Citation
In-plane flow of geosynthetics for landfill drainage

Material Information

Title:
In-plane flow of geosynthetics for landfill drainage
Creator:
Campbell, Robert Paul
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
viii, 137 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Fills (Earthwork) ( lcsh )
Drainage ( lcsh )
Drainage ( fast )
Fills (Earthwork) ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (M.S.)--University of Colorado at Denver, 1992. Civil engineering
Bibliography:
Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Science, Department of Civil Engineering.
Statement of Responsibility:
by Robert Paul Campbell.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
26909569 ( OCLC )
ocm26909569

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

IN-PLANE FLOW OF GEOSYNTHETICS FOR LANDFILL DRAINAGE by Robert Paul Campbell B.S.W.E., Ohio State University, 1977 M.B.A., University of Colorado at Denver, 1985 A thesis submitted to the Faculty of the Graduate School of the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Master of Science Civil Engineering 1992

PAGE 2

This thesis for the Master of Science degree by Robert Paul Campbell has been approved for the Department of Civil Engineering by ,t7,tfjq2 Date

PAGE 3

Campbell, Robert Paul (M.S., Civil Engineering) In-Plane Flow of Geosynthetics for Landfill Drainage Thesis directed by Associate Professor Jonathan T.H. Wu ABSTRACT The in-plane flow of geonets and geocomposites for landfill drainage applications was investigated. Laboratory testing was performed to evaluate the factors affecting the flow rate. The factors investigated were geosynthetic geometry and compressive strength, hydraulic gradient, applied normal stress, flow direction, boundary condition and time. Three geonets and one geocomposite were tested. The boundary conditions used were a geomembrane, geotextile, sand, clay, two bentonite synthetic clay liners and foam rubber. Testing was performed at hydraulic gradients from 0. 03 to 1. 0, with normal compressive stresses up to 19,500 psf. The relationship between the flow rate and the hydraulic gradient was nonlinear, which indicates the flow was probably turbulent for all tests. The flow rate increased, and the transmissivity decreased as the hydraulic gradient increased. The flow rate iii

PAGE 4

increased as the geosynthetic thickness increased. The flow rate for the geonet in the roll direction was approximately double that in the cross-roll direction. The flow rate decreased as the normal stress increased, with the decrease being smaller for geonets with a higher compressive strength. The highest flow rate was obtained with the geonet or geocomposite between two geomembranes. The flow rate was lower for the other boundary conditions due to intrusion of boundary layer into the flow path. The type of soil cover used as a boundary did not have a large effect on the flow rate. Foam rubber is suitable replacement for soil as a cover material during flow testing. The flow rate decreased with time due to creep of the geonet or geocomposite core, and additional intrusion of the boundary layer. The decrease in flow rate with time was larger at higher normal stresses. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. Signed iv

PAGE 5

ACKNOWLEDGEMENTS The University of Colorado at Denver for funding this study. Fluid Systems, Inc, National Seal Co., Exxon Chemical Co., James Clem Corp. and Gundle Lining Systems, Inc. for providing materials and technical assistance. Steve Turner for the statistical analysis, Karen Mighell for compressive testing and Jonathan Wu for guidance and assistance throughout the project. v

PAGE 6

CONTENTS Chapter 1. 2. Introduction 1 1.1 Problem Statement 1 1.2 Research Objectives 2 1.3 Method of Research 3 Background 5 2.1 Geosynthetic Drainage Systems 5 2.1.1 Overview of Geosynthetics 5 2.1.2 Geosynthetics for Drainage 6 2.1.3 Landfill Drainage Applications 7 2.2 Flow Properties of Geosynthetics 15 2.3 2.2.1 Theory . 15 2. 2. 2 Testing Methods . 18 Factors Affecting Flow 2 0 2.3.1 Geosynthetic Type and Geometry 20 2.3.2 Geosynthetic Compressive Strength and Applied Normal Stress . . 21 2.3.3 Gradient and Fluid Properties .. 22 2.3.4 Boundary Conditions . 24 2.3.5 Long-Term Effects 27 2.4 Design Considerations 3 0 vi

PAGE 7

3. Testing 3 4 3.1 Test Apparatus . 34 3.2 Test Materials . 37 3.2.1 Geosynthetics 3 7 3.2.2 Soils 39 3.3 Test Program 43 Compressive Testing . 43 3.3.2 Flow Testing 4 4 4. Results and Discussion .. 50 5. 4.1 Compressive Testing 50 4.1.2 Short-Term Tests 50 4 .1. 2 Long-Term Tests . 52 4.2 Flow Testing . . . 54 4.2.1 Short-Term Testing of Geonets .. 54 4.2.2 Long-Term Testing of Geonets .. 78 4.2.3 Short-Term Testing of Geocomposite ...... 87 4.2.4. Long-Term Testing of Geocomposite ... 4.3 Design Implications Summary and Conclusions 5.1 Summary 5.2 Conclusions 9 0 93 97 97 99 5.3 Recommendations for Further Study 104 vii

PAGE 8

Appendixes A. Test Apparatus Design Drawings 106 B. Comparison with Other Laboratory Results .. 109 c. Short Term Geonet Flow Efficiencies 114 D. Long Term Geonet Flow Efficiencies 120 E. Short Term Geocomposite Flow Efficiencies . 126 F. Long Term Geocomposite Flow Efficiencies .. 130 Bibliography 132 viii

PAGE 9

CHAPTER 1 INTRODUCTION 1.1 Problem statement The disposal of hazardous and solid waste has become a major concern. In order to protect the public from unacceptable exposure, the United States Environmental Protection Agency (U.S. EPA) has promulgated regulations controlling the design and operation of land disposal facilities including landfills. The regulations require liner systems to prevent the migration of harmful chemicals into the environment. The liner is a low permeability layer of clay or a geomembrane. The liquid leachate that collects above the liner must be removed to prevent large hydraulic gradients. The landfill must also have an impervious cover to prevent the infiltration of surface water which also includes a drainage layer. In the past high permeability soils such as sand and fine gravel were used for the drainage layers. Geosynthetics are increasingly used because of performance, constructability and cost advantages. 1

PAGE 10

However, since geosynthetic are relatively new, long term performance data are not available. Also, new geosynthetic products have been developed that may be suitable for landfill drainage applications. Many of the conditions that exist in the field have not been adequately evaluated by laboratory testing. For many products only index type properties are available that do not represent the actual design conditions. 1.2 Research Objectives The objective of this study was to evaluate the some of the factors that affect the in-plane flow of geosynthetics used for landfill drainage. The factors selected for investigation included those that were not evaluated, or were tested under limited conditions in previous research. An additional objective was to confirm the results of research performed by other laboratories, and provide a background for additional research at the University of Colorado at Denver. The objective of the study was not to perform comparison testing of the geosynthetics available from different manufacturers. The geosynthetics tested were selected as being representative of the typical products available for landfill drainage applications. 2

PAGE 11

1.3 Method of Research A laboratory test apparatus was used to measure the in-plane flow of geonets and geocomposites. The apparatus allowed testing with hydraulic gradients, normal compressive stresses, and boundary conditions such as clay that are typically encountered in actual applications. Three geonets and one geocomposite were used for flow testing. The factors investigated were the geosynthetic geometry and compressive strength, hydraulic gradient, normal compressive stress, flow direction, boundary conditions and time. Short and long-term compressive testing was performed to evaluate the stress-strain behavior of the geosynthetics, and the relationship with the flow rate. The maximum long-term flow test was 500 hours with most of the tests lasting 48 to 168 hours. Longer times would be advantageous, but were not permitted by the time schedule. Several replicate tests were performed to evaluate the repeatability of the test apparatus. The results were also compared with those of the manufacturer and another research facility. 3

PAGE 12

Background information regarding regulatory requirements, design configurations, basic flow theory, factors affecting flow and design implications based on previous research is presented in Chapter 2. A description of the test apparatus, test materials and the test program is presented in Chapter 3. The test results and discussion are presented in Chapter 4. Chapter 5 contains_the summary, conclusions and recommendations for further study. The appendixes contain the tabulated results for the individual tests. 4

PAGE 13

CHAPTER 2 BACKGROUND 2.1 Geosynthetic Drainage Systems 2.1.1 overview of Geosynthetics Geosynthetics are man-made materials used for geotechnical applications. Geosynthetics are manufactured from polymeric materials. Some of the commonly used polymers are polyethylene, polyester, polypropylene, polystyrene and polyvinyl chloride. The major types of geosynthetics are: Geotextile -fabric made from fibers. Geogrid -solid longitudinal and transverse ribs. Geonet -net like structure of polymer ribs. Geomembrane -impermeable sheet of polymeric material. Geocomposite -combination of different geosynthetic materials. Commonly a core structure with a geotextile attached. The primary applications and the type of geosynthetic commonly used are: 5

PAGE 14

Reinforcement -geotextile or geogrid Separation -geotextile Filtration -geotextile Drainage -geotextile, geonet or geocomposite Barrier -geomembrane This study is concerned with drainage applications. However, the other applications are related since they are used in conjunction with drainage products. Additional background information on geosynthetics is contained in Koerner (1990) and IFAI (1990). 2.1.2 Geosynthetics for Drainage The geosynthetics commonly used applications geocomposites. are geotextiles, Geotextiles used for drainage geonets and for drainage applications are usually manufactured from nonwoven fibers. The drainage properties of geotextiles were not evaluated in this study, as the flow rates are not sufficient for the applications considered. However, geotextiles were used for separation applications. Geonets are polymer ribs extruded into a net-like structure. The ribs are layered at a constant angle to allow passage of liquid. A geonet with an attached 6

PAGE 15

geotextile is shown in Figure 2.1. Geocomposites are typically a polymer core structure with an attached geotextile to prevent clogging of the core by soil. The core structure can be designed to allow flow on either one or both sides. A geocomposite is shown in Figure 2.2. In this study, a geonet with an attached geotextile is considered a geonet, geocomposite. 2.1.3 Landfill Drainage Applications and not a One important application of geosynthetics is drainage systems for landfills. The following drainage systems are required for hazardous waste landfills: Primary Leachate Collection and Removal (PLCR) Secondary Leachate Collection and Removal (SLCR) or Leak Detection Surface Water Collection and Removal (SWCR) Leachate is water that infiltrates or is generated by the waste, and migrates to the bottom of the landfill. The PLCR system is located directly below the waste and above the primary liner. The SLCR system is located between the primary and secondary 7

PAGE 16

Figure 2.1 Geonet with Attached Geotextile 8

PAGE 17

Figure 2.2 Geocornposite 9

PAGE 18

liner. The SLCR collects any leachate that leaks through the primary liner. Figure 2.3 shows typical geosynthetic drainage system configurations. A double composite liner used for hazardous waste landfills is shown in Figure 2.4. A composite liner consists of a geomembrane with a clay liner underneath. New regulations for solid waste landfills (Federal Register 1991) require a single composite liner and leachate collection system. New regulations (Federal Register 1992) for hazardous waste require a single top liner and a composite bottom liner for hazardous waste landfills. The guidelines for the PLCR system are (U.S. EPA 1985): Maximum leachate depth of 30 em (1 foot) Granular drainage layer, minimum 30 em thick with minimum hydraulic conductivity of cmjsec or an equivalent geosynthetic drainage layer Minimum slope of 2 percent The transmissivity of a 30 em thick drainage layer with a hydraulic conductivity of cmjsec is 3 x m2jsec (0.15 galjmin-ft). 10

PAGE 19

Primary Leachate Collection DRAINAGE AND/OR -..-.-:-:--:--_ ----;- r F>ROTECTIVE SOil L.Al'ER .. . rGEOTEXTilE ---. :._ -/ H l 'iRmo&i&iiZbcQ _-GEONET '-GEOMEMBRANE SElE:T REuSE DR .. , _. ... --,.,rF>ROTECTIVE SOill.lYEf\ '. ' ;__:...._! -GEOTEXTIL.E . .. . . _/ Secondary Leachate Collection : . . . . ... -... .:-.: .. .. .... ................ -... JGEOMEMIIRAHE --GEONET '\_ GEOMDI .. AHE -GEOMEMBRANE I III!IQ!!ilildi!lllilt@/)1/llllll/IIUIII t::::::: ::TING --GEDNET Figure 2.3 Typical Geosynthetic Landfill Drainage Configurations (Lundell and Henoff 1989) 11

PAGE 20

LANDFILL LINER SYSTEM Individual components / Cover Protector (gaotextile or other) Barrier (geomembrane) Major system segmems Figure 2.4 Liner for Hazardous Waste Landfill (Daniel and Koerner 1991) 12

PAGE 21

The requirements for the SLCR System are (Federal Register 1992): Granular drainage layer, minimum 1 foot thick with minimum hydraulic conductivity of cmfsec Or geosynthetic with minimum transmissivity of 3 x 105 m2fsec Minimum slope of 1 percent The SWCR system is placed on top of the completed landfill above the cover liner. The purpose is to prevent surface water from infiltrating the landfill. Figure 2.5 shows a typical SWCR system. The design guidelines for hazardous waste SWCR systems are (U.S. EPA 1989a): Granular drainage layer, minimum 1 foot thick with minimum hydraulic conductivity of cmfsec Or geosynthetic with minimum transmissivity of 3 x 105 m2fsec Minimum slope of 3 percent The advantage of a geosynthetic over granular drainage layers is less thickness resulting in additional waste capacity. Also, there is less chance 13

PAGE 22

<: Figure 2.5 Landfill Cover with Surface Water Collection and Removal System (U.S. EPA 1989b) 14

PAGE 23

for damaging the underlying geomembrane compared to a granular drain. Often a granular soil drainage layer is used for the bottom, and a geonet for the sides of the PLCR system. A geonet is used for the SLCR system. Both geonets and geocomposites are used for the SWCR system. 2.2 Flow Properties of Geosynthetics 2.2.1 Theory There are two directions in which flow can occur in a geosynthetic; cross-plane and in-plane. Only the in-plane flow was evaluated in this study., A diagram of in-plane and cross plane flow is shown in Figure 2. 6. Cross-plane flow occurs normal to the geosynthetic, and is of concern for filtration applicatians. The geosynthetic must permit the flow of water, yet prevent the migration of soil into the drainage layer. Permittivity is used to describe cross-plane permeability of a geosynthetic. Permittivity is defined as: where = k/t k = coefficient of permeability t = thickness of geosynthetic 15

PAGE 24

_j_ T uqn Variables for calculating crossplane flow rates (permittivity). j_t T Variables for calculating inplane flow rates (transmissivity). Figure 2.6 Illustration of Cross and In-Plane Flow for Geosynthetics (U.S. EPA 1989b) 16

PAGE 25

The in-plane flow occurs parallel to the geosynthetic surface and is the critical factor for drainage applications. Transmissivity is used to describe in-plane permeability. Transmissivity (8) is defined as: e = kt Using Darcy's law the flow rate can be calculated as follows: q = = kiA = kiWt e = kt = qjiW where q = flow rate k = hydraulic conductivity i = hydraulic gradient w width L = length Often the flow is expressed as flow rate per unit width as follows: qfW = 8i When i = 1, q/W = 8 When the flow is laminar Darcy's law is valid, and the transmissivity is constant at different hydraulic gradients. The transmissivity is a function of the geosynthetic, fluid and boundary conditions. 17

