International audiencePrestress losses due to creep of concrete is a matter of interest for long-term operations of nuclear power plants containment buildings. Experimental studies by Granger (1995) have shown that concretes with similar formulations have different creep behaviors. The aim of this paper is to numerically investigate the effect of size distribution and shape of elastic inclusions on the long-term creep of concrete. Several microstructures with prescribed size distribution and spherical or polyhedral shape of inclusions are generated. By using the 3D numerical homogenization procedure for viscoelastic microstructures proposed by Šmilauer and Bažant (2010), it is shown that the size distribution and shape of inclusions have no measurable influence on the overall creep behavior. Moreover, a mean-field estimate provides close predictions. An Interfacial Transition Zone was introduced according to the model of Nadeau (2003). It is shown that this feature of concrete's microstructure can explain differences between creep behaviors

International audienceAn extension of the Mori–Tanaka and Ponte Castañeda–Willis homogenization schemes for linear elastic matrix-inclusion composites with ellipsoidal inclusions to aging linear viscoelastic composites is proposed. To do so, the method of Sanahuja (2013) dedicated to spherical inclusions is generalized to ellipsoidal inclusions under the assumption of time-independent Poisson’s ratio. The obtained time-dependent strains are successfully compared to those predicted by an existing method dedicated to time-shift aging linear viscoelasticity showing the consistency of the proposed approach. Moreover, full 3D numerical simulations on complex matrix-inclusion microstructures show that the proposed scheme accurately estimates their overall time-dependent strains. Finally, it is shown that an aspect ratio of aggregates in the range 0.3–3 has no significant influence on the time-dependent strains of composites with per-phase constitutive relations representative of a real concrete

The aim of the present work is to evaluate the effect of nano-silica (NS) on the hydration, the rheology and the strength development of cement pastes. The advance of chemical reactions is monitored by mean of isothermal calorimetry and thermogravimetric analysis: adding nano-silica particles speeds up the hydration of the cement paste but alter its workability. Indeed, the effect of the nano-silica particles on the hydration kinetics can be modelled by accounting for its high specific surface and a flocculation model based on the DLVO theory is proposed so as to investigate the stability of nano-silica suspensions in the fresh cement paste. As a consequence, the dosage of nano-silica can be optimized to promote the early age strength. Lastly, a ternary blend incorporating fly ash can be designed so as to provide an early age strength similar to that of the cement while lowering the induced CO 2 emissions.

The hydration model of Parrot & Killoh (1984) [1] has been extended to blended cements and coupled to a micromechanical scheme similar to that of Pichler & Hellmich (2011) [2] to estimate the Young modulus and the compressive strength of cementitious materials as a function of time. A finite aspect ratio of 7 is introduced to describe the shape of the hydrates and improve the estimate of the early age strength by the micromechanical scheme. Furthermore, accounting for the stress fluctuations in the cement paste partly explains the fact that the compressive strength of a concrete can be lower than that of its cement paste. Finally, the estimated physical properties are compared to numerous experimental measurements from the literature and new experimental measurements on blended cement pastes featuring significant weight fractions of limestone filler, fly ash or silica fume. It is shown that the present model slightly overestimates the dilution effect.

No abstract

No abstract

The stiffness of foam concrete depends primarily on the added porosity. Nevertheless, by performing 3D elastic numerical simulations on artificial unit cells in the frame of periodic homogenization, it is shown that describing foam concrete as a porous material is not sufficient to explain the experimental measurements of the Young modulus for added porosity higher than 40%. Indeed, introducing sand as a third phase enables to recover accurate estimates of the Young Modulus. Furthermore, for highly porous concrete foams, it is shown that the stress concentrates in thin members deprived of stiff sand particles, thus leading to a softer overall stiffness.

International audienceA multi-scale homogenization scheme is proposed to estimate the time-dependent strains of fiber-reinforced concrete. This material is modeled as an aging linear viscoelastic composite material featuring ellipsoidal inclusions embedded in a viscoelastic cementitious matrix characterized by a time-dependent Poisson's ratio. To this end, the homogenization scheme proposed in Lavergne et al. [1] is adapted to the case of a time-dependent Poisson's ratio and it is successfully validated on a non-aging material computed in the Fourier domain. Finally, the new extended estimates of the time-dependent strains of fiber-reinforced concrete are compared to the experimental measurements of Chern and Young [2] and the effects of the volume fraction and aspect ratio of the steel fibers on the time-dependent strains of fiber-reinforced concrete are investigated