T. Matthew Evans

Sand–rubber mixtures: Experiments and numerical simulations

This paper documents the results of laboratory experiments and numerical simulations conducted to examine the behavior of mixtures composed of rubber and sand particles of similar size. Emphasis was placed on assessing the role of loading type on the load-deformation behavior and selecting appropriate parameters for the discrete element modeling of sand–rubber, with relevance to the use of compressible particulate systems for filtration control. Experimental results show that sand–rubber exhibits load–unload hysteresis and residual strains post-unloading due to particle–particle and particle–wall locking effects that arise from sidewall friction. It is shown that the discrete element modeling of sand–rubber requires unconventional schemes because of the stiffness contrast between sand grains and rubber grains. The results have implications in the design of compressible particulate systems for seepage and filtration control and in the development of prediction tools for the field performance soil–rubber, w...

Gradation effects on the mechanical response of crushed stone aggregate

A study was conducted to evaluate the effects of varying the gradation of aggregate base course (ABC) on material performance. Five different gradations consistent with a range of common ABC were selected. Experimental trials were performed to classify the materials and assess the performance differences between the five gradations. The gradation was found to affect the behaviour of the aggregate in that coarser gradations gave better strength and resilience values under conditions most similar to those found in the field. From research and experience during the laboratory testing, it was concluded that the coarsest blends became too difficult to realistically work with and they also lacked the stability of the well-graded combinations. It was also observed that as the amount of fines in the specimens exceeded 8–12% by mass, the fines governed the behaviour of the material. Results are interpreted using previously published work on specific surface, ‘ideal’ gradations and micromechanical models for the response of granular materials.

Interface Behavior of Granular Soils

Interface shear zones between particulate materials and continuum elements of man-made or natural materials have not been traditionally considered as shear bands. However, the results of recent micro-scale experimentation and numerical simulations have shown that they are indeed well developed partial shear bands. Further, these studies have indicated that significant similarities can be identified between these partial shear bands and what are more traditionally considered to be shear bands that evolve wholly within particulate materials. This paper presents results from physical and numerical experiments that suggest there is significant merit to this opinion. The physical experiments include quantitative analysis of the particle deformation following shear adjacent to continuum surfaces of different roughness as well as analyses of sand specimens sheared to different global strain levels. Complimentary 2-D Discrete Element Method modeling of particulate-continuum surface interfaces illustrates this parallel behavior.

A Numerical View into Direct Shear Specimen Size Effects

The ASTM D 3080-04 standard, Standard Test Method for Direct Shear Test of Soils under Consolidated Drained Conditions, indicates that the direct shear test specimen width L must be at least ten times larger than the largest particle in the specimen, L/d ≥ 10. The results documented in this note provide visual insight into specimen size and boundary effects through discrete element simulations conducted with numerical specimens ranging in size from L/d = 6.5 to L/d = 177. Particle translation data indicate that in the tested configuration, localized shear zones located far from the specimen boundaries develop only when the specimen size is L/d > 58. The results provide a description of the microstructure evolution during shearing and suggest that the ASTM specimen size criterion may be inappropriate when localized shear banding in uniform granular materials is sought.

Shear Banding and Microstructure Evolution in 2D Numerical Experiments

A limitation of using laboratory experiments to study the micromechanics of soils is that detailed information about the specimen microstructure is typically available at only one state in a test sequence due to the destructive nature of the forensic process. To study microstructure evolution, characterization of replicate specimens tested to various global axial strain levels has been undertaken, but this procedure presents some practical and theoretical problems. Accordingly, a numerical program was undertaken using the discrete element method to model the micromechanical deformation response of particulate assemblies in two dimensions. The simulated assemblies failed via regions of high localized strain. The microstructures of these assemblies were studied as a function of global axial strain to assess evolution of local and mesoscale void ratio distributions and mean free paths. Local void ratio distributions were modeled statistically and mesoscale measurements were used to assess microstructure inside and outside of the shear bands.

