Issue 78

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

cubes and standard cylinders, and it is the only information contained in the Eurocode [2] regarding the influence of aggregate size on experimental results. The same standard [2] indicates that the tensile strength measured on cubic specimens is about 10% greater than the value obtained for cylindrical specimens. However, the standard does not address the influence of cylinder size on splitting tensile strength. Available literature [8, 31] suggests that a size effect may also be present in splitting tensile tests performed on cylindrical specimens. The effect of specimen diameter on concrete’s splitting tensile strength has been investigated, for example, by Hasegawa et al. [32] and Bažant et al. [33]. The experiments reported in [31] have shown that the length of cylindrical specimens may also influence the measured tensile strength. Nevertheless, a consistent and widely accepted interpretation of tensile strength results obtained from cylinders of varying lengths has not yet been established. The authors’ own experiments were carried out to research in more depth the influence of aggregate parameters on concrete’s tensile strength, particularly the role of the aggregate type on the tensile mode of failure in concrete. Furthermore, the analysis of the failure crack’s patterns arising as a result of the splitting tests could help to better explain the influence of aggregate and cement matrix properties on concrete tensile strength. Materials The experimental program was designed to examine the effect of selected parameters on the splitting tensile strength of concrete. The main variables considered were the cement mortar strength and the type of coarse aggregate. In addition, splitting tensile tests were carried out on two cylindrical specimen sizes: 150 mm in diameter and 300 mm in height, as well as 150 mm in diameter and 150 mm in height. When designing concrete mixtures, a maximum grain size of 16 mm was used for both granite and gravel aggregates. The proportion between fine and coarse aggregate fractions was kept constant for mixtures containing natural gravel and crushed aggregate. The grading curves were selected within the recommended boundary limits specified in the Polish guidelines [34]. More detailed information regarding the concrete mixture design is provided in [31]. Considering the selected variables, the specimens were divided into four series, as summarised in Tab. 1. The following symbols were used to describe the specimens:

H– higher flexural strength of cement mortar, L – lower flexural strength of cement mortar, 16 – maximum aggregate size of 16 mm,

N – natural gravel aggregate, G – crushed granite aggregate. The workability of the fresh concrete mixtures was assessed using the Vee Bee test (see Tab. 1).

Coarse aggregate 2.0–16.0 mm [kg/m 3 ]

Sand 0.0–2.0 mm [kg/m 3 ]

Type of aggregate

Cement [kg/m 3 ]

Water [kg/m 3 ]

Vee Bee time [s]

Consistency class

Series

w/c

H16N L16N H16G L16G

gravel gravel granite granite

0.38 0.55 0.38 0.55

580.7 349.8 580.7 355.6

223.4 192.4 223.4 197.6

421.6 497.1 421.6 481.0

1140.0 1344.0 1140.0 1300.5

12 12 11 13

V2 V2 V2 V2

Table 1: Parameters and composition of concrete mixtures for specimens grouped in series.

Plastic moulds were used for making the specimens. After 24 hours, the cylinders were demoulded and carried to the water tubs. The curing conditions were kept in accordance with the code [28]. After 28 days of curing, the laboratory tests were performed. The beam specimens for testing the flexural strength of the cement mortar were prepared at the same time and in the same conditions as cylindrical specimens. In the meantime, the strength parameters of the aggregate were tested as well. Cement mortar testing Since both the strength of cement mortar and the mechanical properties of hardened concrete are strongly influenced by the water–cement ratio (w/c), two types of cement mortar were prepared to obtain cement matrices with different strengths. The first mixture represented a stronger matrix with w/c = 0.38, while the second one was designed as a weaker matrix with w/c = 0.55. Mortar specimens were produced using Portland cement CEM II 42.5R and fine aggregate (sand in 0–2.0 mm fraction). The flexural properties of cement mortar were determined on prismatic specimens measuring 40 mm × 40 mm

397