Elsevier

Geotextiles and Geomembranes

Volume 40, October 2013, Pages 48-57
Geotextiles and Geomembranes

Analysis of soil-welded steel mesh reinforcement interface interaction by pull-out tests

https://doi.org/10.1016/j.geotexmem.2013.08.002Get rights and content

Abstract

The interface friction coefficient is an important design parameter for reinforced soil structures where the friction between the soil and reinforcement elements is mobilized. The pull-out test is the most commonly adopted method to identify this friction coefficient. In this paper, 18 pull-out tests were conducted on two types of welded steel meshes (normal and dense mesh) embedded in a sand to investigate the soil/reinforcement interaction. The tests were conducted under vertical stresses ranging from 20 to 140 kPa on the reinforcement. The French standard (NF P 94-270, 2009) uses an analytical method to predict the friction coefficient. The comparison with the experimental results agrees reasonably well. Ju et al. (2004) developed a new test method based on a staged pull-out test for extensible reinforcements. In this study and for inextensible reinforcement, this new test method has also been used. In a usual pull-out test, only one vertical stress has been used. For a staged pull-out test, several vertical stresses are used. The results of staged pull-out tests are in good agreement with results of usual pull-out test.

Introduction

The stabilization of soil with inclusions is an ancient technique. For mechanically stabilized earth (MSE), the first setup (Haeri et al., 2000) has been done by Henry Vidal in 1961. In an MSE structure (e. g., reinforced retaining wall, reinforced embankment, or reinforced soil), reinforcements are placed horizontally in granular soil layers that are normally used as backfill or embankment material. Recent research involving MSE and reinforced embankments have been reported by many researchers (e.g., Abdelouhab et al., 2010, 2011; Bathurst et al., 2005; Bergado and Teerawattanasuk, 2008; Chen et al., 2007; Esfandiari and Selamat, 2012; Hufenus et al., 2006; Li and Rowe, 2008; Nouri et al., 2006; Palmeira, 2009; Rowe and Skinner, 2001; Sieira et al., 2009; Skinner and Rowe, 2005a,b; Varuso et al., 2005).

Materials used in reinforcement elements include steel, polymeric materials and even bamboo. Considering theirs modulus, the reinforcement materials can be classified into two types: inextensible materials and extensible materials. Inextensible materials are wire mesh, steel strip, bar mat, and welded wire or steel mesh. Extensible reinforcement includes non-metallic material such as geosynthetics (Tin et al., 2011).

The stability of earth structures depends on the interaction between soil and reinforcement. Soil/reinforcement interaction makes an important role to determine the length of the reinforcement and the distance between the upper and lower reinforcements (Ju et al., 2004). A pull-out test allows to simulate the tensile stress applied on the reinforcement and to define the evolution of the interface parameters during its mobilisation (Khedkar and Mandal, 2009; Lajevardi 2013; Palmeira, 2009; Sieira et al., 2009; Su et al., 2008). Unlike the direct shear test, the pull-out test highlights the evolution of several parameters: shear stress and friction along the soil/reinforcement interface; the soil dilatancy and the reinforcement strain (Abdelouhab et al., 2010). These parameters are taken into account in several design methods of MSE walls (e.g. French standard: NF P 94-270, 2009). The pull-out mechanisms of various inextensible reinforcements have been investigated not only by full-scale and laboratory model tests, but also by numerical methods (e.g., Abdelouhab et al., 2011; Abdi and Arjomand, 2011; Abdi et al., 2009; Alagiyawanna et al., 2001; Desai and Hoseiny, 2005; Dias et al., 2002; Goodhue et al., 2001; Gurung, 2001; Moraci and Cardile, 2009, 2012; Moraci and Gioffre, 2006; Palmeira and Milligan, 1989; Subaida et al., 2008; Sugimoto, 2003; Yin et al., 2008; Zhou et al., 2011).

