This study investigates the behavior of High Strength Concrete (HSC) under uniaxial state of stresses. Emphasis is placed on experimental evaluation of important mechanical and fracture properties. Owing to high brittleness of HSC, experimental results especially on tensile behavior have been largely limited and scarce. In this research, direct uniaxial tension tests are employed for determination of the post-peak tensile softening characteristics of HSC. The softening characteristics of high strength concrete is found to be considerably different than that of normal strength concrete (NSC). Fracture energies evaluated form the descending branch of the stress softening reveal significant drop in the post peak compliance of the high strength concretes. Such relationships of stress-crack separation are vital input for developing a model capable of accurately predicting behavior of HSC in tension.

The obtained softening relationship is incorporated into an non-linear finite element model using ABAQUS program. The model is shown to be successful in predicting the test results of the present study as well as the ones of other researchers. The predictions are of equal degree in accuracy for both the load crack mouth opening displacement (CMOD) and load-Deflection (LPD) responses. Performing of a parametric study as well as development of a methodology that suggests the use of load-CMOD response in beam fracture tests as an alternative method of determining the fracture toughness (G_F) from beam tests are undertaken. Important parameters such as flexural strength, size of process zone of normal and high strength concrete are also determined using the FEM model. It is found that for an increase of about 30% in the fracture toughness G_F and the tensile strength f'_t of HSC, the reduction in the difference between flexural strength and tensile strength is considerable and the size of process zone is also significantly smaller in HSC as compared to NSC. It is shown that to apply Linear Elastic Fracture Mechanics (LEFM) principles, a minimum size (depth) of beam of HSC is about 9.0" whereas for NSC the minimum depth of the beam is almost twice as much i.e. about 18.0". An important recommendation for determining the fracture energy G_F from load-CMOD curves instead from the conventional Load Deflection response is shown to produce lesser variation in G_F values since CMOD measurements are less likely to be affected by experimental setups and errors. Errors that are known to generally affect the load-line deflection (LPD) measurements can cause significant inflated values of fracture energy G_F to be reported. Finally based on the test results of beam bending tests, a recommendation is made regarding a suitable size of beam specimen that can be used as a standardized fracture test specimen. The beam specimen of span depth (S/D) ratio of 4 is found to be more suitable than the RILEM recommended beam size S/D = 8.