On interchangeability and selection of size effect and boundary effect experiments for characterization and prediction of quasi-brittle fracture of concrete

Abstract

Size effect experiments using geometrically similar specimens of different sizes overlap boundary effect experiments using specimens of single size but different notches if sufficient sample designs are tested. This study shows the interchangeability and relevance of size and boundary effect experiments through in-depth analysis of comprehensive size effect experiments. Initially-designed five groups of geometrically similar concrete specimens with notch/size ratios of 0, 0.025, 0.075, 0.15 and 0.3 are rearranged into four groups with constant sizes of 40, 93, 215 and 500 mm. Thus, total nine groups of experimental results, containing abundant information on quasi-brittle fracture of concrete and associated size effect phenomenon, are analysed by three size effect laws (SELs) and a linearized boundary effect model (BEM). It is found that fracture behaviours revealed from comprehensive size effect experiments (using large and small geometrically similar specimens required by SEL) can also be confirmed by simple boundary effect experiments (using specimens of single convenient size). Discussion on SEL and BEM based on principles of mathematical fitting and data processing is provided for more objective quasi-brittle fracture experiments and modelling.