Geotechnik im Bauwesen

Geotechnical Engineering

Univ.-Prof. Dr.-Ing. Martin Ziegler

Investigation of the Confining Effect of Geogrid Reinforced Soil with Plane Strain Model Wall Tests
ABSTRACT Large scale laboratory testing of unreinforced and geogrid reinforced soil has been carried out at RWTH Aachen University to investigate the confining effect of geogrids in reinforced soil in principle. With a series of triaxial tests the development of the additional confining effect due to the reinforcement has been confirmed. This confining effect is necessary for the increase of the load bearing capacity of the soil. Furthermore, to determine the development of shear zones within the geogrid reinforced soil a test apparatus with a transparent sidewall has been constructed to monitor soil particle displacements and rotations during a controlled facing movement. The evaluation of the soil particle movements confirmed the development of the confining effect and explains the development of an earth pressure from the composite material which is far less than the active earth pressure from the unreinforced soil.

1. Introduction
The immense contribution of geogrids to the strength of reinforced soil is well known in both science and practice. The vast amount of research that has been carried out in this area is reflected in the German design standard for geosynthetic reinforced soil (EBGEO, 2009) that has been updated recently. The earth pressure acting on the facing of geogrid reinforced retaining walls is assumed herein to be far less as it used to be in the former version and as suggested by other standards, such as (BS 8006, 1995). The beneficial effect of geogrids to the stability of soil structures as well as economic factors also lead to an increasing acceptance of geogrid reinforced soil as a construction technique. (Ziegler, 2009) reported growth rates of up to 15 % for the use of geogrids in Germany. The mechanism of the geogrids leading to the increased strength of the soil has been identified as confining effect (Ling & Tatsuoka, 1994). To investigate this confining effect in principle and especially in order to visualize it, large scale triaxial and plane strain laboratory testing has been carried out.

2. Triaxial testing
Large scale triaxial tests have been carried out to investigate the influence of the stress level, the type of filling soil and the vertical reinforcement spacing to the reinforcing effect of geogrids. The dimensions of the triaxial cell have been chosen to allow testing at 1:1 scale.

Test Set-up
The soil specimens were 500 mm in diameter and approximately 1100 mm high (Fig. 1). The confining pressure has been applied by vacuum. Two types of soil have been tested, i.e. a crushed base course material (d50 = 12 mm) and a medium sand (d50 = 0.5 mm) that have been compacted to both 95% and 100% proctor density.

Geotechnik im Bauwesen
Geotechnical Engineering

Univ.-Prof. Dr.-Ing. Martin Ziegler

Tests have been conducted with a constant strain rate of 0.1 %/min, while plunger load, geogrid strains and average radial strains have been recorded. The analysis made in this paper is based on results of tests with a biaxial polypropylene geogrid with 30 kN/m nominal strength made of welded pre-stretched flat bars. The aperture size of the grid was 32 mm x 32 mm and the tensile force at 2 % strain was 12 kN/m as stated by the manufacturer.

Test Results
The test results that have been reported earlier by (Ruiken & Ziegler, 2008) indicate a significant increase of the bearing capacity and reduction of deformations due to the geogrid reinforcement. The distribution of average radial strains over the height of specimen with a varying number of reinforcement layers has shown that the development of radial strains is reduced significantly due to the geogrids, even without attaching the geogrids to the elastic cover.

Fig. 1. Principal sketch of triaxial cell.

Fig. 2. Deformed unreinforced and reinforced triaxial specimens at the same load level and corresponding mechanical models.

This confining effect of the geogrids, that has been observed also by (Moghaddas-Nejad & Small, 2003) and (Eiksund et al., 2004), can be explained with the mechanical model given in Figure 2. The effect is similar to an additional confining pressure 3 acting homogeneously over the whole height of the specimen if the vertical spacing between the reinforcement layers is small enough. The magnitude depends on the degree of activation of the geogrids. Corresponding stress paths that take the additional confinement into account have been determined and drawn for the reinforced triaxial tests in (Ziegler & Ruiken, 2009).

