International Topical Meeting on Nuclear Research Applications and Utilization of Accelerators

4-8 May 2009, Vienna

Deterministic Model for the Analysis of YALINA–Booster Experiments with the ERANOS code System

G. Aliberti¹, Y. Gohar¹, F. Kondev¹, A. Talamo¹, H. Kiyavitskaya², V. Bournos², Y. Fokov², C. Routkovskaya², and I. Serafimovich²

¹Argonne National Laboratory, Argonne, Illinois, United States of America
²Joint Institute for Power and Nuclear Research–Sosny, National Academy of Sciences, Minsk, Belarus

Corresponding Author: aliberti@anl.gov

The YALINA experimental programhas been launched by the Joint Institute for Power and Nuclear Research–SOSNY (Belarus) to study the physics of ADS. The present work focuses on the analysis of the YALINA–Booster subcritical assembly by the use of deterministic codes. The YALINA–Booster couples a fast zone of two U-235 enrichments (90% and 36%) in a lead lattice, and a thermal zone with 10% enrichment of U-235 in a polyethylene moderator. The two zones are separated by an absorbing layer of B4C and natural uranium rods. The active core is surrounded radially by a graphite reflector and axially by borated polyethylene. A deterministic model for YALINA–Booster has been created with the ERANOS code package. Cross-sections have been processed with the ECCO code using JEF2.2, JEF3.1 and ENDF/B-VI.8 data. The VARIANT module has been used for reactivity and flux calculations in XYZ geometry and the KIN3D code is used for the analysis of dynamic measurements.

A meticulous analysis was devoted to the existing empty spaces in the original YALINA–Booster configurations (beam tube, experimental channels, etc.), that required to be homogenized with other materials, due to the code difficulties to perform in presence of void regions. Additionally, due to the current limitations of the existing cell codes, another major difficulty associated with the use of deterministic codes consists in the impossibility to reproduce all the details of the reactor geometry. As consequence, local effects can not be simulated as they really are and a physical approach is then required to correctly reproduce the global effects in the selected regions where the cross-sections are processed so that the results can be obtained without any loss of accuracy.

For instance, for YALINA–Booster significant efforts had to be also devoted to process accurate cross-sections for the B₄C absorber rods between the fast and thermal zones. The selection of the reactor zones associated with the material homogenization and cross-section processing is in general the most delicate phase when creating a deterministic model. Details of the deterministic model created for YALINA–Booster with the ERANOS code system will be given in the full paper. The calculated results, showing a good agreement with the measurements, confirmed the ERANOS capabilities for the analysis of complicated systems such as YALINA–Booster.

Full text paper