Geotechnical News • June 2016
43
GROUNDWATER
History: how it was decided to use a soil-bentonite mixture to
seal the pipes passing through compacted clay liners
Robert P. Chapuis
Abstract
Two compacted clay liners were
built to retain household wastewater.
Just after construction, in the 1980s,
full-scale leakage tests, compulsory
in Quebec, were carried out. The total
leaks were five times higher than
target values. The author’s analysis
suggested either leaks around pipes, or
damage to the upper parts of the liners.
Field inspection confirmed the two
suspicions. The liners were repaired
and passed new full-scale leakage
tests: they still have a good perfor-
mance. The solution was to coat the
pipes with a 30-cm thick soil-benton-
ite mixture, to avoid gaps around the
pipes due to different thermal dilation
factors. This solution was subse-
quently used for hundreds of clay and
soil-bentonite liners after this incident
and its successful repair.
Introduction
Compacted clay liners are frequently
used for waste storage facilities. For
example, in Quebec, Canada, about
1500 cells and lagoons have been
built, mostly between 1980 and 2000,
to contain solid or liquid wastes. Most
liners were made of compacted clays
and soil-bentonite mixtures (Chapuis
2002). The total leakage rate of a soil
liner depends upon many details in the
design, construction and field control
of the liner. Several methods have
been proposed to predict the leakage
rate, using laboratory or in situ test
data (Harrop-Williams 1982, 1985;
Bogardi et al. 1989, 1990; Fenton et
al. 2013), but regrettably without mea-
suring the total leakage rate, a test that
is compulsory in Quebec. These meth-
ods usually do not consider that poor
filter materials and minor construction
defects can considerably influence
the leakage rate (e.g., Chapuis 1990a,
2002; Guyonnet et al. 2003).
There is a great need for better
forecasting the total leakage rate,
especially when local bylaws do
not require a total leakage test, or
when there is not enough water to
fill a lagoon. The test was enforced
in Quebec in the early 1980s (e.g.,
SQAE 1985) in addition to other field
tests (Chapuis 1990a, 1995). It is the
last field control required before final
approval of the liner: it is used to com-
pare prediction and reality. In Quebec,
over 1000 lagoons were tested for total
leakage (liners of surface between 10
4
and 5x10
5
m
2
).
Here, the story of two clay liners, built
in the 1980s and immediately tested, is
presented. The total leakage was five
times higher than the target value. A
simple analysis in the 1980s yielded a
correct diagnosis and led to efficient
repairs. The lagoons passed the leak-
age tests after repairs, and have had a
low leakage rate up to now. The total
leak, when a lagoon is in operation,
can be measured in the hydraulic con-
trol structure of the drainage system
(filter sand, drain pipe, solid pipe)
below the liners.
The predicted outcome was correct,
but unpublished. Lessons learned have
not been passed on. New analysis and
predictive methods for compacted clay
liners were proposed later (Chapuis
1990a; Chapuis et al. 2006; Chapuis
2013), long after this case of poor
performance.
In this paper, basic rules for design
and construction are summarized and
then, the full-scale leakage tests of
the 1980s are presented, including
the analysis which yielded correct
diagnosis and successful repairs. The
reader may be curious as to why such
old data are only now being published.
The case resulted in a dispute, which
ended with a confidential out-of-court
settlement, implying that nothing
could be published. This was regret-
table because valuable technical
information was retained or hidden.
Over the past few years, the author
has requested authorization to publish
scientific analyses of old, but still
interesting, cases. As time heals all
wounds, the authorizations to publish
arrived, with the condition that all
names, dates, and legal issues be kept
confidential, and that no photograph
should permit identification of a site or
person.
Design and construction of
compacted clay liners
For lagoons in Quebec, soil-bentonite
liners are 10–25 cm thick, whereas
clay liners are 45–90 cm thick. Soil-
bentonite liners are constructed in
a single lift, whereas clay liners are
constructed in several lifts to avoid
superposing defects and joints. The
total leakage rate must be lower than
some target value, and the liner must
have adequate mechanical properties
(e.g., Goldman et al. 1990; Benson et
al. 1999; Camp et al. 2009).