PAGE 26

For laminar flow, the flow rate as a function of hydraulic gradient is linear. Ling, Tatsuoka and Wu {1990) and Raumann {1982) found the flow to be laminar for geotextiles. The flow through geonets was found to be turbulent by Cancelli, Cazzuffi and Rimoldi (1987) using dye to measure the velocity. Williams, Giroud and Bonaparte {1984) found the flow in geonets to be laminar when the transmissivity was less than 2 x 104 m2jsec (0.1 galjmin-ft). For turbulent flow, the results are normally reported as flow rate per unit width instead of transmissivity. 2.2.2 Testing Methods Laboratory measurement of transmissivity is performed by placing the geosynthetic in a test device, applying the desired normal compressive stress, and measuring the flow rate at different hydraulic gradients. The test device can measure radial flow or parallel flow. The radial devices allow water to enter the circular sample at the center and exit at the outer edge. Parallel devices provide a longitudinal flow path parallel to the sides of the sample. Radial flow devices have been used for low 18

PAGE 27

transmissivity geosynthetics such as geotextiles. Parallel flow devices are normally used to measure the transmissivity of geonets and geocomposites. A constant head or falling head test can be performed, with a constant head being more widely used. Koerner and Bove ( 1983) reviewed different testing devices that have been used. Modified triaxial cells were used by Leclercq et al. (1986); and Ling, Tatsuoka and Wu (1990) to measure transmissivity of geotextiles. ASTM Standard D 4 716 ( 19 8 7) "Standard Test Method for Constant Head Hydraulic Transmissivity (Inplane Flow) of Geotextiles and Geotextile Related Products" was developed for measuring the in-plane flow of geosynthetics. The test uses a constant head parallel flow device. The standard allows index testing with the geosynthetic between solid platens or performance tests with actual field boundary conditions of confining pressure and soil intrusion. ASTM D 4716 was developed to standardize the procedure used for testing. However, due to differences in test device design and test boundary conditions, different results can be obtained with the same geosynthetic. The minimum sample size is 4 in. with 24 in. being the 19

PAGE 28

largest size commonly used. However, the University of Illinois (Dempsey 1989) has a flow channel that can test specimens 24 ft. long by 1.5 ft. wide. Empirical relationships have been developed to estimate flow rates for geotextiles. Raumann (1982), Ionescu and Kellner (1982), and Gourc et al. (1982) developed theoretical values of transmissivity that agreed with the experimental results. However, only solid platen boundary conditions were evaluated. Ling et al. (1990) found the in-plane flow to be related to the geotextile void ratio for boundary conditions of rigid plates, a rubber membrane and soil. The theoretical behavior of geonets and geocomposites is more difficult to model, and flow testing is required. 2.3 Factors Affecting Flow 2.3.1 Geosynthetic Type and Geometry The most important factor affecting in-plane flow is the type of geosynthetic and the geometry. In general, geotextiles have the lowest transmissivity while geocomposites have the highest. The thickness of a geotextile is typically 0.125 in. maximum, while geocomposites range up to 1.5 in. In most cases, 20

PAGE 29

increased thickness results in a larger flow rate. Typical flow rates for geosynthetics are shown in Table 2 .1. Table 2.1 Approximate Values of In-Plane Flow Type of Geosynthetic Geotextile Geonet Geocomposite 12" soil (k=0.01 cmfs) Transmissivity Cgal. /min.-ft.) 0.0001 to 0.1 1 to 10 1 to 70 0.15 Some geosynthetics exhibit preferential drainage directions. Geotextiles and most geocomposites have isotropic flow rates. Geonets are anisotropic and have directional flow properties. The angle between the ribs in the roll or longitudinal direction ranges from 50 to 75. The flow in the cross roll or transverse direction should be smaller since the angle of the flow channels to the flow direction is larger. ASTM D 4 716 specifies testing of both machine and cross-machine direction samples. The effect of anisotropic flow is complicated by the use of a small laboratory test device where some of the flow channels intersect the sides of the device. 21

PAGE 30

2.3.2 Geosynthetic Compressive Strength and Applied Normal Stress In most drainage applications a normal stress is applied to the geosynthetic by a vertical or horizontal overburden pressure. The normal stress compresses the thickness of the geosynthetic resulting in reduced flow. The compressive strength of the geosynthetic depends upon the geometry and type of polymer material. Geonets are manufactured from medium or high density polyethylene. The geonet can be manufactured from extruded solid ribs or extruded foamed ribs. Since the ribs of a foamed geonet are filled with gas, the compressive strength is lower. Geocomposites are typically manufactured from polyester, polyvinyl chloride or polyethylene. The compressive strength also depends upon the geometry of the core. The core is usually dimpled or cuspated. In the past geocomposi tes did not have sufficient compressive strength to be used for landfill leachate collection systems. However, current geocomposites have crush strengths up to 20,000 psf. 22

PAGE 31

2.3.3 Gradient and Fluid Properties For both laminar and turbulent conditions the flow increases with hydraulic gradient. The gradient for the bottom of a landfill with a 2 percent slope is approximately 0.035. The gradient for a sidewall with a 1:2 slope is approximately 0.45. Most testing is performed in the range of 0.03 to 2.0. The fluid viscosity also affects the flow rate. Since water is normally used for the testing, the viscosity is only a function of temperature. For laminar conditions, the flow rate is directly related to the viscosity. ASTM D 4716 requires that the transmissivity be corrected to a water temperature of 20C (68F). The correction is made as follows: 020 Oc = (Jkst temp ( Pkst temp/ P2oOd where v = kinematic viscosity For turbulent conditions, the flow rate is not directly related to the viscosity, and the above correction does not theoretically apply. The exact relationship between viscosity and flow for turbulent flow is unknown. Most studies have corrected the flow rate using the above correction factor, when using water at temperatures other than 2 0C. The water 23

PAGE 32

temperature also affects the mechanical properties of the geosynthetic. The amount of creep that occurs increases with temperature. However, the variation in properties is minor within the temperature range normally used for flow testing. 2.3.4 Boundary conditions The boundary conditions of the drainage system affect the flow properties of the geosynthetic. The common boundary conditions for landfill applications were shown in Figures 2.3 and 2.4. The highest flow rate is obtained between two rigid geomembranes. Williams, Giroud and Bonaparte (1984) found no reduction in flow for a geonet with a reinforced hypalon or high density polyethylene geomembrane. A 50% flow reduction was found with a 0.020 in. flexible PVC geomembrane at 10,000 pounds/ft2 (psf). A geonet or geocomposite core with a geotextile results in a reduced flow rate due to intrusion of the geotextile into the flow channels. The largest reduction is observed when the geotextile is backed with soil. The soil can cause the geotextile to 24

PAGE 33

deform and intrude geosynthetic. further into the drainage The amount of intrusion is dependent upon the geotextile type, thickness and strength. Hwu, Koerner and Sprague (1990) studied the effect of geotextile type and a stiffening treatment on the geonet flow. A 5 percent increase in flow efficiency was found compared to the untreated geotextile. Thick, nonwoven geotextiles result in the least flow. Gallup, Bachus and Luettich (1991) found the resin content of the geotextile to have a large effect on the in-plane flow of a geocomposi te. Increasing the resin from 0 percent to 5 percent resulted in a fivefold increase in flow. In some cases, the adjacent geotextile can have a larger effect than the properties of the drainage geosynthetic. Williams, Giroud and Bonaparte (1984) found the transmissivity of a geonet to decrease by a factor of five for a geotextile and clay on one side, and a factor of 100 to 1000 with a geotextile and clay on both sides at 10, 000 psf. Bonaparte, Williams and Giroud (1985) found the flow with clay on one side to be 10 to 20 percent of the flow between two 25

PAGE 34

geomembranes at 20,000 psf. Hwu, Koerner and Sprague {1990) found the flow with sand on one side to be as low as 26 percent, and kaolinite clay 12 percent of the flow between solid platens at 15,000 psf. Fine particles such as clay can also extrude through the geotextile and clog the drainage path. Extrusion is related to the filtering capacity of the geotextile, which is related to the apparent opening size {AOS). Heavy non-woven fabrics are recommended to prevent extrusion. When testing with a dry bentonite clay; Hwu, Koerner and Sprague {1990) found clay particles penetrating the geotextile and blocking the geonet over time. No problems with extrusion were reported by Shaner and Menoff {1991) when testing the effect of bentonite composites on geonet drainage. Estornell {1991) found no extrusion of clay particles through a 10.5 ozjyd2 nonwoven geotextile into the underlying geonet when testing the permeability of synthetic bentonite mats. The tests were run for 100 days with an overburden stress of 158 psf. GeoServices {Estornell 1991) performed tests on the extrusion of granular bentonite through various geotextiles at 15,000 psf for 24 and 168 hours. No 26

PAGE 35

visible bentonite extrusion was observed for nonwoven geotextiles 5.8 ozjyd2 or heavier. The soils that come in contact with the drainage geosynthetic in a landfill include sand, gravel, protective fill, compacted clay and geosynthetic clay liners. ASTM D 4716 permits the use of closed cell foam for testing where soil is anticipated. The foam eliminates problems with soil compaction, placement, and leakage. However, ASTM D 4716 recommends testing with on-site soils when the end use of the geosynthetic is known. 2.3.5 Long-Term Effects Factors affecting the long-term drainage behavior of geosynthetics are creep, chemical degradation and clogging. Creep is probably the most significant effect. Sustained loading causes deformation of the drainage geosynthetic, while creep of the adjacent geotextile and soil may result in greater intrusion. Both factors result in a reduced flow area. Hwu, Koerner and Sprague (1990) did not find a significant drop in flow rate for a geonet with kaolinite on one side, when tested for 1000 hours at 27

PAGE 36

10, ooo psf. Slocumb, Demeny and Christopher (1986) compared the flow rate for a geonet with creep data for polyethylene. The decrease in flow over 100 days at 10,000 psf days was much less than predicted by compressive strength only. The additional reduction was attributed to initial compression and strand layover. Extrapolation of the flow data with the model predicted zero flow in 3 1/2 years. Correcting for the initial compression and strand layover, the flow would be 24 percent of the initial value after 30 years. Koerner et al. (1986) performed long term creep tests on geocomposite cores. While strain data showed deformation of the core, testing for 300 hours at 1500 psf did not show a significant reduction in flow. Mirafi Inc. (1986) did not find a significant drop in flow for a geocomposite tested for 300 days at 4320 psf. Cancelli, Cazzuffi and Rimoldi (1987) performed creep tests on a geonet and two geocomposites. The strain verses time curves showed a definite secondary creep behavior with the geonet having the lowest rate. Smith and Kraemer (1987) found geocomposites to have volumetric strains of up to 60 percent when tested for 28

PAGE 37

250 hours at 4320 psf. However, the total strain included both primary and secondary creep. The flow at 167 hours at 3600 psf was 25% of the original. Additional testing is required to provide the data necessary to extrapolate the creep effects for a 100 year design life and determine the required laboratory test duration. Due to the effects of creep and intrusion, the U.S. EPA (1989b) recommends that transmissivity tests be run for a minimum of 100 hours at 2 to 3 times the overburden pressure. The actual field cross section should be simulated. Landfill leachate can contain chemical constituents that degrade geosynthetics. Geonets are manufactured from polyethylene which is resistant to most chemicals. Polyethylene is used for geomembranes, so extensive testing has been performed. Polystyrene is commonly used for geocomposite cores. Polystyrene would not be acceptable for most leachate collection systems, but could be used for surface water collection systems. Polyvinyl chloride or polyethylene geocomposites could be used for leachate collection systems. U.S. EPA (1986) Method 9090 describes the testing requirements for compatibility 29

PAGE 38

testing of geosynthetics. U.S. EPA (1989b) recommends the drainage geosynthetic be aged in leachate, but the actual transmissivity testing be performed with water. The leachate collection system can be clogged by dissolved and suspended solids in the leachate, or biological deposits. While the polymers themselves are not attacked, microorganisms, fungi and other growth can attach to the geosynthetic and cause clogging. A multi-million dollar pool fund study on the durability of geosynthetics is currently underway. The study is investigating the long-term behavior of geosynthetics due to construction damage during installation, creep under load, and chemical aging. 2.4 Design considerations The drainage geosynthetic must be designed to handle the quantity of leachate or surface water generated, in addition to the regulatory guidelines. The PLCR system must be designed such that the maximum leachate depth over the liner does not exceed one foot. Several computer programs are available to aid in determining the required flow capacity. The Hydrologic Evaluation of Landfill Performance (HELP) 30

PAGE 39

(U.S. EPA 1984} is a model for water movement across, into, through and out of landfills. The program can be used to determine the drainage requirements of the leachate collection and cover systems. The Leachate Collection System (LC System} (U.S. EPA 1990a} program uses expert system techniques to evaluate the leachate collection system design. The Final Cover Evaluation Advisor (F-Cover} (U.S. EPA 1990b} program can provide assistance in evaluating the final cover system design. The SLCR System must meet the action leakage rate specified in the permit for the landfill. The action leakage rate is the maximum design flow rate that can be removed without the fluid head on the bottom liner exceeding one foot. The EPA (Federal Register 1992} recommends the following formula for calculating the flow originating through a hole in geomembrane liner. Q = k*h*tana*B Where Q = flow rate in SLCR system h = head on bottom liner k = hydraulic conductivity a = slope of SLCR system B = width of flow 31

PAGE 40

The action leakage rate using the minimum design requirements for a 100 ft flow width with a factor of safety of 2 is about 100 gal/acre-day. The geosynthetic must have sufficient compressive strength to provide the required flow at the design normal compressive stress. The unit weight of hazardous waste can range from 30 to 110 pcf, while municipal waste ranged from 18 to 67 pcf (Oweis 1990). With a cell depth of 100 ft, the normal stress can reach 11,000 psf. A factor of safety is then applied to the flow rate, the normal stress, or both to give the required flow rate at the design normal stress. Design examples are provided by Koerner (1990}, and u.s. EPA (1987}. If the flow capacity of a single geonet is not adequate, two geonets can be used. Koerner and Hwu (1989} found a double geonet to give flow efficiencies of 200 to 460 percent over a single geonet when a geotextile and soil were on one side. One factor seldom addressed is the effect of seams on the flow rate. In many cases the drainage system can be designed without seams perpendicular to the flow direction to the collection piping. Zagorski and Wayne (1990} found only a slight decrease in the 32

PAGE 41

flow of geonets when butt jointed with no gap. The decrease for overlapped joints ranged from none to 50 percent, depending upon the test conditions. Plastic ties are normally used to join the geonets, with the spacing dependent upon strength considerations. Geocomposites can be joined with a male to female overlap connection of the core dimples. The reduction in flow capacity should be minor. Hwu and Koerner (1990) investigated the effect of overlap on the seam tensile strength. 33

PAGE 42

3.1 Test Apparatus CHAPTER 3 TESTING A constant head parallel flow testing apparatus was designed and fabricated for this study. The apparatus was constructed of clear acrylic plastic to allow observation of the test specimen. The test specimen size is 9 in. by 9 in. The maximum hydraulic gradient is 2.0. A pneumatic consolidation load frame is used to apply the normal stress. The maximum normal stress is 19,500 psf. A schematic diagram of the apparatus is shown in Figure 3.1. A photograph is shown in Figure 3.2. Appendix A. Design drawings are shown in The test apparatus was designed in conformance with ASTM D 4716 except for the sample aspect ratio. ASTM D 4716 specifies that the specimen length be twice the specimen width for widths up to 12 in. A square sample was used to reduce the edge effects and permit more uniform loading. Previous studies have used square samples (Hwu, Koerner and Sprague 1990; 34