Transmissivity of a nonwoven polypropylene geotextile under suction

A permeameter has been developed for measuring in-plane transmissivity of geotextiles under a nearly constant value of suction along its length. The permeameter is capable of imposing gradients in excess of 10% and normal pressure up to 240 kPa, and permits the monitoring of the suctions within the geotextile during testing. To demonstrate the capability of the permeameter, a series of transmissivity measurements were made on a nonwoven polypropylene geotextile subject to different suction heads, normal pressures, and gradients. Transmissivities were up to two orders of magnitude less than the saturated value depending the on the magnitude of the suction head and whether the geotextile was being wetted or dried (hysteresis). Transmissivity values were independent of the gradient for these measurements. Increasing the applied normal pressure decreased the transmissivity at all values of suction head.

Wave propagation in assemblies of cemented spheres

Wave velocity through granular assemblies is an important measure of system stiffness and an indicator of response at small strains. In the laboratory, shear wave velocity is often measured using piezoelectric bender elements to induce and receive waves across a specimen. Shear wave velocity is well-known to be a function of packing fraction and bulk stress, but it has also been shown to be a strong function of depositional mechanisms – that is, specimens of equal packing fraction and bulk stress may have different shear wave velocities if they were assembled by different means. Specifically, the mechanical response of assemblies of cemented particles will be a strong function of deposition mechanism, stress history, and cementation properties. However, it is often not possible to have a detailed understanding of these three interrelated processes in natural systems. In the current work, measurement of shear wave velocity in numerical assemblies of spheres is used for process monitoring and system identif...

Membrane Effects in Biaxial Compression Tests

The effects of the confining membrane in laboratory tests on soil specimens have been the subject of numerous experimental, analytical, and numerical studies over the past half-century. This technical note expands the existing knowledge base by presenting a methodology and the associated results from an experimental study that has quantified the effect of the confining membrane in biaxial shear tests conducted on medium sand. The applicability of the method of biaxial tests on clay specimens is also presented. The results show that for both tests on sands and clays, the effect of the membrane on the shear stress on the failure plane are significant and should be accounted for in the interpretation of biaxial shear test results where localization occurs.

Particulate Study of Drained Diffuse Instability in Granular Material

Conventionally, the definition of instability is considered to be shear failure. However, there is substantial evidence that instabilities can also occur in a diffuse (homogeneous) manner without the apparent presence of a shear band under both undrained and drained conditions. Compared to undrained diffuse instability, drained diffuse instability of granular soil is less studied and is not yet well understood. This paper presents a discrete element method (DEM) study of drained diffuse instability of a granular soil. Instability and loss of controllability were observed well below the failure line when the DEM specimens were subjected to a constant shear-drained (CSD) stress path, which coincides with experimental and continuum numerical findings in the literature. The preliminary micro-scale examination includes the monitoring of coordination number evolution along CSD stress path. This study shows DEM modeling can well reproduce drained diffuse instability in granular materials and can be used to further investigate the mechanics of this phenomenon from a micro-scale perspective.

Geotextile Drains in Road Sections Subjected to Unsaturated Conditions

In many cases, the intrusion of water (either through rainfall or capillary rise) into the subgrade or aggregate base course in road sections can result in the swelling of the subgrade, decreasing the effective stress and, consequently, the shear strength. To mitigate these adverse conditions, a geotextile drain (GD) can be introduced to intercept upward and downward water flow and divert it towards edge drains, with flow induced under unsaturated conditions. Design of this type of system for the general case of partial saturation, however, is significantly complicated by the significant scale differences in the spatial extent and hydraulic properties of the soil and geotextile layers. In this paper, a road section consisting of aggregate base course (ABC) and subgrade is modeled using the finite element program SIGMA/W. Results show that the suction head in the subgrade and the air entry value of the GD have important effects on downward water flow into the subgrade. Low air entry value geotextiles cause a decrease in suction above the GD leading to a decrease in shear strength and increase in deformation. Guidelines are presented for the efficient design of GD as capillary barriers in roadways to minimize intrusion of water into the subgrade and ABC.