Bonaparte, R., Koerner, R.M., and Daniel, D.E., Assessment and Recommendations for Improving the Performance of Waste Containment Systems, research report published by the U.S. Environmental Protection Agency, National Risk Management Research Laboratory, EPA/600/R-02/099, December 2002.
This broad-based study addressed three categories of issues related to the design, construction, and performance of waste containment systems used at landfills, surface impoundments, and waste piles, and in the remediation of contaminated sites. The categories of issues, the locations in this report where each category is addressed, and the principal investigator for the study of each category are as follows:
- geosynthetic tasks are described in Chapter 2 and Appendices A and B; the principal investigator for these tasks was Professor Robert M. Koerner, P.E.;
- natural soil tasks are described in Chapters 3 and 4 and Appendices C and D; the principal investigator for these tasks was Professor David E. Daniel, P.E.; and
- field performance tasks are described in Chapter 5 and Appendices E and F; the principal investigator for these tasks was Dr. Rudolph Bonaparte, P.E.
Each portion of the report was authored by the identified principal investigator, and individuals working with the principal investigator. However, each principal investigator provided input and recommendations to the entire study and peer-reviewed and contributed to the entire report.
Geosynthetic materials (e.g., geomembranes (GMs), geotextiles (GTs), geonets (GNs), and plastic pipe) have been used as essential components of waste containment systems since at least the early 1980’s. Five separate laboratory and/or analytical tasks were undertaken to address technical issues related to the use of these materials in waste containment systems. The technical issues related to geosynthetics are: (1) protection of GMs from puncture using needlepunched nonwoven GTs; (2) behavior of waves in high density polyethylene (HDPE) GMs when subjected to overburden stress; (3) plastic pipe stress-deformation behavior under high overburden stress; and (4) service life prediction of GTs and GMs. Conclusions are: (1) needlepunched nonwoven GTs can provide adequate protection of GMs against puncture by adjacent granular soils; a design methodology for GM puncture protection was developed from the results of laboratory tests and is presented; (2) temperature-induced waves (wrinkles) in GMs do not disappear when the GM is subjected to overburden stress (i.e., when the GM is covered with soil), rather the wave height decreases somewhat, the width of the wave decreases even more, and the void space beneath the wave becomes smaller; (3) waves may induce significant residual stresses in GMs, which may reduce the GM’s service life; residual stresses induced in HDPE GMs by waves may be on the order of 1 to 22% of the GM’s short-term yield strength; (4) if GM waves after backfilling are to be avoided, light-colored GMs can be used, GMs can be deployed and seamed without intentional slack, GMs can be covered with an overlying light colored temporary GT until backfilling occurs, and backfilling can be performed only in the coolest part of the day or even at night; (5) based on finite element modeling results, use of the Iowa State iv formula for predicting plastic pipe deflection under high overburden stress is reasonable; (6) polypropylene GTs are slightly more susceptible to ultraviolet (UV) light degradation than polyester GTs, and lighter weight GTs degrade faster than heavier GTs; (7) GTs that are partially degraded by UV light do not continue to degrade when covered with soil, i.e., the degradation process is not auto-catalytic; (8) buried HDPE GMs have an estimated service life that is measured in terms of at least hundreds of years; the three stages of degradation and approximate associated durations for each as obtained from the laboratory testing program described in this report are: (i) antioxidant depletion (≈ 200 years), (ii) induction (≈ 20 years), and (iii) half-life (50% degradation) of an engineering property (≈ 750 years); these durations were obtained from the extrapolation of a number of laboratory tests performed under a limited range of conditions; it is recommended that additional testing be performed under a broader range of conditions to develop additional insight into the ultimate service life of HDPE GMs, and other types of GMs as well.