Geotechnik im Bauwesen
Geotechnical Engineering

Univ.-Prof. Dr.-Ing. Martin Ziegler

3. Plane Strain Testing

Additional large scale plane strain testing of geogrid reinforced soil has been carried out to determine the development of shear zones within the geogrid reinforced soil and to investigate the advantageous reinforcement effect of reducing the lateral earth pressure on the facing, which has been suggested various times, e.g. (Clayton et al., 1993) and (Soong & Koerner, 1997), and which has recently been measured by (Yang et al., 2009). The basic mechanism suggested in (Soong & Koerner, 1997) for the stress condition of the backfill soil directly behind the facing corresponds with the effect that has been observed during triaxial testing, i.e. a zone of unconfined soil between the geogrid layers, which is therefore not restricted in deformation by the composite material itself, but has to be supported externally either by the confining pressure at triaxial testing or by the front facing of geogrid reinforced retaining structures, respectively.

Fig 3. Apparatus for plane strain testing of reinforced soil.

Test Set-up
The apparatus shown in Fig. 3 has been constructed to carry out reinforced soil wall model testing under plane strain conditions as well as plane strain element testing under a constant confining pressure. The maximum dimensions of the soil specimen are (H WD) 1m x 1m x 0.45m. Front and back facing can be retracted independently, whereas the fixed sidewalls provide the plane strain conditions. The 106 mm thick glass sidewall is designed for a maximum deflection of 0.1 mm under surcharge loads up to 50 kPa that are being applied uniformly with a compressed air cushion. The geogrids can be installed either without or with connection to the facing, where the connection loads can be recorded. Test have been conducted with the above described sand, which has been deposited with a rainfall technique to more than 100 % proctor density. The friction parameters have been determined with direct shear tests to 7 between sand and glass at the sidewalls and to less than 2 between the facing elements and a thin latex membrane lubricated with silicone grease.

Geotechnik im Bauwesen
Geotechnical Engineering

Univ.-Prof. Dr.-Ing. Martin Ziegler

Test Results
Horizontal earth pressure development The horizontal earth pressure distribution over the height of the facing has been obtained from 20 elements (each 5 cm high). Redundancy was achieved by comparing the sum with the result of one large load cell supporting the whole facing. In Figure 4 the earth pressure development is given for unreinforced and reinforced specimens, where the geogrids were not attached to the facing. The curves start with a difference in the initial values. This is due to slight movement (up to 0.7 mm) of the facing elements during loading, which results in the activation of the geogrids already during the loading sequence. For the comparison always the sum of the lateral stresses acting on the whole facing has been considered.

Fig. 4. Development of the horizontal earth pressure (grids not attached to the facing). The results indicate a significant reduction of the active earth pressure down to 40% of those of the unreinforced soil, even without connecting the geogrids to the facing. The effect increases with an increasing number of reinforcement layers. (Tsukamoto et al., 1999) also carried out plane strain model wall tests with geogrid reinforced sand and observed similar significant reductions for the coefficient of the lateral earth pressure Ka. The distribution of the earth pressure over the height of the facing indicated a significant reduction of the lateral stresses below the upmost geogrid layer. The coefficient of the lateral earth pressure Ka seems to decrease therefore with depths. This observation supports the assumption made in the recently updated (EBGEO, 2009) for the earth pressure acting on the back of the facing of geogrid reinforced retaining structures with deformable facing elements, i.e. Ka/2 at the lower 60 % of the structure.

Geotechnik im Bauwesen
Geotechnical Engineering

Univ.-Prof. Dr.-Ing. Martin Ziegler

Particle displacements and rotations In order to visualize the confining effect of the geogrids and the development of shear zones within the reinforced soil, digital pictures of the specimens have been taken during testing. These pictures have been processed afterwards evaluating particle displacements and rotations (PIV-method). The digital camera used for this purpose was remote controlled to avoid misalignment from touching for the release and had a fixed focal length lens to achieve a minimum distortion. The resolution of the photographs has been identified as being the parameter with the biggest influence for the quality of the PIV-evaluation. Therefore, the camera had a resolution of 15 Mpixel.

