THESIS
2017
xxii, 216 pages : illustrations ; 30 cm
Abstract
Unsaturated loess has a metastable structure, which can be maintained temporarily
by suction. Many theoretical and experimental studies have demonstrated that
volumetric behaviour of soil is also strongly affected by stress and temperature. Volume
changes of unsaturated soils can have a significant influence on many geotechnical
structures. However, in the previous studies, the effects of soil structure (intact versus
recompacted) are neglected. Moreover, only elevated temperatures are studied in the
previous researches and limited data with temperatures lower than the room
temperature can be found in literature.
The principle objective of this research is to investigate volumetric behaviour of
intact and recompacted collapsible loess under thermo-hydro-mechanical loading. A
s...[
Read more ]
Unsaturated loess has a metastable structure, which can be maintained temporarily
by suction. Many theoretical and experimental studies have demonstrated that
volumetric behaviour of soil is also strongly affected by stress and temperature. Volume
changes of unsaturated soils can have a significant influence on many geotechnical
structures. However, in the previous studies, the effects of soil structure (intact versus
recompacted) are neglected. Moreover, only elevated temperatures are studied in the
previous researches and limited data with temperatures lower than the room
temperature can be found in literature.
The principle objective of this research is to investigate volumetric behaviour of
intact and recompacted collapsible loess under thermo-hydro-mechanical loading. A
suction- and temperature- controlled double cell triaxial apparatus, which can fulfil both
heating and cooling, was developed for testing unsaturated soils. By using this modified
apparatus, four series of tests were carried out on intact and recompacted loess at
various suctions and temperatures. Isotropic compression behaviour, thermally induced
volume change, wetting-induced collapse and water retention behaviour were studied.
Moreover, a new hysteretic water retention model considering effects of pore non-uniformity
and pore size distribution evolution was proposed.
Isotropic compression tests show that yield stress of loess decreases with a
decreasing suction (wetting-induced softening) and decreases with an increasing
temperature (thermal softening). The wetting-induced softening of yield stress of
recompacted loess becomes more significant at a higher temperature. The observed
thermal effects on wetting-induced softening of yield stress are likely because with
decreasing suction, the stabilizing inter-particle normal force decreases more at a higher
temperature. For both normally consolidated intact and recompacted loess,
experimental results show that cooling-induced plastic volume change is observed
during a heating and cooling cycle. The observed elastoplastic behaviour during cooling
cannot be captured by using existing thermo-mechanical models, which predict elastic
contraction during cooling. Thus, a new yield surface (temperature decrease) is
proposed. Moreover, it is found that wetting-induced contraction of recompacted loess
at 50℃ is about three times of that at 5℃. The larger contraction at 50℃ is mainly
because the wetting-induced softening of yield stress is larger at a higher temperature.
At a confining stress of 110 kPa, at which both intact and recompacted soils are
normally consolidated, wetting-induced volumetric contraction of intact specimen is
about 30% larger than that of recompacted specimen. This is mainly because the
wetting-induced softening of yield stress of intact loess is more significant than that of
recompacted one.
The proposed water retention model considered effects of pore non-uniformity on
water retention during drying and wetting. Furthermore, pore size distribution evolution
during various stress paths was also incorporated. Based on a single pore size
distribution, both drying and wetting paths of water retention curves of unsaturated soils
are well predicted.
Post a Comment