Use of Geosynthetic materials in solid waste landfi ll design: A review of geosynthetic related stability issues

Geosynthetics used in landfi lls provides a technical and economic advantages over traditional clay liners. It may create stability issue and even lead to landfi ll failure due to its low interface or internal shear strength if improperly designed and/or constructed. The most common failure mechanism in geosynthetic-lined landfi lls is transitional failure involving waste and bottom liner (deep-seated failure) or only fi nal cover system (shallow failure). Shear strengths of geosynthetic-geosynthetic and geosynthetic-soil have a wide range of variations. Shear strengths of interface from literature may be used in preliminary design. For fi nal design, site-specifi c interface shear strengths shall be used. Internal shear strengths of unreinforced geosynthetic clay liner (GCL) are less than those of reinforced GCLs. Unreinforced GCLs are not recommended for slopes steeper than 1:10 (1 Vertical and 10 Horizontal). Peak shear strength of interface and internal GCLs can be used in bottom liner; residual shear strength of interface and internal GCLs shall be used for geosynthetic placed along the slopes. Site-specifi c shear strengths of waste are recommended to be used in the design. Landfi ll failure could be triggered by static loadings including excessive leachate, pore pressure above the bottom liners, gas pressure, and excessive wetness of the geomembrane-GCL, and earthquake loading. The factor of safety of 1.5 is recommended for static loading and 1.0 for earthquake loading. A higher factor of safety is recommended if a failure could have a catastrophic effect on human health or the environment, and if large uncertainty exists in input parameters to calculate the factors of safety. The main objective of this review article is to provide a comprehensive knowledge of slope failure mechanisms, causes, and probable remedies in one place. Review Article Use of Geosynthetic materials in solid waste landfi ll design: A review of geosynthetic related stability issues Lin Zhao1 and MA Karim2* 1Graduate Student, Department of Civil and Construction Engineering, Kennesaw State Universality Marietta Campus, 1100 South Marietta Parkway, Marietta, 30060, Georgia 2Associate Professor and Assistant Department Chair, Department of Civil and Construction Engineering, Kennesaw State Universality Marietta Campus, 1100 South Marietta Parkway, Marietta, 30060, Georgia *Address for Correspondence: MA Karim, Associate Professor and Assistant Department Chair, Department of Civil and Construction Engineering, Kennesaw State Universality Marietta Campus, 1100 South Marietta Parkway, Marietta, Georgia 30060, Tel: 470-578-5078; Email: mkarim4@kennesaw.edu; makarim@juno.com Submitted: 11 June 2018 Approved: 21 June 2018 Published: 22 June 2018 Copyright: 2018 Zhao L, et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Introduction
Geosynthetics are most commonly used in solid waste land ills to protect surface water and groundwater due to their multiple functions, excellent hydraulic properties, ease of installation, and cost saving [1]. Geosynthetic used in land ill include nonwoven geotextile, woven geotextile, geogrid, geomembrane, geocells, geosynthetic clay liner (GCL), geonet, geocomposite, etc, and each geosynthetic type serves as speci ic functions. The functions that geosynthetic system can serve in land ill are separation, drainage, iltration, hydraulic barrier, gas barrier, and protection [1]. Table  1 summarizes geosynthetic types and their functions, and Figure 1 illustrates their use in land ills. Geomembrane, GCL, geopipe, and geotextile are more commonly used geosynthetic materials in land ill applications while geonet, geocomposite, and geogrid are used a lesser extent [1].
While geosynthetic system provides huge economic and technical advantage over traditional liners, it may create stability issue and even lead to land ill failure if improperly designed and/or constructed. One of the most important problems associated with the use of geosynthetics for land ill linings is their stability. This becomes a very important issue by the fact that more and more land ills are designed and constructed with a small footprint and require moderate to steep slopes to raise their capacity [3]. There have been massive failures of land ills related to geosynthetic systems. For example, Koerner and Soong [4] reported ive failed land ill sites that contained geomembrane liners with volumes from 60,000 to 1,200,000 m 3 . These land ill failures caused dramatic damage to the environment, and resulted in litigations and ines. The ive failure scenarios are presenmted in igure 2. In this respect, evaluation of geosynthetic related stability is a critical consideration for land ill design, construction, and operation. As such, this review paper is concerned on failure mechanisms, shear strengths, triggering factors of geosynthetic-lined land ill failures and design criteria.

Landfi ll slope failure mechanisms
Land ill slope failure can be classi ied as two major types: rotational failure and translational failure. Translational failures are more prevalent in land ills containing geosynthetics while rotational failure is more common in land ills without geosynthetics. This is because geosynthetic-geosynthetic interface, internal GCLs, geosynthetic-waste interface and geosynthetic-soil interface are weaker in shear  strengths than waste materials and foundation soils, and thus represent weaker planar failure surfaces in land ills, at which land ill slide can occur [5]. The land ills without geosynthetics contain relatively uniform materials without a weaker prevalent planar surface. Translational failures tend to occur when dissimilar materials are involved while rotational failures tend to occur through a relatively uniform material [5].