The use of inextensible reinforcements (polymeric or steel grids) to stabilize earth structures has grown rapidly in the world. When used for retaining walls or steep slopes, they can be laid continuously along the width of the reinforced soil system (grid or mesh type) or laid at intervals (strip type). Many researchers (Bergado et al., 1987; Khedkar and Mandal 2009; Nernheim, 2005; Palmeira and Milligan, 1989) made clear that geometry of the reinforcement is one of the important factors in pull-out study. The advantage of the steel mesh reinforcement is that the pull-out bearing resistance in the resistant zone is high. However, the total volume (weight) of steel mesh required is still high due to wasted transverse (bearing) bars in the active (unstable) zone (Horpibulsuki and Niramitkornburee, 2010). The pull-out resistance mechanisms of steel mesh reinforcement have been extensively studied by many researchers (Bergado et al., 1988, 1996; Chai, 1992; Horpibulsuki and Niramitkornburee, 2010; Jayawickrama et al., 2012; Jewell et al., 1984; Lawson et al., 2013; Palmeira, 2009; Palmeira and Milligan, 1989; Peterson and Anderson, 1980; Shivashankar, 1991; Suksiripattanapong et al., 2013; Tin et al., 2011). The overall configuration of the test systems used by different researchers remained the same.

Finite element method analyses usually assume mesh (grid) reinforcement as a continuous equivalent rough planar reinforcement. However, the pull-out response of this type of reinforcement is fundamentally dependent on shape and geometrical characteristics of the mesh, with particular reference to the transverse bearing members. A consequence of the influence of the mesh geometrical characteristics, of the interference between transverse members and the assumption of the mesh as a planar continuous element is that the apparent soil/mesh friction coefficient (f*) becomes dependent on the length of the mesh, which complicates the estimate of soil/mesh bond for design purposes (Palmeira, 2009).

Pull-out tests are necessary in order to study the interaction behaviour between soil and reinforcement in the resistant (anchorage) zone; hence these properties have direct implications to the design of reinforced soil structures. In order to analyse the internal stability of reinforced earth structures, it is necessary to evaluate the pull-out resistance of reinforcement, mobilized in the resistant zone. The pull-out resistance in a pull-out test can be described by the following equations (Moraci and Recalcati 2006):PR=2Lσvfbtanφ=2Lσvf=2LσvαFSo:fbtanφ=f=αFwhere PR is the pull-out resistance (per unit width); L the reinforcement length in the anchorage zone; σv the effective vertical stress; φ′ the soil friction angle; fb the soil/reinforcement pull-out interaction coefficient; f* the apparent friction coefficient; F* the pull-out resistance factor; α the scale effect correction factor taking into account for a non-linear stress reduction over the embedded length of highly extensible reinforcements (FHWA, 2001).

The soil/reinforcement pull-out interaction coefficient fb may be determined by means of theoretical expressions (Jewell et al., 1985), whose limits have been investigated by different researchers (Ghionna et al., 2001; Moraci and Montanelli, 2000; Palmeira and Milligan, 1989; Wilson-Fahmy and Koerner, 1993), or by back-calculation from pull-out test results. In this case previous experimental studies (Ghionna et al., 2001; Moraci and Montanelli, 2000; Palmeira and Milligan, 1989) have shown that the values of fb are largely influenced by the choice of the value of the soil friction angle.

According to FHWA (2001) the scale effect correction factor α can be obtained from pull-out tests on reinforcements with different lengths or derived using analytical or numerical load transfer models which have been “calibrated” through numerical test simulations. In the absence of test data, α = 1 for steel reinforcements, α = 0.8 for geogrids and 0.6 for geotextiles can be taken.

In absence of a clear indication regarding the choice of the soil friction angle to be used for the determination of fb and to avoid the use of sophisticated numerical analyses, the problem of the determination of the pull-out resistance may be overcome by the use of the soil/reinforcement interface apparent friction coefficient using the following equation that depends on the vertical stress:f=PR/(2Lσv)

It is important to note that the determination of f* using above equation can be performed without any assumption about the values of the soil friction angle at the interface and with an assuming that the friction occurs on both sides of the reinforcement.

Conservative default values for f* are provided by the French standard guideline (NF P 94-270, 2009) to be used in the absence of pull-out test data (Table 1).

This standard provides for different types of reinforcements, the default value of f* which depends on the type of reinforcement (strip or mesh and extensible or inextensible) and for the geotechnical characteristics of the backfill (for example: Cu and φ).

