216
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
3 SOME NUMERICAL ASPECTS
The isotache concept, proposed
by Šuklje (1957)
, can
conveniently be used for modeling rate dependency of clays.
The isotache concept states that there is a unique relationship
between the current strain rate (change in void ratio), effective
stress state and strain (void ratio).
Under oedometer testing conditions, a direct implication of
the isotache concept is that the experienced
p
′
c
is dependent on
the time between the load increments, or the rate of loading.
Under EOP incremental loading scheme, the implication of the
isotache concept is sketched in Figure 1 for fast and slow
consolidation durations and the experienced
p
′
c
is shown to be
rate dependent. In opposition to this, hypothesis A implies a
unique EOP effective stress-strain relationship irrespective of
consolidation duration (Mesri and Choi, 1985b). Hence, the
distinction between the two hypotheses basically comes down
to whether the resulting
p
′
c
is rate dependent or not.
Effective stress,
log
v
Strain (
)
v1
v2
fast1
fast2
slow2
slow1
EOP
NOTE:
slow1
>
fast1
slow2
≈
fast2
decreasing rate,
(increasing time)
Figure 1. Implication of the isotache concept for EOP states of fast and
slow consolidation times under incremental loadings up to EOP states.
In the isotache concept the strain rate is determined by the
current void ratio and effective stress. In water-saturated soil,
change in void ratio can only take place when water is expelled
from the soil. Therefore the strain rate is indirectly controlled by
the global consolidation process. However, a soil element inside
a soil layer (or sample) has neither any direct information of this
global consolidation process nor remaining time until the EOP
consolidation state is reached. However, for hypothesis A to
hold true, the response in all soil elements must be a function of
this remaining time and its advocates argue that
“no sub layer,
including the drainage face, experiences any secondary
compression until the simultaneous completion of primary
consolidation of all sub layers”
(Mesri and Vardhanabhuti
2006). Such assertion, however, violates some basic axioms of
continuum mechanics such as axiom of
material invariance
and
axiom of
objectivity
(see e.g. Eringen (1967)).
4 SOME MISCONCEPTIONS AND CLARIFICATIONS
ON SUBSTANTIATIONS OF CREEP HYPOTHESIS A
Degago et al. (2009) re-evaluated the EOP experiments
conducted on 127 and 508 mm thick specimens by Feng (1991)
and showed that the experiments actually substantiate
hypothesis B. In addition, Degago et al. (2009) used a numerical
tool based on hypothesis B to analyse the raw data of the tests
as they were originally conducted and showed that they are
explainable using this model. Mesri and Feng (2009) questioned
the validity of hypothesis B and attempted to provide an
explanation for their tests. However, a series of misconceptions
are visible in Mesri and Feng (2009) that needs clarifications
and these are given in the following sections by classifying the
apparent misconceptions into laboratory and field studies.
4.1
Laboratory studies
As illustrated in Figure 1, the specific load increment that starts
below initial
p
′
c
and exceeds it is critical. This has been treated
in greater detail in Degago et al. (2011a). Degago et al. (2009)
focused on this increment and showed that the tests conducted
by Feng (1991), on the 127 and 508 mm thick specimens of
Batiscan and St. Hilaire clays, did not have the same EOP state.
To determine if there is any possibility that hypothesis A has
a practical use, evidences for hypothesis A must be found. This
requires first of all giving an objective definition of time at the
end of “primary consolidation”
. The obvious criterion would be
the remaining excess pore pressure. However, this requires a
detailed knowledge on the excess pore pressure. Mesri and Feng
(2009), referring to Mesri et al. (2005), admit that such an exact
criterion does not exist.
Mesri and Feng (2009) claim that Degago et al. used
“micro
-
management
”
to evaluate the EOP criterion adopted in the test
by Feng (1991). However, a clear criterion is exactly what one
needs for answering this fundamental question, especially to
study the validity of hypotheses A where EOP is a key state.
With this regard, it is worthwhile to mention that EOP
definition is not important for hypothesis B where there is a
smooth transition from primary to secondary consolidation
phases. Still, it is important to understand the nature of excess
pore pressure around EOP state where creep starts to dominate
and governs the dissipation of the remaining excess pore
pressure. At this stage, the soil can continue to deform without a
significant change in excess pore pressure. Consequently, the
EOP criteria can easily be misused and there is a potential of
exposing specimens being compared to unsystematic creep
durations. In such cases, comparisons may end up being not
genuine enough to reflect reality. The excess pore pressures for
the 127 and 508 mm thick specimen of Batiscan clay were 0.1
and 0.8 kPa and for St. Hilaires clay they were 1.0 and 2.2 kPa,
respectively. Under these conditions one cannot claim that the
thin and the thick specimen have had the same EOP state.
One fundamental proof that was overlooked in the discussion
of Mesri and Feng (2009) is time considerations aspects. From
the classical consolidation theory, the ratio of the time needed to
achieve the same degree of consolidation between two
specimens is equal to the square of the ratio of the heights.
However, because of the consolidation time being increased by
creep, a thick specimen would need more time than the one
calculated based on the classical consolidation theory concept.
Accordingly, one can compare the time needed to achieve EOP
state for the 127 mm (
t
127
) and 508 mm (
t
508
) thick specimens
studied by Feng (1991). In fact the ratio
t
508
/
t
127
in the actual
tests of Batiscan and St. Hilaire clay were only 7 and 9 instead
of being larger than 16. Therefore the tests do not even qualify
as tests conducted in accordance to hypothesis A where the ratio
t
508
/
t
127
is expected to be equal to 16 (Ladd et al., 1977).
Based on the final excess pore pressure of the 508 mm thick
specimen, Degago et al. (2009) established the time that
corresponds to the same EOP state of the 127 mm thick
specimen. This gave a ratio
t
508
/
t
127
of 19 and 20 for both clays
(>16), and most importantly an EOP strain that increases with
specimen thickness. Figure 2 shows details of the excess pore
pressure and volumetric strain development of St. Hilaire clay,
for the step of interest, for both sample thicknesses. It is seen
that a small variation in excess pore pressure gives significant
difference in the
“primary consolidation”
duration and the
corresponding strains. To achieve same EOP criterion with the
508 mm sample, the 127 mm sample should have been loaded
for 14 days instead of the actual 33 days adopted in the tests.