Case Histories
Richardson et al. [9], reported a shallow translational slide occurred during construction of one-six-hare inal cover project. The slope on which the slide occurred is inclined at 14 degrees with an 1V:4H (1 Vertical and 4 Horizontal), and the slope is 60 feet high with no benches. The cover system consisted of topsoil, vegetative soil, drainage sand, polyvinyl chloride (PVC), geomembrane, GCL, and gas relief system from top to bottom. Failure of the cover system took place when the drainage sand layer. The sand above the geomembrane and the geomembrane moved downslope along the geomembrane/GCL interface [9]. The cover system failure was induced by a combination of low interface shear strength of geomembrane-GCL interface, excess pore pressure, and gas pressure below the GCLs [9].   The Kettleman Hill Land ill failure is one of the most famous deep-seated translational slide land ill failures, which has been extensively studied by various authors [3]. The Kettleman Hills Unit B-19, Phase 1-A land ill had an area of about 120,000 m 2 and was part of a waste treatment and storage facility at Kettleman Hills, California. The land ill has straight-sided liner with an oval-shaped bowl, which had a nearly level base. The solid waste placement and soil cover construction began in 1987 and progressed at an essentially steady rate. The land ill slid on March 19, 1988, resulting in a lateral movement of the ill of 35 feet towards the southeast. The vertical displacement was up to 14 feet along the back of the sliding mass. Surface cracking and tears and displacement of the geosynthetic liner were clearly visible. The sliding mass slid along the liner system. There were no rain, earthquakes, or other triggering events during the failure. Many studies believe that the failure occurred because the waste ill placed on the top of land ill reached a height that created a marginal stability of land ill slope. Major reasons that could contributed this land ill failure include that (1) the friction angle could be as low as 8 degrees between layers of geosynthetic materials; the interface between geomembranes and compacted clay could have only a few hundred pound per square foot; (2) the over-wetness yielded a very ow shearing strength at geomembrane/clay interfaces; and (3) the conditions at which the liner interface strengths were tested were different from the ield conditions, and the interface shear strengths could not represent the site-speci ic shear strength [3].

Shear strengths of interface and waste and internal shear strength of GCL
Slope failure in geosynthetic lined land ills, as discussed in the previous sections, could occur along the following weak planes: 1. Soil-Geosynthetics Interface; 2. Geosynthetic-Geosynthetic Interface; and 3. Geosynthetic-Waste Interface.

Interface shear strengths
Bouazza et al. [1], summarized the ranges of interface strengths between geosynthetics collected from the literature ( Table 2). Table 2 shows a very wide range of variations in interface shear strengths, which is due to different types of the geosynthetics materials, different testing conditions, testing protocols and testing equipment [1]. As such, interface shear strength from the published values are not recommended to be used in the land ill design; however, they can be used in a preliminary design. The site-speci ic testing shall be performed to obtain interface shear strength on a site-speci ic basis for design. Figure 6 presents comparison of failure envelope for smooth and textured geomembrane-geotextile interfaces, and peak shear strength and residual strength. Textured geomembrane/geotextile has greater friction angle than smooth geomembrane-geotextile. If a higher shear resistance is desired for the design, textured geomembrane can be used to replace the smooth geomembrane.
One major issue with geosynthetics placed in slopes is their strain softening behavior when subjected to shear forces [1]. When sheared, the peak interface shear strength is mobilized within few millimeters, and then its shear strength decreases to a residual strength, which is signi icantly less than the peak shear strength. This raises one question for design engineers: if the peak shear strength or residual shear strength shall be used for the interface and internal geosynthetic materials during the land ill design. Jones and Dixon [9] and Gilbert [12] stressed the importance of residual strength resistance and its implication on design. Gilbert [12] proposed that both peak and residual strengths are used for land ill design. The peak shear strength can be used to ind slippage location, while the residual strengths can be used to evaluate the stability of geosynthetic lined land ill. It is common practice to use the peak shear strength for the interface at the loor and the residual shear strength value shall be used for the geosynthetic placed on side slope [5].

