Abstract

The rock mass behavior can be studied using the complete stress–strain curve obtained from laboratory tests. To capture the complete curve and prevent violent failure, a stiff servo-assisted testing machine is required. Mainly due to the challenges in performing these tests and the fact that rock mass behavior is typically studied prior to failure, there is a lot of uncertainty in post-peak behavior. In fact, several studies have been developed to estimate pre-peak mechanical properties based on geological characteristics such as grain size and mineral composition but not to estimate post-peak properties. This study uses geological characteristics, index test results and pre-peak mechanical properties to estimate post-peak properties. For this purpose, uniaxial compression tests were conducted on ten different lithologies of brittle rocks. The procedure applied is based on standardized recommendation, photography techniques at hand scale have been used while for geological analysis. Three models were adjusted. The first model predicts UCS using the Dynamic Young’s modulus, the second predicts the likelihood of capturing the post-peak behavior, and the third is a multivariable model that estimates the drop modulus considering mechanical properties and porosity. Regarding geological analyses at the scale of photography, no conclusive information was found. Literature suggests that porosity is an important property for estimating pre-peak mechanical properties, and this study provides evidence that porosity can be helpful in predicting the drop modulus. It is recommended to strengthen the models with additional tests, reduce biases associated with LVDTs location and the representativeness of certain lithologies compared to others, and evaluate other testing methodologies for performing tests faster. © The Author(s), under exclusive licence to Springer-Verlag GmbH Austria, part of Springer Nature 2024.