PAGE 43

PLATEN PLATEN SPECIMEN DETAIL NORMAL STRESS H I PLATEN I OVERFLOW TEST SECIMEN
PAGE 44

( ( Figure 3.2 Photograph of Test Apparatus 36

PAGE 45

Williams, Giroud and Bonaparte 1984; Cancelli, Cazzuffi and Rimoldi 1987). Koerner et al. (1986) used a sample that was shorter than the width. 3.2 Test Materials 3.2.1 Geosynthetics A solid geonet, two foamed geonets, and a dimpled core geocomposite were used for the testing. The manufacturer's specifications for the three geonets are shown in Table 3.1 and the geocomposite in Table 3 2 The products were selected based on a typical representation of available drainage products. Geonet A is typical of the product supplied by several manufacturers. Geonet B which is intended for cover applications, has a similar thickness and flow capacity, but is foamed. Geonet C is thicker and is foamed, but to a lesser extent than geonet B. The geocomposite has a polystyrene center core with high density polyethylene laminated on both sides to provide improved chemical resistance. 37

PAGE 46

Table 3.1 Geonet Material Specification (from manufacturer) Geonet A Geonet B Geonet C Manufacturer Fluid Systems,Inc.jNational Seal Co. Product PN-3000 PN-3000-CN PN-4000 Core Thickness (in) Polymer Manufacturing Density (gjcm3 ) Geotextile Manufacturer Product Thickness (in) Weight (ozjyd2 ) Polymer Grab Strength(lb) AOS (US sieve) Manufacturing TN-3001/1125 with GT 0.220 0.220 Polyethylene Solid Foamed 0.937 0.937 Trevira 1125 0.110 7.5 Polyester 300 70 None Nonwoven Needlepunched 0.300 Foamed 0.937 None Table 3.2 Geocomposite Material Specification (from manufacturer) Manufacturer Exxon Chemical Co. Model Battle Drain Plus Core Thickness (in) Polymer Yield Point (psf) Geotextile Manufacturer Product Thickness (in) Weight (ozjyd2 ) Polymer Grab Strength (lbs) AOS (sieve) Manufacturing 38 0.380 HDPE/PolystyrenejHDPE 15,000 min. Exxon Chemical Co. GTF140EX 0.060 4.5 Polypropylene 120 80 Nonwoven

PAGE 47

3.2.2 Soils The effect of different boundary conditions on the flow rate of the geosynthetics was evaluated with sand, clay, and two geosynthetic clay liners. The properties of the sand are given in Table 3.3. The grain size distribution is shown in Figure 3.3. The sand was placed on the geonetfgeotextile or geocomposite and tampered into a 1 in. thick layer of approximately 70 percent relative density. Table 3.3 Properties of Sand (Wu 1991) Specific Gravity Maximum Dry Unit Weight (pcf) Minimum Dry Unit Weight (pcf) 2.65 112.19 97.52 The clay was an sand-clay mixture with 5% sodium bentonite added. The clay is representative of an admixed clay liner material. The properties of the clay are given in Table 3.4. The grain size distribution for the clay prior to adding the bentonite powder is shown in Figure 3.4. The Volclay sodium bentonite powder has a particle size of less than #200 mesh. The clay was compacted 2% wet of optimum which is typical for liners to give minimum permeability. The clay was compacted directly on top of the geonet or 39

PAGE 48

1CIO .. tD = c u .. L 4 D Gravel Sand I Coarse to I Fine meaeum I I I U.S. standara sieve siZes I I 2 0 I ci ci ci ci ci z z z %. z _l_ I I _l_ i II II I ljiN I lr ! !ill I! I II I I, I ! I I I I!! II ji ,, I I 1 I II ;: I I Ill I 111 I J I I II I I i I I I i : I I I I I I i 1 I i i I I I I I I I''!!:' i I I "' .. ... I I 1!1 I i I WII I I !:1 I !II I 1!11 I i:l I : I iii I! I I u 1 :1 l I!! II !1-I I! li lll l I !!1.1111 1!1111 II I : i II !r 1 .. I ,1.' II I i I IIIII I I I I. iA. .II I I jJI ll!'l I Wil 1!1 :i .1. I I I I -s 2 i-: --: q 0 0 0 0 Grain diameter. mm Silt Clay I I I I I I I I I' II i I I -I I .I I I I iII II I I I I I 1111 .l I i I i i IIIII I J I I I II I I i I I I I !II I I .lj' i i II I I I I II I I II ,, I II II I II I I q ... 0 0 Figure 3.3 Sand Grain Size Distribution (Wu 1991) 40

PAGE 49

' Q) 0 E (/) ...... c Q) u Q) 0... Note: -#200 sieve portion obtained from o hyrdometer test 100 ----------------------------------I I 1111111 I 1111111 1111111 I 11111111 I 11111111 I 11111111 --r r rn n r --r r rn n r--r r rn n I 11111111 I 11111111 I 1111 BO __ L _I_ L LI.J U L __ L _I_ L LI.J U L --L -'-L L I I 1111111 I 1111111 I I 111111 I 11111111 I 11111111 1111 I 1111111 --r r rn n r--r r rn n r-r r rn n r--r r rn n I 11111111 I 11111111 11111111 I 1111111 60 ___ L _I_ L LI.J UL __ L -'-L LLJ UL _ -'-L LI.J UL __ L -'-L LU U I I I IIIII I I I IIIII I I 111111 I I IIIII I I I 111111 I I I I IIIII I I 111111 I I I IIIII --r r rn n r --r r rn --r r rn n r --r r rn n I I 1111111 I II I I 1111111 I I I IIIII 40 --L _I_ L LU U L __ L _I LJ U L __ L -'-L LI.J U L __ L -'-L LLJ U I 1111111 I I 1111111 I 111111 II I I 111111 I I 1111111 I I 111111 nr--r r rn nr--r r rn nr-r r rn n I I I I IIIII I I I I I II II I I I I I Ill 20 _I_ L LU U L-_ L _I_ L LU U L __ L -'-L LIJ U L __ L _I_ L LU U 11111111 11111111 11111111 1111111 I 11111111 I 11111111 I 11111111 I 1111111 ---r r rn n r--r r rn nr--r r rn nr--r r rn n I I I IIIII I I I IIIII I I 111111 I I I IIIII 0.001 Figure 3.4 0.01 0.1 Millimeters Clay Grain Size Distribution Without Bentonite (Wu 1991) 41 10

PAGE 50

geocomposite in a 9 in. by 9 in. mold using a standard Proctor compaction ram. The clay was compacted in a single one-inch thick layer. The number of ram blows was corrected for the volume to provide the same compaction energy as the standard compaction test. Table 3.4 Properties of Clay Liquid Limit Plastic Limit 52 19 Placement Water Content Maximum Dry Unit Weight (Standard Proctor) 22% (2% wet of optimum) 102 pcf The manufacturer's specification for the CLAYMAX geosynthetic clay liner is given in Table 3.5 and the Gundseal in Table 3.6. Table 3.5 Specification for CLAYMAX Geosynthetic Clay Liner (from manufacturer) Manufacturer Sodium Bentonite Composition Sizing Thickness Cover Fabric Material Weight Primary Backing Filler Fiber Substrate Weight Grab Strength Permeability Coef. James Clem Corporation 1. 0 lbs. /ft2 90% Montmorillonite (min.) Specially graded, 6 mesh and 30 mesh granules 1/4 in. Spunlace polyester 1 oz. fyd2 Nylon Woven polypropylene 4 oz.fyd2 95 lbs. 2 x 10-10 cmjsec. @35 ft. 42

PAGE 51

Table 3.6 Specification for Gundseal Geosynthetic Clay Liner (from manufacturer) Manufacturer Sodium Bentonite Loading Composition Particle Size Thickness Backing Material Thickness Yield Strength Bentonite Hydraulic Conductivity Membrane Coef. of Permeability 3.3 Test Program Gundle Lining Systems, Inc. 1 lb. /ft2 80-90% Montmorillonite 20-50 mesh 0.125 in. HPDE geomembrane, Gundline HD 0.020 in. 2300 psi 3. 7 x 1010 cmjsec 2. 7 x 10-n cmjsec 3.3.1 compressive Testing Short-term compressive tests were performed on the geonets and geocomposite. A 6.5 in. by 4.5 in. sample was placed between two rigid aluminum plates, and loaded at a strain rate of 0.01 injmin for the short-term compressive test. Geonet A was also tested with a strain rate of o. 002 in/min. The test was performed in accordance to ASTM D 1621 (1973), Procedure A; except a rectangular sample was used instead of a square, and the height of a single layer was tested rather than the one-inch minimum. The load was applied perpendicular to the geosynthetic. 43

PAGE 52

Long-term compressive tests were performed on geonet A and the geocomposite. A 6.5 in. by 4.5 in. sample was loaded on a pneumatic consolidation load frame between two rigid aluminum plates. A constant load was applied and the displacement was measured at time intervals. The tests were run for 170 hours. The geonet was tested at 10, 000, 15, 000 and 19, 500 psf, while the geocomposite was tested at 7500, 10,000 and 15,000 psf. Due to inconsistencies in the data, four addi tiona! geonet samples were tested at each load for one hour. 3.3.2 Flow Testing Flow testing was performed to evaluate the following effects: 1. Geosynthetic geometry and compressive strength 2. Hydraulic gradient 3. Applied normal stress 4. Flow direction 5. Boundary conditions 6. Time The tests performed are shown in Table 3.7. 44

PAGE 53

Table 3.7 Test Number Designation of Tests GEONET GN-GM 3 GNB-GM 1 GNC-GM 1 GN-GM-Cross 1 GN-GM-Parallel 1 GN-GT-GM 2 GN-GT-Sand 2 GN-GT-Clay 2 GN-Gundseal 1 GN-GT-CLAYMAX 1 GN-CLAYMAX 1 GN-GT-Foam Rubber 1 GEOCOMPOSITE Testing Program Configuration (Top to Bottom) PlatenjGM/GN-A/GM/Platen PlatenjGM/GN-B/GM/platen PlatenjGM/GN-C/GM/Platen PlatenjGM/GN-A/GM/Platen II II PlatenjGM/GT/GN-A/GM/Platen PlatenjSand/GT/GN-A/GM/Platen PlatenjClayjGT/GN-A/GM/Platen PlatenjSCL/GN-A/GM/Platen PlatenjSCL/GT/GN-A/GM/Platen PlatenjSCL/GN-A/GM/Platen Platen/Foam/GT/GN-A/GM/Platen GC-GM(GT Removed) 1 PlatenjGM/GC/Platen GC-GT 1 PlatenjGT/GC/GM/Platen GC-GT-Sand 1 PlatenjSand/GT/GC/GM/Platen GC-GT-CLAY 1 Platen/Clay/GT/GC/GM/Platen GN=Geonet GC=Geocomposite GT=Geotextile GM=HDPE Geomembrane (0.060 in. Top, 0.040 in. Bottom) SCL=Synthetic Clay Liner cross=Cross Roll Direction Parallel=Parallel Direction All geonet tests were performed with Geonet A in the roll (longitudinal or machine) direction unless specified otherwise. The roll direction is that with the smaller angle between the geonet ribs. The cross roll (transverse or cross machine) test was performed with the geonet rotated 90 degrees from the roll direction. The parallel test was performed with the geonet rotated 30 degrees from the roll direction, such that one set of the geonet ribs was parallel to 45

PAGE 54

the direction of flow. With a direct flow path the flow rate may be higher than the roll direction were both sets of ribs are at an angle to flow direction. The Gundseal was tested with the 0. 020 inch integral geomembrane directly against the geonet without a geotextile. This is the manufacturer's recommended configuration for a double composite liner. The CLAYMAX was tested both with the thin cover fabric against the geonet with a geotextile, and with the thicker primary backing fabric directly against the geonet without a geotextile. The manufacturer specifies that the CLAYMAX be install with the primary backing side up, to protect the bentonite. Applications may exist where the CLAYMAX could be used without an additional geotextile. A one-inch thick layer of closed cell neoprene sponge was used for the foam rubber test. The geonets and geocomposite were tested at six hydraulic gradients ranging from 0.03 to 1.0. ASTM D 4716 suggests a maximum gradient of 1.0 for gravity flow and 0.1 for pressure flow. A minimum gradient is not suggested. The applied normal stress ranged from 46

PAGE 55

the 500 psf minimum specified in ASTM D 4716 to the fixture capacity of 19,500 psf. The tests were started at the lowest confining pressure. The flow rate was measured at each hydraulic gradient starting with the highest. The next confining pressure was applied, and the process was repeated. The flow rate was calculated from the time required for a known discharge volume. The short-term tests were performed after the 15 minute minimum seating period specified by ASTM D 4716. The long-term tests were performed by maintaining a constant applied normal stress and measuring the flow rate over time. The purpose of the long-term tests was to observe the effect of creep and intrusion on the flow rate. Longer test times would be required for accurate prediction of ultimate flow values. The long-term tests performed are shown in Table 3.7. All tests were performed at a water temperature of 19C to 20C. The flow was not adjusted for temperature, since the flow was turbulent and the exact correction is unknown. The variation in viscosity from 19C to 20C is only 3 percent. 47

PAGE 56

Table 3.7 Long Term Flow Tests Test GN-GT GN-GT-Sand GN-GT-Sand GN-GT-Clay GN-GT-Clay (a) GN-GT-Gundseal GN-GT-CLAYMAX GN-GT-Foam GC-GN GC-Sand GC-Clay Normal Stress (osf} Time (hours} 10000, 15000 and 19500 48 10,000 and 19,500 48 15,000 120 10,000 and 19,500 48 15,000 120 19,500 48 19,500 48 19,500 48 7500 48 7500 48 7500 500 (a) 1000 hour test in progress Recirculated tap water was used for the testing to maintain a constant temperature and to minimize air bubble formation. The 10C water directly from the tap caused numerous visible air bubbles in the geonet and reduced flow rates with time. With the recirculated water, air bubble were visible only in stagnant areas, and the flow did not decrease with time. ASTM D 4716 specifies the use of deaired water, but does not specify a method for deairing or acceptable dissolved oxygen level. Studies with geonets and geocomposites either used tap water (Hwu 1990; Gallup, Bachus and Luettich 1991; Shaner and Menoff 1991) or did not state whether deaired water was used. Several studies with geotextiles have used vacuum deaired water. The use of vacuum deaired water 48

PAGE 57

would have a greater effect on geotextiles which have much lower flow rates. The use of deaired water was found by Halse, Lord and Koerner (1988) to have a significant effect on the cross plane permittivity of geotextiles. 49

PAGE 58

CHAPTER 4 RESULTS AND DISCUSSION 4.1 Compressive Testing 4.1.1 Short-Term The results for the short-term compressive tests are shown in Figure 4.1. The geonets have a bend in the stress-strain curve where the strain increases with little increase in stress. The bend in the curve is caused by layover of the geonet ribs, and is often considered the compressive strength. However, there is still a visible flow channel through the geonet, so failure has not occurred. The layover occurred at approximately 28,000 psf for geonet A, 6000 psf for geonet B and 16, 000 psf for geonet C. The foamed geonets had a lower compressive strength than the solid geonet A. A lower strain rate resulted in about 2000 psf lower compressive strength for geonet A. The geocomposite had a definite yield point near 16, ooo psf where the core began to collapse. The manufacturer's specified minimum is 15,000 psf. 50

