By Richard B.

Erickson,
Designer’s Forum Richard Thiel, P.E.,
and Gregory N. Richardson, P.E., Ph.D.

GCL design series—Part 1:

GCL performance as a fluid barrier
Over the past 18 months, GSE in con- sign issues and GCL installation options, a reinforced configuration (outer geotextiles
junction with industry design and acade- and discusses design principles related to are stitched or needlepunched together). To
mic professionals have developed the in- hydraulic performance and leakage. Part 2 create a composite liner system, these types
dustry’s first comprehensive GCL design focuses on GCL slope stability, followed by of GCLs are overlain by a continuous
guidance document, the GSE GundSeal De- Part 3 which expands on GCL installation geomembrane, as depicted in Figure 1.
sign Manual (Thiel et al. 2001). Although and durability. The geomembrane-supported GCL (GM-
the manufacturer of one particular type of GCL) is comprised of a thin layer of ben-
GCL product sponsored the manual, it fea- Types of GCL compos- tonite mixed with a water-based adhesive
tures design methodologies and procedures that attaches it to a polyethylene geomem-
for evaluating all types of GCLs in a wide ite lining systems brane. This product has been used as a one-
range of composite liner applications. It pre- Conventional composite-liner applications product composite liner in bottom liner and
sents state-of-the-practice design principles in waste-containment industries require a cap applications, effectively replacing both
related to hydraulic performance evalua- geomembrane that overlays a low-permeabil- the geomembrane and compacted clay com-
tion, slope stability analyses, construction ity compacted clay liner. Alternately, GCLs ponents of traditional prescriptive compos-
and durability issues in utilizing GCLs in are now commonly used as an alternative to ite liners. There are two general design con-
bottom liner systems, caps, ponds, and sec- replace all or part of the lining system. figurations for the GM-GCL product:
ondary containment lining applications. Fabric-supported GCLs have a thin layer
An overview of the fundamental design of bentonite (typically a sodium-based mont- Single composite mode
issues presented in the manual is summa- morillonite clay) carried between various In this installation, the bentonite side of
rized in this three-part GFR series, with im- combinations of woven and nonwoven the material is generally installed face down
plications for making a design utilizing needlepunched geotextiles. Products are and the geomembrane side face up, to form
GCLs simpler, quicker and more effective. available in a non-reinforced configuration a one-product composite (geomembrane-
Part 1 presents an overview of the GCL de- (bentonite glued between the fabrics), or in clay) liner. Normally, the overlaps are not
mechanically joined, but are overlapped for
self-sealing, as shown in Figure 2. It is also
Figure 1: Composite liner consisting of a primary possible to weld the geomembrane compo-
geomembrane and a reinforced or unreinforced fabric- nents together utilizing conventional
encased GCL. geomembrane welding techniques, includ-
ing either dual-track hot-wedge welding or
extrusion welding procedures (Figure 3).
Geomembrane
Encapsulated mode
Fabric-encased GCLs
geotextile/bentonite/geotextile In this installation, a supplemental
geomembrane is installed against the ben-
Unreinforced bentonite or needlepunch reinforced bentonite tonite side of the material, as shown in Fig-
ure 4. In this application, the GM-GCL
product is usually installed with the ben-
June/July 2002 • www.gfrmagazine.info

