@article{oai:niigata-u.repo.nii.ac.jp:00028232,
author = {吉田, 昭治},
journal = {新潟大学積雪地域災害研究センター研究年報, 新潟大学積雪地域災害研究センター研究年報},
month = {Mar},
note = {The author has shown in his previous paper (YOSHIDA, 1978) that a set of mean pressure and deviatoric stress (which is not identical to principal stress difference) can be adopted reasonably to describe the stress-strain relationship for soils; furthermore, the author’s new idea that dilatancy is defined as the volumetric strain induced by deviatoric stress led to the conclusion that diIatancy is a phenomenon caused by anisotropy in deformation. And two types of coefficients of deformation, D and G, are introduced for “shear” are assumed to be expressed by the parameters of mean pressure and principal stress ratios (Eq.2). Five constants n,(a, α) and (b, P) are used to define these coefficients of deformation and were determined directly from the results of conventional triaxial compression tests. The two typical results of triaxial tests on dilatancy of sands are well accounted for by the simplified, nonlinear and anisotropic stress-strain relationship for soils (Eqs. (1) and (2)) in this paper as follows: 1) The samples of a sand contract first and subsequently expand until failure is reached under constant mean pressure (TATSUOKA and ISHIHARA, 1971). This experimental evidence has been expressed by the relationship between volumetric strain and stress ratio q/σm in many investigations. The relationship is well accounted for by Eq.(3) or (4), by which the results calculated agree well with TATSUOKA’s data as shown in Fig.1. 2)For dense sands, the volumetric change due to the increase of mean pressure, σm, is generally a contraction, but volumetric change may become an expansion at higher principal stress ratio, R=σ_1/σ_8, than a certain value of R (El-SOHBY, 1969). This can be explained by Eq. (6) as follows. Under the condition a>b and α<β, one gets D