Issue 78

M. S ł owik et alii, Fracture and Structural Integrity, 78 (2026) 393-408; DOI: 10.3221/IGF-ESIS.78.23

I NTRODUCTION

W

hen describing concrete as a construction material, at least two basic facts should be considered. The first concerns the composite structure of the material. The second is related to the brittle character of concrete failure. To estimate the cracking resistance of concrete structures and to analyse fracture processes in concrete, we need to take into account the tensile concrete strength and other fracture properties of concrete, such as fracture energy   or stress intensity factor K I . Concrete’s internal structure is different from that of other engineering materials. It consists of different sizes of coarse aggregate and a hardened cement matrix. The internal structure of the concrete matrix also contains water, unhydrated cement grains, voids, pores and microcracks. The random distribution of aggregate grains and microdefects is a direct cause of the heterogeneity of concrete’s internal structure. As aggregates constitute the bulk of concrete (60-80% of the concrete volume), they significantly influence concrete properties. Both the type of aggregate and its granulation are of importance. The fact that concrete is a heterogeneous material means that its properties should be tested experimentally and evaluated statistically. When testing concrete properties, the type and size of tested specimens are crucial, and therefore experiments should be performed at standardised conditions. Concrete compressive strength is tested in a uniaxial compression test (according to the procedure given in the European code [1]), and a uniaxial tensile test would be the best for testing tensile strength and other fracture properties of concrete, such as fracture energy GF and stress intensity factor KI. However, indirect methods are usually recommended due to the difficulties involved in performing direct tensile tests. The Brazilian splitting test, described in the European code [2], is a standard method for testing the tensile strength of concrete, whereas flexural tests are proposed for testing fracture energy [3] and stress intensity factor [4]. Although the splitting tensile test may be considered a traditional method, its simplicity and robustness make it still highly valuable for present-day studies, especially when larger datasets and statistical approaches are required. Unlike many direct tension methods, it allows practical testing of numerous specimens without sophisticated equipment, which makes it particularly suitable for comparative studies such as the one presented in the paper. Concrete properties are sensitive to the specimen’s size. The decrease of a material property is usually observed with the increase of the specimen’s dimensions. The experimental results, for example [5-10], suggest that the shape and dimensions of a specimen affect the tensile strength of concrete. Unfortunately, there is still insufficient information in the code [2] on the size effect when testing the tensile strength of concrete using the Brazilian method. The latest research performed by the authors and described in [11] has shown that the aggregate granulation influences the tensile strength of concrete significantly. Experimental results presented in the professional literature showed how important it is, when describing concrete properties, the role played by the aggregate. For example, Golewski demonstrated that various types of coarse aggregate significantly affect the strength and fracture parameters of concrete [12]. Similarly, Saini et al. [13] showed that variations in concrete strength characteristics are influenced not only by aggregate type but also by its mineralogical composition and mechanical properties. Consistent findings were reported by Nadera et al. [14], who also confirmed the influence of the aggregate type on concrete’s strength. In this paper, an analysis based on the experimental research is presented on how the aggregate type - gravel and granite - influences the splitting tensile strength of concrete. Among available aggregate types, natural gravel and crushed granite remain widely used worldwide in structural concrete due to their availability and mechanical reliability. Their ongoing prevalence in construction makes it important to continue studying their influence on tensile behaviour, in order to provide data directly relevant for both engineering practice and code development. Although previous studies have demonstrated that both aggregate characteristics and cement matrix properties influence the tensile behaviour of concrete, the interaction between these factors has received considerably less attention. Most available studies focus primarily on the influence of aggregate characteristics or report tensile strength values without relating the measured mechanical response to the corresponding fracture mechanisms. Consequently, the combined role of cement matrix strength and aggregate type in governing tensile failure remains insufficiently clarified. To address this research gap, the present study investigates the combined influence of cement matrix strength and aggregate type on the splitting tensile behaviour of concrete by analysing crack propagation and fracture surfaces after failure. In particular, different patterns of a failure crack propagation, with aggregate grains cracking or only through an interfacial transition zone, were examined. Even though the effect of cement matrix and aggregate type on tensile strength has been investigated for decades, it remains an active research topic because of the complexity of fracture processes and the continuous refinement of standards and predictive models. The novelty of this study lies in combining classical Brazilian testing with statistical modelling and fracture surface analysis to quantify the joint effect of aggregate type and cement matrix strength on splitting tensile strength. This integrative approach provides

394