms and, where available, previously extracted values for Jrms [3] for comparison. Nucleus αI Jrms (present) Jrms (previous) 83 Se >1.2 >7 90 Rb 0.2 − 0.5 3.1 − 4.2 119 Cd < 0.3 < 5.8 121 Cd >2 > 11.8 123 Cd 1.1 − 1.2 9 123 In > 0.8 > 6.5 125 In > 0.7 >6 127 Sn > 0.9 > 8.5 128 Sn < 0.4 < 5.1 130 Sb 0.5 − 1.1 5.5 − 7.5 9.2±1.5; 8.6±0.9 131 Te 0.5 − 0.8 8 − 8.2 5.5±1.2; 5.4±1.0 133 Te 0.45 − 0.95 6−8 4.6±0.4;4.7±8; 4.9±0.8 135 Xe 0.75 − 1.15 7.5 − 9 5.4±1.0; 5.4±1.5;5.65±0.70 138 Cs 0.62 − 1.2 6.5 − 9.5 10.0±1.1
III.
SUMMARY AND CONCLUSION
We have investigated the role of initial angular momentum distribution and nuclear structure information in the description of prompt neutrons and photons emitted following the thermal neutron-induced fission of 235 U. We have compared our simulations against newly obtained experimental data, as well as select isomeric ratios, which allowed us to target properties that involve a relatively small number of nuclides. The average angular momentum obtained using constraints from isomeric ratios are not always in good agreement with other extractions from experimental data. We demonstrated that the accuracy of the available nuclear data, in this case known discrete levels, plays an important role in our ability to describe the isomeric ratios. More quantities similarly sensitive to experimental and theoretical information, which can shed light on the fission process will be considered in the future.
extracted, as demonstrated in Table I, and the calculations are quite sensitive to the choice of αI . However, because the fragments produced are short-lived, the simulations could be also influenced by the poor knowledge of experimental data. Thus, an indication of a problem induced by the insufficient knowledge of nuclear structure is already displayed in Fig. 2(b), where we plot the isomeric ratio production for 135 Xe, as a function of αI , using two slightly different sets of experimental discrete levels. As one can see immediately, the original RIPL-3 set [11] produces counterintuitive results. By increasing αI , we increase the initial angular momentum of the fragment which should induce an increased production of the isomeric state with high spin. One thus expects a monotonic increase of the isomeric ratio with αI , as observed in Fig. 2(a). However, if discrete high-spin states are missing, one can follow a path that ends up in the low-spin
Acknowledgments: This work was performed at Los Alamos National Laboratory, under the auspices of the US DOE.
[9] A. Koning and J. Delaroche, Nucl. Phys. A 713, 231 (2003). [10] J. Kopecky and M. Uhl, Phys. Rev. C 41, 1941 (1990). [11] R. Capote et al., Nucl. Data Sheets 110, 3107 (2009), ISSN 0090-3752, Special Issue on Nuclear Reaction Data. [12] R. Capote, “RIPL-3 updated data base,” (2012), private communication. [13] M. Jandel et al., “Prompt γ-ray emission in neutron induced fission of 235 U” (2013), to be submitted. [14] G. Rudstam, in Proceedings of a Specialists’ Meeting on Fission Product Nuclear Data (Tokai, Japan, 25-27 May, 1992) p. 271.
[1] P. Talou et al., Phys. Rev. C 83, 064612 (2011). [2] B. Becker et al., Phys. Rev. C 87, 014617 (2013). [3] H. Naik, S. Dange, R. Singh, and T. Datta, Nucl. Phys. A 587, 273 (1995). [4] H. Naik, S.P. Dange, and R.J. Singh, Phys. Rev. C 71, 014304 (2005). [5] J.B. Wilhelmy, et al., . Phys. Rev. C 5, 2041 (1972). [6] M.M. Hoffman, Phys. Rev. 133, B714 (1964). [7] N.D. Dudey and T.T. Sugihara, Phys. Rev. 139, B896 (1965). [8] W. Hauser and H. Feshbach, Phys. Rev. 87, 366 (1952).
232