This issue of the "Technical Bulletin" summarizes a report entitled "Comparison of PVC and HDPE Geomembranes - Interface Friction Performance, " published in November 1996. The report is written by Dr. Shobha K Bhatia and Gautam Kasturi of Syracuse University. A complete copy of this report can be obtained by contacting the PVC Geomembrane Institute: 217-333-3929.
Modern solid-waste landfills and hazardous landfills in the USA are required to have a low hydraulic conductivity liner and drainage system, consisting of geosynthetic materials (geomembranes, geotextiles, geonets and geocomposites) and compacted clay. A cross section of a typical modem landfill consists of several layers of soils and geosynthetic products. The stability of these 'slopes' is controlled by the shear strength of the various interfaces in such a composite liner. Critical interfaces include soil vs. geomembrane, soil vs. geotextile, geomembrane vs. geotextile and geomembrane vs. geonet. The strength of each of these interfaces has to be determined after careful, site specific material testing (Koerner, 1994). The experience and confidence gained from these tests on different materials and soils is valuable to designers. Such data give the basis for better judgement in design.
Geomembranes are critical components of modem landfill design, performing moisture barrier functions in the containment system. Today, a variety of geomembranes are in use. The basic difference between them is the material and/or method of manufacture. The most commonly used material types are PVC (Polyvinyl Chloride) and HDPE (High Density Polyethylene).
Geomembrane interface frictional failure has been identified as the cause of numerous geosynthetic lined slope failures. As a result, the interface frictional strength of any geomembrane interface has to be determined with the utmost care. It is recommended that wherever possible, the interface frictional strength for a geomembrane-soil combination be determined experimentally, without sorting to use of generalized values for similar soils from published data (Koerner, 1994). Direct shear, pullout and ring shear tests have been performed extensively, mainly on soil-geomembrane interfaces, to characterize their strengths (Koemer et al, 1986; Seed et al., 1988; O'Rourke et al., 1990; Takasumi et al., 1991; Stark and Poeppel, 1994).
There has been a limited number of testing programs that have attempted to draw a general comparison between PVC and HDPE geomembranes. O'Rourke (1990) reported that the higher the stiffness or hardness of the geomembrane, like HDPE, the lower the friction angle, as compared to a flexible membrane, like PVC. Martin et al., (1984) tested geotextile vs. geomembrane interface friction using a very soft, flexible geomembrane like Ethylene Propylene Diene Monomer (EPDM), a medium stiffness PVC geomembrane and a tough geomembrane, like HDPE.
PVC geomembranes have unique interface friction behavior when compared to other geomembranes due to their flexibility. However, a systematic comparison between PVC and HDPE geomembranes is necessary to expand the existing knowledge base.
PVC and HDPE have different mechanical and physical properties, as well as, field applicability. However, depending on specific application, both have been widely used in many applications. Some of the major differences between the two materials are:
- PVC geomembranes are flexible and relatively easy to handle, while HDPE Geomembranes is tough and non-flexible.
- HDPE geomembranes tend to exhibit a sharp peak in their stress-strain curve and therefore, tend to undergo a relatively abrupt failure whereas PVC undergoes a very large amount of elongation before failure.
- It is universally recognized that the field seaming is potentially the most problematic aspect for liner construction. Due to its flexibility, it is possible to do a majority of the PVC seams under controlled factory-conditions because they can be folded easily. HDPE geomembranes, however, still need to be seamed in the field. A PVC liner may require as low as 20 percent the field seams required by a HDPE liner (Peggs, 1992).
For the proposed study three different types of PVC Geomembranes were compared with two types of HDPE geomembranes. These geomembranes were tested with sand, sandy loam and silty clay soils, as well as, a geotextile.
For all the tests the bottom box was filled with compacted silt clay at a dry density of 2081 kg/g3 and at an optimum moisture content of 9.78 percent. The dry, silty clay was crushed using a jaw crusher and sieved in a #40 sieve. It was then mixed with water to get the optimum moisture content. The clay was then placed in the bottom box in three layers and compacted using a wooden tamper. After the soil was placed in the bottom box, the top was leveled using a straight edge and covered with plastic wrap until the geomembrane was placed, to avoid loss of moisture. At the beginning of each test, a fresh sample of geomembrane, cut into dimensions 330 mm x 460 mm (13"x 18"), was placed on top of the compacted clay in the bottom box. The bottom box was moved to the starting position. The top box was then placed on top of it. For soil interfaces, the soil (mixed with the appropriate amount of water to simulate field moisture conditions) was placed in the top box, in three layers, and compacted. On average, the height of the soil in the top box was 20-mm above the geomembrane. After the soil was placed in the top box, it was raised slightly and tightened in this position to ensure full contact between the geomembrane and the soil.
For tests with geotextile interfaces, the geotextile was clamped to both ends of the top box. To ensure that the geotextile did not tear out of the clamping before the interface failed, holes were made at the ends of the geotextile and the clamping screws were inserted through them and tightened.
The loading plate was placed on top of the soil or geotextile in the top box, along with the loading ball. The loading yoke was lowered so that it sat exactly on the ball. Pressure was increased to the calculated level to exert the required amount of normal stress for each run of the test The load cell and vertical and horizontal LVDRs were reset for zero. The exerted normal stress was allowed to settle fully on the interface for about five minutes. Testing was performed at displacement of 64 mm (2.5"). Shear force was exerted on the interface by pulling the bottom box relative to the top box. The gear system controlling the motion of the bottom box was automated through a computer. Data collection was also done using a computerized data acquisition system.
