Computer Aided Design and Analysis of Geosynthetic Landfills

Abstract

Several huge failure of soil cover, geosynthetics lined side slopes and within the waste mass itself have occurred. Investigations of the some prominent failures have shown that sliding may occur at the weakest interface between the components of the liner system and a lateral translational failure can also occur with the solid waste when sliding occurs immediately above, within, or immediately beneath the liner system at the base of the waste mass. It is also generally agreed that significant tension can develop in the geosynthetic layers when they are subjected to unbalanced interface friction forces. Therefore interface friction values between geosynthetics/soil and geosynthetics/geosynthetics were obtained by large scale direct shear test. Several design models for stability of the overlying materials have been done by several authors. But not many and appropriate efforts have been done in the direction to develop the software based on all these design models. Authors developed the user friendly software with technical guidelines. Since limit equilibrium approach ignores displacement compatibility so a simple composite column method is also reported in this present paper.

Keywords

Landfill; Veneer stability; Shear strength; Wedges; Computer program, Design.

1. Introduction

Critical interfaces of composite geosynthetic liner include soil vs. geomembrane (GM), geomembrane vs. geotex tile (GT) and geomembrane vs. geonet. Various researchers like Martin et al. (1984), Mitchell et al. (1990), Natraj et al. (1995), Jones et al. (1998), Stark et al. (1996), Lee et al. (2000), Dhani et al. (2003), and Bergado et al. (2004), had carried out the interface friction properties between different geosynthetics and soils. However, these values are based on a wide variety of materials, test conditions, and test methods, they should not be used for design of specific projects (Giroud et al. 1989; Koerner et al. 1991; Qian et el. 2003). Because either these values are potentially over conservative and lead to uneconomical design or they unnecessary introduce a potential risk of unacceptable performance or failure of the system. Several authors (Martin et al., 1984, 1985; Giroud and Beech, 1989; Koerner and Hwu, 1991; Long et al, 1993; Druschel and Underwood, 1993; Koerner and Soong, 1998 and Mandal et al. 2000) presented Limit equilibrium methods for analyzing the unreinforced and reinforced cover soil on geosynthetic slope. Recently translational waste mass failure analysis by two part wedge method was proposed by Qian et al. (2003) and it is helpful in stability analysis of landfills in every respect.

But not many and appropriate efforts have been made in the direction to develop the software which can include stability analysis of unreinforced and reinforced cover soil, effect of seepage over the stability of cover soil, design of runout anchor trench, estimation of tensile load, waste mass failure analysis. With this in view, an attempt has been made to develop a computer program for stability analysis of landfills and design of runout anchor trench. The developed program is very user friendly and has a lot of advantages like it speeds up the calculations for the long equations and advantage of making no mistake in solving the rigorous equations.

2. Experimental investigation and testing program

Two different types of HDPE geomembranes (Smooth and Textured), two types of nonwoven geotextile (Thermally bonded and Needle punched) and one high strength woven geotextile were selected for experimentation purpose.

2.1 Testing materials

2.1.1 Properties of Sand (Ganga sand)

Test results of sand are tabulated in Table 1.

Table 1: Property of sand sample

2.1.2 Properties of Geosynthetics

(1) Geomembrane

Physical properties

The two types of geomembranes are used for conducting the experiments. The thickness and mass of smooth high density polyethylene geomembrane (HDPE) are 1.61 mm and 1550 gm/m2 respectively. The thickness and mass of the textured high density polyethylene Geomembrane (HDPE) are 1.5 mm and 1550 gm/m2 respectively.

Tensile Behavior

Procedure suggested by ASTM D412, D638 is followed for specimen preparation in dumbbell shape and testing. The dumbbell shape specimen gives one dimensional behavior because of drawing in the central region only is shown in Fig. 1(a), which is not experienced in the field situations (Koerner 1997, Koerner 1991, ASTM D5321). Therefore uniform and wider specimen test (ASTM D4885) is conducted [see Fig. 1(b)], which is suitable for plain strain condition and much more design oriented than dumbbell tensile behavior. ASTM D4885 test gives performance strength of geomembrane and determines the ability of a geomembrane to sustain the stresses and strains under design conditions.

Fig. 1 (a). Dumbbell shape
behavior during test
Fig. 1(b). Wide width shape
behavior during test

The summary of tensile behavior of HDPE geomembrane is depicted in Table 2.

