By James L. Hanson and Nazlı Yeşiller
Composite liner systems consisting of geomembranes (GMs) placed over geosynthetic clay liners (GCLs) commonly are used for containment applications. Such systems are used in base liners for ponds, reservoirs, basins and canals, and bottom and cover liner systems for landfill facilities. Even though the use of GCL-GM composites has become commonplace in containment systems, the long-term field performance of these systems is not well documented or understood (e.g., NRC 2007, Benson et al. 2011).
Timely cover of the geosynthetics is critical for performance of liner systems (Koerner 2012). Multiple problems have been identified for geosynthetic liner components under exposed conditions in the field. For GMs, thermal stresses resulting from cyclic temperature variations cause development of wrinkles, and UV exposure and oxidation cause aging and degradation of the materials (e.g., Rowe 2005, Koerner 2012). Specifically, for high-density polyethylene (HDPE) geomembranes, the service life of the materials was estimated to decrease significantly with increasing temperature ranging from up to 900 years at 68°F (20°C) to less than 20 years at 140°F (60°C) (Rowe 2005). For GCLs, shrinkage of GCL panels and separation of GCL panels underneath exposed geomembrane liners have been reported from landfills (e.g., Koerner and Koerner 2005a and b, Thiel and Rowe 2010). Panel separations of 2 to 47 inches (50 to 1,200 mm) corresponding to 5%–28% strain were reported for GCLs that were originally overlapped over a distance of 6 to 12 inches (150 to 300 mm) under geomembranes along slopes with 2°–34° angles (Gassner 2009, Thiel and Rowe 2010). The composite barriers in these cases were left exposed without placement of overlying layers for durations between 2 and 60 months. In addition, in a test plot with an exposed GM-GCL, migration of bentonite in the GCL from locations near the top of slopes toward the toe of slopes was reported (Take et al. 2015), resulting in reduced/nonuniform bentonite quantity in the GCL. All of these adverse effects compromise the performance of the composite liner systems for effective containment. Observations are presented herein on the condition of a GM-GCL composite liner system that was left uncovered at a municipal solid waste landfill located in the United States. The liner system was constructed and then left exposed without placement of overlying materials for more than a decade.
Field site
A GM-GCL bottom liner system was constructed for a cell at a municipal solid waste landfill in California. The landfill is situated at a location characterized as being in a temperate, dry summer, warm summer climate zone (Csb), according to the Köppen-Geiger classification system (Peel et al. 2007). The cell was constructed with 2H:1V side slopes. The length of the slopes ranged from 79 feet (24 m) for the north-facing slope (i.e., south slope) to 98 feet (30 m) for the west-facing slope (i.e., east slope). The geomembrane and the underlying GCL were placed over a compacted subgrade of native soils. The geomembrane consisted of a 60-mil (1.5-mm) thick black HDPE geomembrane, which was textured on the bottom side against the underlying GCL and smooth on the top exposed side. The GCL consisted of a needlepunched nonwoven product with the light-colored side oriented downward and placed against the underlying subgrade and the dark-colored side oriented upward and placed against the textured surface of the overlying geomembrane. The GCL was manufactured with granular bentonite. The composite liner system was constructed in 2004. A leachate collection and removal system, separation layer, or waste were not placed in the cell at any time after construction of the liner system and the GCL-GM liner system was left uncovered for more than 12 years, through June 2016. The annual mean temperature at the site varied between 58.6°F and 63.1°F (14.8°C and 17.3°C) between 2004 and 2016, with an average value of 60.4°F (15.8°C) for the 12-year period. The average mean minimum and mean maximum temperature were 48.4°F (9.1°C) and 72.5°F (22.5°C), respectively, and the average annual precipitation was 16.3 inches (414.7 mm) for the 12-year exposure period. In 2016, for expansion of the landfill cell, the exposed liner system was removed to accommodate the revised site geometry for the construction of a new cell. The authors were allowed access to the site during the liner removal operations. The condition of the geosynthetic components was observed and assessed, and samples were collected from both the geomembrane and the GCL for further evaluations in the laboratory.
