Affordable Access

CYCLIC STRESS-STRAIN AND LIQUEFACTION CHARACTERISTICS OF SANDS. (VOLUMES I AND II)

Authors
Publisher
Purdue University
Publication Date
Keywords
  • Engineering
  • Civil
Disciplines
  • Design
  • Engineering

Abstract

Liquefaction of saturated sand and silty sand deposits has been recognized as a major cause of damage during earthquakes. However, in spite of many research studies during the past two decades, there are still conflicting opinions on critical aspects of the phenomenon of soil liquefaction, including the definition of the term "liquefaction." For these reasons, a research program on the behavior of cohesionless soils under cyclic loading was initiated at Purdue University.^ An important effort was devoted to delineate and understand the physical factors controlling the response of cohesionless soils to cyclic loading and consequently, the factors involved in the different phenomena which have been described with the term "liquefaction" such as steady-state flow, a condition of zero effective stress, or cumulative residual strains due to cyclic loading.^ A new hybrid resonant column/torsional shear apparatus was designed and built as a part of this research program. The new apparatus permits the determination of dynamic soil properties on a single solid or hollow cylinder specimen over the entire range of shear strain amplitudes of engineering interest, i.e. from 10('-4)% to 10%. Torsional shear tests were performed with the new apparatus on reconstituted specimens of Ottawa 20-30 sand.^ Among other things, it was found that: (a) the maximum shear modulus, G(,max), is relatively insensitive to important factors affecting the undrained behavior of sands such as stress ratio (and hence fabric), stress history, and cyclic prestraining. Consequently, a direct correlation between G(,max) and liquefaction potential is not possible; (b) if staged testing includes the determination of shear moduli by means of resonant column tests, which imply the application of a considerable number of cycles, it can lead to significant overestimations of shear modulus for sands with no prior cyclic strain history even at shear strain amplitudes as low as 1.1 x 10('-2)%; and (c) at a given relative density, the effective stress path corresponding to an undrained monotonic loading tests defines a "state boundary" or "collapse" surface marking the initiation of strain softening behavior of loose sands under cyclic loading at different shear stress amplitudes smaller than the peak monotonic strength. ^

There are no comments yet on this publication. Be the first to share your thoughts.