Shape and size effects on the compressive strength of high-strength concrete J.R. del Viso, J.R. Carmona, G. Ruiz ⁎ E. T. S. de Ingenieros de Caminos, Canales y Puertos, Universidad de Castilla-La Mancha, 13071 Ciudad Real, Spain Received 6 February 2007; accepted 13 September 2007 Abstract In this paper we investigate the influence of the shape and of the size of the specimens on the compressive strength of high-strength concrete. We use cylinders and cubes of different sizes for performing stable stress–strain tests. The tests were performed at a single axial strain rate, 10 - 6 s - 1 . This value was kept constant throughout the experimental program. Our results show that the post-peak behavior of the cubes is milder than that of the cylinders, which results in a strong energy consumption after the peak. This is consistent with the observation of the crack pattern: The extent of cracking throughout the specimen is denser in the cubes than in the cylinders. Indeed, a main inclined fracture surface is nucleated in cylinders, whereas in cubes we find that lateral sides get spalled leading to the so-called hour-glass failure mode. The remaining cube core gets fragmented due to crushing, in some cases exhibiting a dense columnar cracking in the bulk of the specimen. Finally, we investigate the relationship between the compressive strength given by both types of specimen for several specimen sizes. © 2007 Elsevier Ltd. All rights reserved. Keywords: Concrete compressive strength; Size effect; Specimen shape; High-strength concrete 1. Introduction By far the most common test carried out on concrete is the compressive strength test. The main reason to understand this fact is that this kind of test is easy and relatively inexpensive to carry out [1]. Testing standard requirements use different geometries of specimens to determine the compressive concrete strength, f c . The most used geometries are cylinders with a slenderness equal to two and cubes. Shape effect on compression strength has been widely studied and different relationships between the compressive strength obtained for these geometries have been proposed, mainly from a techno- logical standpoint. Such approach eludes the fact that there is a direct relation between the nucleation and propagation of fracture processes and the failure of the specimen. Indeed, experimental observations confirm that a localized micro- cracked area develops at peak stress [2] or just prior to the peak stress [3]. For this reason compressive failure is suitable to be analyzed by means of Fracture Mechanics. The failure in uniaxial compression is due to a localization of the damage in a certain zone as it was shown by Van Mier [2]. To study the process of localization in compression he used a method in which the pre-peak deformation was subtracted from the total deformation of the specimen. Later on Markeset and Hillerborg [4] developed a model to study damage localization in compression based on the observations by Van Mier, but added an additional factor, namely the volume energy dissipation, which was mentioned independently by Willam et al. [5] in a conference paper around the same time. Nevertheless, the volume dissipation contribution is rather small. In addition to all that, RILEM TC 148 “Strain Softening of Concrete” focussed the attention on localization in compression. The papers produced by this committee [6,7] show that two effects interact during localization: the slenderness of the samples plus the boundary restraint between the loading platens and the specimen. The first effect, as mentioned above, was observed in 1984 by Van Mier [2]. The second effect was in depth discussed and convincingly shown by Kotsovos [8]. In parallel to the program run by the RILEM TC 148, Van Vliet and Van Mier [9] extended the round robin test program developed by RILEM TC 148 to prisms whose slenderness was between 0.25 and 2.0, and Jansen et al. [10,11] varied the slenderness between 2.0 and 5.5. The results presented Available online at www.sciencedirect.com Cement and Concrete Research 38 (2008) 386 – 395 ⁎ Corresponding author. ETSI Caminos, C. y P., UCLM. Avda. Camilo José Cela s/n, 13071 Ciudad Real, Spain. Tel.: +34 926 295 398; fax: +34 926 295 391. E-mail address: gonzalo.ruiz@uclm.es (G. Ruiz). 0008-8846/$ - see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.cemconres.2007.09.020