PAGE 59

<.5 0 CD ZN gg
PAGE 60

4.1.2 Long-Term Tests The compressive creep results for geonet A are shown in Figure 4.2. The creep at 10,000 psf resulted in an additional 3 percent strain in 170 hours for a total of 9 percent. The creep at 15,000 psf resulted in an additional 8 percent strain in 170 hours for a total of 13 percent. The five tests had an error range of plus or minus 2 to 3 percent strain, so error bars are not shown. At 10, 000 and 15, 000 psf the material appears to be in a stage of secondary creep with a constant rate of deformation. The creep at 19,500 psf was much higher, with a total strain of 40 percent in 170 hours. The sample tested for 170 hours, which is plotted in Figure 4.2, had the slowest deformation rate of the five samples. All of the samples leveled off at a strain of 35 to 40 percent. The strain is comparable to that where rib layover occurred during the short-term compressive test, but the stress is lower. At 19,500 psf variable results would be expected during flow testing. With a 15 minute waiting period, the strain in the geonet could range from 15 to 40 percent. 52

PAGE 61

e;e -z < a: IU1 en I..) -o19,500 psf -15,000 psf 50 EAAOA BAAS FOA 5 TESTS OF 1 HOUR DURATION ----..... 40 30 20 10 I ,// _,. _,. __.-L __ ;. ----o-o 0 0 0.001 ' '"I I I I I I 1 I I I I II a( I I I I II d 0.01 0.1 1 10 Tl ME (hours) -o-1 O,OOOpsf GEONET A o--D-D .. --.i.-.6 o---Q-0 I I I I 11 I I I I Ill 100 1000 Figure 4.2 Compressive Creep Results for Geonet A

PAGE 62

The compressive creep results for the geocomposite are shown in Figure 4. 3. The geocomposi te failed after 3 0 minutes at 15, 000 psf which is the minimum compressive strength. The failure was rapid going from visible deformation to complete collapse in one minute. At 10,000 psf the geocomposite failed sometime between 48 and 120 hours. At 7 500 psf there was an addi tiona! strain of 3 percent in 168 hours for a total of 13 percent. The manufacturer performed creep testing at 3750 psf for 1000 hours, and found less than 5 percent total strain. 4.2 Flow Testing 4.2.1 Short-Term Testing of Geonets The flow rate per unit width verses confining pressure for test GN-GM #1 with a geomembrane on both sides is shown in Figure 4.4. The flow rate increases with hydraulic gradient and decreases with normal stress. The flow dropped off more rapidly at 19,500 psf, which is below the 28,000 psf short-term compressive strength. A plot of flow verses hydraulic gradient is shown in Figure 4.5. The curves are not linear which indicates the flow is probably not 54

PAGE 63

-'* ........ z <( a: U1 I-U1 en A-15,000psf --A-1 O,OOOpsf 80 70 60 50 40 30 rp GEOCOMPOSITE I I I ) L/ / / / ..... ..... ..... 20 Ia -A--A---I&--A---1::---A 10 0 0.1 .0. -c.. J!> .A --&' I I I 1 111 1 -o--G' I I I I I II I 10 TIME (HOURS) --0-7500psf I I I I I I I I I I I .... J.. -J --0-B I I I 1 1 1l I I I I I I 1 1 100 1000 Figure 4.3 Compressive Creep Results for Geocomposite

PAGE 64

U1 0\ --i-1.0 i-o.s -i-0.25 o u-0.12 -6-i-o.oe -oo-o.oa I c: -E ::::::-tU C) I 1-0 3: -w 1-< a: 3: 0 .....J LL ___ 7 st --. __ < 21---I -. -u ------{:j,---------JI.. 4 ..... .... .... -------- 'I> ............. ............. --....... 3 -D D D D 2 &---6-----&---------&---------. ... __ ... '0 1 0 0 ..... ----::::::,. 0 0 ---0-=----= -0----.......... ... o -------o I I 5 10 NORMAL STRESS (psf) I 15 20 (Thousands) Figure 4.4 Flow Rate Verses Normal Stress for Test GN-GM #1

PAGE 65

0 A. 0 \ 0 \ \ .. w-I 0) ? \ \ \ 0 ..... \ I :1:1: I ::E I \ CD A. \ (!) 0 0 0 I ct z ., \ I \ (!) w\ \ \ "" \ I 0 I I 0 Q) .... \ \ \ \ I-I \ \ z .... -\ \ \ w 0 (0 -A. \ \ 0 -0 \ \ \ 0 c: 0 \ \ \ < Q) ct I a: \ \ "C 0 C) as -1.0 .... tJ 0 0 (.) (!) \ \ CJ \\ \ ....J 0 I Q) \ \ I ::J 0 I \ \\ \ < .... Q) ,, a: A. 0 > \\ \ \ 0 0 \ >Q) 0 I -0 I as ., C') a: \ 0 I 0 0 u. -\ C\1 Ll) \ A. \ 0 'lit 0 \ 0 Q) 10 q .... C\1 :::1 0 I 0 u.
PAGE 66

laminar. The flow rate does not increase in a linear fashion due to greater frictional losses at higher velocities. At a given confining pressure the flow area is constant, so the velocity increases with flow rate. A plot of the transmissivity verses confining pressure is shown in Figure 4.6. The shape is similar to Figure 4. 4, except the transmissivity decreases with hydraulic gradient. Transmissivity verses hydraulic gradient is shown in Figure 4.7. Since the flow is most likely turbulent, the transmissivity is not constant. The transmissivity was not constant for any of the tests performed; therefore, the flow rate per unit width was used throughout this study. The relationship between the flow rate and gradient is linear when plotted on a log-log scale as shown in Figure 4.8. This implies that at a given normal stress the flow rate, q, and the hydraulic gradient, i, can be related by the following equation: q = c*id Considering the effect of normal stress, a, the flow rate can be described by the equation: log (q) = a + b1*log ( i) + b2* (a) 58

PAGE 67

-I c -E :::::-ro C) .......... >-I> CIJ CIJ lTI :::?! \0 CIJ z <( a: I---i .o ll-oo.s -i.25 o i.12 -A-io.oe -oiO.os ' ', I ---. ...... ...... <:............ ...... "!J. 15 [o ... D. D ... o......... D... ',, _. ___ ... -;. 1 o L ----------o 0 ---=-=-::...:---=-::-::::::::::-: -1l1 s r ----o 0 0 5 10 NORMAL STRESS (psf) 15 20 (Thousands) Figure 4.6 Transmissivity Verses Normal Stress for Test GN-GM #1

PAGE 68

0'1 0 -----500 psi t:.-2500 psi -5000 par o 10,000 psi -.a.-15,000 pal-o-19,500 psf ........ I c E co C) ........., >-1> CIJ CIJ CIJ z < a: 140 35 30 25 20 0 15 \ b"'-, 10 5 -'---, .. __ ::::-_ :=-:==:-:::-=== ----____ 0 _______ --------------0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 HYDRAULIC GRADIENT Figure 4. 7 Transmissivity Verses Gradient for Test GN-GM #1

PAGE 69

Q. 0 0 10 ,. ... 6 I Q. 8 .,; ... I ... I a. 0 ... Q. 0 0 0 10 I Q. 0 0 = I -VJ J: Q) VJ ._ Q) > Q) -co a: == 0 u:: CD v Q) ._ ::::J C) 0 u:::

PAGE 70

The coefficients obtained from a regression analysis using the above equation are given in Table 4.1. The high R2 (correlation coefficient) shows the equation fits the data well. Table 4.1 Regression Analysis Coefficients Test g 121 122 B2 GN-GM 0.94 0.54 -0.000015 0.982 GN-GT 0.54 0.63 -0.000019 0.996 GN-SAND 0.45 0.69 -0.000020 0.993 GN-CLAY 0.56 0.65 -0.000028 0.995 q in gpmjft a in psf The repeatability of the test procedure was evaluated by performing replicate tests with a new geonet. The range of values for tests GN-GM, GN-GT, GN-Sand and GT-Clay is shown in Figures 4.9 through 4.11. The results are plotted on a semi-log scale so that the relative variability is the same for all flow rates. The coefficient of variation was calculated at each combination of gradient and normal stress. The results are shown in Table 4.2. Table 4.2 Coefficient of Variation Test GN-GM GN-GM excluding 19,500 psf GN-GT GN-Sand GN-Clay Average 0.065 0.034 0.065 0.11 0.034 62 Range 0.001-0.39 0.001-0.043 0.034-0.12 0.017-0.22 0.0001-0.085

PAGE 71

(I') 0 0 .! I 0 I CD e 0 I I I "' .... 0 I = I 0 I 10 "' 0 .! I I 10 0 I I < C) ...,. (J) .... ::J CJ) u..

PAGE 72

C') 0 C;) C\1 rn 0 --T "'C I I I I c: I I ca I I I I I rn 0 I I I ::J I 0 I I i I ..r::: CD I I I-II I I I I I 0 j I I CJ) I + f z t-+ 1 LO CJ) I w I ,.. fl) I .. .... t-I 0 I I fl) I C\1 I I CD N I 0 I 1-t-I ..... .,... I I 0 Q (!) I 0 I I rn I I c. I z I I -CD en (!) I I I I en c::: I I I I I en as 0 I I f 1 w a: I + f 1 0 a: "'0 an I-c::: I I I ,.. en as C'! I I I CD 0 I I ...J I <( .. !. as I I I a: I I I a: :1: I 0 I I I 0 ..Q I I z u.. I 0 I I I CD an I t en I 0 + as 0 + 1 1. ..... .! CD I > I I <( I I i I <] I 0 I J ,... I -1+ -1-+ 0 CD I i I ,' ..... .,... ::::J I I I .2l I 1 1 u.. I ..L j_ _j_ 0 I 0 0 ...... ,.. 0 (U-U!W/Ie6) HlOIM/31 'Itt MOl.:f 64

PAGE 73

-I c E :::::-co C) ........ I I-0 0\ lTI w I-< a: 0 _J LL --i.0 !::.-i.5 --i.25 -0i.12 --6.08 -o-6.03 4.0 .-----------------------------, 1.0 0.1 0 GN-SAND 2 TESTS ---+---------t-----------1 ____ -+-------+-----------+------------1 1---i---t---t--1 "i---+-----+----_, -----+-----+--------I--------I 5 10 15 NORMAL STRESS (psf) 20 (Thousands) Figure 4.11 Average Flow Rate and Range for Tests G N-SAN D

PAGE 74

-I c E ::::::-co C) -:I: 0 0\ w 0\ <( a: 0 ....J LL --1.0 -6.--e-1.25 -01.12 -A-1.08 -0-1.03 4.0 1.0 0.1 0 GN-CLA Y 2 TESTS ...... __ -..1 -L ......._ -------------------. ..._ -+-+------..__ __ +--_ --1 r-____ __ k -----+---------------------. 5 10 15 NORMAL STRESS (psf) 20 (Thousands) Figure 4.12 Average Flow Rate and Range for Tests GN-CLAY

PAGE 75

The results for the geomembrane test were very consistent except for the 19,500 psf normal stress. One of the samples had a much larger displacement, and the flow rate was approximately half that of the other two samples. The difference is probably due to variability in the geonet creep. The higher variability of the sand tests may have been caused by difficulty in controlling the compaction. An analysis of variance (ANOVA) was performed to compare the effect of the test variables on the flow rate. The results are shown in Table 4.3. Factor Gradient Boundary Test Error Pressure Table 4.3 Analysis of Variance Mean Squares 33.4 35.2 13.4 6.3 F-Value 1424 1215 573 271 Probability 0.00 0.00 0.00 o.oo The F-value shows the amount of variability accounted for by each factor. The probability is for the factor effect being caused by chance. The gradient had the largest effect while pressure had the smallest. A comparison of the results of this study (UCD) with those obtained by the manufacturer and the Geosynthetics Research Institute (GRI) (Koerner and 67

PAGE 76

Hwu 1989) are shown in Table 4.4. The results for each hydraulic gradient are shown in Appendix B. The comparisons are approximate, since the testing conditions and materials were not identical. The results for test GN-GM were in agreement except for the 20,000 psf normal stress where the UCD flow rate was 23 percent lower compared to the manufacturer. At high normal stress, UCD found a higher flow rate for geonet B and lower flow rate for geonet c. For test GN-Sand the manufacturer and GRI obtained higher flows at low normal stress. At 20,000 psf the manufacturer's flow rate was 83 percent lower than UCD. For test GN-GT-Clay the manufacturer's results were high except at 20,000 psf, while the GRI results were close to UCD. The hydraulic gradient did not appear to affect the difference between the manufacturer and UCD. However, the GRI results were lower as the gradient decreased. The flow efficiency for each test was calculated using the flow rate for the GN-GM test at each combination of hydraulic gradient and confining pressure as 100 percent. The low flow values for the one GN-GN sample at 19,500 psf were not used in the 68

PAGE 77

Table 4.4 Comparison with Other Laboratory Results Average Percent Difference* Test Normal Stress Man.vs UCD GRI vs UCD GN-GM 2000 psf 7.2 NA 5000 -1.8 -14.0 10 000 -0.9 -10.9 15 000 -12.3 -6.6 20,000 23.4 NA GNB-GM 2000 8.3 NA 5000 -1.3 NA 10,000 -46.6 NA 15,000 -62.0 NA 20,000 -66.8 NA GNC-GM 2000 -10.4 NA 5000 -4.8 NA 10,000 20.3 NA 15,000 123.9 NA 20,000 339.0 NA GN-GT-SAND 2000 71.5 NA 5000 85.7 46.0 10,000 52.9 15.1 15.000 32.6 -8.2 20,000 -82.8 NA GN-GT-CLAY 2000 25.5 NA 5000 23.0 -4.6 10,000 53.4 14.1 15,000 47.2 -5.8 ?n nnn .d Mll Pos1t1ve value means Man. or GRI had h1gher flow rate than UCD 69

PAGE 78

calculation. The average values for the six gradients at each normal stress are shown in Table 4.5. Values greater than 100 percent mean the flow rate for the test at the given normal stress was higher than the flow rate for geonet A between two geomembranes. Geonets B and c had efficiencies greater than 100 percent at low normal stresses and less than 100 percent at high normal stresses. Flow directions other than the roll direction, or boundary conditions other than a geomembrane resulted in efficiencies less than 100 percent. The flow efficiencies for each hydraulic gradient are shown in Appendix c. Comparison plots of flow rate verses normal stress for the tests were made for a gradient of 1.0. The effect of test direction is shown in Figure 4.13. The roll direction gave the highest flow rate while the cross roll direction gave the lowest. The cross direction flow rate was about 40 percent of the roll direction, while the parallel direction flow rate was about 75 percent. 70

PAGE 79

Table 4.5 Average Geonet Flow Efficiency Using Test GN-GM as 100 Percent (a=psf) 500 2500 5000 10000 15000 19500 GN-GM 100 100 100 100 100 100 GN-GM-CROSS 39 39 40 41 44 56 GN-GM-PARL 75 74 71 74 77 86 GNB-GM 116 110 100 55 11 8 GNC-GM 147 139 134 98 29 14 GN-GT 40 35 33 31 30 33 GN-GT-SAND 33 28 26 23 22 28 GN-GT-CLAY 37 32 29 23 21 21 GN-GUNDSEAL 88 78 72 61 49 41 GN-CLAYMAX 60 46 36 17 12 NA GN-GT-CLYMX 35 30 28 24 22 22 GN-GT-FOAM 45 33 28 23 19 18 The effect of geonet type and compressive strength is shown in Figure 4.14. At low confining pressures the thicker geonet c had 50 percent more flow. Geonet B had a higher flow rate due to a slightly greater nominal thickness. The flow rate for geonets B and C dropped more rapidly as the normal stress increased due to lower compressive strengths. The flow rate was significantly reduced by adding a geotextile to the geonet as shown in Figure 4.15. The flow efficiency with a geotextile ranged from 40 71