Observation during field exhumation
Initially, the geomembrane was removed at the site, followed by the removal of the underlying GCL. In days prior to the exhumation operations, the geomembrane was observed to undergo significant dimensional changes between cool morning hours and warm afternoon hours, leading to the formation of wrinkles (Figures 2a and 2b). Space between the geomembrane with wrinkles and underlying GCL were observed throughout the cell after the onset of the geomembrane removal processes. The height of the cavity between the geomembrane and the GCL due to wrinkling was measured to be on the order of 6 to 8 inches (150 to 200 mm) along the base of the cell, with less space present between the geosynthetic liners along the slopes. The condition of the parent geomembrane was relatively uniform throughout the cell. Similarly, the condition of the seams also was uniform in the uncovered cell. Significant defects such as holes, tears, puncture marks or seam separation/irregularities were not observed at the site during the investigation.
After the removal of the geomembrane, the condition of the GCL liner was assessed starting with an evaluation of the seams. GCL seams along the east and south slopes of the landfill cell were investigated during exhumation. A scaled drawing of the GCL panels along the east and south slopes is presented in Figure 3. The length of the seams and dimensions of the seam gaps (length along the slope direction and width perpendicular to the slope direction) were measured for each seam. Of 43 panel seams, seam separation was observed at eight locations, along seven separate seams (i.e., one seam had two distinct zones of separation). The minimum seam separation was 0.8 inch (20 mm) in width, whereas the maximum separation was 8.7 inches (220 mm) in width. The length of the seam separation zones along the direction of the slopes ranged from 5.6 to 55.8 feet (1.7 to 17 m). The zones of seam separation were along the upper half of the slope length, with distances from the crest of the slope to onset of seam separation ranging from 3 to 46 feet (0.9 to 14.0 m). A photograph of one panel with a wide seam separation is presented in Figure 4.
While the condition of the geomembrane appeared relatively uniform in the cell, the condition of the underlying GCL was observed to be variable. The GCL was dry near the top and relatively wet at the bottom of the east slope. Less variation of the moisture content of the GCL was present along the south slope. In all cases for upper portions of the slopes, the bentonite in the GCL was generally partially hydrated, maintaining a granular consistency. The bentonite core of the GCL had macrocracks in these cases. At lower elevations along the slopes where the GCLs were wetter, the bentonite generally had a gel structure.
Migration of bentonite and accumulation of bentonite between the geomembrane and the GCL was observed at the cell during exhumation. Thin stripes of relatively dry bentonite were observed on the upper surface of the GCL throughout the length of the slopes at the site, likely corresponding to bentonite deposited from rivulets of water flowing downslope. In addition, significant accumulation of bentonite was present along the toe of the east slope, forming a wedge shape. The wedge feature consisted of accumulation of bentonite between the geomembrane and the GCL along the slope length, with minimal bentonite mass along the upper regions of the slope and maximum depth of bentonite at the toe of the slope (Figure 5). The wedge feature extended approximately 295 feet (90 m) along the width of the east slope. At the deepest section (i.e., 3.5-inch [90-mm] depth) of the feature, the wedge extended 9.6 feet (2.93 m) up the slope from the toe. Figures 6a and 6b present a schematic and a photograph of the section of the wedge with the maximum depth near the toe. Samples obtained from several locations along the upper parts of the slopes and from near the toe location along the south slope indicated significant loss of bentonite with limited and, in some cases, nearly no bentonite present between the two nonwoven geotextile sheets (Figure 7).