A total of 101 tests were conducted on the 20 interfaces (five different geomembranes against three different soils and one geotextile). Initially, tests were repeated to verify their reproductibility. Since good reproductibility was observed in the initial tests, subsequent tests were not repeated.
The interface friction angles based on stress obtained at about I0 percent strain (25.4 mm) from the stress displacement curves and given in table 1. In the case of the HDPE and textured HDPE geomembranes, the stress-displacement response of the interface is such that after reaching peak stress, further shearing to a larger strain causes stabilization of the stress. Hence, the shear stress at I0 percent strain for the rigid geomembranes is less than the peak. However, for the flexible geomembranes (PVC), due to stretching during the tests, the strength at higher strain is greater than at lower strain. This was observed with all PVC interfaces, with all other interface materials except fine sand. It can be seen that under field conditions, if the PVC geomembranes are stressed beyond the yield stress for the interface, the material stretches under the load without any loss of strength or material damage.
The failure of an interface can take place in any of the following ways:
- By shear failure at the predetermined horizontal interface between the interface geomembrane (this was clearly the failure mode for the tough and rigid HDPE geomembranes).
- When the entire interface slides over the base material. (In this testing program, the base was of compacted silty clay, but it could have been a steel or wooden plate. This behavior is typical of PVC interfaces, where the flexibility of the geomembrane is so tight that after initial failure of the interface, the continued increase in the shear stress causes failure with respect to the base. This facilitates the stretching of the sample. This was the main reason why frictional strength of PVC interfaces could not be characterized by a peak failure stress. Instead, the interface friction angles had to be calculated based on stress at a particular strain).
- Failure within the interface material. (This was always observed in all the interfaces with soils fine sand, silty clay and sandy loam. After the interface fails, further shearing causes failure to occur within the soils layer because the internal friction of the soil is reached. This is manifested in the form of a thin layer of about 2 mm of soil left on the top of surface of the membrane as shearing is continued to higher strain).
- Considerable stretching of all PVC geomembranes was observed. .
- Wrinkles were observed to have formed in the PVC geomembranes with sandy loam due to the large grain-size of sandy loam.
- Geotextiles experienced stretching with textured geomembranes both HDPE and PVC.
- Both textured HDPE and textured PVC had significant amount of soil embedded in the geomembrane at the end of the test.
The relatively large particles of the sand and the sandy loam get embedded in the flexible surface of the PVC geomembranes, giving additional friction. This is not observed with silty clay, however, because the particles of silty clay are too small to cause effective embedment in the PVC geomembrane surface.
Textured HDPE was found to have a higher interface friction angle than smooth PVC in all cases except with the geotextile. This is due to the reduction in contact area since the rough side of the geotextile and the rough surface of the geomembrane are in contact at texture projections.
Textured geomembranes give better interface friction values than their corresponding smooth geomembranes.
The file-finish in the PVC had no useful influence on the strength of the interface in the case of the sand interface. This is probably due to the incompatibility of the grain size of the sand particles with the size of the file-grid. It is felt that the size of the file-finish grid was too fine for sand particles to be embedded within it and caused improvement in frictional characteristics of the interface. At the same time, interface friction angle was higher or nearly the same for the file-finish PVC against the sandy loam and silty clay. This was due to the fact that these soils have a greater portion of finer particles than the sand. This leads us to the conclusion that File-Finish PVC improves frictional characteristics with only specific types of soils, depending on the particle size and the size of the file-finish grid.
Similarly, textured PVC does not always give a higher interface friction angle than smooth PVC. With large particle sizes (like sand and sandy loam), interface friction is improved, while with smaller particles and with the geotextile, interface friction is reduced. This is probably due to the rolling of the smaller particles around the smooth edges of the texturing projections. This confirms the fact that interface friction is very specific to the soil, as well as, the geomembrane.
The stress strain behavior of PVC is much different from that of HDPE. Even after reaching yield stress of the interface, PVC interfaces will not fail but maintain stability by stretching of the membrane material without loss of strength or material damage.
Koerner, R. M., (1994), "Designing with Geosynthetics," Third Edition, Prentice Hall, Englewood Cliffs, NJ.
Koerner, R. M., Martin, J.P. and Koerner, G.R., (1986), "Shear Strength Parameters Between Geomembranes and Cohesive Soils," Geotextiles and Geomembranes, Vol 44, No. 1, pp 21-30.
O’Rourke, T.D. Druschel, S.J. and Netravali, A.N., (1990), "Shear Strength Characteristics of Sand Polymer Interfaces," Journal of Geotechnical Eng., ASCE, Col 116, No. 3, pp 451-469.
Peggs, I.G., (1992), PVC and HDPE Geomembranes in Municipal Landfill liners and Covers-The Facts", Report for the PVC Geomembrane Institute.
Takasumi, D.L., Green, K.R., and Holtz, R.D., (1991), "Soil-Geosynthetics Interface Strength Characteristics: A Review of State of the Art Testing Procedures," Proc. Geosynthetics 91’, Atlanta, Georgia, USA, pp 87-100.
Environmental Protection, Inc provides this PGI Technical Bulletin for your information.
For more information call 800-OK-LINER today!