PAGE 80

B 7 I c -E 6 :::::-cu C) 5 ......... I I-0 4 -..1 w 1\J I3 < a: 2 0 ....J LL 1 0 --!ROll r:,.-IP'ARAlLEL --CROSS ROlL --- -------GRADIENT=1.0 0 {j.--f:::, -----/),. .... -I 5 ------------------u--------f),.._ ..... ..... ..... ..... ..... /),. ---------------------I I 10 15 NORMAL STRESS (psf) I 20 (Thousands) Figure 4.13 Effect of Test Direction on Flow Rate for Test GN-GM

PAGE 81

I I m 1w z: 0 w CJ I . en Iw ... c: (J) 0 Ic: 0 en (J) _.1 CJ < -::::!: 0 c: 0 0 (J) :t: z w Ll) (J) .... :::1 C) u. 0

PAGE 82

I <] :1 z I I 0 II ..... z w Q <( a: CJ j I I . ..,..... 1.... en ca _J "C c < :::::1 0 a: aJ 0 0 z 0 Q1 -LO w LO ..... v Q1 .... :::::1 Cl u::: 0 0

PAGE 83

percent at low normal stress to 30 percent at high normal stress. The o. 020 inch-thick geomembrane backed by bentonite on the Gundseal resulted in a lower flow rate than with the 0.060 inch-thick geomembrane boundary condition. The thin Gundseal geomembrane intruded into the geonet at higher confining pressures. The tests with a geotextile in combination with a sand, clay, bentonite or foam rubber cover all had similar flow rates as shown in Figure 4.16. The flow efficiency ranged from 21 to 37 percent. The flow efficiency with the soil or foam cover was only 5 to 10 percent lower than the efficiency with a geotextile backed by a geomembrane. The primary backing geotextile on the CLAYMAX was thinner than that used for the other tests. Less intrusion occurred at low confining pressures, but the geotextile completely blocked the geonet at 19,500 psf. There was no evidence of extruded clay blocking the geonet during the testing. Some clay leakage was observed as the water turned slightly cloudy with finely suspended clay particles. However, it could not be determined whether leakage occurred through the 75

PAGE 84

..... 0 C!J I C\1 -z rn C!J I I I J 1J I' c: b I I // a:J rn I / j I 1 ::J 4D 0 0 I u .t::. cu i I J 1-a: I ..-I I J -;: ""'"c II I I I I 0 qz 1z I I LO u. (!JCI) w ,..... il I 4D I -c: 0 I I I 0 I I < I {/ -4D a: // rn . I < .... cu "'0 1):( a: c: :J q> 0 0 Z...J z a:J C!JO 0 I LO (J 4D --w I I-:::I co ,..... zo "=t C!J&L 4D .... I :J 4 0) u. 0 LO "=t ('f) C\1 ,..... 0 )( ::I I> Z...J (U-U!W/IeB) H.lOIM/3.1Vt:f MOl.:f C!JO I 76

PAGE 85

geotextile or at the edges of the sample. The effect of the gradient on the flow efficiency is shown in Table 4.6. The average flow efficiency for the six confining pressures tested at each gradient are shown. For the tests with a geotextile, the flow efficiency increased with the gradient. The other tests showed no consistent pattern. Table 4.6 Average Geonet Flow Efficiency Using Test GN-GM as 100 Percent 0.03 0.06 0.125 0.25 0.5 1 GN-GM 100 100 100 100 100 100 GN-GM-CROSS 48 45 43 42 41 41 GN-GM-PARAL 79 78 77 76 72 76 GNB-GM 63 64 66 67 69 70 GNC-GM 87 96 93 91 90 91 GN-GT 29 30 32 36 37 39 GN-GT-SAND 20 23 25 28 31 34 GN-GT-CLAY 22 24 26 29 31 32 GN-GUNDSEAL 64 64 65 65 66 66 GN-CLAYMAX 32 32 34 35 35 37 GN-GT-CLYMX 22 18 27 30 32 34 GN-GT-FOAM 22 23 26 30 32 34 Average of GT Tests 23 23 27 30 33 35 77

PAGE 86

4.2.2 Long-Term Testing of Geonets The results for a typical long-term test are shown in Figure 4. 17. The flow rate for each hydraulic gradient decreases with time. The long-term flow efficiency was calculated using the flow rate at 15 minutes for each confining pressure and gradient as 100 percent. This allows better comparison of the time dependency, since the tests started at different initial flow rates. The long term flow efficiency for each test is shown in Appendix D. A plot of the average long-term flow efficiency at 10,000, 15,000 and 19,500 psf is shown in Figures 4.18 through 4.20. At 10,000 psf the flow decreased approximately 10 percent in 100 hours. If the graph is extrapolated out to 106 hours (114 years) the flow decreases a total of 15 percent. These are only general trends since extrapolation of the limited data for 5 log cycles is not reliable. The geotextile had a slightly smaller reduction than the sand or the clay. At a normal stress of 15,000 psf the flow rate drops more rapidly with time. The efficiency is approximately 80 percent for the geotextile, 60 78

PAGE 87

(') 0 0 0 Cl) 0 I 0 Cl) .,.... CD ._ I -CIJ 0 -m I E CD ._ 0 0 z 0 -.! Cl) a. I . .! I I ca I I I .......... (.) o I I I Cl) I I i ..... 1--I I :::J (!J I I 0 I I i : 0 .s:::. z It) I .,.... (!J C\1 I I w 0 I 0 I I -Cl) I I I I I CD I I I I-1--I I I I I ._
PAGE 88

......... ... c: Q) 0 .... Q) c. >() z w () CXI 0 u. u. w 0 _, u. 100 90 BO 70 60 50 40 t 30 20 10 0 0.1 --GIN-Gl 6-GNI-SANJD--aNJ-CI!.AY ...... ............ =-= -......... --_.,._ -- ..... 10,000 psf I I 1 10 100 Tl ME (hours) Figure 4.18 Long Term Geonet Flow Efficiency at 10,000 psf Normal Stress

PAGE 89

> :s () I z Cl I I c z < rn I . 0 0 0 .c. c T""" -CD w "(j -t-W == 0 u::: E ... CD T""" Cl c 0 ...J 0) ...CD ... ::::J Cl T""" u::: 0

PAGE 90

N _._ ONBT 100 90 ........ -80 c CD 0 ... 70 CD Q. >60 () z w 50 () LL 40 LL w 3: 30 0 _J 20 LL 10 0 0.1 -fl.-OINI-011'---QN-OTo QN-OUND--BN-GT -o-ON-GTSAND CLAY CLYMX FOAM (:)......... 19,500 psf .... . ---. /:)., .... ..}------{:). .... ""'-:.. ---........... "-,,. . . . ----. "-, o "-, ''-...., 'o I I 1 10 TIME (hours) 100 Figure 4.20 Long Term Geonet Flow Efficiency at 19,500 psf Normal Stress

PAGE 91

percent for the clay, and 50 percent for the sand at 100 hours. The larger decrease for the sand and clay relative to the geotextile indicates that soil intrusion is a factor in addition to geonet creep. Extrapolation gives an efficiency of 50 percent in 106 hours compared to zero for the clay and sand in 105 hours. The efficiency values at a normal stress of 19,500 psf show inconsistencies with the other normal stresses. The efficiency for sand is higher than the geotextile. It is likely that the inconsistencies are due to variability in the geonet creep. The decrease in flow rate at 19,500 psf was approximately the same as that at 15,000 psf. It would be expected that the decrease would be larger at the higher normal stress. The decrease in flow rate with time for the geotextile, sand and clay at the three normal stresses is shown in Figures 4.21 through 4.23. At 15,000 psf the 100 hour flow rate drops below the 15 minute flow rate at 19,500 psf, showing the impact of time relative to normal stress. The rate of decrease is similar at 15,000 and 19,500 psf for all three conditions. 83

PAGE 92

.......... ... I c E ::::::-cu C) .......... I I-0 ()) w I-< a: 0 __. LL 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.1 ----10,000 psf .. b---.__ ------6-15,000 psf --19,500 psf GN-GT GRADIENT=1.0 -6-_____ --A,---t:r-----------6 ... __ ....._ __ ----..._ ______ ---------I I 1 10 100 TIME (hours) Figure 4.21 Effect of Normal Stress on Long Term Flow Rate for Test GN-GT

PAGE 93

0) U1 3.0 2.5 I c E cu 2.0 C) -I 1-0 -w 1-<( a: 0 .....J LL 1.5 1.0 0.5 0.0 0.1 101000 psf -A-151000 psf -191500 psf GN-SAND GRADIENT=1.0 ----b..._ ...... ...... ...... .... --A----...._:_:._--------.. ----::.....-___--. ---E:J, I I I I I I I II I I I I 1 l LLI I I I _l 1 1 l tJ I I I I I 1 L 1 1 10 100 1000 Tl ME (hours) Figure 4.22 Effect of Normal Stress on Long Term Flow Rate for Test GN-SAND

PAGE 94

-I c: E ::::::::-cu C) -I I-0 ;: -w 00 I-0\ < a: ;: 0 _J lL -----1 0,000 psf -t::.--15,000 psf _.__ 19,500 psf 2.00 ,.---------------------------, GN-CLAY 1. 50 A. .................... .......... ...... 'b..-...... :6-------------n .__ __ 1.00 ----------------__.. 0.50 1----------------------------l 0 0 0 L._______J__.J._---l,.._l____L..Lj__.__.__ _.____..._____._---'-'--.J._J_J_J-------'-----'---'---'---'---'--L--'-'----'---..._____._---l.......I...L.I....L.J 0.1 1 10 100 1000 Tl ME (hours) Figure 4.23 Effect of Normal Stress on Long Term Flow Rate for Test G N-CLA Y

PAGE 95

4.2.3 Short-Term Testing of Geocomposite The average flow efficiencies for the short-term geocomposite tests are shown in Table 4.7. The flow efficiencies for each hydraulic gradient are shown in Appendix E. A comparison plot of the tests at a hydraulic gradient of 1.0 is shown in Figure 4.24. Table 4.7 Average Geocomposite Flow Efficiency Using Test GC-GM as 100 Percent 500psf 2500 5000 7500 GC-GM 100 100 100 100 GC-GT 85 85 85 85 GC-SAND 69 67 65 68 GT-CLAY 75 70 64 60 The flow rate dropped rapidly above 10,000 psf. At 19,500 psf the core collapsed during the 15 minute wait. The flow rate was highest for a geomembrane without a geotextile and lowest with a sand and clay cover. The flow rates with the sand and clay cover were relatively close. As with the geonet tests, a clay cover resulted in a higher flow rate at low normal stress, while a sand cover resulted in a higher flow rate at high normal stress. The boundary condition effect for the geocomposite is smaller than 87

PAGE 96

20 18 .......... .... I 16 c -E -14 ttl C) .......... I 12 1-0 10 ;: -B 0) w 0) 1-< 0:: 6 ;: 0 4 ....J LL 2 0 --BC-OM -6-BC-BT ---------rr--/:::;.---!::;._--0. e, -----o .... .... -----.-... .. 'l!!!'l. -o ' I ' ' 0 5 -.-OC-QT-SAND o OC-BT-CLAY .... /:::;., .... .... .... .... I ' 10 GRADIENT=1.0 .... ' .... _/\ ""' ' ' I 15 ' '\ ' /:::;. ' 20 (Thousands) NORMAL STRESS (psf) Figure 4.24 Effect of Boundary Condition on Geocomposite Flow Rate

PAGE 97

for the geonet due to the much larger flow channel. The minimum flow efficiency for the geocomposite with a clay cover was 60 percent compared to 20 percent for the geonet. The effect of the gradient on the flow efficiency is shown in Table 4. 8. There was only a slight increase in efficiency with larger gradients. Table 4.8 Geocomposite Average Flow Efficiency Using Test GC-GM as 100 Percent 0.03 0.06 0.13 0.25 0.5 1 GC-GM 100 100 100 100 100 100 GC-GT 83 84 85 84 85 87 GC-SAND 65 67 66 66 67 71 GC-CLAY 67 67 67 67 68 71 A comparison with the manufacturer's results is shown in Table 4.9. The 15.8 galjmin-ft was near the capacity of the test apparatus, and may have been lower than the manufacturer due to higher frictional losses. Table 4.9 Comparison with Manufacturer's Results for Test GC-GT Normal Gradient Man. UCD Man. vs UCD Stress Cal Cqpm/ftl Cgpm/ft) (percent> 500 1.0 17.5 15.8 +11 500 0.1 5.5 5.6 -1 (a) Man. chart shows constant flow verses stress up to 15,000 psf. 89

PAGE 98

4.2.4 Long-Term Testing of Geocomposite The long-term flow efficiency for each test at 7500 psf is given in Appendix F. The change in flow rate with time is shown in Figure The sand cover resulted in the smallest decrease while the clay resulted in the largest. It is not known why the reduction in flow rate was greater with the geomembrane boundary compared to the sand cover. The extrapolated efficiency at 106 hours ranges from 90 percent with the sand to 40 percent with the clay. The clay had more visible intrusion into the core than the sand. The clay intrusion was greatest at the inlet and outlet of the sample where the geotextile was not anchored as well. A higher strength geotextile could be used to reduce the intrusion. The geotextile has a grab strength of 120 lbs compared to 300 lbs for the geotextile used with the geonet tests. 90

PAGE 99

........ .... c: CD 0 .... CD c. >() z w () \0 1-' LL LL w ;: 0 .....J LL --GC-GM !:l-GC-GT-SANIIO--GC-GT -CLAY 100 90 BO 70 60 50 40 30 20 10 0 0.1 -' -. ---...-...-...-.... ... ' "I 1 --b. 7500 psf -----------............. _. ---. ---a ....... -. ' I I I II ' I I I II I I I I 10 100 1000 TIME (hours) Figure 4.25 Long Term Geocomposite Flow Efficiency at 7500 psf Normal Stress

PAGE 100

,....... - c E ::::::-cu C) ......... I 0 -w \0 "' < a: 0 ...J LL --GC-GM 6-GC-GT-SAND --GC-GT-CLA Y 17 16 15 14 13 12 1 1 10 9 8 7 0.1 I -------=-7500 psf GRADIENT=1.0 -- 6------6-------6.----------1::,. --------------------.. ... '-I I I I I I II I I I I I I I II I I I I I I I II I I 1 ... Ill 1 10 100 1000 T I ME (hours) Figure 4.26 Long Term Geocomposite Flow Rate at 7500 psf Normal Stress

PAGE 101

4.3 Design Implications All of the short-term geonet tests, except for the CLAYMAX without a geotextile, met the regulatory minimum transmissivity requirement of .145 gpm/ft at 19,500 psf. Since the transmissivity decreased with the gradient, the most critical gradient is 1.0. This assumes the flow is in the roll direction and seams do not cause any reduction. If the flow testing is performed at a normal stress of 19,500 psf with a factor of safety of 2 to 3 as recommended by the U.S. EPA (1989b), the maximum design overburden stress is 6500 to 9750 psf. This is approximately a height of 65 to 100 feet for 100 pcf waste. Some applications could require testing at normal stresses as high as 30,000 psf for a 100 ft landfill with a factor of safety of 3. If the flow is taken at 100 hours, as recommended by the U.S. EPA ( 1989b), the geonet A tests would still meet the regulatory minimum at 19,500 psf. If the extrapolated flow for the landfill design life is used, geonet A with a soil cover would not meet the minimum 93

