Mechanical performance and limited embodied-carbon assessment of concrete incorporating desert sand and surface-modified waste rubber

The rapid depletion of natural river sand resources and the increasing accumulation of waste tyres have intensified the demand for sustainable alternative materials that can simultaneously reduce resource consumption and environmental burdens in the construction industry. Although desert sand and waste rubber have each attracted considerable research attention as alternative aggregates in concrete, their combined effects on the mechanical performance and environmental sustainability of cement-based composites remain insufficiently understood. In particular, the synergistic influence of desert sand and surface-modified rubber particles on mechanical properties and limited carbon emissions has not been fully elucidated. In this study, rubberised desert sand concrete (RDSC) was developed by simultaneously incorporating desert sand and surface-modified waste rubber as partial replacements for conventional fine aggregates. A total of 16 concrete mix proportions were designed, with desert sand replacement ratios ranging from 0% to 60% and modified rubber contents ranging from 0% to 15%. The mechanical properties of the prepared composites were evaluated through cubic compressive strength, axial compressive strength, flexural strength, and static elastic modulus tests. In addition, the embodied carbon contributions of each constituent material were estimated within the defined raw material stage in accordance with GB/T 51,366–2019, and empirical relationships between the mix proportion parameters and the measured mechanical responses were established. The results indicate that replacing natural river sand with 20%–40% desert sand can enhance the compressive strength, while an appropriate amount of surface-modified rubber can increase the measured peak flexural strength. The R-DS20-10 mixture achieved the highest measured peak flexural strength. Through entropy-weighted TOPSIS evaluation of three strength-normalised carbon intensity indicators, R-DS20-0 was found to rank highest within the defined decision matrix with C i = 0.918, and maintained the top rank under equal-weight conditions. The selected empirical regression models all exhibited coefficients of determination higher than 0.85 within the investigated dataset and parameter ranges. The difference in embodied carbon among the mixtures was relatively small, with a maximum variation of approximately 0.17%. Therefore, the core value of this material system lies in its potential to conserve natural river sand and utilise waste rubber, rather than in achieving substantial carbon reduction. These findings provide a quantitative basis for selecting RDSC mixtures according to mechanical performance and carbon intensity priorities.