Fig. 5. Shear zone development from particle rotations (PIV) in unreinforced and reinforced specimens (grids not attached to the facing). The evaluation of soil particle rotations after a facing retraction of 10 mm is shown for an unreinforced specimen as well as for specimens reinforced with two and five geogrid layers in Fig. 5. The soil particles in the slip surface rotate counter-clockwise indicating a downward movement of the failure wedge behind the facing. It can clearly be seen that an increasing degree of reinforcement results in a decreasing distance between slip surface and front facing. The small failure wedges of the reinforced specimens cause less pressure on the facing as those of the unreinforced soil, which corresponds to the results of the lateral pressure presented in Fig. 4. The confining effect of the reinforcements is of major concern in this study. Therefore, by focusing on a detail section of the specimen (0.3 m x 0.2 m) at the height of the upper geogrid at the facing high resolution photographs of the sand were achieved. The total particle displacements within the same detail section are shown in Fig. 6 for the unreinforced and for the reinforced soil. The results clearly confirm the closer position of the slip surface to the front in the reinforced soil.

Geotechnik im Bauwesen
Geotechnical Engineering

Univ.-Prof. Dr.-Ing. Martin Ziegler

Fig. 6. Total particle displacements of unreinforced and reinforced soil within the detailed section at the facing. In Fig. 7 the horizontal displacements (left plot) and the soil particle rotations (right plot) are given for the detail section of the reinforced specimen after 10 mm facing retraction. With regard to the total displacements and the rotations of the sand particles, it is clear that in contrast to the development of a single failure plane in the unreinforced soil, various failure planes have been developed due to the geogrid. However, it has to be noted that the various slip surfaces do not appear at once but develop one after the other, always at the position with the lowest resistance of the soil. Once a sliding plane reaches a geogrid it is subsequently activated resulting in a local increase of the strength of the composite material. The failure plane then jumps to another position, wherever the resistance of the composite material of soil and geogrid is lower. Remarkable in this content is the fact that the failure plane seems to be orthogonal to the geogrid, trying to avoid its activation. In other words, in the vicinity of the geogrids the soil is confined horizontally, which becomes very apparent with regard to the evaluation of the horizontal displacements in the left plot of Fig. 7.

Fig. 7. Horizontal particle displacements and particle rotations within the detailed section at the facing. Considering the evaluation of the rotations of the sand particles in the right plot of Fig. 7, the different colours of the slip surfaces indicate a different sense of rotation of the soil particles. Primarily, this indicates relative downward movement of the soil between the pink major shear zone (A) and the facing. It is the main failure wedge. Furthermore, the orange/red (B) colour indicates a secondary sliding plane within the main failure wedge, which is leading from the free

Geotechnik im Bauwesen
Geotechnical Engineering

Univ.-Prof. Dr.-Ing. Martin Ziegler

end of the geogrid down- and inwards. The clockwise sense of rotation indicates a relative downward movement of the soil between (A) and (B), which means that the soil underneath (B) is not confined by the geogrid and therefore pushed towards the facing. For the whole model wall the mechanism observed in the detail section corresponds with an arching effect of the soil between the geogrid layers. This effect that has been assumed already from the presented results of the triaxial testing, from FE-calculations for reinforced plane strain compression tests (Peng et al., 2000) and reinforced retaining structures (Pachomov et al., 2007) can be seen in the sliding planes in Fig. 5 that have been developed in the specimen reinforced with five geogrid layers. The described mechanisms and the results of the horizontal displacements led to the idealized mechanical model presented in Fig. 8 for the reinforced soil with geogrids that are not attached to the facing. It is analogue to those derived from the triaxial tests (Fig. 2). Together with the fact that the primary sliding plane is shifted towards the facing due to the confining effect of the geogrids, it delivers the explanation why the lateral pressure of geogrid reinforced soil has to be far less than those of the unreinforced soil.

Fig. 8. Mechanical model for the reinforced soil (geogrids not attached).

