The awareness of the impact of high temperatures on rock properties is essential to the design of deep geotechnical applications. The purpose of this research is to assess the influence of heating and cooling treatments on the physical and mechanical properties of Egyptian granodiorite as a degrading factor. The samples were heated to various temperatures (200, 400, 600, and 800 °C) and then cooled at different rates, either slowly cooled in the oven and air or quickly cooled in water. The porosity, water absorption, P-wave velocity, tensile strength, failure mode, and associated microstructural alterations due to thermal effect have been studied. The study revealed that the granodiorite has a slight drop in tensile strength, up to 400 °C, for slow cooling routes and that most of the physical attributes are comparable to natural rock. Despite this, granodiorite thermal deterioration is substantially higher for quick cooling than for slow cooling. Between 400:600 °C is ‘the transitional stage’, where the physical and mechanical characteristics degraded exponentially for all cooling pathways. Independent of the cooling method, the granodiorite showed a ductile failure mode associated with reduced peak tensile strengths. Additionally, the microstructure altered from predominantly intergranular cracking to more trans-granular cracking at 600 °C. The integrity of the granodiorite structure was compromised at 800 °C, the physical parameters deteriorated, and the rock tensile strength was negligible. In this research, the temperatures of 400, 600, and 800 °C were remarked to be typical of three divergent phases of granodiorite mechanical and physical properties evolution. Furthermore, 400 °C could be considered as the threshold limit for Egyptian granodiorite physical and mechanical properties for typical thermal underground applications.

Purpose. This study investigates the feasibility of using calcined marble cutting residues as a partial substitute for fine aggregates in concrete to improve material sustainability while maintaining mechanical performance. The research aims to evaluate the effects of incorporating marble waste on strength properties and to determine an optimal substitution ratio suitable for practical applications. Methods. Marble cutting residues were collected, dried, and calcined at 750°C, then ground into a fine powder. The processed material was used to replace natural sand at substitution levels of 0, 5, 10, and 15% by weight. Standard concrete specimens were prepared and cured for 7, 14, and 28 days. Compressive, flexural, and splitting tensile strength tests were conducted in accordance with relevant standards. Microstructural observations were performed to analyze particle distribution, interfacial bonding, and pore structure. Findings. The results indicate that concrete containing 5% calcined marble waste achieved a 28-day compressive strength of 38.6 MPa, compared with 39.2 MPa for the control mixture. Flexural and tensile strengths at this replacement level varied by less than 4%. In contrast, higher replacement ratios of 10 and 15% reduced strength by up to 12%, primarily due to increased porosity and reduced packing efficiency. Originality. This study provides systematic experimental evidence on the mechanical and microstructural behavior of concrete containing thermally treated marble residues. Practical implications. The findings confirm that limited substitution of natural sand with calcined marble waste can be safely used in concrete production, supporting waste valorization, resource conservation, and environmentally sustainable construction practices.

Rock mechanical properties are crucial inputs for rock engineering applications in mining projects. While these properties can be determined through well-established laboratory testing methods, such destructive tests are often expensive and time-consuming. As a result, indirect (non-destructive) methods (e.g., P-wave ultrasonic velocity) can be used when a strong correlation with destructive test results exists. Traditional indirect approaches generally rely on single-variable predictor relationships; thus, they require extensive datasets. This study introduces multivariate predictors derived from geotechnical laboratory test results, developed using an Adaptive Neuro-Fuzzy Inference System (ANFIS) to predict the mechanical rock strength properties. Twenty-four NX-size diorite core samples collected from three mining sites were tested. P-wave velocity, bulk density, and uniaxial compressive strength (UCS) were used as inputs to predict Poisson’s ratio (ν) and elastic modulus (E) using different numbers of rules and membership functions (MFs). Similarly, P-wave velocity and bulk density were used to predict UCS with two and three rules and MFs. The developed ANFIS models were utilized as predictors and successfully reproduced the experimental dataset with high accuracy, achieving RMS values exceeding 90%. This approach can enhance the applicability of rock property indexing at remote sites and during early stages of mining project pre-feasibility studies, where testing access is limited.