Dilatant Granular Materials Research Articles

Initial planar deformation of dilatant granular materials

A theory for the initial planar deformation of dilatant granular materials based on a kinematic proposal of R. Butterfield and R.M. Harkness (1972) is presented. The theory introduces an additional parameter called the angle of dilatancy into the traditional structure of plasticity theories for granular materials and soils. When the angle of dilatancy is zero, the present theory reduces to the theory introduced by A.J.M. Spencer in 1964. When the angle of dilatancy is equal to the angle of internal friction, the present theory reduces to the planar form of the theory introduced by D. C. Drucker and W. Prager in 1952. The properties of the theory presented here include coincidence of the stress and velocity characteristics, realistic energy dissipation predictions, and, in general, non-coincidence of the principal axes of stress and strain-rate. However, the angle of dilatancy is assumed to be a constant in this analysis and it does not decrease to zero with increased monotonic shearing deformation as experiment requires that it should, the theory therefore being limited to the initial deformation of dilatant granular materials.

Volume Changes During Undrained Triaxial Tests on Saturated Dilatant Granular Materials

Synopsis The results from a set of simulated undrained tests on saturated samples of lead shot, in which the volume of the pore-water is maintained constant for part or all of the test by increasing the cell pressure, are shown to be very similar to those obtained from undrained tests. They are used to show that considerable volume expansion of the sample may occur during an undrained test in which cavitation does not occur, and that this expansion is balanced by penetration of the rubber membrane into the sample under the increasing effective stress. From a further set of tests, it is concluded that if the same sample dimensions and rubber membranes as those herein are used, the error in the “cohesion” corresponding to the φu = 0 condition (Bishop and Eldin, 1950)1 is likely to be small for materials finer than coarse sand. Further, this error may be reduced by employing thicker membranes and samples of larger dimensions. The fact that the tangent to the Mohr circles from the undrained tests intercepts the abscissa at a value below the measured limiting negative pore pressure is also discussed. Les résultats d'une série d'essais simulés à teneur en eau constante sm des échantillons saturés de petits plombs, dans lesquels le volume de l'eau interstitielle est maintenu constant durant une partie ou l'ensemble de l'essai par l'augmentation de la pression cellulaire, s'avèrent très similaires à ceux obtenus lors des essais à teneur en eau constante. On les utilise pour montrer qu'une expansion considérable de volume de l'échantillon peut se produire lors d'un essai à teneur en eau constante au tours duquel la cavitation ne se produit pas et pour montrer que cette expansion est équilibrée par la pénétration de la membrane de caoutchouc dans l'échantillon sous l'effort effectif croissant. D'une autre série d'essais, on conclut que si l'on employe les mêmes dimensions d'échantillon et de membranes de caoutchouc que celles utilisées cidessus, l'erreur de la “cohésion” correspondant à la condition φu = 0 (Bishop et Eldin, 1950) sera apparamment faible pour des matériaux plus fin que du gros sable. D'autre part, cette errem peut être reduite en employant des membranes plus épaisses et des échantillons de plus grandes dimensions. On discute le fait que la tangente des cercles Mohr des essais à teneur en eau constante intercepte l'abscisse en un point de valeur inférieure à la pression interstitielle négative limite mesurée. 1The references are given on p. 182.