One of the shortcomings of the Mohr-Coulomb failure criterion is that it will
predict the failure load as being the peak load which is usually accurate when concrete is
highly confined but it will not work for low confinement concrete because most of the
ductility of such concrete is produced after reaching the peak strength. Highly confined
concrete is usually produced when relatively high amounts of FRP confinement
reinforcement are used.
Other failure criteria have been used to determine when failure of confined
concrete occurs. Richart's failure criterion is similar to the Mohr-Coulomb failure
criterion as expressed in Equation 2-1 with k equal to 4.1 (Xiao and Wu 2003; Li and
Hadi 2003; Xiao and Wu 2000; Samaan et al. 1998). Tsai-Wu failure criterion is a failure
criterion that focuses on the FRP laminate failure and takes into account the interaction of
the axial compressive and tensile hoop stresses that produces an earlier failure of the FRP
encasement compared with ignoring the axial compressive stress and accounting only for
the hoop tensile stress (Mandal et al. 2005; Cho et al. 2005; Barbero 1998). The Tsai-Hill
failure criterion is similar to the Tsai -Wu failure criterion with the difference that the
stress values used can only be tensile or compressive but not both (Cho et al. 2005;
Barbero 1998). For example if pure tension stress state is assumed in all directions the
Tsai-Hill criterion yields the same results as the Tsai-Wu criterion.
Confinement models to predict the maximum stress or the whole stress-strain curve
of confined concrete have been developed (Campione and Miraglia 2003; Li and Hadi
2003; Li et al. 2003; Xiao and Wu 2000, 2003; Fam and Rizkalla 2001). Most models
that describe the whole stress-strain curve of FRP confined concrete employ a bilinear
curve (second portion of the curve ascending) that in most cases works well. Most