High-strength mortar (HSM) gradually has wide applications due to its exceptional strength, micro-expansion properties, and excellent fluidity. Behavior deterioration of structures in saline soil areas is primarily attributed to freeze-thaw cycles and sulfate attack. In this study, the coupling effect of freeze-thaw cycles and sulfate attack on the appearance, mass loss, and relative dynamic elastic modulus of HSM was investigated during erosion. Then, compressive experiments were conducted to assess the mechanical properties of HSM subjected to both freeze-thaw cycles and sulfate attack. The influences of coupling freeze-thaw cycles and sulfate attack on the compressive properties of HSM were quantified through regression analysis of experimental results. Empirical models for compressive stress-strain curves and damage constitutive behavior of HSM were developed, taking the coupled adverse effect into account. The results indicate that the coupled effect of freeze-thaw cycles and sulfate attack causes performance deterioration of HSM. The empirical models reproduce the compressive behaviors of HSM subjected to freeze-thaw cycles and sulfate attack.

Deep rock is under a complex geological environment with high geo-stress, high pore pressure, and strong dynamic disturbance. Understanding the dynamic response of rocks under coupled hydraulic- mechanical loading is thus essential in evaluating the stability and safety of subterranean engineering structures. Nevertheless, the constraints in experimental techniques have led to limited prior investigations into the dynamic compression behavior of rocks subjected to simultaneous high in-situ stress and pore pressure conditions. This study utilizes a triaxial split Hopkinson pressure bar (SHPB) system in conjunction with a pore pressure loading cell to conduct dynamic experiments on rocks subjected to hydraulic-mechanical loading. A porous green sandstone (GS) was adopted as the testing rock material. The findings reveal that the dynamic behavior of rock specimens is significantly influenced by multiple factors, including the loading rate, confining stress, and pore pressure. Specifically, the dynamic compressive strength of GS exhibits an increase with higher loading rates and greater confining pressures, while it decreases with elevated pore pressure. Moreover, the classical Ashby-Sammis micromechanical model was augmented to account for dynamic loading and pore pressure considerations. By deducing the connection between crack length and damage evolution, the resulting law of crack expansion rate is related to the strain rate. In addition, the influence of hydraulic factors on the stress intensity factor at the crack tip is introduced. Thereby, a dynamic constitutive model for deep rocks under coupled hydraulic-mechanical loading was established and then validated against the experimental results. Subsequently, the characteristics of introduced parameter for quantifying the water- induced effects were carefully discussed. (c) 2025 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).

The existing literature suggests that natural aggregate concrete demonstrates the least shrinkage, followed by recycled aggregate concrete (RAC) prepared using natural sand, with RAC prepared using recycled sand (RS) from the weathered residual soil of granite demonstrating the greatest shrinkage. Internal incorporation of a MgO expansion agent (MEA) effectively compensates for the excessive shrinkage of the latter; however, the influence of the MEA on the strength development of RAC prepared using RS after natural curing, rather than accelerated carbonation curing, remains unclear. In this study, compression tests of RAC prepared using RS at different stages of natural curing were performed and the corresponding material compositions of RAC were determined and quantified via X-ray diffraction and thermogravimetry-differential thermogravimetry. The soluble carbonate content in RS was determined by ion chromatography, and the morphology of RAC was observed using scanning electron microscopy. The mechanism of strength development of RAC during aging was determined. Furthermore, compressive tests of recycled lump-aggregate concrete (RLAC) were performed to investigate the influencing degree of RAC as fresh concrete on the compressive properties of RLAC. The following key results were noted: (a) the MEA impairs the compressive strength of concrete, but the degree of impairment decreases with curing, and this is attributed to the transformation of Mg(OH)2 to MgCO3. (b) The presence of soluble carbonates in RS (7.2 %) is the main source of carbonate in the conversion of Mg(OH)2 to MgCO3. Mg(OH)2 particles adhere to the surface of RS particles and react with soluble carbonate to generate MgCO3. (c) At 56 days of curing, the addition of 6 % MEA or increasing the replacement ratio of RS impaired the compressive strength of RLAC to a certain extent. However, even with 100 % RS, the compressive strength and elastic modulus of RLAC were impaired by only 7.4 % and 5.8 %, respectively. With 6 % MEA, the impairments were even smaller and negligible.

The disposal of massive amounts of demolished concrete and excavated soil, and the consumption of large amounts of cement and natural aggregates for concrete have restricted urban development. Recycled lumpaggregate concrete (RLAC), containing recycled lumps (RLs), recycled coarse aggregates (RCAs), and recycled sand (RS), is a potential solution. However, the higher crushing index and clay content, along with the lower fineness modulus and apparent density of RS derived from weathered residual soil of granite (WRSG) compared to natural sand, have raised concerns regarding the performance of RLAC. In this study, the compressive behavior and permeability of RLAC were investigated. The positive effect of WRSG-derived RS on RLAC permeability was quantitatively revealed for the first time, and the underlying mechanisms were elucidated. The quantitative effect of RS on the compression behavior and its engineering acceptability were also clarified. The results showed that: (a) The use of WRSG-derived RS had a slight effect on the compressive strength of RLAC. However, the presence of clay minerals with poor elastic modulus in RS led to a reduction in the elastic modulus of RLAC, with a maximum decrease of 6.1%, which is acceptable for practical applications. (b) The promotion of RS to cement paste hydration, adequate filling of RS in the cement paste, and tight mechanical interactions between RS and the cement paste contributed to a denser mortar microstructure. The permeability coefficient of RLAC and the porosity of harmful pores (>100 nm) in fresh mortar were reduced by 86% and 55%, respectively, with 70% RS utilization. (c) When the water-cement ratio of fresh concrete was 0.36 or 0.45, although the impermeability grade of the RLs was only P4, the RLAC still had an impermeability grade of P12, which can satisfy almost all the engineering impermeability requirements.

The use of both recycled coarse aggregates (RCAs) and recycled sand (RS) derived from weathered residual soil of granite (WRSG) into concrete has the potential to greatly enhance the recycling of construction and demolition waste. However, the characterization of RS from WRSG and the compressive and flexural performance in fresh concrete containing RCAs and RS have not been thoroughly investigated. In this study, clay content, fineness modulus, chemical compositions, mineral compositions, and pore structure of RS from WRSG were tested. On this basis, the optimized preparation parameters of RS were suggested. The compressive behavior, flexural behavior, and cement hydration degree of recycled aggregate concrete (RAC) simultaneously containing RS and RCAs were investigated comprehensively. A stereological model was proposed to explain the results related to cement hydration. The results showed that: (a) the optimized preparation could substantially lower the clay content of RS; (b) RS was more porous than natural sand (NS), resulting in a higher water absorption during mixing; (c) the compressive strength of concrete containing RS developed faster than the concrete with NS; (d) at day 90, the compressive and flexural strength of the concrete containing RS were not less than those of the concrete with NS; and (e) RS was shown to have a greater influence on the hydration degree of cement paste than RCAs, due to RS significantly reducing the average value of inter-aggregate spacing in concrete, making the cement paste more susceptible to the internal curing effect induced by the water in aggregate pores.