Internal shear strength of GCL
When GCLs are used in a land ill, one slope stability concern is mid-plane shear through the bentonite layer. Peak shear strengths for the unreinforced GCL products are similar to those for sodium bentonite, which has very low shear strength, and makes them prone to sliding. Furthermore, shear strength of hydrated bentonite could have a friction angle as low as 6 degrees [13]. Because of this low friction angle, unreinforced GCLs are usually not recommended for slopes steeper than 1V:10H [14].
To increase the internal shear strength of GCLs, GCL manufacturers have created "reinforced" GCLs in which the two outer layers of geotextile are either needlepunched or stitched together through the bentonite layer [13]. Reinforced GCLs have greater internal peak strength due to the presence of iber reinforcements. The peak shear strength of different types of reinforced GCLs (needle-punched, thermal bonded, stitch-bonded) may differ signi icantly [15].  Figure 7 shows the results of the internal shear strengths of both reinforced and unreinforced GCLs [15]. The frictional angles of reinforced GCLs have a wide range from 10 to 45 degrees, while unreinforced GCLs have frictional angles typically less than 15 degrees.

Shear strengths of waste
The measured shear strengths obtained from the literature have a wide range. The reported friction angles range from 15 degrees to 42 degrees with cohesion ranging 0 to 28 kPa [8]. This large scatter is due to the large variety and heterogeneous of wastes, which results in the dif iculties to obtain representative shear strength. Jones and Dixon [8] suggested municipal solid waste (MSW) shear strength envelopes for design as shown in igure 8.
Due to large variation of shear strengths, site-speci ic data is recommended to be used in the design of land ill. However, if no site speci ic is available the approach proposed by Van Impe and Bouazza [16] can be used as a starting point in design. This approach de ines design values of cohesion (c) and friction (φ) into three distinct zones:  • "Zone A: corresponding to very low stress (0 kPa ≤ σ v < 20 kPa) where the MSW behavior can be described as being only cohesive. In this case, c = 20 kPa." • "Zone B: corresponding to low to moderate stresses (20 kPa ≤ σ v < 60 kPa). In this case, c = 0 kPa and φ ≈ 38°." • "Zone C: corresponding to higher stresses (σ v ≥ 60 kPa). In this case, c ≥ 20 kPa and φ ≈ 30°."

Triggering mechanism
Land ill failures are often induced by processes that either increase the driving force or decrease the resistant force along the weak plane, typically an interface between two geosynthetics, or an interface between a geosynthetic or both [5]. Major contributing causes of land ill failures include the following: 1. Rise in leachate level within the waste mass that exceeds the maximum allowable level; 2. Excessive buildup of leachate level due to liquid waste/leachate injection that exceeds the maximum allowable level; 3. Excessive pore pressure buildup in the inal cover system that exceeds the maximum allowable level; 4. Excessive gas buildup below inal cover system; 5. Excessive wetness of the Geomembrane-GCL interface resulting in a lower shear strength; and 6. Earthquake or blasting.
The above causes can be grouped into two loading cases: static loading (Cause No. 1 thru 5) and seismic loading (Cause No. 6).

Stability design criteria
According to Ohio EPA [5], the following factors of safety (FS) should be used to evaluate land ill stability: 1. Static analysis: FS > 1.50 2. Seismic analysis: FS > 1.00 The use of higher factors of safety is recommended if a failure could create catastrophic effect on the environment and human health, and if large uncertainty exists in input parameters to calculate the factors of safety [5].

Conclusions and Recommendations
A number of technical papers related to stability of geosynthetic-lined land ill have been reviewed, and the following conclusions and recommendations were made: 1. Geosynthetics used in land ills provides an economic and technical advantages over traditional liners. But it may create stability issue and even lead to land ill failure due to its low interface or internal shear strength if improperly designed and/or constructed.
2. The most common failure mechanism in geosynthetic-lined land ills is transitional failure involving waste and bottom liner (deep-seated failure) or only inal cover system (shallow failure).
3. Shear strengths of geosynthetic-geosynthetic and geosynthetic-soil have a wide range of variations. Shear strengths of interface from literature may be used in preliminary design. For inal design, site-speci ic interface shear strengths shall be used. 4. Internal shear strength of unreinforced GCLs are less than those of reinforced GCLs. unreinforced GCLs are usually not recommended for slopes steeper than 1V:10H.
5. Peak shear strength of interface and internal GCLs can be used in bottom liner; residual shear strength of interface and internal GCLs shall be used for geosynthetic placed along the slopes.
6. The measured shear strengths of waste obtained from the literature have a wide range with reported friction angles ranging from 15 degrees to 42 degrees and cohesion ranging 0 to 28 kPa. Due to large variation of shear strengths, site-speci ic data is recommended to be used in the design of land ill. If no site speci ic is available the approach proposed by Van Impe and Bouazza [16] can be used as a starting point.
7. Land ill failure could be triggered by static loadings including excessive leachate, excessive pore pressure above the bottom liners and excessive gas pressure and excessive wetness of the geomembrane-GCL, and earthquake and blasting loading.
8. The factor of safety of 1.5 is recommended for static loading and 1.0 for earthquake loading. A higher factor of safety is recommended if a failure could create a catastrophic effect on the environment or human health, or if large uncertainty exists in input parameters to calculate the factors of safety.