The total stress tensor as the average stress within a triphasic granular medium is formally derived from micromechanics where internal forces associated with the solid phase, the two immiscible fluid phases and the associated three interfaces are explicitly accounted for. It is demonstrated that for rigid solid particles, the contributions of all local solid-fluid surface tensions to the total stress are eventually zero. The present work gives the total stress expression as a function of a solid-phase specific stress tensor and a fluid mixture stress contribution that is related to the material's microstructure. A generally non-spherical fluid mixture stress is obtained in contrast to an averaged hydrostatic fluid pressure usually associated with standard thermodynamics. The tensorial nature of this fluid mixture stress contribution is highlighted through numerical experiments pertaining to an idealized granular material in the pendular regime at low wetting saturations. Numerical simulations providing full access to microstructural information are conducted using the Discrete Element Method (DEM), which describes internal forces using resultant forces that clearly deviate from the distributed nature of internal forces in triphasic granular media, e.g., fluid pressures. Nevertheless, this micro-scale representation is demonstrated to be indeed valid for macro-scale stress description in the pendular regime.

We investigate the macroscopic mechanical influence of the local liquid-solid contact angle that governs the fluid distribution in granular soils under unsaturated conditions. To this end, a Discrete Element Method (DEM) based implementation that accommodates for any contact angle is proposed and applied to an idealized granular material in the pendular regime. The DEM model includes resultant capillary forces as well as a comprehensive description of the capillary bridges (volume, surface, orientation tensor) by solving the Laplace-Young equation in a general case, instead of using any unnecessary phenomenological relation. Macroscale mechanical simulations for different constant contact angle values reveal that granular assemblies are less sensitive to unsaturated conditions for higher contact angles, which is in line with the contact angle influence at the microscopic capillary bridge

The potential effects of a calcium channel blocker (nifedipine) on epiphyseal growth plate and bone remodeling have been investigated in growing rabbits. The treated group received 6 mg/kg/day nifedipine twice daily by gavage for 10 weeks. An untreated group was used as control; with this dose, neither toxic effects nor decrease in the body weight have been observed. No modifications of blood phosphocalcic parameters have been found. In the treated group there is a significant lower cancellous bone volume, lower osteogenesis, shorter labeled perimeters, and lower mineral apposition rate than in the control group. Epiphyseal growth plate thickness is lower than in the untreated animals and considerable morphological changes are observed in the growth zone compared with the control group. A decrease in the growth of humerus length was found. In conclusion, nifedipine affects bone physiology, especially with consequences on bone growth. These effects appear to be quantitatively important, and there is the possibility of bone side effects on therapeutic use in humans, especially in young subjects.

a b s t r a c tAn incrementally nonlinear constitutive relation is formulated to describe the mechanical behaviour of infilled rock joints. This relation is calibrated using a discrete element model, which is validated by experimental data. Since the phenomenological relation fully combines the normal and tangential directions of a rock joint, it can reproduce rock joint features as the dilatancy process and the contribution of compression on tangential stress. To take into account the hardening of the material, the influence of the previous shear loading history on the mechanical response of the rock joint is considered. Finally, the performance of the constitutive relation is verified by showing the good agreement between the responses predicted by the relation and those obtained by the discrete model for different loading paths.