Page 44 - GN-JUNE-2014

Basic HTML Version

44
Geotechnical News • June 2014
www.geotechnicalnews.com
THESIS ABSTRACTS
accurate specification of vapour pressure or relative humid-
ity at ground surface. Flux of water at the soil boundary is
also a boundary condition of moisture for the analysis of the
problem of the flow of water at the ground surface.
The evaporation of water from water surface known as
potential evaporation is quite well understood. However, the
evaporation of water from a saturated-unsaturated surface
known as actual evaporation needs to be re-evaluated.
Several methods of estimating evaporation from unsaturated
soil surfaces can be found in the literature. According to
these methods, the actual rate of evaporation has been cal-
culated on basis of the total suction or relative humidity pre-
dicted at the soil surfaces. Total suction not only depends on
the character of the soil matrix but also on salt concentration
in the pore-water. Currently, the accuracy of these methods
can be questioned since they overestimate the actual rate
of actual evaporation. To overcome this deficiency, either
a variety of adjustment factors of total suction or modified
relative humidity at the soil surface have been used to com-
pute evaporative flux from the soil surface.
In this thesis’s work, the fundamental physics of water trans-
fer from a soil surface are reconsidered. The mechanism of
mass and heat transfer and the derivation of the equation
of evaporation are also re-visited. At the end, a theoretical
model (i.e., new soil-atmosphere flux equation) is developed
for prediction of evaporation rate from a soil surface using
the concept of “surface resistance’’ to vapour water diffu-
sion from the soil surface to atmosphere.
Soil suction and the corresponding water content at which
the actual rate of evaporation begins to depart from the
potential rate of evaporation during drying process are
re-assessed using a series of laboratory data (i.e., thin soil
section drying tests and soil column drying tests) collected
from the research literature. It is observed that soil suction at
which the actual rate of evaporation begins to reduce from
potential one for soil columns may be different from thin
soil sections. For example, the value of suction at evapora-
tion-rate reduction point appears to be approximately 3,000
kPa for the thin soil sections regardless of the soil texture.
However, it is observed that this suction appears to be in
between the air-entry value and residual soil suction for the
soil columns. As a result, a formula to determine soil suc-
tion at evaporation-rate reduction point is derived for soil
columns. A new set of equations related to the coefficient of
surface moisture availability, vapour pressure at soil surface
and soil surface resistance is then proposed.
The effect of pore-water salinity on the evaporation rate
from salinized soils was also considered. A function of
osmotic suction which depends on initial salt content and
volumetric water content at soil surface is derived for thin
soil layers during drying process and verified using data of
osmotic suction measured in the laboratory testing program.
Drying tests on thin soil layers as well as thick soil layers
were conducted using the non-saline and salinized soils (i.e.,
the selected sand and silt). The obtained results were utilized
to verify the proposed equations. Good agreement was
generally found between the computed and measured rate of
evaporation. In addition, these equations were also veri-
fied using the evaporative data collected from the research
literature. The findings throughout this thesis will help solve
the challenge of predicting evaporation from non-saline and
salinize soil surfaces with which the geotechnical engineers
are facing in many practical problems.
Supervisor: D. Chan, Geotechnical and Geoenvironmental Group,
Department of Civil & Environmental Engineering, University
of Alberta, 3-133 Markin/CNRL Natural Resources Engineering
Facility, Edmonton, Alberta T6G 2G7, Tel: 780-492-2176,
Fax: 780-492-8198
Understanding and Predicting Excavation
Damage in Sedimentary Rocks: A
Continuum Based Approach
Matthew A Perras
Matthew A Perras, Research Associate and Lecturer in
Engineering Geology, Institute of Geology, ETH Zurich,
Sonneggstrasse 5, 8092 Zürich, Switzerland, T
el. (direct): +41 44 633 3865, email: mperras@ethz.ch
The most widely accepted approach to long-term storage of
nuclear waste is to construct a deep geological repository,
where the geological environment acts as a natural barrier
to radio nuclide migration. Sedimentary rocks, particularly
argillaceous formations, are being investigated by many
countries. Underground construction creates a damage
zone around the excavation. The depth of the damage zone
depends on the rock mass properties, the stress field, and the
construction method. This research investigates the fracture
development process in sedimentary rocks and evaluates
continuum modelling methods to predict the depth of the
damage zone.
At the laboratory scale, a complete classification system for
samples of carbonate and siliciclastic rocks has been devel-
oped, with geotechnical considerations. Using this system,
crack initiation (CI) shows the most uniform range in each
class, particularly for mud rocks. Tensile strength was found
to be higher for the Brazilian method than Direct method of
testing. Brazilian reduction to Direct values was found to be
rock type dependent.
Bedding was found to influence the excavation behavior;
observed at the Niagara Tunnel Project and various excava-
tions in the Quintner limestone in Switzerland. A concep-
tual damage development process and potential fracture
networks in sedimentary rocks are used to summarize the
understanding of excavation damage developed in this
thesis.
Using a continuum modelling approach, a set of predictive
damage depth curves were developed for different damage