979
Technical Committee 104 /
Comité technique 104
3.2 Flexible-wall permeability cell with bender elements
A flexible-wall permeability cell was provided with bender
elements (one in the base pedestal and the other on the top cap)
to enable the simultaneous monitoring of
G
0
and
k
of a cement
bentonite sample. Moreover, height changes during permeation
could also be monitored through a cathetometer placed next to
the cell (Fig. 2).
The parameter
k
was evaluated out of a falling-head test
performed in a conditioned room at constant temperature (18°C)
and at an isotropic effective stress of 30 kPa. The sample was
first permeated with deionized water for about 1 month (1.7
pore volumes of flow). Next, the deionized water was replaced
with the 25 g/L solution of Na
2
SO
4
and the test was continued
for a period of about 250 days (about 10 pore volumes of flow).
Figure 2. Flexible-wall hydraulic conductivity cell provided with bender
elements and a cathetometer
4 RESULTS
4.1 Small-strain shear modulus
Figure 3 summarizes all
G
0
measurements carried out in the
benchtop bender element setup and in the modified flexible-wall
hydraulic conductivity cell. As expected, the results of the
benchtop bender element setup, where the CB sample was cured
in pure water, showed a gradual increase of
G
0
in time due to
normal cement hydration. Verastegui-Flores et al. (2010)
showed that the
G
0
increasing trend of clay treated with blast
furnace cement could be fairly-well characterized through a
logarithmic function. Clearly, all measurements up to a sample
age of 90 days were in excellent agreement with such rule.
0
50
100
150
200
250
300
350
400
450
0
50
100
150
200
250
Small-strain shear modulus (MPa)
Age (days)
Benchtop setup measurement
During permeation with water
During permeation with sulphates
Prediction of G
0
cured
in pure water
Figure 3. Impact of sulphate attack on the small-strain shear modulus of
a cement-bentonite sample
Similary,
G
0
measurements in the flexible-wall permeability
cell are in agreement with measurements out of the benchtop
bender element setup during the first phase of the tests, when
the sample was permeated with pure water for about one month.
However, when the permeation with the 25 g/L Na
2
SO
4
solution
started, the normal cement hydration process was clearly
disrupted (Fig. 3).
G
0
measurements up to 250 days of
permeation, when compared to the expected
G
0
trend in contact
with pure water, show that the stiffness of the CB sample was
significantly reduced due to contact with sulphates. Such
reduction is the result of interparticle cementation degradation
and it could also indicate severe fissuring affecting the original
structure of the CB sample. Clearly, a decrease in
G
0
suggests a
decrease of strength as well as both parameters are strongly
linked to interparticle cementation.
Deterioration of the cement hydration products by sulphates
is a well-known durability problem in cement mortars exposed
to high concentrations of sulphate ions. The most common
manifestations of sulphate attack in concrete are expansion,
caused by formation of ettringite and gypsum within the matrix
of a specimen, and loss of strength. A similar phenomenon was
observed in CB samples.
4.2 Hydraulic conductivity
Figure 4 summarizes all
k
measurements as well as sample
height changes during permeation with water (for one month)
followed by permeation with a 25 g/L Na
2
SO
4
solution for a
total period of about 250 days.
1,E-10
1,E-09
1,E-08
1,E-07
1,E-06
0
50
100
150
200
250
Hydraulic conductivity (m/s)
Age (days)
Water
Sulphates
Prediction of
permeability to water
(a)
60
61
62
63
64
65
0
50
100
150
200
250
Sample height (mm)
Age (days)
(b)
Figure 4. Impact of sulphate attack on the (a) hydraulic conductivity and
(b) the height of a cement-bentonite sample during permeation with
water and a 25 g/L Na
2
SO
4
solution in a flexible wall cell.