Fig. 1 shows the variation of maximum friction coefficient f0* at the top of the wall (where σv = 0) and the maximum friction coefficient f1* at a depth of 6 m and below (where σv = 120 kPa). The soil and the welded steel mesh reinforcement used in this figure are a granular soil (sand) with a normal (t = 0.8 cm and St = 50.5 cm) and dense (t = 0.8 cm and St = 14 cm) welded steel mesh respectively (presented later in this paper). It can be seen that steel strips reinforcement are the soil reinforcement type with most pull-out resistance. Welded steel mesh reinforcement, on the other hand, offers the least reinforcement.

On a construction site where rigid inclusions reinforcements were used in Poland (Gdansk), two types of welded steel mesh were installed in sand. In order to validate the characteristics of the interface between sand and welded steel mesh considered for the improved soil mass design a qualification survey was carried out. The pull-out tests have been carried out with two types of welded steel meshes (normal and dense) in a Polish sand. The purpose of the work was the qualification of the apparent friction coefficient and a comparison with the results obtained from the French standard. This standard has been compared with another one. Two pull-out test methods have been used. Different levels of vertical stress were simulated (20 kPa–140 kPa) to recreate the on site stress conditions of an embankment with a variable height (from 0 to 7 m). The influence of the type of reinforcement and the influence of vertical stress were studied based on the pull-out tests.

Section snippets

Pull-out tests

The anchoring behaviour of a welded steel mesh under tension is studied experimentally. This experiment, described as a pull-out test, consists in applying a tensile force on the studied mesh by a jack. Beforehand the mesh was equipped with force, pressure and displacement sensors and was set in a tank filled with soil. Three physical quantities were monitored: Head tensile force, displacements at different locations in the mesh and levels of vertical stress in the soil.

The results analysis was

Test carried out

Altogether, 18 tests were carried out. The sensibility study was about:

  • -

    The level of vertical stress on the reinforcement level (from 20 to 140 kPa),

  • -

    The width of the mesh (from 0.21 to 0.48 m),

  • -

    The type of mesh (Mesh 1: normal and Mesh 2: dense, see Fig. 3).

Stresses distribution in the test tank

In order to simulate the vertical stresses equivalent to the weight of the soil in an embankment, an air cushion is placed at the surface of the tank. An uncertainty remains concerning the vertical stress applied on the mesh in the centre of

Conclusion

In this work, 18 pull-out tests on welded steel mesh were presented. The sensibility study was about:

  • -

    The vertical stress applied on the reinforcement level (from 20 to 140 kPa),

  • -

    The mesh width,

  • -

    The mesh type (mesh more or less dense).

From the pull-out tests, the tensile force and the displacements at some points of the welded steel mesh have been determined. The maximal apparent friction coefficient mobilized along the mesh was determined from the results.

The values of f* obtained with the usual

References (63)

  • J. Esfandiari et al.

    Laboratory investigation on the effect of transverse member on pull out capacity of metal strip reinforcement in sand

    Geotext. Geomemb.

    (2012)
  • N. Gurung

    1-D analytical solution for extensible and inextensible soil/rock reinforcement in pull-out tests

    Geotext. Geomemb.

    (2001)
  • S. Haeri et al.

    Effect of geotextile reinforcement on the mechanical behavior of sand

    Geotext. Geomemb.

    (2000)
  • R. Hufenus et al.

    Full-scale field tests on geosynthetic reinforced unpaved roads on soft subgrade

    Geotext. Geomemb.

    (2006)
  • O. Jenck et al.

    Soft ground improvement by vertical rigid piles two-dimensional physical modeling and comparison with current design methods

    Soil. Found.

    (2005)
  • M.S. Khedkar et al.

    Pullout behaviour of cellular reinforcements

    Geotext. Geomemb.

    (2009)
  • N. Moraci et al.

    Deformative behaviour of different geogrids embedded in a granular soil under monotonic and cyclic pullout loads

    Geotext. Geomemb.

    (2012)
  • N. Moraci et al.

    Influence of cyclic tensile loading on pullout resistance of geogrids embedded in a compacted granular soil

    Geotext. Geomemb.

    (2009)
  • N. Moraci et al.

    A simple method to evaluate the pullout resistance of extruded geogrids embedded in a compacted granular soil

    Geotext. Geomemb.

    (2006)
  • N. Moraci et al.

    Factors affecting the pullout behaviour of extruded geogrids embedded in a compacted granular soil

    Geotext. Geomemb.

