The role of particle shape in the mechanical behavior of granular soils: A state-of-the-art review

• Particle shape significantly influences the mechanical behavior of granular soils. • Qualitative and quantitative methods of particle shape characterization are reviewed. • The challenges posed by particle shape on soil behavior analysis are explored. • Potential future research avenues for incorporating particle shape effects are identified. • ML models are developed to predict soil properties by including the effect of particle shape. Shape of soil grains has a major role in the characterization of its behavior. This important feature has an appreciable impact on all mechanical properties of granular soils. In order to accurately capture such an important contribution, the grain morphological features should be carefully analyzed and described in a systematic manner. The present study aims to review the literature concerning the qualitative and quantitative characterizations of the soil particle shape and its substantial impact on the mechanical behavior of granular soils. Qualitative characterization of the particle shape involves descriptive assessments of the level of angularity or roundness, such as the terms “rounded”, “sub-rounded”, “sub-angular”, and “angular”, while quantitative characterization refers to measurable parameters, like aspect ratio, regularity, sphericity, and circularity, which provide numerical data on the shape of soil grains. This study specifically examines the influence of particle shape on the compression characteristics, monotonic shear response, shear strength properties, and critical state behavior of granular soils, as well as the mechanical behavior of soil-solid interfaces and stabilized earthen materials. The potential challenges associated with the investigation of the effect of particle morphology on the mechanical characteristics of granular soils are also summarized, and a roadmap is outlined for future research. The findings of this study show that as the soil grains become more angular, the at-rest coefficient of earth pressure ( K 0 ) and the maximum dilation angle ( ψ m a x ) decrease whereas the peak friction angle ( φ p ), critical state friction angle ( φ c s ), intercept of critical state line ( e Γ ), and slope of critical state line ( λ ) all increase. With this broad perspective and in an attempt to put such crucial effects into practice, comprehensive sets of experimental data records are compiled from the previous studies in the literature based upon which new practical machine learning (ML) models are developed for the prediction of various mechanical properties of granular soils by accounting for the substantial contribution of particle shape. The study provides practicing geotechnical engineers with a profound insight into the macro- and micro-scale impacts of grain morphology on the mechanical characteristics of granular soils while offering new horizons in the incorporation of particle shape into predictive models for such properties. With these practical models, engineers can readily estimate the compression and strength-related parameters of granular soils by simply examining their particle shape, grain size distribution, density, and overburden pressure.