Computational Modeling of Fiber Reinforced Cement Composites as a Complex Material

J OLIVER; D.F. MORA; A.. HUESPE

Barrcelona

Conferencia; Computational Plasticity XI Fundamentals and Applications, Barcelona, Spain, September 2011.; 2011

A complex body theory is a mechanical/mathematical tool able to describe materials whichcontain a complex substructure [1]. This substructure is endowed with its own properties and itinteracts with the macrostructure and influences drastically its behavior. Under thismathematical framework materials such as cement composites can be seen as a continuum witha microstructure. Therefore, the whole continuum damage mechanics theory, incorporating anew microstructure, is still applicable.A formulation initially based on the theory of complex bodies of Capriz [2] has been developedto model the mechanical behavior of the high performance fiber cement composites witharbitrarily oriented fibers. This formulation is a two scale approach, in which the microstructuretakes into account the fiber-matrix interface bond/slip processes, which have been recognizedfor several authors [3-4] as the principal mechanism increasing the ductility of the quasi-brittlecement response. These processes are taken into account by considering a micro field thatrepresent the slippling fiber-cement displacement. The conjugate generalized stress to thegradient of this microfield verifies a balance equation having a physical meaning.The paper concerns the computational modeling of those fiber reinforced cement composites(FRCC). To simulate the composite material, a finite element discretization is used to solve theset of equations given by the complex bodies theory for this particular case. Moreover, a twofield discretization: the standard macroscopic and the microscopic displacements, is proposedthrough a mixed finite element methodology. Furthermore, a splitting procedure for uncouplingboth fields is proposed, which provides a more convenient numerical treatment of the discreteequation system. Validation of the obtained numerical results with a selected set of experimentsproves the viability of this approach.