Geosynthetic clay liners (GCLs) are a relatively new type of liner material, having first been used in a landfill in 1986. One of the key issues with respect to field performance of GCLs is their stability on permanent slopes, such as found on landfill final cover systems. Fourteen test plots, designed to replicate typical final cover systems for solid waste landfills, were constructed to evaluate the internal and interface shear strength of GCLs under full-scale field conditions on 2H:1V and 3H:1V slopes. Five different types of GCLs were evaluated, and performance was observed for over four years. All test plots were initially stable, but over time, as the bentonite in the GCLs became hydrated, three slides (all on 2H:1V slopes) that involved the GCLs have occurred. One slide involved an unreinforced GCL in which bentonite that was encased between two GMs unexpectedly became hydrated. The other two slides occurred at the interface between the woven GTs of the GCLs and the overlying textured HDPE GM. Conclusions are: (1) at the low normal stresses associated with landfill final cover systems, the interface shear strength is generally lower than the internal shear strength of internally-reinforced GCLs; (2) interfaces between a woven GT component of the GCL and the adjacent material should always be evaluated for stability; these interfaces may often be critical; (3) significantly higher interface shear strengths were observed when the GT component of a GCL in contact with a textured HDPE GM was a nonwoven GT, rather than a woven GT; (4) if bentonite sandwiched between two GMs has access to water (e.g., via penetrations or at exposed edges), water may spread laterally through waves or wrinkles in the GM and hydrate the bentonite over a large area; (5) if the bentonite sandwiched between two GMs does not have access to water, it was found that the bentonite did not hydrate over a large area; (6) current engineering procedures for evaluating the stability of GCLs on slopes (based on laboratory direct shear tests and limit-equilibrium methods of slope stability analysis) correctly predicted which test plots would remain stable and which would undergo sliding, thus validating current design practices; and (7) based on the experiences of this study, landfill final cover systems with 2H:1V sideslopes may be too steep to be stable with the desired factor of safetydue to limitations with respect to the interface shear strengths of the currently available geosynthetic products.
To evaluate the field performance of compacted clay liners (CCLs), a database of 89 large-scale field hydraulic conductivity tests was assembled and analyzed. A separate database for 12 soil-bentonite admixed CCLs was also assembled and analyzed. In addition, case histories on the field performance of CCLs in final cover test sections were collected and evaluated. Conclusions are: (1) 25% of the 89 natural soil CCLs failed to achieve the desired large-scale hydraulic conductivity of 1 × 10-7 cm/s or less; (2) all of the 12 soil-bentonite admixed CCLs achieved a large-scale hydraulic conductivity of less than 1 × 10-7 cm/s; however, all of these CCLs contained a relatively large amount (more than 6%) of bentonite; soil-bentonite admixed CCLs will not be discussed further; (3) the single most common problem in achieving the desired low level of hydraulic conductivity in CCLs was failure to compact the soil in the zone of moisture and dry density that will yield low hydraulic conductivity; (4) the most significant control parameter of CCLs was found to be a parameter denoted “Po”, which represents the percentage of field-measured water content-density points that lie on or above the line of optimums; when Po was high (80% to 100%) nearly all the CCLs achieved the desired field hydraulic conductivity, but when Po was low (0 to 40%), fewer than half the CCLs achieved the desired field hydraulic conductivity; (5) practically no correlation was found between field hydraulic conductivity and frequently measured soil characterization parameters, such as plasticity index and percentage of clay, indicating that CCLs can be successfully constructed with a relatively broad range of soil materials; (6) hydraulic conductivity decreased with increasing CCL thickness, up to a thickness of about 1 m; and (7) analysis of CCLs constructed in the final cover test sections generally showed that CCLs placed without a GM overlain by soil tended to desiccate and lose their low hydraulic conductivity within a few years.
