Abstract

The out-diffusion of radionuclides from activated material in case of a fire may represent a non-negligible contribution to the radiological source term of such an event. In order to assess the contribution of this phenomenon, a software package has been designed and implemented. In the present document we briefly introduce the numerical treatment used to tackle the problem prior to the explanation of the software’s logic. The document ends with an exemplary simulation and a study carried out to validate the implementation of the algorithm.The presented tool has been named SOLIDUSS, it is mainly written in C++ and uses a Monte Carlo based approach to simulate the diffusion of radioisotopes within solid materials. It is designed to run coupled with CERN-FLUKA, taking advantage of its geometry kernel to carry out diffusion calculations in arbitrarily complex geometries. The user can provide 3D temperature maps along with many other parameters that allow the program to target a wide range of different scenarios. As results SOLIDUSS provides 3D radionuclide concentration maps as well as the amount of radionuclides out-diffused from the selected materials. So far, this software has undergone numerical validation which will be discussed in this paper. Benchmarking against experimental data is currently ongoing.

Highlights

  • A fire in an activated area, such as those of the CERN (European Organization for Nuclear Research) accelerator complex, could result in the release of radionuclides to the environment

  • The requirement of exhaustive radiological source terms for this type of scenarios has revealed the need to evaluate the contribution of a second mechanism: the out-diffusion of radionuclides due to the high temperatures reached in the fire and surroundings

  • Numerical treatment In Appendix A it is shown that random walks emulate the diffusion process under certain assumptions. This well known result suggests that we can simulate diffusion processes using stochastic methods, avoiding the need for deterministic solutions of the diffusion equation often based on the use of Finite Elements Methods that for us would pose a number of difficulties – especially considering diffusion of radionuclides produced in an arbitrary geometry following the interaction of the primary particle beam with material – such as:

Read more

Summary

Introduction

A fire in an activated area, such as those of the CERN (European Organization for Nuclear Research) accelerator complex, could result in the release of radionuclides to the environment. This paper explains the problem and the tool we have designed which aims at more accurately estimating the contribution of radioisotope out-diffusion to radiological source terms in case of fires in arbitrary geometries This was done in the context of the FIRIA (Fire Induced Radiological Integrated Assessment) project (FIRIA, 2018; Gai, 2021) at CERN, which aims to develop an integrated approach to quantitatively assess potential discharges of radioactive substances induced by a fire accident in particle physics’ experimental facilities. To this end, state-of-the-art tools and methods are used to study the radioactive source term, fire dynamics, evacuation, intervention time, environmental dispersion and dose assessment

Theoretical summary
Diffusion
Diffusion coefficient
Numerical treatment
Wiener process
SOLIDUSS: implementation and logic
How to use the software
Further considerations
Outputs post-processing: out-diffusion fraction estimation
Exemplary calculation
Bootstrapping test
Findings
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call