PAGE 102

requirements. However, at the actual load of 10,000 psf, the decrease in the flow rate with time is much less. The adequacy of the geonet depends upon the required factor of safety and how it is applied. Is geonet flow tested at the design overburden stress with the factor of safety applied to the flow; or is the geonet tested at a factor of safety times the actual overburden, with an additional factor of safety for the flow? Is the effect of creep included in the factor of safety, or is the long-term flow extrapolated out to the design life? The testing parameters recommended by the EPA appear to be a good method of providing an adequate long-term design without requiring unreasonably long test times. The decrease in flow at the design stress for the facility life should be less than that at the 2 to 3 times higher test stress for 100 hours. The 15-minute time specified in ASTM D 1621 is insufficient to give an indication of the long-term behavior. Based on the test results, geonet A would be acceptable for both the PLCR and SLCR systems for 94

PAGE 103

landfill applications with the boundary conditions tested, including the SCLs when used in accordance with the manufacturer's recommendations. Testing at higher normal stresses would be required under some conditions. Geonet B is intended for cover applications, and would not be used for leachate collection systems. The suitability of geonet c for leachate collection systems can not be determined, since boundary condition and long term behavior were not evaluated. However, due to the lower compressive strength, geonet c would be restricted to landfills of lower height than geonet A. Geonets B and c are well suited to the SWCR system where low normal stresses are encountered. While not investigated in this study, the products can be supplied with different geotextiles which can produce significantly different results. A heavier weight geotextile should reduce soil intrusion and extrusion at high normal stress, but the flow will be reduced at low normal stress due to intrusion of the thicker fabric. The final design is a tradeoff between the two factors. The geotextile should be 95

PAGE 104

The geocomposite provided much higher flow rates than the geonet at normal stresses below the 16,000 psf compressive strength. The long-term compressive creep strength is the limiting factor. The maximum long-term design strength is probably in the range of 5000 to 7500 psf. Failure occurred at 10,000 psf in a short time period. This agrees with manufacturers' recommendations of long-term strengths, approximately one half the compressive strength. If tested to the EPA recommendations, the maximum normal stress would be about 2500 to 5000 psf, depending upon factor of safety and the actual 100 hour compressive strength. The use of the geocomposite for leachate collection systems is limited to shallow landfills where a geonet does not provide adequate flow. The geocomposite is well suited to SWCR systems where the normal stress is low and the flow requirements can be high. 96

PAGE 105

5.1 summary CHAPTER 5 SUMMARY AND CONCLUSIONS The in-plane flow behavior of geosynthetics for landfill drainage applications was evaluated. The study primarily focussed on geonets and geomembranes, as geotextiles do not have adequate flow capacity. Geosynthetics are used for the primary (PLCR), secondary (SLCR) and surface water (SWCR) collection and removal drainage systems. The regulatory requirement for hazardous waste landfills is a minimum transmissivity of 3 x lo-s m2fsec (0.15 gpm/ft). A higher transmissivity may be required depending on the design flow rates for the specific facility. A factor of safety is normally applied to the flow rate, the normal compressive stress, or both in evaluating the adequacy of the design. The objective of this study was to evaluate the factors affecting the transmissivity or in-plane flow rate of geosynthetics. Actual field conditions were simulated as close as possible. The materials and 97

PAGE 106

range of variables typically encountered in landfills were used for the testing. often conducted for short geosynthetic between two Laboratory testing is time periods with the rigid platens. The configuration used for actual applications may include a geotextile and soil adjacent to the drainage geosynthetic. The soil can intrude into the geosynthetic resulting in reduced flow. A constant head, parallel flow, testing apparatus was designed and built for measuring in-plane flow. The test specimen size was 9 in. by 9 in. Three geonets and one geocomposite were tested with normal compressive stress up to 19,500 psf at hydraulic gradients ranging from 0.031 to 1.0. The effects of geosynthetic geometry and compressive strength, hydraulic gradient, flow direction, applied normal stress, boundary conditions and time on the flow rate were investigated. The boundary conditions used were a geomembrane, sand, clay, two synthetic clay liners (SCL} and foam rubber. Long-term flow tests up to 500 hours were performed. Compressive strength and compressive creep tests were performed to investigate the stress-strain behavior. 98

PAGE 107

Replicate geonet flow tests were performed to evaluate the repeatability of the test procedure. The flow results were compared with those from the manufacturer and another laboratory. A regression analysis was performed to mathematically model the geonet flow behavior. An analysis of variance was performed to determine the effect of the test variables on the flow rate. 5.2 Conclusions 1. The relationship between the flow rate and the hydraulic gradient was not linear for any of the tests. Also, the transmissivity was not constant and decreased as the hydraulic gradient increased. This indicates that the flow is probably turbulent. 2. Thicker geosynthetics result in higher flow rates. The maximum flow was 18.7 gpmjft for the geocomposite, 11.2 for the thick foamed geonet and 7.9 for the solid geonet, at a normal stress of 500 psf with a hydraulic gradient of 1.0. 3. The flow rate in the geonet was greatest in the roll direction. The flow rate in the cross-roll 99

PAGE 108

direction was approximately 40 percent of that in the roll direction. 4. The decrease in flow rate as the normal stress increases is related to the geosynthetic compressive strength. The two foamed geonets had a lower compressive strength than the solid geonet, resulting in lower flow rates at high normal stresses. The geocomposite collapsed when the compressive strength was exceeded. 5. The flow rate for a geonet between two rigid platens or geomembranes is not indicative of the flow rate with a geotextile or soil cover boundary condition. The flow rate reduced to approximately 35 percent of that between two geomembranes when a geotextile was added on one side of the geonet. 6. The type of soil cover did not have a large effect on the geonet flow rate. Sand, clay, foam rubber, or the CLAYMAX SCL above the geonet with a separator geotextile all resulted in approximately the same-short and long-term flow rates. The flow rate was slightly lower than that with the separator geotextile backed by a 100

PAGE 109

geomembrane, with the difference increasing as the normal stress increased. 7. The Gundseal SCL resulted in a lower flow rate in the geonet than that with a geomembrane backed by a rigid platen. The thin, integral geomembrane extrude? into the geonet at high normal stresses. The CLAYMAX SCL should not be used at high normal stress without a separate geotextile. The backing fabric completely intruded into the geonet at 19,500 psf. 8. Foam rubber is a suitable replacement for soil as a cover material for flow testing. At high normal stresses, the flow rate in the geonet was identical to that with a clay cover. Foam is conservative as the long-term flow was lower than the other boundary conditions. Foam is much easier to use and may be more repeatable than soil tests. 9. The flow rate for a geonet with a geotextile or soil cover decreased relative to that with a geomembrane as the hydraulic gradient decreased. 10. The boundary condition had less effect on the geocomposite flow rate. The flow rate with a 101

PAGE 110

geotextile backed by a geomembrane was 85 percent compared to that without the geotextile. The flow rate with a sand or clay cover was about 65 percent of that with a geomembrane boundary. The flow rate with a sand cover was constant, while the flow rate with a clay cover decreased relative to that with the geomembrane boundary as the normal stress increased. 11. The creep rate increased as the normal stress increased. The solid geonet creep was stable at 10, ooo and 15, ooo psf. At 19, 500 psf strand layover occurred at times ranging from 6 to 60 minutes. Creep of the geocomposite was stable at 7500 psf. The core collapsed at 10,000 psf. 12. The 15 minute minimum wait period specified in ASTM D 4716 does not adequately account for the time dependent reduction in flow at high normal stresses. The geonet flow decreased by up to 50 percent in 100 hours from the 15 minute value at 15,000 and 19,500 psf. The geocomposite core can collapse at loads for which there is only a slight reduction in the 15 minute flow. 102

PAGE 111

13. A regression analysis found the data to fit the equation: log (q) =a+b1*log ( i) +b2* (a) 14. The hydraulic gradient was found to have the greatest effect on the flow rate followed by the boundary condition, test error and normal stress. 15. The geonet flow results were repeatable with a average variability of plus or minus 3.4 percent with a clay cover or geomembrane boundary to 11 percent with a sand cover. 16. There was a large variation among the geonet results of the three different laboratories. For a geomembrane boundary on both sides, the results were in agreement at 15,000 psf and lower normal stress. For other test conditions the difference ranged from plus 339 percent to minus 83 percent. The geocomposite flow results were within 10 percent of the manufacturer's results. Even following ASTM D 4716, different laboratories can obtain a wide variation in results. 103

PAGE 112

5.3 Recommendations for Further Study 1. Determine the effect of geomembrane thickness and polymer type on the geonet flow rate when backed by clay. Evaluate polyvinyl chloride and textured high density polyethylene geomembranes. 2. Determine the effect of normal compressive stresses up to 30,000 psf on the flow of geonets. 3. Correlate geonet strain with flow rate. Obtain relationship between normal stress, time, strain and flow rate. 4. Test geocomposites with higher compressive strengths. 5. Test geocomposites manufactured from other polymers including high density polyethylene and polyvinyl chloride. 6. Determine compressive stress-time relationship for geocomposite failure. 7. Determine the effect of the separator geotextile type and properties on the flow rate. 8. Determine the effect of leachate chemicals on long-term compressive strength and flow rate. 104

PAGE 113

9. Perform tests with the fill or protective soil typically used above the PLCR system. 10. Perform flow and compressive tests of 1000 hours or longer to better determine long term creep and intrusion effects. 11. Determine the cause of the variability in flow test results obtained by the different laboratories. 105

PAGE 114

APPENDIX A Test Apparatus Design Drawings 106

PAGE 115

ai I I r 1 Kl I l / / I I I 8 I!! N I s. s. I I ]_ I : 1/ I I 19 ... _j_ 8 T ci I 19 I I ... I 8 I I I I J I I r--. I I J 1-I I L... I tg 2 I & 'S I te I ,.; l I I T I ,, 1 8 -8 ,. L 1-107 I!! I!! en te 10 Kl I!! 3 w 1---i > Q_ D I--

PAGE 116

1-' 0 OJ SECTION A NOTES: I. ALL MATERIAL 3/4 IN. CELL CAST ACRYLIC SHEET UNLESS SPECIFIED OTHERWISE 2. ALL JOINTS ADJACENT TO 9 BY 9 HIGH STRESS TEST REGION TO BE GLUED WITH TWO PART WELDON 40 OR PS 30 P[l_YHERIZABLE CEI'ENT. TAKE SPECIAL PRECAUTIONS TO PREVENT CRAZING OR OTHER DEFECTS. LOAD PLATE IS HYORAULICLLY LOAOEO TO 10,000 LBS. < 125 PSI> MAX 3. ALL TOLERANCES +/1/8 IN UNLESS SPECIFIED OTHERWISE. 4. ALL JOINTS TO BE WATERTIGHT .------------. _,.--LOAD SPACERS I IN THICK I EA 8 X 8, 7 X 7, 6 X 6, 5 X 5 AND 4 X 4 ACRYLIC 3/4 IN. SLIP PVC UNICJI 9 X 9 X I IN. ACRYLIC PLATE TO FIT INSIDE 9 X 9 OPENING WITH l/32 +/1/64 IN. CLEARANCE TRIH PLATE TO FIT ANO HARK ORIENTATION LOAD PLATE CUT IN HALF Atl GLUED TO BASE TO BE SUPPLIED BY BUYER 2 IN. ID ACRYLIC Tl.IE PROVIDE g X g IN SPACER PLATES IN I l/4, I. 3/4 AND 1/2 IN THICKNESS TO FIT IN BOTTOM OF g X g AREA

PAGE 117

APPENDIX B COMPARISON WITH OTHER LABORATORY RESULTS Table 8.1 Percent Difference for Test GNB-GM Normal Gradient UCD Man. Man. vs Stress gpm/ft gpm/ft UCD ( \) 2000 1 8.9 9.66 8.6 psf 0.5 6.01 6.52 8.5 0.25 4.1 4.59 11.9 0.125 2.95 3.08 4.4 5000 1 7.49 7.73 3.2 0.5 5.15 4.83 -6.2 0.25 3.5 3.38 -3.4 0.125 2.39 2.42 1.1 10000 1 4.21 2.22 -47.2 0.5 2.78 1.50 -46.1 0.25 1.83 0.97 -47.2 0.125 1.23 0.66 -46.0 15000 1 0.78 0.29 -62.8 0.5 0.47 0.19 -58.9 0.25 0.29 0.11 -62.5 0.125 0.2 0.07 -63.8 20000 1 0.14 0.41 -64.6 0.5 0.1 0.26 -62.8 0.25 0.05 0.16 -69.8 n 1?r:; n n1 n -69 8 109

PAGE 118

Table 8.2 Percent Difference for Test GNC-GM Normal Gradient UCD Man. Man. VB Stress gpm/ft gpm/ft UCD (%) 2000 1 10.6 9.42 -11.1 0.5 7.2 6.52 -9.4 0.25 5.1 4.59 -10.0 0.125 3.6 3.20 -11.1 5000 1 9.52 9.18 -3.6 0.5 6.45 6.28 -2.6 0.25 4.61 4.35 -5.7 0.125 3.26 3.02 -7.4 10000 1 6.25 7.73 23.7 0.5 4.29 5.31 23.9 0.25 3.03 3.62 19.6 0.125 2.22 2.54 14.2 15000 1 1.8 3.86 114.7 0.5 1.22 2.66 111 .a 0.25 0.77 1.81 135.3 0.125 0.53 1.21 127.9 20000 1 2.56 0.6 326.7 0.5 1.71 0.41 318.3 0.25 1.15 0.27 324.9 0.125 0.88 0.18 386.4 110

PAGE 119

Table 8.3 Percent Difference for Teet GN-GM Normal GradUCD Man. GRI Man. VB GRI VB Stress ient gpm/ft gpm/ft gpm/ft UCD(%) UCD(%) 2000 1 7.7 8.70 NA 12.9 NA pef o.s 5.3 5.68 NA 7.1 NA 0.25 3.7 3.86 NA 4.5 NA 0.125 2.6 2. 72 NA 4.5 NA 5000 1 7.31 7.25 6.98 -0.9 -4.5 0.5 5.06 5.07 4.75 0.2 -6.1 0.25 3.53 3.38 3.21 -4.2 -9.1 0.125 2.47 2.42 2.12 -2.2 -14.2 0.062 1.72 NA 1.35 NA -21.5 0.031 1.18 NA 0.84 NA -28.8 10000 1 6.81 6.76 6.6 -0.7 -3.1 0.5 4.66 4.59 4.44 -1.5 -4.7 0.25 3.21 3.14 3.02 -2.2 -5.9 0.125 2.22 2.23 1.98 0.6 -10.8 0.062 1. 57 NA 1.28 NA -18.5 0.031 1.07 NA 0.83 NA -22.4 15000 1 5.84 5.07 5.94 -13.1 1.7 0.5 3.96 3.62 3.98 -8.5 0.5 0.25 2.69 2.29 2.65 -14.7 -1.5 0.125 1.87 1.63 1. 74 -12.8 -7.0 0.062 1.31 NA 1.13 NA -13.7 0.031 0.91 NA 0.73 NA -19.8 20000 1 3.8 4.69 NA 23.3 NA 0.5 2.6 3.26 NA 25.4 NA 0.25 1. 75 2.17 NA 24.2 NA 0 125 1 25 1 51 NA 20 8 NA 111