Table 2: Tensile behavior properties of HDPE geomembranes

(2) Properties of geotextile

Three types of geotextiles were used to conduct the test.

(i) The properties of nonwoven needle punched polypropylene geotextile are:- mass per unit area =183.19 gm /m2; thickness =1.88 mm; grab tensile strength =63.67 kN and wide width tensile strength =17.84 kN/m.
(ii) The properties of nonwoven thermally bonded polypropylene geotextile are:- mass per unit area=212.75 gm/m2; thickness =0.88 mm, grab tensile strength =117.67 kN and wide width tensile strength =14 kN/m.
(iii) The properties of woven polyester geotextile are;- mass per unit area =289.5 gm/m2and wide width tensile strength =80 kN/m.

2.1.3 Large scale direct shear test

All tests were performed using procedure standardized within the laboratory and consistent with ASTM D5321.The low normal stresses were applied during the test due to simulate typical soil cover thickness in the field i.e. 10 kN/m2 to 40 kN/m2. Test is conducted to a relatively large displacement to determine the residual behavior. All tests were performed at a displacement rate of 0.125 mm/min. To see the effect of strain rate on interface friction parameters some tests were also conducted at strain rate 1.26 mm/min. All materials were tested in dry and wet condition (ASTM D5321). The tests were run without submerging the sample. All the tests are repeated three times and final results are averaged. Residual shear strength parameters for typical interfaces are shown in the Fig.2.

Fig.2.Normal Stress versus Shear stress relationship for typical interfaces

The summaries of interface shear strength parameters under dry and wet conditions are listed in Tables 3 to 8. Tables present the values of peak and residual strengths corresponding to each interface and show the effect of strain rates in dry and wet condition on the interface shear resistance. Additionally a statistical coefficient of regression (R2) and efficiency ratio (E) (in parentheses) are mentioned in the tables. Efficiency ratio shows the relative amount of mobilized soil strength that the geomembrane and geotextile gives.

Table 3: Geomembrane (GM) versus Geotextile (GT) interface friction angles at 0.125 mm/min in wet condition.
(GT-1: Non woven needle punched polypropylene geotextile, GT-2: Nonwoven heat bonded polypropylene geotextile).

Table 4: GM to GT interface friction angles at 0.125 mm/min in dry condition.

Table 5: Soil to Geosynthetics interface friction angles at 0.125 mm/min in dry
condition.

Table 6: Soil to Geosynthetics interface friction angles at 0.125 mm/min in wet
condition.

Table 7: Soil to Geosynthetics interface friction angles at 1.26 mm/min in wet
condition.

Table 8: Geosynthetics to Geosynthetics interface friction angles at 1.26 mm/min
in dry condition.

2.1.4 Results and discussions

The following sections discuss the test results:

1. The general prediction from Table 5 is that the friction between soil and geotextiles or smooth geomembranes is less than that of soil itself. Soil/geotextile friction generally exceeds Soil/Smooth geomembrane friction. Therefore placement of a geotextile over or under a smooth HDPE liner will tend to allow a steeper slope, provided that both fabrics are securely anchored. If anchor fails, then the safe slope angle obviously decreased.

2. Geotextile (GT) and Geomembrane (GM) interface. Tables 3, 4 and 8 illustrate the most critical interfaces with the lowest frictional resistance in the landfill were the GT1/Smooth GM and GT2/smooth GM. These combinations had the interface friction angles of about 120 to 110 respectively. These results are comparable to the results obtained by Martin et al. (1984), Mitchell et al. (1990), Long et al. (1993), and Bergado et al. (2004). Therefore this interface is usual location of the potential slip failure. The peak Shear stress were mobilized at displacement at less than or equal to 3 mm. The difference between dry and wet condition for GT/ Smooth GM had been observed ± 10 that are comparable to Stark et al. (1996), Mitchell et al. (1990).

The interface angle also depends on the thickness of the geotextile. According Stark et al. (1996), generally lower unit weight and thinner geotextile gives greater peak shear stresses for geotextile/geomembrane interface compared to the thicker geotextile, However, the frictional resistance obtained by Stark et al. (1996) using thicker and thinner geotextiles with smooth HDPE GM almost yield similar results with the difference of less than (0.5 to 1.50), this result also depicted in Table 4, by comparing the GT1(1.88 mm thick)/smooth HDPE interface which show 12.10 interface angle and GT2 (0.88 mm thick)/smooth HDPE interface shows 10.60 interface angle in dry condition.