    (2006)
  • H. Nouri et al.

    Development of horizontal slice method for seismic stability analysis of reinforced slopes and walls

    Geotext. Geomemb.

    (2006)
  • E.M. Palmeira

    Soil-geosynthetic interaction: modelling and analysis

    Geotext. Geomemb.

    (2009)
  • E.M. Palmeira et al.

    Large scale direct shear tests on reinforced soil

    Soil. Found. Japan. Soc. Soil Mech. Found. Eng.

    (1989)
  • A.C.C.F. Sieira et al.

    Displacement and load transfer mechanisms of geogrids under pullout condition

    Geotext. Geomemb.

    (2009)
  • G.D. Skinner et al.

    Design and behaviour of a geosynthetic reinforced retaining wall and bridge abutment on a yielding foundation

    Geotext. Geomemb.

    (2005)
  • E.A. Subaida et al.

    Experimental investigations on tensile and pullout behaviour of woven coir geotextiles

    Geotext. Geomemb.

    (2008)
  • C. Suksiripattanapong et al.

    Pullout resistance of bearing reinforcement embedded in coarse-grained soils

    Geotext. Geomemb.

    (2013)
  • N. Tin et al.

    Factors affecting kinked steel grid reinforcement in MSE structures

    Geotext. Geomemb.

    (2011)
  • R.J. Varuso et al.

    Geosynthetic reinforced levee test section on soft normally consolidated clays

    Geotext. Geomemb.

    (2005)
  • AASHTO

    LRFD Bridge Design Specifications

    (1998)
  • M.C. Alfaro et al.

    Soil geogrid reinforcement interaction by pullout and direct shear tests

    Geotech. Test. J.

    (1995)
  • Cited by (38)

    • Interface pullout resistance of polymeric strips embedded in marginal tropical soils

      2022, Geotextiles and Geomembranes
      Citation Excerpt :

      Theoretical (e.g. Palmeira, 2004; Abdelouhab et al., 2010; Cardile et al., 2017), and numerical (e.g. Abdelouhab et al., 2011; Wang et al., 2016; Noorzad and Badakhshan, 2017) methods have been used to simulate the mechanism of interface pullout resistance that develops under field conditions. Pullout tests provide insight into the responses of reinforcements embedded in reinforced soil structures (Palmeira, 2009), facilitating the quantification of the pullout resistance between fill soils and different reinforcement types, including geogrids (e.g. Palmeira, 1987; Teixeira et al., 2007; Sieira et al., 2009; Moraci and Cardile, 2009; Abdi and Arjomand, 2011; Ezzein and Bathurst, 2014; Bathurst and Ezzein, 2016; Wang et al., 2016; Abdi and Mirzaeifar, 2017; Kayadelen et al., 2018), fibers (e.g. Zornberg, 2002; Yang et al., 2003; Tang et al., 2016), cellular confinement systems (e.g. Khedkar and Mandal, 2009), steel meshes (e.g. Lajevardi et al., 2013; Suksiripattanapong et al., 2013), and polymeric strips (e.g. Abdelouhab et al., 2010, 2011; Razzazan et al., 2018, 2019; Cui et al., 2019; Zhang et al., 2020). The scarcity of adequate fill soils nearby construction sites may lead to considerable additional costs.

    • Pullout behavior of geosynthetic reinforcement in biocemented soils

      2021, Geotextiles and Geomembranes
      Citation Excerpt :

      In recent years, the reinforced retaining wall, as an earth structure to enable the design of steep slopes, has gained research attention in geotechnical engineering (e.g., Khedkar and Mandal, 2009; Won and Kim, 2007; Chen et al., 2007; Chen and Chiu, 2008; Yoo and Kim, 2008; Yang et al., 2019). Many different types of reinforcement, which can take up tensile forces (e.g., geotextiles, geogrids, and metal strips) are utilized to improve the strength of reinforced retaining walls (e.g., Tang et al., 2010; Lajevardi et al., 2013; Bilgin and Mansour, 2014; Sukmak et al., 2015). When such reinforcements are used, the pullout of the reinforcement from the soil is one of the critical failure modes, which needs to be considered in the design and safety assessment of reinforced soil structures.

    View all citing articles on Scopus
    View full text