Accumulation of the subgrade soil between the geomembrane and the GCL forming a mountain shape was observed near the toe of the slope at the cell corner between the east and south slopes. The mountain feature had a peak thickness (6.3 inches [160 mm]) with lower thickness zones above and below the peak along the slope. The length of the mountain along the slope was 8.6 feet (2.63 m). Overall, the lateral extent of subgrade soil accumulation in the mountain feature was more localized, and the total amount of soil accumulated was lower compared to the bentonite accumulation in the wedge feature. Figure 8 presents a schematic of the mountain.
The observations made during the exhumation process were in part similar to those reported in literature (e.g., Koerner and Koerner 2005a and b, Thiel and Rowe 2010, Take et al. 2015). The seam separations likely were caused by shrinkage of the bentonite in the GCLs and tensioning (i.e., necking) of the GCLs along the long length of the slopes. In general, the subgrade soil underneath the GCL at the site was observed to be dry, with the exception of near-toe locations. The GCL was in intimate contact with the underlying soil, with no separation between the GCL and the foundation soil observed during the exhumation process. While it is expected that moisture equilibrium between the GCL and the underlying soil (i.e., relatively constant water contents for both materials at all locations) would have been reached over the 12-year period that the liner system remained in place at the site, moisture content of the GCL was variable at the site. This variation in moisture content likely resulted from the movement of water as a result of condensation that formed between the geomembrane and the GCL due to diurnal temperature variations and also from moisture accumulation near the sump location at the landfill cell. Bentonite likely was carried down with condensate water. The migration of bentonite reduced the amount of bentonite in the GCL in the upper parts of the slopes. This bentonite was accumulated on the GCL at lower elevations (in the wedge as a distinct mass and on the GCL panels deposited as stripes). The presence of subgrade soil between the geomembrane and the GCL (at the mountain) prevented intimate contact of the geomembrane and GCL, forming a potential pathway for fluid transfer. The bentonite loss toward the toe of the slope was near the sump; it is expected that significant wetting/drying and accompanying changes in level of standing water occurred at this location due to annual precipitation cycles. Bentonite between the geotextile sheets of the GCL may have been eroded due to hydrodynamic forces associated with precipitation water entering the uncovered cell at the site. Low bentonite content in GCLs near toes of slopes has not previously been reported in the literature. The seasonal ponding at the south end of the cell also may have influenced the accumulation of subgrade soil between the geomembrane and GCL (the mountain feature).
Extensive quantitative analysis has been conducted on the samples of the geomembrane and GCL materials collected from the field. Results for oxidative induction time (OIT) for the geomembrane are presented in Tian et al. (2019), with tests on mechanical properties to be presented elsewhere. For the GCLs, preliminary data on mass per area, water content, swell index and hydraulic conductivity are presented in Williams et al. (2018). Analysis on cation exchange capacity and mineral composition of the bentonite in the GCLs, membrane behavior of the GCLs, and temperature and precipitation cycles at the site are to be presented elsewhere.
Conclusion
Condition of an uncovered single composite bottom liner system that consisted of two geosynthetic layers (geomembrane over GCL) is provided. The liner system was exhumed after 12 years of exposure to the environment at a municipal solid waste landfill cell. The sides of the cell were relatively steep at 2H:1V slopes. The geomembrane was observed to undergo diurnal cycles of expansion and contraction due to temperature changes with 6- to 8-inch (150- to 200-mm) high gaps between the geomembrane and the GCL due to wrinkling along the base of the cell with less space present between the geosynthetics along the slopes. The condition of the geomembrane was uniform based on visual observation alone, with testing required to evaluate the properties of the material. In contrast, the condition of the GCL was observed to be highly variable. The GCL was dry near the top (with granular bentonite) and relatively wet at the bottom (with hydrated gel bentonite). Significant migration of bentonite from the GCL with accumulation between the geomembrane and the GCL, deposition of bentonite over the GCL, and erosion of the bentonite within the GCL were observed. At locations near the slope crest and in the vicinity of a sump near the toe, nearly all of the bentonite had eroded away from between the cover and carrier geotextile sheets of the GCL. The GCL panels underneath the geomembrane were separated at multiple places, where separation was observed at eight locations along seven separate seams (from a total of 43 seams at the cell) with gaps approximately 0.8 to 8.7 inches (20 to 220 mm) in width and 5.6 to 55.8 feet (1.7 to 17 m) in length. Degradation of the condition of the GCL was apparent for the exhumed liner system, and the observations herein highlight the vulnerability of GCLs under exposed conditions. Covering the liner system could have reduced the extent of the deterioration of the material condition. Accumulation of subgrade soil between the geomembrane and GCL near sump locations identified in the investigation constitutes a new concern for GM-GCL liner systems. This mechanism also may be applicable to covered conditions and requires further investigation.
