Effect of fiber content on tensile strength of non-proprietary ultra-high performance concrete

Abstract

Ultra-high performance concrete (UHPC) is being used to solve challenging concrete structural applications where conventional concrete would otherwise deteriorate and improved structural properties can be utilized. UHPC is preferred for its high compressive and tensile strengths, which are achieved by a low water-cementitious materials ratio (w/cm), optimized particle gradation, and adding materials such as silica fume and steel fibers to the matrix. Steel fibers also increase other mechanical properties, such as splitting tensile (indirect tensile) and flexural (modulus of rupture) strengths. The present study uses varying steel fiber contents as well as direct tensile, splitting tensile, compressive and flexural strengths of 60 specimens to develop preliminary empirical models that correlate the direct tensile first crack strength, post-cracking strength, and fiber content with the splitting tensile, flexural (modulus of rupture), and compressive strengths at 28 days for non-proprietary UHPC. The accuracy of the developed models is verified by comparing their predictive capabilities with the actual strengths obtained from the laboratory tests. Literature was surveyed to determine significant information related to this study to develop an advanced understanding of non-proprietary UHPC and correlation modeling. The laboratory experiments that were conducted in this study include flowability, compressive strength, and direct tensile strength tests. Steel fiber contents of 0%, 1%, 2%, 4%, and 6% were considered in this study. Statistical analysis software was used to analyze the strength test results and develop regression models. The root mean square error method was used to analyze the accuracy between the values predicted by the regression models and the actual laboratory values. The regression equations exhibit small errors when compared to the experimental results, which allow for efficient and realistic predictions of the direct tensile and flexural strengths.