American Nuclear Society
Home

Home / Publications / Journals / Fusion Science and Technology / Volume 72 / Number 3

Study on the Thermally-Induced Stress and Relaxation of Ceramic Breeder Pebble Beds

Mahmoud Lotfy, Alice Ying, Mohamed Abdou, Yi-Hyun Park, Seungyon Cho

Fusion Science and Technology / Volume 72 / Number 3 / October 2017 / Pages 255-262

Technical Paper / dx.doi.org/10.1080/15361055.2017.1330637

Received:September 5, 2016
Accepted:February 22, 2017
Published:August 25, 2017

Ceramic breeder pebble beds undergo complex thermally-induced stress build-up and relaxation processes during reactor operations due to the pebble bed thermal expansion and creep deformation. Understanding such processes can facilitate the evaluation of a solid breeder performance, including bed stress/strain equilibrium status, which will guide the design of stable blanket operation and assessment of lifetime. The efforts of this study cover both experimental testing and numerical modeling for this purpose. Measured stresses in pebble beds show a decreasing trend with thermal cycles, until ultimately reaching a saturated state. This stress relaxation is mainly caused by the combined effect of bed plastic rearrangement and accumulation of creep deformation under compressive stresses and high temperatures. As bed stress is reduced, the creep deformation becomes less significant and further cyclic operation would not alter the pebble bed mechanical state. To validate the thermally-induced stress and its variation with cycles, experiments of thermal stress measurement have been designed and conducted for pebble beds heated by both continuous and pulsed power sources. Moreover, the effects of mechanical pre-compaction were investigated with emphasis on understanding the relationship between the bed stress-state evolution and maintaining adequate levels of thermal contact between the pebbles and the coolant structure. The results of this study presents valuable data to serve as a basis for validation of the most recent pebble bed numerical models.