Liquids management data were evaluated for 187 double-lined cells at 54 landfills to better understand the field performance of landfill primary liners, leachate generation rates, and leachate chemistry. Conclusions are: (1) average monthly active-period leak detection system (LDS) flow rates for cells with HDPE GM primary liners constructed with construction quality assurance (CQA) (but without ponding tests or electrical leak location surveys) will often be less than 50 lphd, but occasionally in excess of 200 lphd; these flows are attributable primarily to liner leakage and, for cells with sand LDSs, possibly construction water; (2) average monthly active-period LDS flow rates attributable to leakage through GM/GCL primary liners constructed with CQA will often be less than 2 lphd, but occasionally in excess of 10 lphd; (3) available data suggest that average monthly active-period LDS flow rates attributable to leakage through GM/CCL and GM/GCL/CCL primary liners constructed with CQA are probably similar to those for GM/GCL primary liners constructed with CQA; (4) GM liners can achieve true hydraulic efficiencies in the 90 to 99% range, with higher efficiencies occasionally being achievable; (5) GM/GCL, GM/CCL, and GM/GCL/CCL composite liners can achieve true hydraulic efficiencies of 99% to more than 99.9%; (6) GMs should not be used alone in applications where a hydraulic efficiency above 90% must be reliably achieved, even if a thorough CQA program is employed, except perhaps in situations where electrical leak location surveys or ponding tests are used to identify GM defects and the defects are repaired; (7) GM/CCL and GM/GCL/CCL composite liners are capable of substantially preventing leachate migration over the entire period of significant leachate generation for typical landfill operations scenarios without leachate recirculation or disposal or liquid wastes of sludges; (8) leachate collection and removal system (LCRS) flow rates were highest at the beginning of cell operations and decreased as waste thickness increased and daily and intermediate covers were applied to the waste; leachate generation rates decreased on average by a factor of four within one year after closure and by one order of magnitude two to four years after closure; within nine years of closure, leachate generation rates were negligible for the landfill cells evaluated in this study; (9) municipal solid waste (MSW) cells produced, on average, less leachate than industrial solid waste (ISW) and hazardous waste (HW) cells; for cells of a given waste type, rainfall fractions were highest in the northeast and lowest in the west; the differences in leachate generation rates are a function of type of waste, geographic location, and operational practices; (10) in general, HW landfills produced the strongest leachates and coal ash landfills produced the weakest leachates; MSW ash leachate was more mineralized than MSW leachate and the other ISW leachates; (11) the solid waste regulations of the 1980s and 1990s have resulted in the improved quality of MSW and HW landfill leachates; and (12) the EPA Hydrologic Evaluation of Landfill Performance (HELP) computer model, when applied using an appropriate simulation methodology and an appropriate level of conservatism, provides a reasonable basis for designing LCRSs and sizing leachate management system components; due to the complexity and variability of landfill systems, however, the model will generally not be adequate for use in a predictive or simulation mode, unless calibration is performed using site-specific measured (not default) material properties and actual leachate generation data.
Waste containment system problems were identified at 74 modern landfill and surface impoundment facilities located throughout the U.S. The purpose of this aspect of the project was to better understand the identified problems and to develop recommendations to reduce the future occurrence of problems. Conclusions are: (1) the number of facilities with identified problems is relatively small in comparison to the total number of modern facilities nationwide; however, the search for problems was by no means exhaustive; (2) the investigation focused on landfill facilities: 94% of the identified problems described herein occurred at landfills; (3) among the landfill problems, 70% were liner system related and 30% were cover system related; however, the ratio of liner system problems to cover system problems is probably exaggerated by the fact that a number of the facilities surveyed were active and did not have a cover system; (4) based on a waste containment system component or attribute criterion, the identified problems can be grouped into the following general categories: (i) slope instability of liner systems or cover systems or excessive deformation of these systems (44%); (ii) defectively constructed liners, leachate collection and removal systems (LCRSs) or LDSs, or cover systems (29%); (iii) degraded liners, LCRSs or LDSs, or cover systems (18%); and (iv) malfunction of LCRSs or LDSs or operational problems with these systems (9%); (5) considering a principal human factor contributing to the problem criterion, the identified problems are classified as follows: (i) design (48%); (ii) construction (38%); and (iii) operation (14%); (6) the main impacts of the problems were: (i) interruption of facility construction and operation; (ii) increased maintenance; and (iii) increased costs; (7) problems detected at facilities were typically remedied before adverse environmental impacts occurred; (8) impact to groundwater or surface water was only identified at one facility, where landfill gas migrated beyond the edge of the liner system and to groundwater; (9) all of the identified problems can be prevented using available design approaches, construction materials and procedures, and operation practices; (10) although the environmental impact of problems has generally been negligible thus far, the landfill industry should do more to avoid future problems in order to: (i) reduce the potential risk of future environmental impact; (ii) reduce the potential health and safety risk to facility workers, visitors, and neighbors; (iii) increase public confidence in the performance of waste containment systems; (iv) decrease potential impacts to construction, operation, and maintenance; and (v) reduce costs associated with the investigation and repair of problems.