-
Improvement of Geant4 Neutron-HP package: Unresolved Resonance Region description with Probability Tables
Authors:
M. Zmeškal,
L. Thulliez,
P. Tamagno,
E. Dumonteil
Abstract:
Whether for shielding applications or for criticality safety studies, solving the neutron transport equation with good accuracy requires to take into account the resonant structure of cross sections in part of the Unresolved Resonance Region (URR). In this energy range even if the resonances can no longer be resolved experimentally, neglecting them can lead to significant numerical biases, namely…
▽ More
Whether for shielding applications or for criticality safety studies, solving the neutron transport equation with good accuracy requires to take into account the resonant structure of cross sections in part of the Unresolved Resonance Region (URR). In this energy range even if the resonances can no longer be resolved experimentally, neglecting them can lead to significant numerical biases, namely in flux-based quantities. In Geant4, low energy neutrons are transported using evaluated nuclear data libraries handled by the Neutron High-Precision (Neutron-HP) package. In the version 11.01.p02 of the code, the URR can only be described by average smooth cross sections that do not take into account the statistical resonant structure of the cross sections. To overcome this shortcoming, the treatment of the URR with the use of the probability table method has been implemented in Geant4 and successfully validated with the reference Monte Carlo neutron transport codes MCNP6 (version 6.2) and Tripoli-4 (version 12). These developments will be taken into account in the next release of Geant4. All the validations of Geant4 have been performed with probability tables generated from both the NJOY and CALENDF pre-processing tools. Therefore Geant4 now has this unique feature to study the relative impact of the strategies involved during the production of probability table by the two pre-processing codes. This has been used to show that self-shielding is important also for inelastic cross sections in the example of 238U. The tool to generate probability tables usable by Geant4 either from NJOY or from CALENDF is made available on a dedicated GitLab repository and will be included in Geant4.
△ Less
Submitted 25 April, 2024;
originally announced April 2024.
-
Tensor factorization in ab initio many-body calculations: Triaxially-deformed (B) MBPT calculations in large bases
Authors:
M. Frosini,
T. Duguet,
P. Tamagno
Abstract:
Whether for fundamental studies or nuclear data evaluations, first-principle calculations of atomic nuclei constitute the path forward. Today, performing \textit{ab initio} calculations (a) of heavy nuclei, (b) of doubly open-shell nuclei or (c) with a sub-percent accuracy is at the forefront of nuclear structure theory. While combining any two of these features constitutes a major challenge, addr…
▽ More
Whether for fundamental studies or nuclear data evaluations, first-principle calculations of atomic nuclei constitute the path forward. Today, performing \textit{ab initio} calculations (a) of heavy nuclei, (b) of doubly open-shell nuclei or (c) with a sub-percent accuracy is at the forefront of nuclear structure theory. While combining any two of these features constitutes a major challenge, addressing the three at the same time is currently impossible. From a numerical standpoint, these challenges relate to the necessity to handle (i) very large single bases and (ii) mode-6, \textit{i.e.} three-body, tensors (iii) that must be stored repeatedly. Performing second-order many-body perturbation theory(ies) calculations based on triaxially deformed and superfluid reference states of doubly open-shell nuclei up to mass $A=72$, the present work achieves a significant step forward by addressing challenge (i). To do so, the memory and computational cost associated with the handling of large tensors is scaled down via the use of tensor factorization techniques. The presently used factorization format is based on a randomized singular value decomposition that does not require the computation and storage of the very large initial tensor. The procedure delivers an inexpensive and controllable approximation to the original problem, as presently illustrated for calculations that could not be performed without tensor factorization. With the presently developed technology at hand, one can envision to perform calculations of yet heavier doubly open-shell nuclei at sub-percent accuracy in a foreseeable future.
△ Less
Submitted 9 July, 2024; v1 submitted 12 April, 2024;
originally announced April 2024.
-
Calculation and verification of neutron irradiation damage with differential cross sections
Authors:
Shengli Chen,
David Bernard,
Pierre Tamagno,
Jean Tommasi,
Stephane Bourganel,
Gilles Noguere,
Cyrille De Saint Jean
Abstract:
The Displacement per Atom (DPA) rate is conventionally computed with DPA cross sections in reactor applications. The method of direct calculation with energy-angular distributions given in the Center of Mass (CM) frame is proposed and recommended in the present work. The methods for refining and verifying the calculations of DPA cross sections are proposed: (i) Gauss-Legendre-Quadrature-based Piec…
▽ More
The Displacement per Atom (DPA) rate is conventionally computed with DPA cross sections in reactor applications. The method of direct calculation with energy-angular distributions given in the Center of Mass (CM) frame is proposed and recommended in the present work. The methods for refining and verifying the calculations of DPA cross sections are proposed: (i) Gauss-Legendre-Quadrature-based Piecewise Integration (GLQPI) for ensuring the numeric convergence of integral over emission angle due to the discontinuity of integrand; (ii) verification of the convergence for trapezoidal integration over the secondary energy; (iii) interpolation of double-differential cross sections. For 56Fe of JEFF-3.1.1, the current numeric integration over emission angle is shown not convergent, whereas the direct trapezoidal over the secondary energy and the direct interpolation of energy-angle-integrated damage are shown accurate. On the other hand, it is shown that the DPA cross sections are overestimated if isotropic angular distributions are assumed. However, the DPA cross section is not sensitive to the high-order Legendre polynomials because the former is an angle-integrated quantity. Numerical results of neutron elastic scattering show that 2 orders of Legendre polynomials can give the DPA rates of 56Fe within 0.5% overestimation for fission reactors, while 4 orders are required for fusion reactors. For neutron inelastic scatterings-induced DPA, the first order Legendre polynomial is sufficient for both fission and fusion reactors.
△ Less
Submitted 19 July, 2019; v1 submitted 8 February, 2019;
originally announced February 2019.
-
Calculation and Verification of Irradiation Damage Cross Section with Energy-Angular Distribution
Authors:
Shengli Chen,
David Bernard,
Pierre Tamagno,
Cyrille De Saint Jean
Abstract:
To complete the computation of Displacements per Atom (DPA) cross sections, the present work shows the methods of calculating DPA cross sections with the nuclear data of energy-angular distribution in both the laboratory and the Center-of-Mass (CM) frames. The method of direct calculation with data in the CM frame is proposed and recommended to decrease the computation burden and keep all informat…
▽ More
To complete the computation of Displacements per Atom (DPA) cross sections, the present work shows the methods of calculating DPA cross sections with the nuclear data of energy-angular distribution in both the laboratory and the Center-of-Mass (CM) frames. The method of direct calculation with data in the CM frame is proposed and recommended to decrease the computation burden and keep all information. Theoretical analyses reveal that more than 7-point Gauss-Legendre Quadrature (GLQ) should be used to ensure the convergence of the angular integration for DPA computations. Numerical results show that 8-point GLQ is sufficient for the continuum inelastic neutron scattering, while 64-point GLQ is implemented in NJOY. Because the integrand over secondary energy is not derivable in the whole domain of the secondary energy, the trapezoidal integration is used to perform the numerical integration. The numerical calculations show that the trapezoidal integration is suitable to perform the integration over the secondary energy on the fine grid given by nuclear data files at least for 56Fe. The present work reveals that the direct interpolation of energy-angular-integrated damage can give the same results computed with standard interpolated energy-angular distributions. The DPA cross sections will be overestimated if isotropic angular distributions are assumed. However, the first-order Legendre polynomial can give DPA cross sections within 0.4% deviation, while 12 orders are required to describe the anisotropic angular distribution.
△ Less
Submitted 8 February, 2019;
originally announced February 2019.