tonite side face up and the geomembrane

Figure 2: Overlapped GM-GCL with shingle seams. side facing against the subgrade, with a sup-
plemental geomembrane installed over the
bentonite surface. This configuration, how-
Geomembrane backing (smooth or textured)
0.4 mm thru 2.0 mm ever, can also be reversed so that the GM-
GCL is deployed on top of a previously de-
B
ployed geomembrane, with the bentonite
side face down. Note that this configura-
t tion can also be used with fabric-encased
GCLs by deploying the GCL between two
Bentonite coating
12 in. (300 mm)*
geomembranes. The encapsulated design
*Overlap length dependent on subgrade condition and anticipated settlement
mode offers the following distinct perfor-
GFR •
mance advantages: tion 1. For hydraulic head ≥ 3 m, the em- [For hw ≥ 3 m, and defect diameter a ≤
• improved fluid containment; pirical equation takes the form of Equation 5x10-4 m2 (25 mm dia.)]
• improved bentonite durability during con- (2) (Thiel et al. 2001): where QGM = rate of leakage through a
struction by preventing pre-hydration of Equation (1) defect (m 3/s), C = a constant related
the bentonite; and QGM = C [1 + 0.1(hw / t)0.95] a 0.1 hw 0.9 ks 0.74 to the quality of the intimate contact
• improved slope stability. [For hw< 3 m, and defect diameter a ≤ between the geomembrane and its under-
5x10-4 m2 (25 mm dia.)] lying clay liner, hw = head of liquid on top
Performance as a of the geomembrane (m), t = thickness
Equation (2) of the soil component of the composite
fluid barrier QGM = C [1 + 0.1(hw / t)0.95] a 0.1 hw 0.9375 ks 0.74 liner (m), a = area of defect in geomem-
The hydraulic performance of GCLs can
be evaluated using standard accepted
leakage models. Leakage through a com-
posite liner may be caused by: Figure 3: Extrusion-welded GM-GCL seams.
• defects in the primary geomembrane
resulting primarily from installation Approx. 3 in. (75 mm)

damage; Bentonite free geomembrane

• coincident defects in the upper and edge for welding

lower geomembranes in encapsulated Geomembrane backing (smooth or textured) Extrusion weld (grind
30 mil (0.75 mm) thru 80 mil (2.0 mm) geomembrane prior
GCL installations (geomembrane-clay- to welding
geomembrane); and
• seepage at overlapped seams when the
geomembrane seams are not welded for
the GM-GCL product. Bentonite coating
The design equations and general ap- *Overlap length dependent on subgrade
proach for evaluating composite liner leak- 6 in. (150 mm) – 12 in. (300 mm)* condition and anticipated settlement

age, and environmental protection com-

parisons between the GM-GCL and an
equivalent geomembrane-compacted clay
composite liner are summarized below.
Figure 4: Encapsulated GM-GCL.
Leakage modes Geomembrane backing (smooth or textured)
30 mil (0.75 mm) thru 80 mil (2.0 mm)
Defects in composite-liner geomembranes Overlying geomembrane

Empirical modeling and field obser-

vations (Giroud and Badu-Tweneboah
1992; Giroud 1997) have resulted in the
“Giroud equation” for estimating leak- Bentonite coating
age through a hole in the geomembrane
portion of a composite liner. The em-
pirical equation takes the form of Equa- 6 in. (150 mm) – 12 in. (300 *Overlap length dependent on sub-
mm)* grade condition and anticipated settle-
ment GFR
•
June/July 2002 • www.gfrmagazine.info

Figure 5: Drawing illustrating the factors taken into account by the “Giroud equation”
for estimating leakage through a hole in the geomembrane part of a composite liner.

