Full Program »

Behavior of ultra-high-performance-concrete at early age: experiments and simulations

Ultra-high performance concrete (UHPC) is a cementitious composite material with high level of performances reached by a low water-binder ratio, an optimized gradation curve, and with the use of thermal activation. In the past decades, various UHPC's have been developed and utilized, however, very limited research is conducted studying their early age strength development, which is crucial for construction planning and reliability assessments. Hence, a comprehensive mathematical model that is capable of describing the early age effects of UHPC cannot be found in the available literature. The literature will show the experimental characterization as well as the results of subsequent aging simulations utilizing and coupling the Hygro-thermo-chemical model and the Lattice Discrete Particle Model with aging effects (HTC-A-LDPM) for a UHPC at various early ages. Investigated tests include unconfined compression, cylinder Brazilian, and beam 3-point-bending. The HTC component of the computational framework allows taking into account any form of curing conditions as well as known material constituents and predicts the level of concrete aging. The LDPM component, which is a discrete mechanical model, simulates the failure behavior of concrete at the coarse aggregate mesoscale level. Connecting the two components, the aging functions are developed with great simplicity to compute accurately the material properties. The presentation will demonstrate that the proposed hygro-thermo-chemo-mechanical coupled early age framework (HTC-A-LDPM) can comprehensively capture cement hydration effects as well as strength development for the UHPC investigated and predictively other types of cement-based concrete.

Furthermore, with the model comprehensively calibrated and validated, size effect experiments and simulations as well as predictions for broader size range and early ages will also be exhibited and discussed. In addition, size effect type mathematical models data fitting is as well included.

Author(s):

Lin Wan    
Northwestern University
United States

Roman Wendner    
IKI-BOKU Vienna
Austria

Gianluca Cusatis    
Northwestern University
United States