New insights into the relevance of the length-to-diameter ratio on the compressive behaviour of rock specimens

Abstract The precise determination of uniaxial compressive strength (UCS) is crucial for assessing the load-bearing capacity of geological materials. To ensure the reliability and reproducibility of results, various standards and recommendations have been established. Although indirect measurement techniques exist, the uniaxial compression test (UCT) remains the benchmark for precise UCS characterization. However, discrepancies between the theoretical assumptions underlying the test and the observed outcomes—such as variations in the accepted failure patterns and sensitivity to specimen and platen properties—highlight the need for deeper analysis. In this context, numerous standards recognize the major influence of the specimen’s length-to-diameter ratio ( $$\:L/D$$ ), yet they recommend different $$\:L/D$$ values. To elucidate its relevance in the stress field generated within the specimen, this study investigates the stress distribution within cylindrical rock specimens across a range of L/D ratios (0.1, 1.0, 3.0, and 5.0), using a validated modelling framework supported by previously obtained experimental results. Findings reveal that the use of standardized steel platens slightly alters the stress ratio at the failure initiation point, deviating the stress state from the desired uniaxial compression condition. To address this, a method for identifying an optimal L/D ratio is proposed, enabling failure to initiate under conditions that closely approximate true uniaxial compression. Additionally, it is demonstrated that the deviations produced by the standardized platens can be mitigated by employing platens with stiffness comparable to that of the tested specimen.