hW Defect with area a

Geomembrane

Contact
C-Value
t Clay with ks m/s

QGM
 brane (m2), and ks = hydraulic conduc- coincident defects through an encapsulated tornell and Daniel (1992) and the Cincin-
tivity of the underlying clay liner (m/s) liner system (m2) (see Figure 6). nati U.S. EPA GCL test plot “P” exhumed
(see Figure 5). after 4.5 years of performance.
The basis for Equation (1) is referenced Seepage at overlapped (unwelded) Leakage per unit length due to seepage
in the U.S. EPA Technical Manual (1993), GM-GCL seams along a saturated GM-GCL overlap would
and is incorporated into the latest versions In the case of overlapped GM-GCL be calculated in accordance with Darcy’s
of the HELP computer model (U.S. EPA seams (Figure 2), liquid will seep directly law as follows:
1994) used for predicting landfill leachate into and possibly through the overlaps. Equation (4)
generation and leakage. Therefore, the seepage rate through over- Qolap = ks (hw/B) t
Defects in encapsulated bentonite system lapped GM-GCL seams must be quantified
For an encapsulated design (Figure 4), in a leakage evaluation. where Qolap = flow rate per unit length
the size of the defect in the lower geomem- Due to the weight of its bentonite coat- (m3/s•m), ks = hydraulic conductivity of the
brane would control leakage, and leakage ing, an installed GM-GCL lays flat on the bentonite (m/s), hw = hydraulic head on top
would occur when an event caused coinci- subgrade. This virtually eliminates wrinkles of the liner (m), B = width of overlap (m),
dent defects in the upper and lower and results in excellent contact between and t = thickness of the bentonite (m) (see
geomembranes. In this case, Darcy’s law overlapped panels at their seam areas. For a Figure 7).
controls the advective flow rate through a typical overlap distance of 300 mm, it would To determine leakage due to seepage at
defect of a given size. The leakage equation take more than 5 years before seepage would overlap seams, the total linear length of
would take the following form: begin through the GM-GCL overlap with seam for a given project must be calculated.
Equation (3) a fluid buildup of up to 300 mm. Steady- The general length of overlap seams (S) in
Qenc = ks i a = ks [(hw + t) / t] a state leakage would most likely take several an installation area (A) is:
more years to develop as documented by Equation (5)
where Qenc = leakage (m3/s), ks = hydraulic Dr. David Daniel in Thiel el al. (2001). This S = A (1/L + 1/W)
conductivity of the bentonite, i = hydraulic seam performance is based on data provided where L = average length of panels less over-
gradient [(liquid head hw + t) / t, where t = by the large-scale tank tests reported by Es- lap (typically 51.2 m), W = average width of
bentonite (m/s) thickness], and a = area of panels less overlap (typically 5.0 m). Applying
Equation (5) to a typical GM-
GCL installation thus results in
approximately 2200 m of over-
Figure 6: Drawing illustrating the factors taken into account by lap seam per hectare of lined area.
the equation, derived from Darcy’s law, for estimating leakage The actual length of overlap
through a hole in an encapsulated bentonite system. seam would increase slightly if
the complexity of the installa-
tion increased due to structures,
Coincident defects with area a
hw for example, or irregularities.
Total leakage at overlapped seams
t
is subsequently determined by
multiplying Qolap by the length
of seam S for a given lined area A.
Q enc Geomembranes
Bentonite with ks m/s
Factors affecting
leakage
June/July 2002 • www.gfrmagazine.info

Intimate contact “C-Value”

Figure 7: Drawing illustrating the factors taken into account by The Giroud equation con-
the equation, derived from Darcy’s law, for estimating seepage tains the factor C which ac-
at overlapped (unwelded) GM-GCL seams. counts for the degree of inti-
mate contact between the
geomembrane and adjacent
clay. In Thiel et al. (2001), Dr.
hw J.P. Giroud evaluates the con-
Geomembrane
tact C-factor between the ben-
t
tonite component of a GM-
Bentonite with ks m/s
GCL and an adjacent
Q olap geomembrane by analyzing the
•

