In the present work, dynamic tensile strength of concrete is experimentally investigated by means of spalling tests. Based on extensive numerical simulations, the paper presents several advances to improve the processing of spalling tests. The striker is designed to get a more uniform tensile stress field in the specimen. Three methods proposed in the literature to deduce the dynamic strength of the specimen are discussed as well as the use of strain gauges and a laser extensometer. The experimental method is applied to process data of several tests performed on wet microconcrete at strain rates varying from 30 to 150/s. A significant increase of the dynamic tensile strength with strain-rate is observed and compared with data of the literature. In addition, post-mortem studies of specimens are carried to improve the analysis of damage during spalling tests.

Dynamic fragmentation is observed in brittle materials such as ceramics, concrete, glass or rocks submitted to impact or blast loadings. Under such loadings, high-stress-rate tensile fields develop within the target, and produce fragmentations characterized by a high density of oriented cracks. To improve industrial processes such as blast loadings in open quarry or ballistic efficiencies of armours or concrete structures against impact loadings, it is essential to understand the main properties of such damage processes (namely, characteristic time of fragmentation, characteristic density, orientation and extension of cracking, ultimate strength) as functions of the loading rate, the size of the structure (or the examination volume), and the failure properties of the brittle material concerned. In the present contribution, the concept of probability of non-obscuration is developed and extended to predict the crack density for any size, shape of the loaded volume, stress gradients, and stress-rates. A closed-form solution is used to show how a brittle and random behaviour under quasi-static loading becomes deterministic and stress-rate-dependent with increasing loading rates. Two definitions of the tensile strength of brittle materials are proposed. As shown by Monte-Carlo simulations, for brittle materials, the "ultimate macroscopic strength" applies under high loading rate or in a large domain whereas the "mean obscuration stress" applied in a small domain or under low stress rate. Next, a multi-scale model is presented and used to simulate damage processes 3 observed during edge-on impact tests performed on an ultra-high strength concrete. Last, the fragmentation properties predicted by modelling of six brittle materials (dense and porous SiC ceramics, a micro-concrete, an ultra-high strength concrete, a limestone rock and a soda-lime silicate glass) are compared.

For one decade, spalling techniques based on the use of a metallic Hopkinson bar in contact with a concrete sample have been widely employed to characterise the dynamic tensile strength of concrete at strain rates ranging from a few tens to hundreds of s−1. However, the processing method based on the use of the velocity profile measured on the rear free surface of the sample (Novikov formula) remains quite basic. In particular, the identification of the whole softening behaviour of the concrete material is currently out of reach. In the present paper, a new processing technique is proposed based on the use of the virtual fields method (VFM). First, a digital ultra‐high‐speed camera is used to record the pictures of a grid bonded onto the specimen. Then, images of the grid recorded by the camera are processed to obtain full‐field axial displacement maps at the surface of the specimen. Finally, a specific virtual field has been defined in the VFM equation to use the acceleration map as an alternative ‘load cell’. This method applied to three spalling tests with different impact parameters allowed the identification of Young's modulus during the test. It was shown that this modulus is constant during the initial compressive part of the test and decreases in the tensile part when microdamage exists. It was also shown that in such a simple inertial test, it was possible to reconstruct average axial stress profiles using only the acceleration data. It was then possible to construct local stress–strain curves and derive a tensile strength value.

No abstract