Table 3 demonstrates that the interface in wet condition displayed slightly higher peak stress than the corresponding dry condition (Bergado et al., 2004).

As expected, the interface friction angle of geomembranes increases with an increase in roughness resulting from intentional scratches or surface texture. Therefore textured GM/GT interface shows higher interface angle (140, dry interface, see Table 4) in comparison of smooth GM. Specially in the case of wet condition textured GM/GT1 shows relatively higher angle 18.50 as per Table 3. This behavior is because of stretching and strong contact with GT1 (NW geotextile) in wet condition. Hence textured GM is good option for liner where geotextile is placed over the geomembrane.

3. Geotextiles and Sand interface: In general the Mohr-coulomb failure envelopes for the sand with geotextiles pass near the origin or shows slight adhesion as per Fig. 2. The interface friction angle for sands with geotextiles is less than that of the angle of internal friction angle of sand alone. Table 5 also reveals that sand to geotextile friction generally exceeds sand to smooth geomembrane friction. The relative efficiencies of all geotextiles and particularly, the needle punched nonwoven fabric, are particularly high. Unlike all other cases, the peak shear stress is mobilized between 5.4 mm to 8 mm deformation. This behavior may be due to the entrapment of the sand particles in to the surface openings of the geotextile. In geotextile/sand interface, the geotextile was stretched due to sand particle interlocking with the geotextile during the test. These results are comparable to those obtained by other researchers, Martin et al. (1984), Williams (1987) and Nataraj (1995). No strain softening was observed in tests with the nonwoven needle punched polypropylene geotextiles. It is shown that geotextiles can mobilize large amount of friction.

4. Geomembranes: All the Mohr-Coulomb envelopes for smooth geomembranes in contact with sand pass through the origin with no measurable adhesion the range is 0.045 to 0.055 Kg/cm2. Therefore adhesion is neglected unless otherwise specified (Koerner 1998 and Martin 1984). In smooth geomembrane peak shear resistance is mobilized at deformations less than 3 mm in dry interface case and less than 1 mm in wet interface case. This behavior suggests the slip phenomenon between smooth surfaces of geomembrane with the fine grains of sand. Surface roughness is not induced by normal stress on the smooth HDPE to soil interface, and low friction angles and relative efficiency indexes result in comparison of geotextile/sand and textured geomembrane/sand interfaces. It would appear that it is necessary to anchor a geotextile over the material in order to build a steep slope with smooth HDPE. Fig. 2, shows that textured geomembrane interface provide higher interface friction angle than others. This may be due to the embedment of the sand particles on the surface of the geomembrane. Much larger displacements are required to mobilize peak shear stresses (Tables 5 and 6), which are higher than smooth geomembrane. This behavior proves that textured surface greatly improves slope stability by increasing friction between the geomembrane and soils, geotextiles and other geosynthetics. This performance of textured geomembrane enables design engineers to improve factor of safety on slopes of varying steepness. However with textured geomembrane a greater decrease in shear stress from peak (strain softening) is experienced in experiments by increasing horizontal displacement like Russell (1998).

3. Computer program

Following Methods are included in computer program:

1. Stability analysis of reinforced and unreinforced cover soil;

• Model proposed by Koerner (1991) and Soong (1998).
• Tensile stresses induced in the geomembrane (Koerner and Hwu 1991; Bourdeau and Ludlow 1993).

2. Design of anchor trench;

• Model proposed by Koerner (1997) and Qian (2000).

3. Stability analysis of cover soil in seepage condition by Soong model (1998).
4 Translational failure analysis of waste mass (Qian 2003).

Use of proposed computer program are advantageous in reference of that it is very much user friendly, provides technical guidelines for user when necessary, help in selecting the input values. It also shows that computed magnitude of forces which is used in receiving the final results. It includes waste mass failure analysis and tensile stresses induced in the geomembrane due to unbalanced frictional forces which are not available in HELP model. It was tried that limitations pertaining to the HELP model probably overcome by the proposed program. The proposed IITB model gives the following various options as per flow chart given in Fig. 3.

Option A of computer program deals with unreinforced and reinforced veneer slope stability problem and produce factor of safety (FS) value against cover soil sliding on the geomembrane. The results are matching with HELP model which is proposed by USEPA. When the FS value less than one, reinforcement is required to prevent the sliding of the cover soil and FS value must be greater than one with same reinforcement. In HELP model trial and error procedure are required to get the desired value of the FS (at least greater than one) with reinforcement and sometimes it generate negative value of the FS with reinforcement. Sometimes it produces negative or less value in reinforced case than unreinforced case. The example is given below.

