Meso-Scale Modeling of Irradiated Concrete in Test Reactor A. Giorla 1 a 2 Y. Le Pape 1 P. ˇ Stemberk 2 M. Vaitov´ 1 Oak Ridge National Laboratory, 2 Czech Technical University, Prague November 5, 2015 Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 1 / 35
Outline Why numerical modelling? 1 Elleuch’s experiment 2 Numerical model 3 Preliminary sensitivity analysis 4 Conclusion 5 Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 2 / 35
Outline Why numerical modelling? 1 Elleuch’s experiment 2 Numerical model 3 Preliminary sensitivity analysis 4 Conclusion 5 Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 3 / 35
Motivation NPP licence renewal - condition of bological shield RIVE of aggregate - first order degradation mechanism Irradiation effect on concrete assessment - experiments Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 4 / 35
Differences between PWR and Test Reactor PWR Test Reactor / Gamma facility fast neutron flux (n.cm − 2 s − 1 ) 5 × 10 11 to 2 × 10 14 1 to 2 × 10 10 fast neutron fluence (n.cm − 2 ) < 6 × 10 19 < 10 20 gamma flux (kGy h − 1 ) 5 to 20 0 . 02 to 200 gamma dose (MGy) 50 to 200 0 . 1 to 1000 gamma heating rate (W g − 1 ) 0.02 0 . 04 to 1 . 3 temperature ( ◦ C) < ≈ 65 (design) 40 to 250 internal RH (-) strong gradient often pre-dried mechanical boundary conditions structural restraints free materials concrete mortar time scale (years) 60 to 80 < 1 year Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 5 / 35
Exposure time difference short time experiments higher flux of irradiation to get total fluence(dose) role of creep PWR Test Reactor / Gamma facility fast neutron flux (n.cm − 2 s − 1 ) 5 × 10 11 to 2 × 10 14 1 to 2 × 10 10 fast neutron fluence (n.cm − 2 ) < 6 × 10 19 < 10 20 gamma flux (kGy h − 1 ) 5 to 20 0 . 02 to 200 gamma dose (MGy) 50 to 200 0 . 1 to 1000 time scale (years) 60 to 80 < 1 year Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 6 / 35
Temperature difference higher flux of irradiation to get total fluence(dose) cooling of the samples temperature gradient PWR Test Reactor / Gamma facility fast neutron flux (n.cm − 2 s − 1 ) 5 × 10 11 to 2 × 10 14 1 to 2 × 10 10 gamma flux (kGy h − 1 ) 5 to 20 0 . 02 to 200 gamma heating rate (W g − 1 ) 0.02 0 . 04 to 1 . 3 temperature ( ◦ C) < ≈ 65 (design) 40 to 250 Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 7 / 35
RH difference moist walls pre-dried samples different gradients PWR Test Reactor / Gamma facility internal RH (-) strong gradient often pre-dried Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 8 / 35
Scale difference material mechanical boundary conditions different gradients T, RH PWR Test Reactor / Gamma facility mechanical boundary conditions structural restraints free materials concrete mortar Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 9 / 35
Meso-scale model Understand experiments Transpose the experimental data to assess the actual structure Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 10 / 35
Model Expansion and degradation of concrete Function with parameters of environmental conditions and composition of material Calibrate model with known experiment - environmental conditions, concrete composition Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 11 / 35
Outline Why numerical modelling? 1 Elleuch’s experiment 2 Numerical model 3 Preliminary sensitivity analysis 4 Conclusion 5 Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 12 / 35
Experimental data Experiment dimensional changes mechanical properties components of concrete and concrete itself Data for simulation material parameters of cement paste and aggregate (strength, modulus of elasticity, thermal expansion coefficient etc.) environmental conditions: relative humidity, thermal and fluence history Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 13 / 35
Elleuch’s experiment Environment temperature history - setting, thermal treatment, storing, irradiation neutron fluence relative humidity exposure period Materials cement paste - aluminous cement aggregate - serpentine mechanical properties physical properties Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 14 / 35
Experimental data Conditions dependent on the position in the rig temperature neutron fluence Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 15 / 35
Outline Why numerical modelling? 1 Elleuch’s experiment 2 Numerical model 3 Preliminary sensitivity analysis 4 Conclusion 5 Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 16 / 35
Modeling Strategy Aggregates - Fluence - Radiation-induced Conditions - Temperature expansion/softening - Humidity - Thermal expansion - Brittle failure Results Mechanisms Finite Elements - Macro. expansion Cement paste - Micro. damage - Drying shrinkage - Thermal expansion - Brittle failure - Creep Microstructure Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 17 / 35
Modeling Strategy Implementation in AMIE C++ finite element library initially developed at EPFL Random generation of concrete microstructures Robust creep-damage algorithm . 2D simulations Relatively small computational time ( ≈ 20 h) More crack percolation than in 3D? ITZ is neglected Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 18 / 35
Cement paste Visco-elastic Logarithmic creep curve (recoverable/irrecoverable creep strains) Depends on T, RH (no drying creep) . Quasi-brittle Non-local damage model (no mesh sensitivity) Linear softening post-peak behavior . With imposed deformations Thermal expansion ( ∝ T, reversible) Drying shrinkage ( ∝ 1-RH, irreversible) Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 19 / 35
Aggregates Elastic Homogeneous circular inclusions Young’s modulus is reduced by irradiation No failure in the aggregates (cement paste breaks first) . With imposed deformations Thermal expansion ( ∝ T, reversible) Radiation Induced Volumetric Expansion (RIVE) (irreversible, using [Zubov and Ivanov, 1966] sigmoid) � e δ Φ − 1 � ǫ Φ = κ ǫ max ǫ max + κ e δ Φ Parameters calibrated by [Le Pape et al, 2015] Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 20 / 35
Microstructure F¨ uller PSD based on the experimental mix design (fine and coarse) Particles are randomly placed from largest to smallest Particles smaller than element size are ignored Very large aggregate size compared to sample size Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 21 / 35
Environmental conditions Pre-drying Irradiation Cool-o ff 5 different irradiations duration 3 different temperature/fluence Cure Expansion T emperature measurements Simulate the experimental protocol Fluence Use intermediate data to calibrate unknown material parameters Humidity 0 100 200 300 400 500 Time [d] Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 22 / 35
Example of simulation result Stages of damage process: Debonding of large aggregates 1 Crack bridging between large 2 aggregates Debonding of small aggregates 3 . Very high damage in the cement paste Loss of sample integrity Overestimation of damage (homogeneous aggregates) Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 23 / 35
Reference results Good agreement between simulation and experiments Slightly lower expansion Less aggregates Scatter in expansion Temperature, flux Anisotropic expansion for very high fluences Experimental sample is not representative Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 24 / 35
Outline Why numerical modelling? 1 Elleuch’s experiment 2 Numerical model 3 Preliminary sensitivity analysis 4 Conclusion 5 Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 25 / 35
Preliminary sensitvity analysis Additional shrinkage in cement paste? Role of creep? Role of aggregate loss of stiffness? Role of damage model? Giorla, Vaitov´ a et al (ORNL, CTU) Meso-Scale Modeling of Irradiated Concrete November 5, 2015 26 / 35
Recommend
More recommend