B
approaches developed and ex-
GFR

pounded by Rowe (1998),

 Designer’s Forum
Foose et al. (2001), and Harpur et al. load range from 10 kPa to 1000 kPa, re- The hydraulic analysis and design
(1993). Using the results published in those spectively (Thiel et al. 2001). methodology presented above can be
references, Giroud recommends a conser- adapted to all design applications in order
vative value of C = 0.01 for contact be- to evaluate the environmental performance
tween the bentonite component of the Leakage rate compar- and equivalency of a GM-GCL composite
GM-GCL and its geomembrane. The sec- isons liner to conventional geomembrane-com-
ond author uses a value of C = 0.05 for fab- In evaluating hydraulic performance, each pacted clay composite liner systems for a
ric-encased GCLs (NWNP side against a liner system is analyzed by utilizing the pro- project-specific set of design parameters.
geomembrane) to represent “excellent” ject-specific design criteria outlined above
contact conditions. and applying the applicable leakage equa-
Summary
tions. The total potential leakage for the Whether a GCL is used in a bottom liner
Hydraulic conductivity composite liner system is calculated by com- or cover, as a single-composite liner or with
The hydraulic conductivity ks of sodium bining the leakage through the assumed fre- a separate geomembrane, three fundamen-
bentonite in GCLs is affected by the level quency of geomembrane defects with the tal design issues should be considered for
of normal stress applied to the GCL, and leakage at the overlapped seams, if the GCL applications, including: performance as
chemical alterations caused by different geomembrane seams are not welded. The a fluid barrier; slope stability; and installa-
permeating liquids that may increase the methodology for deriving a design leakage tion and durability. Part 1 of this series on
hydraulic conductivity of sodium bentonite. rate for composite liners is presented in GCL design guidance focused on the gen-
Guidance to selecting the appropriate hy- Thiel et al. (2001) with examples demon- eral approach and design equations required
draulic conductivity value(s) for a project- strating various design assumptions and per- for evaluating GCL hydraulic performance
specific GCL application and liquid is pre- formance criteria in each design chapter for based on standard accepted leakage models.
sented in Chapter 2 of Thiel et al. (2001) bottom liners, caps, ponds, and secondary The three potential modes of leakage
as compiled by Dr. David Daniel. containment applications. through a geomembrane-clay composite liner
Figure 8 (p.20) presents an example of and leakage analyses were presented, in-
Project-specific design assumptions leakage rate comparison between the var- cluding (1) leakage through geomembrane
• Liquid head buildup, hw — The buildup ious GM-GCL seam configurations (over- defects, (2) geomembrane defects in encap-
may vary from less than 25 mm for cap ap- lapped seams, welded seams, and encapsu- sulated bentonite (geomembrane-bentonite-
plications, up to 300 mm for regulated al- lated bentonite alternatives) and a geomembrane) liners, and (3) seepage at
lowable buildup above bottom liners, and prescriptive U.S. EPA Subtitle D geomem- overlapped (unwelded) GM-GCL seams.
elevated liquid head for secondary con- brane-compacted clay composite liner for These leakage mechanisms, combined with
tainment leakage events and impound- a typical landfill bottom liner application project-specific design factors (including in-
ment applications. (Erickson and Thiel 2002). For the calcu- timate contact, hydraulic conductivity, liquid
• Defect area, a, and frequency of defects lations in this comparison, the design as- head, frequency and size of geomembrane
per unit area—Industry average standards sumptions were: liquid head hw = 300 mm; defects, and clay liner thickness) provide the
for estimating defects in an installed bentonite thickness tbent = 5 mm; assumed basis for evaluating global project leakage.
geomembrane assume that approximately area of defects a = 0.0001 m2; defects per Finally, a methodology for comparing po-
two to ten 100 mm2 holes per ha exist after hectare n = 10; and overlap distance B = tential lining system leakage rates through a
a geomembrane is deployed and covered 300 mm. Design assumptions for the pre- geomembrane-compacted clay composite
with soil. The number and size of these de- scriptive compacted clay liner included liner vs. a GM-GCL composite liner was pre-
fects can be reduced through more thor- thickness tccl = 600 mm and hydraulic con- sented. This leakage analysis allows the de-
ough CQA procedures, such as the use of ductivity kccl = 1 x 10-9 m/s. sign practitioner to effectively evaluate en-
an electric defect-detection survey after the As shown in Figure 8, the simple-over- vironmental performance and equivalency
overlying soil has been placed. The quality lap design with the one-product composite of a GM-GCL compared to conventional
GFR

of installation and the assumed size and fre- liner will environmentally out-perform a compacted clay-geomembrane composite
•

quency of geomembrane defects should be liners for a given set of design parameters.
June/July 2002 • www.gfrmagazine.info