Example 1: Given a cover soil slope of ß = 18.4 (i.e. 3 to 1), length of slope L = 100 m, thickness of cover soil H = 0.9 m, unit weight of cover soil = 18 kN/m3, cohesion of cover soil c = 0 kPa, adhesion between cover soil and geomembrane Ca = 0, friction angle of cover soil = 320, interface angle between cover soil and geomembrane = 14, determine the resulting FS.

Example 2: Given a cover soil slope of ß = 18.4 (i.e. 3 to 1), L = 30 m, H = 0.3 m, = 18 kN/m3, c = 0 kPa, Ca = 0, = 30, = 22, determine the resulting FS.

Option B suggests that run out length of rectangular anchor trench is straightforward to obtain the anchorage ratio one. This will give balance design with satisfying the criteria of both pullout and rupture of geomembrane. It also indicates the limiting value of tensile stresses of geomembrane when run out length is not required. On the other hand HELP model requires various trials to achieve anchorage ratio one.

Option C gives FS value in seepage conditions against sliding of cover soil in both cases when horizontal seepage build up and parallel-to-slope seepage build up. This option provides three ways of input the data for particular problem, which makes suitable working of users. This option is not available in HELP model.

Option D is the waste mass failure analysis by two part wedge method and this is not available with HELP model.

Tensile load in geosynthetic components above plane of slippage was computed by simple composite method (Gilbert et al. 2003) which exhibits comparable results with field and rigorous finite elements results.

Conclusions

Based on research work following observations have been made:

• Based on large scale direct shear test it has been observed that the weakest interface in the geosynthetic liner is in between smooth geomembrane with geotextiles; and the strong interface is in textured geomembrane with sand and/or geotextile with sand. The interface friction angle with sand has found to be mainly dependant on seepage conditions resulting into at least 40 to 50% variation in its values. It also depends on texture of the geomembrane. Interface angle with respect to sand of textured geomembrane has found to increase up to 60 to 70% compared with smooth geomembrane; hence it can provide higher stability against steep slopes. Effect of rate of strain has also been compared with interface friction angle. It can be concluded that the high rate of strain shows higher value of interface friction angle.

• Proposed computer program is user friendly and can be used efficiently for analyzing and designing the landfill components than that of HELP model as per given options A, B, C and D. It provides more information than HELP model and no trial is required to get the desired output. It is also helpful in parametric studies.

References

1. Bergado D. T., Youwai S., and Voottipreux P., (2004) “Interface shear strength of different Geosynthetics in landfill liner syatem: A case of Sakaew landfill, Thailand” Geoasia-204, 507-514.

2. Blumel, W. and Stoewahse, C., (1998) “Geosynthetic Interface Friction Testing in Germany- Effect of Test Setups,” Proc. 6 th IGS, IFAI, Roseville, MN, pp.447-452.

3. Bourdeau, P. L., Ludlow, S. J., and Simpson, B.E. (1993) “Stability of soil covered Geosynthetics-Lined Slopes: A parametric Study” proc.,Geosynthetics’93 conference, Vancouver, Canada, IFAI, Roseville, Minn., 1511-1521.

4. Dhani, Narejo B. (2003). “A simplified tilt table device to measure index friction angle of geosynthetics”, j. Geotextile and Geomembrane, (21), 49-57.

5. Druschel, S. J. and Underwood, E.R.,(1993): “Design of Lining and Cover System Side slopes”, Proceedings of Geosynthetics Confrence’93 , IFAI, Vol. 3, Vancouver, B.C., Canada, March 1993, pp. 1341-1355.

6. Eid, H. T., Stark, T. D., Evans, W. D. and Sherry, P. E., (2000) “Municipal Solid Waste Slope Failure. I: Waste and Foundation Soil Properties.” J. of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 126, No. 5, May, pp. 397-407.

7. Giroud, J. P. and J. F. Beech, (1989) “Stability of soil layers on geosynthetic lining systems”, Geosynthetics’89 Conference Proceedings , IFAI, San Diego, CA, pp.35-46.

8. Gilbert, R. B., Long, J.H., Daly, J.J.,(1993) “Structural integrity of composite geosynthetic lining and cover systems”, Proceedings of Conference Geosynthetics ’93, Vancouver, Canada , pp. 1389-1401.