Summary and Discussion

The confining effect of the geogrids has been shown with large scale triaxial tests and has been visualized in plane strain model wall testing. Both types of testing have shown a considerable reduction of deformations in the vicinity of the geogrids. The confining effect led to a significant increase of the bearing capacity of the triaxial specimen. During plane strain model wall testing, the horizontal earth pressure on the facing has been reduced by the geogrids. An arching effect and the corresponding zone of the soil between the geogrids that cannot be supported has been identified. From a soil mechanical point of view the assumption of a very low to moderate horizontal earth pressure on the back of the facing of reinforced retaining structures with deformable facings is therefore justified.

Geotechnik im Bauwesen
Geotechnical Engineering

Univ.-Prof. Dr.-Ing. Martin Ziegler

Acknowledgement
The Authors would like to thank the Naue GmbH & Co. KG and Colbond B.V. for providing the geogrids and for their financial support. Special thanks go to the Geosynthetic Institute that also supported part of this work under its GSI Fellowship Program.

References
BS 8006 (1995): Code of Practice for Strengthened/Reinforced Soils and other Fills. BSi, Milton Keynes, UK. EBGEO (2009): Berechnung und Dimensionierung von Erdkrpern mit Bewehrungseinlagen aus Geokunststoffen. DGGT, Essen, Germany, Draft 02/2009 (in German). Clayton, C.R.I.; Milititsky, J. & Woods, R.I. (1993): Earth Pressure and Earth-Retaining Structures. Blackie Academic & Professional, Glasgow, UK, 2nd Edition. Eiksund, G.; Hoff, I. & Perkins, S. (2004): Cyclic Triaxial Tests on Reinforced Base Course Material. Proc. EuroGeo3, DGGT & igs, Munich, Germany, Vol. 2, 619-624. Ling, H.I. & Tatsuoka, F. (1994): Performance of Anisotropic Geosynthetic-Reinforced Cohesive Soil Mass. J. of Geotechnical Engineering, Vol. 120, No. 7, pp. 1166-1184. Moghaddas-Nejad, F. & Small, J.C. (2003): Resilient and Permanent Characteristics of Reinforced Granular Materials by Repeated Load Triaxial Tests. Geotechnical Testing Journal, ASTM, Vol. 26, Issue 2. Pachomov, D.; Vollmert, L. & Herold, A. (2007): Der Ansatz des horizontalen Erddrucks auf die Front von KBE-Konstruktionen. J. geotechnik, special issue 2007, pp. 129-136 (in German). Peng, F.-L.; Kotake, N.; Tatsuoka, F.; Hirakawa, D. & Tanaka, T. (2000): Plane Strain Compression Behaviour of Geogrid-Reinforced Sand and its Numerical Analysis. J. Soils and Foundations, Japanese Geotechnical Society, Vol. 40, No. 3, pp.55-74. Ruiken, A. & Ziegler, M. (2008): Effect of Reinforcement on the Load Bearing Capacity of Geosynthetic Reinforced Soil. Proc., 4th European Geosynthetics Conf., igs, Edinburgh, UK, Proc. on CD. Soong, T.-Y. & Koerner, R.M. (1997): On the Required Connection Strength of Geosynthetically Reinforced Walls. J. Geotextiles and Geomembranes, Vol. 15, pp. 377-393. Tsukamoto, Y.; Ishihara, K.; Higuchi, T. & Aoki, H. (1999): Influence of Geogrid Reinforcement on Lateral Earth Pressures against Model Retaining Walls. J. Geosynthetics Int., Vol. 6, No. 3, pp. 195-218. Yang, G.; Zhang, B.; Lv, P. & Zhou, Q. (2009): Behaviour of Geogrid Reinforced Soil Retaining Wall with Concrete-Rigid Facing. J Geotextiles and Geomembranes 27 (2009), pp. 350356. Ziegler, M. (2009): Bodenbewehrung der Markt mit den grten Steigerungsraten. J. tis Tiefbau Ingenieurbau Straenbau, 9/2009, p. 32-34. Ziegler, M. & Ruiken, A. (2009): Materialverhalten des Verbundbaustoffs "geogitterbewehrter Boden" aus groen triaxialen Druckversuchen. J. Geotechnik 32 (2009), No. 3, p. 148155.