Shear behavior of dowel reinforced slurry infiltrated fiber concrete (SIFCON) under monotonic and reversed cyclic loading.

Abstract

The two main objectives of this research are: (1) to investigate experimentally the shear strength and energy absorbing capability of Slurry Infiltrated Fiber CONcrete (SIFCON) and dowel reinforced SIFCON (DRSIFCON) under monotonic and reversed cyclic shear loading, and (2) to develop a mathematical model to predict the complete shear stress-slip relationship of SIFCON and DRSIFCON under monotonically increasing shear loading. The experimental investigation consisted of testing eighteen double shear specimens to determine their shear strength, ductility, and energy absorbing capability under monotonically increasing as well as cyclically reversing shear loading. The eighteen specimens consisted of four control specimens, six SIFCON specimens, and eight DRSIFCON specimens. The experimental results showed an outstanding shear strength and an excellent energy absorbing capability. The shear strength averaged 4000-4500 psi for SIFCON and 4500-5000 psi for DRSIFCON, about three times higher than that of reinforced concrete. The energy absorbing capability of SIFCON and DRSIFCON were found 2-3 times and 5-12 times higher than that of reinforced concrete. The mathematical modelling consisted of predicting the shear stress-slip response of SIFCON and DRSIFCON specimens under monotonically increasing shear loading. The modelling of SIFCON was based on the pull-out behavior of steel fibers; and the modelling of DRSIFCON was based on the pull-out behavior of steel fibers and the dowel action of reinforcing bars. The steel fibers were modeled as a beam on a linearly elastic foundation to determine the fiber-matrix pressure, the fiber deflection, fiber bending, and fiber shear forces along their embedded length. The dowel reinforcing bars were modeled as laterally loaded free headed piles in cohesive soil to determine their contribution to the shear strength. The mathematical models predicted very accurately the slope of the ascending branch, the peak shear stress, and the descending branch of the shear stress-slip responses. The estimated errors between predicted and experimental results were about 2-5% for the slope of the ascending branch, 5-7% for the peak shear stress, and 10-15% for the post-peak response.