PAGE 120

Table 8.4 Percent Difference for Test GN-GT-Sand Normal GradUCD Man. GRI Man. VB GRI VB Stress ient gprn/ft gprn/ft gprn/ft UCD(\) UCD(\) 2000 1 2.82 4.44 NA 57.6 NA psf 0.5 1. 75 3.02 NA 72.5 NA 0.25 1.12 2.05 NA 83.3 NA 0.125 0.7 1.21 NA 72.5 NA 5000 1 2.39 4.11 3.56 71.8 49.0 0.5 1.51 2.78 2.2 84.0 45.7 0.25 0.94 1.81 1.36 92.7 44.7 0.125 0.59 1.15 0.81 94.5 37.3 0.062 0.36 NA 0.48 NA 33.3 0.031 0.23 NA 0.27 NA 17.4 10000 1 2.04 2.90 2.63 42.1 28.9 0.5 1.28 1.93 1.63 51.0 27.3 0.25 0.78 1.21 0.94 54.8 20.5 0.125 0.48 0.79 0.54 63.5 12.5 0.062 0.3 NA 0.32 NA 6.7 0.031 0.19 NA 0.18 NA -5.3 15000 1 1.67 2.27 1.69 36.0 1.2 0.5 1.04 1.33 1.03 27.7 -1.0 0.25 0.62 0.85 0.61 36.4 -1.6 0.125 0.37 0.48 0.34 30.6 -8.1 0.062 0.23 NA 0.2 NA -13.0 0.031 0.15 NA 0.11 NA -26.7 20000 1 1.6 0.29 NA -81.9 NA 0.5 1 0.19 NA -80.7 NA 0.25 0.6 0.11 NA -81.9 NA 0 125 _n .3 7 0 OS NA -86.9 NA 112

PAGE 121

Table 8.5 Percent Difference for Test GN-GT-CLAY Normal GradUCD Man. GRI Man. vs GRI vs Stress ient gpm/ft gpmjft gpm/ft UCD(\) UCD(\) 2000 1 3 4.11 NA 36.9 NA psf 0.5 1.96 2.29 NA 17.1 NA 0.25 1.3 1.57 NA 20.8 NA 0.125 0.83 1.06 NA 27.3 NA 5000 1 2.6 3.38 2.84 30.1 9.2 0.5 1. 69 1.93 1.8 14.3 6.5 0.25 1.07 1.33 1.08 24.2 0.9 0.125 0.71 0.88 0.64 23.3 -9.9 0.062 0.44 NA 0.37 NA -15.9 0.031 0.27 NA 0.22 NA -18.5 10000 1 1.93 2. 71 2.23 40.2 15.5 0.5 1.26 1.81 1.45 43.8 15.1 0.25 0.8 1.21 0.94 51.0 17.5 0.125 0.49 0.88 0.57 78.7 16.3 0.062 0.32 NA 0.35 NA 9.4 0.031 0.19 NA 0.21 NA 10.5 15000 1 1.47 2.03 1.5 38.0 2.0 0.5 0.96 1.35 0.96 40.9 o.o 0.25 0.59 0.89 0.58 51.5 -1.7 0.125 0.37 0.59 0.34 58.3 -8.1 0.062 0.23 NA 0.2 NA -13.0 0.031 0.14 NA 0.12 NA -14.3 20000 1 1.03 0.97 NA -5.7 NA 0.5 0.65 0.58 NA -10.8 NA 0.25 0.39 0.34 NA -13.3 NA n 1?c; n.?1 n 17 -?F; c; 113

PAGE 122

APPENDIX C Short Term Geonet Flow Efficiencies Table C.1 Flow Efficiency (Percent) at Gradient=0.031 (Flow in gpm/ft) 500psf 2500 5000 10000 15000 GN-GM Flow 1.25 1.25 1.18 1.07 0.91 Efficiency 100 100 100 100 100 GN-GM-CROSS 41.8 41.5 43.7 45.0 48.4 GN-GM-PARL 78.1 73.6 77.0 76.3 77.1 GNB-GM 110.9 104.9 101.2 47.3 8.4 GNC-GM 158.0 147.7 146.1 108.3 27.0 GN-GT 34.5 30.3 29.1 26.8 24.5 GN-GT-SAND 24.7 21.4 19.7 17.8 16.3 GN-GT-CLAY 32.2 26.2 22.8 17.8 15.5 GN-GUNDSEAL 91.9 74.4 71.8 58.8 45.3 GN-CLAYMAX 59.3 41.1 31.4 16.5 10.8 GN-GT-CLYMX 31.0 24.8 23.2 19.7 17.1 GN-GT-FOAM 39.5 25.4 21.7 16.3 13.0 114 19500 0.56 100 65.0 93.9 5.0 14.1 29.6 22.1 17.3 42.5 NA 18.0 13.6

PAGE 123

Table C.2 Flow Efficiency (Percent) at Gradient = 0.062 (Flow in gpm/ft) SOOpef 2500 5000 10000 15000 19500 GN-GM Flow 1.83 1. 78 1.72 1. 57 1.31 0.86 Efficiency 100 100 100 100 100 100 GN-GM-CROSS 39.0 39.1 40.1 42.8 46.3 60.3 GN-GM-PARL 76.2 76.6 75.0 74.2 77.6 85.6 GNB-GM 116.7 107.0 97.6 51.0 9.4 5.0 GNC-GM 151.0 145.0 139.5 101.7 27.3 12.9 GN-GT 35.9 31.1 28.6 27.0 25.5 29.9 GN-GT-SAND 28.8 21.9 21.0 18.9 17.5 23.3 GN-GT-CLAY 32.6 28.0 25.5 20.3 17.5 17.9 GN-GUNDSEAL 87.4 79.2 71.5 59.2 47.7 38.7 GN-CLAYMAX 58.3 44.0 32.3 16.3 11.2 NA GN-GT-CLYMX 23.4 20.1 18.5 15.5 13.8 15.1 GN-GT-FOAM 40.9 27.4 22.4 18.3 14.0 15.1 115

PAGE 124

Table C.3 Flow Efficiency (Percent) at Gradient = 0.125 (Flow in gpm/ft) 500psf 2500 5000 10000 15000 19500 GN-GM Flow 2.66 2.56 2.47 2.22 1.87 1.27 Efficiency 100 100 100 100 100 100 GN-GM-CROSS 38.6 39.1 39.0 41.4 44.5 55.7 GN-GM-PARL 73.3 74.4 74.5 75.5 77.8 84.6 GNB-GM 115.0 111.6 96.7 55.5 10.5 8.1 GNC-GM 145.7 140.3 132.2 100.1 28.1 13.9 GN-GT 39.3 33.3 31.9 30.0 28.4 31.3 GN-GT-SAND 31.3 26.1 24.0 21.5 19.6 25.8 GN-GT-CLAY 36.2 31.3 28.6 22.1 19.6 19.5 GN-GUNDSEAL 87.2 78.1 72.0 62.1 50.9 39.1 GN-CLAYMAX 58.6 45.9 36.0 16.5 11.4 NA GN-GT-CLYMX 35.4 30.0 27.8 23.7 21.0 21.3 GN-GT-FOAM 44.5 33.2 26.2 21.7 17.1 10.2 116

PAGE 125

Table C.4 Flow Efficiency (Percent) at Gradient = 0.25 (Flow in gpm/ft) 500psf 2500 5000 10000 15000 GN-GM Flow 3.76 3.65 3.53 3.21 2.70 Efficiency 100 100 100 100 100 GN-GM-CROSS 38.3 38.2 38.4 39.9 42.9 GN-GM-PARL 73.9 74.1 73.6 73.6 76.4 GNB-GM 117.7 111.4 99.1 57.2 10.6 GNC-GM 145.3 135.0 130.5 94.4 28.6 GN-GT 41.8 36.4 35.2 33.2 32.4 GN-GT-SAND 34.5 29.4 26.6 24.4 23.0 GN-GT-CLAY 38.6 33.8 30.3 25.0 22.0 GN-GUNDSEAL 88.2 78.3 72.0 61.5 50.5 GN-CLAYMAX 59.9 47.6 37.4 17.5 12.7 GN-GT-CLYMX 37.9 33.4 30.6 27.4 24.1 GN-GT-FOAM 46.3 35.6 29.5 24.9 20.7 117 19500 1.85 100 54.1 83.8 8.5 14.5 34.6 29.2 22.3 42.0 NA 24.3 20.6

PAGE 126

Table C.5 Flow Efficiency (Percent) at Gradient = 0.5 (Flow in gpm/ft) 500psf 2500 5000 10000 15000 GN-GM Flow 5.48 5.28 5.06 4.66 3.96 Efficiency 100 100 100 100 100 GN-GM-CROSS 38.4 37.9 38.6 39.0 41.2 GN-GM-PARL 73.3 73.5 51.3 73.1 75.4 GNB-GM 117.4 112.4 101.9 59.6 11.9 GNC-GM 140.4 132.4 127.5 92.1 30.7 GN-GT 43.1 38.7 35.8 34.6 34.5 GN-GT-SAND 36.6 31.9 29.8 27.4 26.2 GN-GT-CLAY 39.8 35.8 33.4 27.0 24.3 GN-GUNDSEAL 86.5 78.9 73.7 62.2 51.4 GN-CLAYMAX 60.5 48.2 39.5 18.7 10.1 GN-GT-CLYMX 39.5 35.2 32.9 28.6 26.2 GN-GT-FOAM 47.9 37.5 32.3 27.7 23.2 118 19500 2.79 100 51.2 82.4 9.3 14.8 36.8 33.0 24.5 42.2 NA 26.6 24.1

PAGE 127

Table C.6 Flow Efficiency (Percent) at Gradient = 1.0 (Flow in gpm/ft) ;m-------.Q_ 500psf 2500 5000 10000 15000 GN-GM Flow 7.92 7.59 7.31 6.81 5.84 Efficiency 100 100 100 100 100 GN-GM-CROSS 38.9 39.3 39.1 39.4 41.8 GN-GM-PARL 74.3 74.2 74.4 73.2 75.1 GNB-GM 118.6 115.1 102.5 61.8 13.3 GNC-GM 141.1 136.5 130.3 91.8 30.9 GN-GT 45.3 41.2 39.4 36.6 35.5 GN-GT-SAND 40.0 35.6 32.7 30.0 28.6 GN-GT-CLAY 42.0 37.7 35.6 28.3 25.1 GN-GUNDSEAL 87.4 78.9 73.9 62.5 51.1 GN-CLAYMAX 61.9 51.2 40.8 18.9 13.7 GN-GT-CLYMX 42.5 37.3 34.8 30.9 27.4 GN-GT-FOAM 49.5 40.2 35.1 28.6 25.1 119 19500 4.19 100 50.2 83.5 9.8 14.2 37.9 35.5 25.6 41.5 NA 28.3 25.0

PAGE 128

APPENDIX D Lonq Term Geonet Flow Efficiencies Table 0.1 Flow Efficiency (Percent) for GN-GT at 10,000 psf (Flow in gpm/ft) Time (hours Gradient Flow 0.25 1.2 4 70 1 2.50 100 97.6 95.5 94.0 0.5 1. 61 100 78.9 92.8 90.9 0.25 1. 07 100 96.6 92.2 90.4 0.125 0.67 100 94.0 92.2 89.5 0.062 0.43 100 96.7 92.0 89.2 0.031 0.29 100 94.8 91.6 89.9 Average 93.1 92.7 90.6 Table 0.2 Flow Efficiency (Percent) for GN-Sand at 10,000 psf (Flow in gpm/ft) Time (hours) Gradient Flow 0.25 1 4 45.7 1 2.22 100 97.3 93.7 91.9 0.5 1.43 100 94.9 91.7 90.8 0.25 0.88 100 96.1 91.0 91.0 0.125 0.53 100 96.2 91.6 90.4 0.062 0.35 100 92.0 92.0 89.7 0.031 0.23 100 92.3 85.8 84.1 Average 94.8 91.0 89.7 120

PAGE 129

Table 0.3 Flow Efficiency (Percent) for GN-Clay at 10,000 psf (Flow in gpm/ft) Time (hours) Gradient Flow 0.25 0.75 3.5 69 1 1. 89 100 88.2 87.8 87.6 0.5 1.25 100 87.3 87.7 86.5 0.25 0.80 100 87.0 88.9 87.0 0.125 0.49 100 90.4 91.0 88.1 0.062 0.32 100 86.9 90.3 87.9 0.031 0.19 100 90.4 95.7 94.7 Average 88.3 90.2 88.6 Table 0.4 Flow Efficiency (Percent) for GN-GT at 15,000 psf (Flow in gpmjft) 'I'ime (hours) Gradient Flow 0.25 1 3.8 6.7 46.7 1 2.08 100 94.3 88.4 84.8 81.2 0.5 1. 37 100 95.8 89.8 84.9 80.2 0.25 0.88 100 94.5 89.5 85.7 81.1 0.125 0.53 100 94.5 88.7 84.8 78.8 0.062 0.33 100 97.0 90.1 88.3 84.7 0.031 0.22 100 88.3 81.2 76.7 69.1 Average 94.1 87.9 84.2 79.2 121

PAGE 130

Table 0.5 Flow Efficiency (Percent) for GN-SANO at 15,000 psf (Flow in gpmjft) Time (hours) Gradient Flow 0.25 1 4 118.5 1 1. 67 100 84.8 74.7 53.0 0.5 1. 04 100 85.2 75.0 50.0 0.25 0.62 100 85.4 76.1 47.6 0.125 0.37 100 87.6 79.4 48.6 0.062 0.23 100 88.0 77.1 43.8 0.031 0.15 100 85.5 76.1 43.6 Average 86.1 76.4 47.8 Table 0.6 Flow Efficiency (Percent) for GN-CLAY at 15,000 psf (Flow in gpmjft) Time (hours) Gradient Flow 0.25 1.5 3 69 119 1 1. 47 100 83.9 78.6 66.5 62.4 0.5 0.98 100 82.8 77.4 62.1 58.5 0.25 0.60 100 82.6 76.4 59.8 56.3 0.125 0.36 100 82.1 77.5 63.2 59.1 0.062 0.23 100 83.8 74.7 60.3 58.1 0.031 0.14 100 84.1 77.5 71.7 63.8 Average 83.2 77.0 63.9 59.7 122

PAGE 131

Table 0.7 Flow Efficiency (Percent) for GN-GT at 19,500 psf (Flow in gpmjft) Time (hours) Gradient Flow 0.25 1 4 70 1 1. 68 100 87.9 74.2 59.4 0.5 1.10 100 88.8 70.4 56.1 0.25 0.68 100 88.9 72.9 56.8 0.125 0.43 100 87.7 72.8 55.3 0.062 0.28 100 82.6 69.5 55.0 0.031 0.18 100 90.3 73.9 57.4 Average 87.7 72.3 56.7 Table 0.8 Flow Efficiency (Percent) for GN-SANO at 19,500 psf (Flow in gpmjft) Time (hours) Gradient Flow 0.25 1 4.3 8.2 49 1 1.49 100 91.5 85.2 79.7 76.5 0.5 0.92 100 86.9 79.3 75.1 73.6 0.25 0.54 100 91.7 84.7 80.5 77.7 0.125 0.33 100 93.9 84.8 81.0 77.5 0.062 0.20 100 96.6 85.0 80.3 82.4 0.031 0.12 100 96.7 88.7 80.0 79.3 Average 92.9 84.6 79.4 77.9 123