James L. Hanson, Ph.D., P.E., is a professor in the Civil and Environmental Engineering Department at California Polytechnic State University in San Luis Obispo, Calif.
Nazlı Yeşiller, Ph.D., is director of the Global Waste Research Institute at California Polytechnic State University in San Luis Obispo, Calif.
Acknowledgments
Waste Connections Inc. and Cold Canyon Landfill are acknowledged for allowing site access and sampling of the liner system. Dr. Amro El Badawy, Kyle O’Hara, John Buringa, Sean Herman and Spencer Jemes assisted with sampling. Anthony Trujillo assisted with graphics.
References
Benson, C., Albright, W., Fratta, D., Tinjum, J., Kucukkirca, E., Lee, S., Scalia, J., Schlicht, P., and Wang, X. (2011). Engineered covers for waste containment: Changes in engineering properties & implications for long-term performance assessment, NUREG/CR-7028, Office of Research, U.S. Nuclear Regulatory Commission, Washington, D.C.
Gassner, F. (2009). “Field observation of GCL shrinkage at a site in Melbourne Australia.” Geotextiles and Geomembranes, 27(5), 406–408.
Koerner, R. M. (2012). Designing with geosynthetics (Volumes 1 and 2), 6th ed., Xlibris.
Koerner, R. M., and Koerner, G. R. (2005a). In-situ separation of GCL panels beneath exposed geomembranes. GRI White Paper #5, Geosynthetic Institute, Folsom, Pa.
Koerner, R. M., and Koerner, G. R. (2005b). In-situ separation of GCL panels beneath exposed geomembranes. GFR, 23(5), 34–39.
National Research Council (NRC). (2007). Assessment of the performance of engineered waste containment barriers, The National Academies Press, Washington, D.C.
Peel, M. C., Finlayson, B. L., and McMahon, T. A. (2007). Updated world map of the Köppen- Geiger Climate Classification, Hydrology and Earth System Sciences, 11: 1633–1644.
Rowe, R. K. (2005). “Long-term performance of contaminant barrier systems,” 45th Rankine Lecture. Geotechnique, 55(9), 631–678.
Take, W. A., Brachman, R. W. I., and Rowe, R. K. (2015). “Observations of bentonite erosion from solar-driven moisture migration in GCLs covered only by a black geomembrane.” Geosynthetics International, 22(1), 78–92.
Thiel, R., and Rowe, R. K. (2010). “Technical developments related to the problem of GCL panel shrinkage when placed below an exposed geomembrane.” Proc., GBR-C 2k10: 3rd Int. Symposium on Geosynthetic Clay Liners, 93–102.
Tian, K., Benson, C. H., Hanson, J. L., and Yesiller, N. (2019). “Antioxidant depletion in a HDPE geomembrane from a composite liner after 12-yr of atmospheric exposure.” Proc., Geosynthetics Conf. 2019, 1–9.
Williams, T., Benson, C. H., Tian, K., Yesiller, N., and Hanson, J. L. (2018). “Hydraulic conductivity of a geosynthetic clay liner (GCL) from a composite liner after 12-yr of atmospheric exposure.” Proc., 11th Int. Conf. on Geosynthetics, International Geosynthetics Society, 1–7.