Geosynthetics June/July 2020 - 46

GSI NEWS

however, this could very well be due to
the GCL being beneath the GM in the
lab tests. Such a GM/GCL composite
liner was not considered in these three
theoretical papers, only the GM by itself.

Summary and conclusions

FIGURE 4 Comparison of theoretical and
experimental leakage through 0.08 inch (2
mm) hole in the GM of a composite GM/GCL
composite liner system containing a wrinkle

46

hydraulic head was varied by creating a
pressure differential across the composite
liner system. The pressure head differential was modulated in the following
steps: 0, 3, 14, 35, 70 and 98 kPa. This
corresponds to 0, 0.3, 1.4, 3.5, 7.0 and
10 meters of hydraulic head across the
liner system. The normal pressure of the
apparatus was systematically varied. It
was increased per the following schedule:
0, 7, 69, 138, 276 and 552 kPa. It should
be clearly stated that it was incrementally increased through the entire normal
pressure sequence at each hydraulic head
to obtain the five laboratory curves of
Figure 4. Therefore, no bentonite was
lost beneath the defect in advance of collecting a data set over a normal pressure
range for each gradient.
In Figure 4, the five laboratory-generated leakage curves behave well with
the lower normal pressure, producing
the highest leakage and progressively
lower leakage at higher normal pressures.
Additionally, three theoretical leakage
models were used for 0.08-inch (2-mm)
holes; they were Rowe (1998), Giroud
and Bonaparte (1989) and Forcheimer
(1930). All three have significantly higher
flow rates than the laboratory curves;

This article describes a large-scale laboratory device used to measure leakage
through a GM or GM/GCL barrier system in an accurate simulation of the
field. The targeted hydrostatic pressure
applied to the test specimen should
simulate the equivalent of the design
pressure, which will be applied to the
test specimen to determine if the design
value can be sustained. A wide variety of
subgrade conditions are capable of being
assessed. For example, different soil conditions, different subgrade placement
conditions, effect of sharp objects, effect
of a geotextile protection layer, wrinkles,
folds, waves, etc., are all possible options.
Obviously, each of these field situations
can have punctures, holes, tears or cracks
in the GM. Used in this preliminary
study is a GM/GCL composite liner with
a 0.08-inch (2-mm) hole in the underside
of a wrinkled LLDPE GM. The following
was observed:
*	 Experimental leakage is lower with
decreasing hydraulic head.
*	 Swelling bentonite from the GCL may
be filling the hole, limiting leakage.
*	 Leakage is less with increasing normal pressure.
*	 Suspended solid particles from the
overburden sand may be migrating
into GM defects causing a reduction in
leakage rates through any such defects.
	
All three theoretical solutions for calculating leakage through GMs overpredict
the actual leakage rates obtained in our
experiments, although none evaluated
flow through a GM/GCL composite liner.

Geosynthetics | June July 2020

0620GS_p38-cv4.indd 46

5/20/20 10:04 PM
Geosynthetics June/July 2020

Table of Contents for the Digital Edition of Geosynthetics June/July 2020
