Deep sampling and testing in soft stratified clay

A 37-m thick layer of stratified clay encountered during a site investigation at Swann's Bridge, near the sea-coast at Limavady, Northern Ireland, is one of the deepest and thickest layers of this type of material recorded in Ireland. A study of the relevant literature and stratigraphic evidence obtained from the site investigation showed that despite being close to the current shoreline, the clay was deposited in a fresh-water glacial lake formed approximately 13 000 BP. The 37-m layer of clay can be divided into two separate zones. The lower zone was deposited as a series of laminated layers of sand, silt, and clay, whereas the upper zone was deposited as a largely homogeneous mixture. A comprehensive series of tests was carried out on carefully selected samples from the full thickness of the deposit. The results obtained from these tests were complex and confusing, particularly the results of tests done on samples from the lower zone. The results of one-dimensional compression tests, unconsolidated undrained triaxial tests, and consolidated undrained triaxial compression tests showed that despite careful sampling, all of the specimens from the lower zone exhibited behaviour similar to that of reconstituted clays. It was immediately clear that the results needed explanation. This paper studies possible causes of the results from tests carried out on the lower Limavady clay. It suggests a possible mechanism based on anisotropic elasticity, yielding, and destructuring that provides an understanding of the observed behaviour.Key words: clay, laminations, disturbance, yielding, destructuring, reconstituted.

Material effects on stress-induced defect generation in trenched silicon-on-insulator structures

We have investigated the influence of the material properties of the silicon device layer on the generation of defects, and in particular slip dislocations, in trenched and refilled fusion-bonded silicon-on-insulator structures. A strong dependence of the ease of slip generation on the type of dopant species was observed, with the samples falling into three basic categories; heavily boron-doped silicon showed ready slip generation, arsenic and antimony-doped material was fairly resistant to slip, while silicon moderately or lightly doped with phosphorous or boron gave intermediate behavior. The observed behavior appears to be controlled by differences in the dislocation generation mechanism rather than by dislocation mobility. The introduction of an implanted buried layer at the bonding interface was found to result in an increase in slip generation in the silicon, again with a variation according to the dopant species. Here, the greatest slip occurred for both boron and antimony-implanted samples. The weakening of the implanted material may be related to the presence of a band of precipitates observed in the silicon near the bonding interface. (C) 2001 The Electrochemical Society.