ABSTRACT The erosional resistance of soils dominantly influences the safety and serviceability of hydro-geotechnical structures such as dikes, dams, canals, riverbanks, piers, etc. In the absence of detailed erosion testing, it is prudent to have an educated estimate of erosion resistance based on simple physical/mechanical properties of soils. This study systematically investigates the interrelation between soil physical properties and erosion propensity of clay-sand mixed soils. Jet erosion tests were performed on carefully controlled, laboratory-substantiated mixtures of a clay soil with different amounts (10 %, 20 %, 30 %, 40 %, and 50 %) of silica sand, and erosion parameters, that is, critical shear stress (τc) and erodibility/detachment coefficient (kd) were measured. Experimental findings confirmed that the amount of fines, clay content, plasticity index, and compaction water content positively increased the erosion resistance, whereas the dry density and unconfined compressive strength were not directly influential in erodibility. Moreover, results suggested that the soil fabric significantly influenced erosional behavior insofar as the soil reflected a sharp drop in erosion resistance upon becoming sand dominated. Regarding practical applications, seven equations available in the literature for prediction of critical shear stress (τc) were compared with test measurements, and their accuracies were assessed.
ABSTRACT This work presents a practical glauconite content test using magnetic separation to determine gravimetric (mass) content of intact glauconite-bearing grains to in clean sand, which can be used to represent the crushable proportion of a soil sample. The test provides an additional level of soil characterization for glauconite, for which the currently used methods measure either the glauconite mineral content or provide only an estimate of glauconite grain content using a microscope. The simple test was designed to be used in the field or in a laboratory on an ocean-going vessel, relevant to offshore energy infrastructure. It uses a rare-earth magnet and a sample jar to determine the concentration by mass of glauconite grains, which are green, paramagnetic, and friable. They tend to crush under load, which alters its soil behavior, transitioning from coarse grained to fine grained. Typical measurement methods for determining glauconite content are visual grain counting, X-ray diffraction, and tabletop electromagnetic separators. None of these methods can measure the gravimetric content of glauconite-bearing grains in a soil and they can be costly to run or require expert knowledge to use. The variability in the gravimetric concentration with this method is presented by testing a series of prepared samples with Ottawa sand and glauconite magnetically collected from the Poricy formation. A comparison between the magnetic separation test, digital image analysis, and X-ray diffraction testing is provided. It was found that, for the soil tested, the magnetic separation test has a variability within 0.4 % of the actual concentration in prepared samples. These results show that the magnetic separation test is an effective and efficient option for determining the gravimetric glauconite concentration of a soil sample.
ABSTRACT Collapsible soils, commonly found in arid and semi-arid regions, present significant geotechnical challenges: moisture induces collapse of their metastable structure, causing sudden settlement that jeopardizes infrastructure integrity. This study investigates the effectiveness of a composite binder consisting of 3 % lime and 4–8 % finely ground ceramic waste for stabilizing reconstituted collapsible soils representative of semi-arid and arid regions worldwide, using locally sourced North African materials. Specimens composed of 80 % Boussaâda sand and 20 % Djebel Debbagh kaolin were prepared, cured for 7 d under sealed conditions at 20°C ± 2°C, and evaluated using Atterberg limits, Standard Proctor compaction, one-dimensional oedometer flooding, and unconsolidated–undrained triaxial tests. The addition of 3 % lime combined with increasing ceramic-waste content progressively reduced soil plasticity, with the plasticity index decreasing from 8.66 % for untreated soil to 3.19 % at 8 % ceramic waste + 3 % lime, whereas the 6 % ceramic waste combination restored the maximum dry density to 1.91 g cm−3 at a reduced optimum moisture content. Collapse potential, under one-dimensional oedometer flooding at 200 kPa, was reduced to less than 0.05 % at 8 % ceramic waste with 60-blow compaction energy, reclassifying the treated soil as non-collapsible. Cohesion increased substantially from 16 to 95 kPa, whereas the friction angle varied within approximately 35°–40° across the tested mixtures, suggesting cohesion was the primary driver of the strength gain, consistent with a contribution from cementation and bonding mechanisms. Microstructural analyses using scanning electron microscopy–energy-dispersive X-ray spectroscopy provide evidence consistent with the formation of calcium silicate hydrate (C–S–H) and calcium aluminate hydrate (C–A–H) phases, resulting in denser matrices, thicker interparticle bonds, and reduced porosity. The 3 % lime dosage consistently provided the alkalinity and calcium supply necessary to activate pozzolanic reactions in the ceramic fines, offering an effective, locally valorizable ground-improvement approach.
Laboratory-scale vane tests were performed on iron ore tailings (IOT), which are fine-grained tailings with higher permeability than natural clays, to investigate the effect of shear rate on measured shear strengths. IOT was prepared at different void ratios by slurry consolidation, with vane shear tests then conducted at peripheral velocities ranging from 20 to 200 mm/min. The post-peak strength was measured at rotation angles of 180° and 360°, which are relatively small compared to multiple revolutions (i.e., 3,600° or more) typically required for full remolding. The peak and post-peak shear strengths were significantly influenced by the peripheral velocity, transitioning from being viscous-dominated to partially drained behavior. Viscous effects predominantly dominated the shear strength at higher peripheral velocities and lower rotation angles, whereas partial drainage effects became more prominent when using lower shear rates, higher rotation angles, and lower void ratios. A transition from viscous-dominated to partially drained behavior was observed with decreasing void ratio, associated with an increased coefficient of consolidation. The Blight/Chandler approach was shown to be effective in assessing the effects of drainage parameters on measured strength. These findings highlight the critical need to select material-specific rates of rotation to accurately characterize the shear strength of fine-grained tailings.
ABSTRACT Reliable estimation of critical shear stress (CSS or τc) is essential for accurate prediction of erosion and scour around bridge piers with cohesive soils. The CSS is the shear stress required to initiate erosion, typically assumed at a rate of 0.1 mm/h. This assumption, although convenient, oversimplifies the erosion process and introduces significant variability across different testing techniques and flow regimes. This study compares critical shear stress values obtained from two erosion testing methods: the erosion function apparatus (EFA) and a portable scour testing device (PSTD), using undisturbed CL soil samples from two different bridge sites in Illinois. Two analytical methods are proposed: Method A, which determines τc as the range between the onset of erosion and the 0.1 mm/h erosion rate threshold for individual tests; and Method B, which aggregates data across multiple sites to account for soil and site-specific variability. Results show that EFA generally reports higher τc values than PSTD, likely because of differences in flow regimes. Turbulent conditions in PSTD consistently produced lower τc values, emphasizing the importance of flow regime-specific assessments in scour depth design. Comparison of τc determined in this study with τc determined from an empirical correlation presented in previous literature revealed substantial scatter, with prior studies often underestimating τc. The proposed methods offer a more field-representative and standardized approach to determining τc, which is imperative for the development of realistic and site-appropriate scour prediction models.