prescriptive Subtitle D liner (geomembrane

evaluated on a project-specific basis. over 600 mm compacted clay layer) even The subsequent parts of this GCL design
• Clay liner thickness, t —The thickness when its bentonite’s hydraulic conductiv- guidance series will focus on GCL slope sta-
of compacted clay liners is generally given ity is increased to kbent = 1 x 10-9 m/s. The bility (Part 2) and installation and durabil-
by prescriptive requirements. The thickness environmental performance of the encap- ity (Part 3).
of the GCL bentonite layer is based on the sulated GM-GCL design is exceptional,
mass loading of bentonite (standard 3700 with estimated leakage rates between 100 References
g/m2 at 0% moisture) at the design normal and 100,000 times lower than the prescrip- Erickson, R.B., and Thiel, R. 2002.
load. The thickness of the hydrated ben- Design and application of the geomem-
tive geomembrane-compacted clay liner
tonite component of the GM-GCL as a brane supported GCL in one-product
(showing as nearly zero leakage on the
function of effective compressive stress and encapsulated composite liner sys-
graphical scale in Figure 8), depending on
ranges from 8.5 mm to 3 mm for a normal tems. Geosynthetic Clay Barriers In-
the hydraulic conductivity of the bentonite.
 Figure 8: Comparison of hydraulic performance of example composite bottom liner systems
with (a) GM-GCL alternative liners and (b) U.S. EPA Subtitle D composite bottom liners.

Estimate Leakage vs. Hydraulic Conductivity Expected leakage rate for

a
3 GM-compacted clay liner
Overlapped GM-GCL 0 head h w = 3 0 0 m m
GM-GCL welded seams
Encapsulated GM-GCL G M defects n = 1 0 / h a
Leakage Rate (liters per hectare per day)

2
U.S.EPA Subtitle D k ccl = 1 x 1 0 -9 m / s
5

2 b Expected leakage rate for

encapsulated GM-GCL
a 0

c Typical k-value for bentonite

1
exposed to ash leachate
b d 5

d Range of k-values for ben-

1.00E-08 1.00E-09 1.00E-10 1.00E-11 1 tonite under 200 kPa load
Bentonite Hydraulic Conductivity (m/s) 0
c

ternational Symposium. Nuremberg, contact between geosynthetic clay ria. Office of Solid Waste and Emer-
Germany, April. liners and geomembranes in terms of gency Response, U.S. EPA. Report U.S.
transmissivity. Proceedings of the EPA530-R-93-017, November.
Estornell, P., and D.E. Daniel. 1992. 7th Annual GRI Seminar. Drexel Uni-
Hydraulic conductivity of three versity, Philadelphia, PA, United U.S. EPA. 1994. The Hydrologic Eval-
geosynthetic clay liners. ASCE Jour- States. 138–149. uation of Landfill Performance (HELP)
nal of Geotechnical Engineering, vol. Model, User’s Guide and Engineering
118, no. 10: 1592–1606. Koerner, R.M., D.A. Carson, D.E. Documentation for Version 3. Office
Daniel, and R. Bonaparte. 1996. Cur- of Research and Development, United
Foose, G.J., C.H. Benson, and B.E. rent status of the Cincinnati GCL test States Environmental Protection
Tuncer. 2001. Predicting leakage plots. Proceedings of the 10th GRI Agency, Report U.S. EPA/600/R-
through composite landfill liners. Seminar. Drexel University. Philadel- 94/168b.
Journal of Geotechnical and Geoenvi- phia, PA, United States.
ronmental Engineering, June: 510-
–520. Rowe, R.K. 1998. Geosynthetics and
Giroud, J.P., and K. Badu-Tweneboah. the minimization of contaminant mi-
1992. Rate of leakage through a com- gration through barrier systems be-
June/July 2002 • www.gfrmagazine.info

posite liner due to geomembrane de- neath solid waste. Proceedings of the
fects. Geotextiles and Geomembranes, Sixth International Conference on
vol. 11:1–28. Geosynthetics. Industrial Fabrics As-
sociation International, Roseville, MN,
Giroud, J.P. 1997. Equations for cal- United States. 27–102.
culating the rate of liquid migration
through composite liners due to Thiel, R., D. Daniel, R. Erickson, E.
geomembrane defects. Geosynthet- Kavazanjian, and J.P. Giroud. 2001.
ics International, vol. 4, nos. 3-4: The GSE GundSeal GCL Design Man-
335-348. ual. GSE Lining Technology Inc., Hous-
ton, TX, United States.
Harpur, W.A., R.F. Wilson-Fahmy, and
•

R.M. Koerner. 1993. Evaluation of the U.S. EPA. 1993. Technical Manual for
GFR

Solid Waste Disposal Facility Crite-