9. Gilbert, R. B., Liu, C. -N., (2003) “Simplified method for estimating geosynthetic loads in landfill liner side slopes during filling,” Geosynthetics international,10,No.1, 24-33.

10. Jones, D. Russel V. and Dixion, N., (1998) “Shear strength properties of geomembrane/geotextile interfaces”, J. Geotextile and Geomembrane , “(16), 45-71.

11. Koerner, R. M., Hwu, B.L., (1991) “Stability and Tension Considerations Regarding Cover Soils on Geomembrane Lined Slopes”, Journal of Geotextiles and Geomembranes, Elsevier science publishers Ltd., England, Vol. 10, pp. 335-355.

12. Koerner, R. M., and Soong, T.-Y., (1996) “Seepage Induced Slope Instability”, Geotextiles and Geomembranes, 14,pp.425-445.

13. Koerner, R. M. (1997): “Designing with Geosynthetics”, 4th Ed. Prentice Hall Inc, Englewood Cliffs, NJ, U.S.A.

14. Koerner, R. M., and Soong, T.-Y., (1998) “Analysis and design of veneer cover soils,” Proc. 6 th IGS, IFAI , Roseville, MN, pp.1-26.

15. Lee, M. K. and Manjunath, R. V., (2000) “Soil-geotextile interface friction by direct shear test,” Canadian Geotechnical Journal, 37: 238-252.

16. Liu, C. N., Gilbert,R. B., Thiel, R. S., and Wright, S. G., (1997) “What is an appropriate factor of safety for landfill cover slopes,” Geosynthetics ’97 Conference Proceedings, 1, 481-496.

17. Long, J. H., Gilbert, R.B. and Daly, J. J.,(1993) “Geosynthetic Loads in Landfill Slopes: Displacement Compatibility”, Journal of Geotechnical Engineering, ASCE, Vol. 120, No. 11, pp. 2009-2025.

18. Mandal N. J. and Suresh V., (2000) “Design of Veneer cover soils on landfill”, Indian Geotechnical journal, (31), pp. 60-87.

19. Martin, J. P., Koerner, R. M., and Whitty, J. E., (1984) “Experimental friction evaluation of slippage between geomembranes, geotextiles and soils,” Proc., Int. Conf. on Geomembranes, Denver, IFAI, Colorado, 191-196.

20. Martin, J. P., and Koerner, R. M., (1985) “Geotechnical Design Considerations for Geomembrane Lined Slopes: Slope Stability” Geotextiles and Geomembranes, 2 (1985) 299-321.

21. Mitchell, J. K., Seed, R.B., and Seed, H. B. (1990) “Kettlement Hills Waste landfill Slope Failure I: Liner-System Properties”, J. Geotech Engg ., ASCE 116(4), pp. 647-668.

22. Natraj, M. S., Maganti, S. R., and Mcmanis, L. K., (1995) “Interface Frictional Characteristics of Geosynthetics,” Geosynthetics’95, pp.1057- 1069.

23. Qian, X., Koerner, R. M., and Gray, D. H. (2001) Geotechnical aspects of landfill design and construction, Prentice-Hall, Upper Saddle River, N.J.

24. Qian, Xuede, Koerner, R. M., and Gray, D.H., (2003) “Translational Failure Analysis of Landfills,” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Volume 129, No.6, pp. 506-519.

25. Snow, M., Mansour, R., Swan Jr., H. R. and Kavazanjian, E., (1998) “ Variability of Interface Shear Strengths,” Proc. 6 th IGS, IFAI , Roseville, MN, pp.439-444.

26. Stark, T. D., Williamson, T. A. and Eid, H. T., (1996) “HDPE Geomembrane/Geotextile interface shear strength”, J. of Geotechnical and Geoenvironmental Engineering, ASCE, (122),197-203.

27. Williams, N. D. and Houlihan, M. F., (1987) “Evaluation of interface friction properties between geosynthetics and soils”, Proceedings of Geosynthetics’ 87 , New Oreleans, pp. 616-627.


Special thanks to Professor Ronald Mersky and the The Journal of Solid Waste Technology and Management Department of Civil Engineering, Widener University for granting permission to reprint this paper from the 20th International Conference on Solid Waste Technology and Management, held 3-6 April 2005 in Philadelphia, PA.

0

Leave a Reply

Your email address will not be published. Required fields are marked *