PAGE 132

Table 0.9 Flow Efficiency (Percent) for GN-CLAY at 19,500 psf (Flow in gpmjft) Time (hours) Gradient Flow 0.25 1 3 46 1 1. 07 100 82.8 71.6 63.7 0.5 0.68 100 84.2 74.0 61.0 0.25 0.41 100 84.5 73.5 63.1 0.125 0.25 100 87.5 76.2 63.3 0.062 0.15 100 83.8 76.6 66.2 0.031 0.10 100 84.5 81.4 64.9 Average 84.5 75.6 63.7 Table 0.10 Flow Efficiency (Percent) for GN-GUNOSEAL at 19,500 psf (Flow in gpmjft) Time (hours) Gradient Flow 0.25 1 4 45 1 1. 74 100 81.4 67.6 53.2 0.5 1.18 100 81.6 68.5 53.1 0.25 0.78 100 83.7 70.0 52.0 0.125 0.50 100 81.5 69.6 53.8 0.062 0.33 100 83.5 71.5 52.3 0.031 0.24 100 78.6 66.8 47.1 Average 81.7 69.0 51.9 124

PAGE 133

Table 0.11 Flow Efficiency (Percent) for GN-CLAYMAX at 19,500 psf (Flow in gpm/ft) Time (hours) Gradient Flow 0.25 1. 75 4.5 44 1 1.19 100 80.3 70.3 61.9 0.5 0.74 100 80.6 70.4 58.3 0.25 0.45 100 81.1 73.1 59.2 0.125 0.27 100 82.6 72.2 60.7 0.062 0.17 100 83.3 69.0 56.3 0.031 0.10 100 82.2 75.2 59.4 Average 81.7 71.7 59.3 Table 0.12 Flow Efficiency (Percent) for GN-FOAM at 19,500 psf (Flow in gpm/ft) 'I' ime (hours) Gradient Flow 0.25 1.2 4 70 1 1. 05 100 82.2 64.2 39.4 0.5 0.67 100 79.8 62.6 35.0 0.25 0.38 100 86.1 70.1 36.2 0.125 0.22 100 89.6 74.2 36.2 0.062 0.13 100 91.5 76.2 36.2 0.031 0.08 100 96.1 81.6 38.2 Average 87.5 71.5 36.8 125

PAGE 134

APPENDIX E Short Term Geocomposite Flow Efficiencies Table E.1 Flow Efficiency (Percent) at Gradient=0.031 (Flow in gpm/ft) 500psf 2500 5000 7500 GC-GM Flow 3.86 3.84 3.83 3.49 Efficiency 100 100 100 100 GC-GT 90.4 81.6 79.5 82.0 GC-SAND 68.0 65.2 60.8 67.6 GC-CLAY 77.6 67.3 61.5 59.8 Table E.2 Flow Efficiency (Percent) at Gradient=0.062 (Flow in gpmjft) Test _____ -----.Q" 500psf 2500 5000 7500 GC-GM Flow 5.26 5.05 4.94 4.68 Efficiency 100 100 100 100 GC-GT 84.9 83.7 84.4 81.4 GC-SAND 70.2 68.3 65.3 66.3 GC-CLAY 73.3 70.7 64.3 58.1 126

PAGE 135

Table E.3 Flow Efficiency (Percent) at Gradient=0.125 (Flow in gpm/ft) -------Test -----C1_ 500psf 2500 5000 7500 GC-GM Flow 7.38 7.11 6.73 6.31 Efficiency 100 100 100 100 GC-GT 85.2 85.2 83.7 84.8 GC-SAND 66.4 65.1 65.8 68.3 GC-CLAY 74.0 68.9 64.5 58.8 Table E.4 Flow Efficiency (Percent) at Gradient=0.25 (Flow in gpm/ft) -----Test -----Q 500psf 2500 5000 7500 GC-GM (gpm/ft) 10.26 9.95 9.52 8.70 (Efficiency) 100 100 100 100 GC-GT 79.6 83.7 85.1 85.8 GC-SAND 67.2 65.9 65.5 67.4 GC-CLAY 72.7 68.8 64.1 61.1 127

PAGE 136

Table E.5 Flow Efficiency (Percent) at Gradient=0.5 (Flow in gpm/ft) T;;t------------"-500psf 2500 5000 7500 GC-GM Flow 14.20 13.76 13.13 12.41 Efficiency 100 100 100 100 GC-GT 83.8 85.5 86.7 84.7 GC-SAND 68.3 66.4 66.5 68.6 GC-CLAY 75.5 70.5 65.4 60.6 Table E.6 Flow Efficiency (Percent) at Gradient=1.0 (Flow in gpm/ft) --500psf 2500 5000 7500 GC-GM Flow 18.74 17.76 17.47 16.24 Efficiency 100 100 100 100 GC-GT 84.4 88.1 88.0 88.9 GC-SAND 72.4 70.5 68.1 71.1 GC-CLAY 76.5 74.2 67.1 64.2 128

PAGE 137

Table E.7 Flow (gpm/ft) for Test GC-GT at High Normal Stress 10000psf 15,000 19,500 0.031 2.41 1.41 0.09 0.062 3.23 1. 90 0.14 0.125 4.66 2.83 0.23 0.25 6.52 4.00 0.37 0.5 9.06 6.05 0.57 1.0 12.60 8.12 0.88 129

PAGE 138

APPENDIX F Long Term Geocomposite Flow Efficiencies Table F.1 Flow Efficiency (Percent) for GC-GM at 7500 psf (Flow in gpm/ft) Time (hours) Gradient Flow 0.25 1 4 43.3 1 1. 684 100 98.1 96.3 94.4 0.125 6.31 100 97.2 96.4 90.2 0.031 16.24 100 94.8 92.7 88.4 Average 96.7 95.1 91.0 Table F.2 Flow Efficiency (Percent) for GC-SAND at 7500 psf (Flow in gpm/ft) Time (hours) Gradient Flow 0.25 1 5 46.5 1 11.55 100 100.0 100.0 97.9 0.125 4.31 100 98.5 97.5 95.5 0.031 2.36 100 100.9 98.4 96.3 Average 99.8 98.6 96.6 130

PAGE 139

Table F.3 Flow Efficiency (Percent) for GC-CLAY at 7500 psf (Flow in gpm/ft) Time (hours) Gradient Flow 0.25 1 2.75 49 1 10.42 100 95.3 91.3 81.8 0.125 3.711 100 95.7 92.0 79.8 0.031 2.083 100 93.7 88.6 76.0 Average 94.9 90.6 79.2 T1me (hours) Gradient 97 170 269 337 503 1 78.8 76.9 74.7 73.5 69.3 0.125 76.6 73.9 71.9 70.9 66.0 0.031 71.2 67.4 65.0 62.8 58.5 Average 75.5 72.7 70.5 69.1 64.6 131

PAGE 140

ASTM BIBLIOGRAPHY (1973), D 1621-73, "Standard Compressive Properties of Plastics," American Society Materials, Philadelphia, PA. Test Method for Rigid Cellular for Testing and ASTM (1987), D 4716-87, "Standard Test Method for Constant Head Hydraulic Transmissivity (In-Plane Flow) of Geotextiles and Geotextile Related Products," American Society for Testing and Materials, Philadelphia, PA. Bonaparte, R., Williams, N., and Giroud, J.P. (1985), "Innovative Leachate Collection Systems for Hazardous Waste Containment Facilities," Proceedings, Geotechnical Fabrics Conference '85, IFAI, Cincinnati, OH, pp. 10-34. Bright, D.G. (1991), "Polymeric Behavior of Geosynthetic Materials," presented at Structural Synthetic Geogrids for Waste Facility Applications, Denver, co, March 19, 1991, Tensar Corporation, Morrow, GA. Cancelli, A., Cazzuffi, D., and Rimoldi, P. (1987), "Geocomposite Drainage Systems: Mechanical Properties and Discharge Capacity Evaluation," Proceedings, Geosynthetic '87 Conference, IFAI, New Orleans, LA, pp. 393-404. Daniel, D. E. and Koerner, R. M. ( 1991) "Landfill Liners from Top to Bottom," Civil Engineering, Vol. 61, No. 12, December, pp. 46-49. Dempsey, B.J. (1989), "Hydraulic Requirements of Geocomposite Fin-Drain Materials Utilized in Pavement Subdrainage," Geotextiles and Geomembranes, Vol. 8, pp. 191-215. 132

PAGE 141

Estornell, P.M. (1991), Conductivity Tests Materials for Liner Thesis, University of "Bench-Scale Hydraulic of Bentonitic Blanket and Cover Systems, M. s. Texas Austin. Federal Register (1991), "Solid Waste Disposal Facility criteria; Final Rule," Vol. 56, No. 196, October 9, pp. 50978-51119. Federal Register (1992), "Liners and Leak Detection Systems for Hazardous Waste Land Disposal Units; Final Rule," Vol. 57, No. 19, January 29, pp. 3462-3497. Gallup, R.A., Bachus, R.C., and Luettich, S.M. (1991), "Comparative Hydraulic Performance of Resinated Geotextile overwraps," Proceedings, Geosynthetics '91 Conference, IFAI, Atlanta, GA, pp. 653-664. Gourc, J.P., Faure, Y., Rollin, A., and Lafleur, J. (1982), "Structural Permeability Law of Geotextiles," Proceedings, Second International Conference on Geotextiles, IFAI, Las Vegas, NV, pp. 149-154. Halse, Y.H., Lord, A.E., and Koerner, R.M. (1988), "Effect of Dissolved Oxygen (and Bubbles) on the Measured Permittivity of Geotextiles," Geotechnical Testing Journal, Vol. 11, No. 2, June 1988, pp. 158-160. Hwu, B.L. (1990), "Design, Evaluation and Behavior of Geosynthetic Drainage Systems," Ph.D. Dissertation, Drexel University. Hwu, B.L., Koerner, R.M., and Sprague, C.J. (1990), "Geotextile Intrusion into Geonets," Proceedings, 4th International Conference on Geotextiles, Geomembranes and Related Products, the Hague, Netherlands, pp. 351-356. Hwu, B.L., and Koerner, R.M. (1990), Geocomposite Sheet Drain Joining," Geotextiles and Geomembranes, Vol. 9, No. 4-6, pp. 501-506. 133

PAGE 142

IFAI (1990), A Design Primer: Geotextiles and Related Materials, International Fabrics Association International, St. Paul, MN. Ionescu, A., and Kellner, L. (1982), "About Longitudinal Permeability and Draining Capacity of Non-woven Geotextiles," Proceedings, Second International Conference on Geotextiles, IFAI, Las Vegas, NV, pp. 127-131. Koerner, R.M. (1990}, Designing with Geosynthetic, 2nd ed, Prentice-Hall, Englewood Cliffs, NJ. Koerner, R.M., and Bove, J.A. (1983), "In-Plane Hydraulic Properties of Geotextiles," Geotechnical Testing Journal, Vol. 6, No.4, December, pp.190-195. Koerner R.M., and Hwu, B.L. (1989}, "Behavior of Double Geonet Drainage Systems," Geotechnical Fabrics Report, Vol. 7, No. 5, September/October, pp. 39-44. Koerner, R.M., Luciani, V.A., Freese, J.S., and Carroll, R.G. (1986}, "Prefabricated Drainage Composites: Evaluation and Design Guidelines," Proceedings, Third International Conference on Geotextiles, IFAI, V1enna, Austria, pp. 551-556. Leclercq, B., Fayoux, D., Faure, Y. and Millet, F. (1986), "Determination of Transmissivity of Geotextiles, Comparison of Tests with Samples Unconfined or Confined in Soil," Proceedings, Third International Conference on Geotextiles, IFAI, Vienna, Austria, pp. 1257-1262. Ling, H. I., Tatsuoka, F., and Wu, J .T.H. (1990), "Measuring In-Plane Hydraulic Conductivity of Geotextiles," Geosvnthetic Testinq for Waste Containment Applications, ASTM STP 1081, R. M. Koerner, editor, ASTM, Philadelphia, PA, pp. 257-272. 134

PAGE 143

Ling, H.I., Tatsuoka, F., and Wu, J.T.H., and Nishimura, J. ( 1990) "Measuring Flow Behavior of Geotextiles Subject to Stress Confinement," Submitted for publication in Journal of Geotechnical Engineering. Lundell, c .M., and Menoff, s. D. ( 1989) "The Use of Geosynthetics as Drainage Media at Solid Waste Landfills," Proceedings, Geosynthetics, '89, IFAI, San Diego, CA, pp. 10-17. Mirafi Inc. (1986), "Long-Term Testing of Miradrain 6000," Mirafi Technical Note 86-02, Charlotte, NC, July. Oweis, I.S. (1990), Geotechnoloav of Waste Management, Butterworth, Boston, MA. Raumann, G. (1982), "Inplane Permeability of Compressed Geotextiles," Proceedings, Second International Conference on Geotextiles, IFAI, Las Vegas, NV, pp. 55-60. Shaner, K.R., and Menoff, S.D. (1991), "Impacts of Bentonite Geocomposites on Geonet Drainage," Proceedings, Geosynthetics in Filtration, Drainage and Erosion Control, 5th Annual GRI Seminar, Geosynthetic Research Institute, Philadelphia, PA, pp. 153-166. Slocumb, R.C., Demeny, D.O., and Christopher, B.R. (1986), "Creep Characteristics of Drainage Nets," presented at Superfund '86, Washington, DC. Smith, A.D. and Kraemer, s. R. ( 1987) "Creep of Geocomposite Drains," Proceedings, Geosynthetic '87 Conference, IFAI, New Orleans, LA, pp. 422-433. U.S. EPA (1984), The Hydrologic Evaluation of Landfill Performance (HELP) Model -Volume I. User's Guide for Version 1, EPA/530-SW-84-009, u.s. Environmental Protection Agency, Office of Solid Waste and Emergency Response, Washington, DC. 135

PAGE 144

u.s. EPA (1985), Minimum Technology Guidance on Double Liner Systems for Landfills and Surface Impoundments; Design, construction and Operation, EPA/530/SW-85-014, U.S. Environmental Protection Agency, Office of Solid Waste and Emergency Response, Washington, DC. U.S. EPA (1986), Test Methods for Evaluating Solid Waste, EPA SW-846, u.s. Environmental Protection Agency, Office of Solid Waste and Emergency Response, Washington, DC. U.S. EPA (1987), Geosynthetic Design Guidance for Hazardous Waste Landfill Cells and Surface Impoundments, EPA/600/2-87/097, u.s. Environmental Protection Agency, Office of Research and Development, Cincinnati, OH. u.s. EPA (1989a), Technical Guidance Document: Final Covers on Hazardous Waste Landfills and Surface Impoundments, EPA/530-SW-89-047, u.s. Environmental Protection Agency, Office of Solid Waste and Emergency Response, Washington, DC. U.S. EPA (1989b), Requirements for Hazardous Waste Landfill Design. Construction, and Closure, Seminar Document, EPA/625/4-89/022, U.S. Environmental Protection Agency, Office of Research and Development, Cincinnati, OH. u.s. EPA (1990a), User's Guide to LC System. Leachate Collection System, Version 1.0, u.s. Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH. U.S. EPA (1990b), User's Guide to F-Cover, Final Cover Evaluation Advisor. Version 1.0, U.S. Environmental Protection Agency, Risk Reduction Laboratory, Cincinnati, OH. Williams, N., Giroud, J.P., and Bonaparte, R. (1984), "Properties of Plastic Nets for Liquid and Gas Drainage Associated with Geomembranes," Proceedings, International Conference on Geomembranes, '84, IFAI, Denver, co, pp. 339-404. 136

PAGE 145

Wu, J.T.H. (1991), "Predicting Performance of Geosynthetic-Reinforced Retaining Walls," Proceedings, International Symposium on Geosynthetic-Reinforced Soil Retaining Walls, Denver, co, pp. 1-54. Zagorski, G.A., and Wayne, M.H. (1990), Seams," Geotextiles and Geomembranes, No. 4-6, pp. 487-499. 137 "Geonet Vol. 9,