Observations from a parametric studyof the seismic design of soil nailing&1 Felipe A. Villalobos MSc, DPhil, CEng

Assistant Professor, Laboratory of GeoMaterials, Department of CivilEngineering, Catholic University of Concepción, Concepción, Chile(corresponding author: avillalobos@ucsc.cl)

&2 Sergio A. Villalobos CEngResearch Assistant, Department of Civil Engineering,Catholic University of Concepción, Concepción, Chile

&3 Paulo L. Oróstegui CEngGeneral Manager, OITEC Engineering, Geotechnics, Hydraulics andSurveying, Concepción, Chile

1 2 3

A parametric study of the seismic design of soil-nailed walls is performed. Values of a set of geometrical andmechanical parameters for the nail and soil of a slope have been varied to assess their effects on the global factor ofsafety FSG. Equations of limit equilibrium of forces based on the Coulomb criterion for a two-block-failure mechanismare adopted. Pseudo-static force analyses with a horizontal seismic coefficient of 0·15 are carried out, which isnormally used in engineering practice for temporary soil-nailing projects. It was found that FSG clearly increases withsoil cohesion and friction as well as nail length, diameter and inclination, although nail inclinations higher than 15°can become detrimental. It was found that there is a combination of nail spacing and wall inclination for which thetrend of FSG changes from decreasing to increasing. This situation is caused by a software mathematical optimisationroutine, which does not take into account whether it is possible for the failure surface shape to occur or not.Additionally, perforation diameters larger than 100 mm are necessary for FSG≥1·1 and soil–nail shear stress rs valuesof more than 200 kPa are not significantly beneficial.

NotationAb nail steel bar cross section areaah horizontal accelerationC soil shear resistance forcec cohesionc′ drained cohesionDd drilling diameterE active earth lateral forcefy yield stressg acceleration of gravitygd grouting pressureH seismic forcek soil coefficient of permeabilityka Coulomb’s active lateral earth pressure coefficientkh, kv horizontal seismic or acceleration coefficientL nail lengthLab distance between a and bN normal forcen number of nailsQ overburden forceq uniformly distributed overburden pressurers soil-nail shear stressSH, SV horizontal and vertical spacing between anchorsT nail tension force

Ta nail steel bar allowable tension capacityTpi resistance force added by each nail crossing the

failure surfacet shotcreted wall thicknessW soil weightα nail inclinationβ wall slopeγ unit weight of soilϕ′ angle of frictionθ block base angle

1. IntroductionSoil nailing has become a preferred solution for stabilising cutsand slopes in residual soils in Greater Concepción, Chile. Thisis mainly due to lack of available flat urban space forcing con-structions on challenging steep slopes. Projects of housing devel-opments and widening of motorways located in hilly groundsrender necessary the re-profiling of slopes or simply verticalcuts to create flat surfaces for construction. For instance, in2000 several landslides occurred due to the excavations forroads and buildings in the housing development project BarrioModelo in Concepción (Figure 1). The slope failures jeopar-dised not only the stability of the buildings and roads, but also

112

Cite this articleVillalobos FA, Villalobos SA and Oróstegui PL (2018)Observations from a parametric study of the seismic design of soil nailing.Proceedings of the Institution of Civil Engineers - Ground Improvement 171(2): 112–122,https://doi.org/10.1680/jgrim.17.00027

Ground Improvement

Research ArticlePaper 1700027Received 31/03/2017;Accepted 07/08/2017Published online 06/11/2017

ICE Publishing: All rights reserved

Keywords: anchors & anchorages/geotechnical engineering/retaining walls

Downloaded by [ Universidad Católica de la Santísima Concepción] on [15/05/20]. Copyright © ICE Publishing, all rights reserved.

the stability of two drinkable water tanks of 1000 and 2000 m3

on top of a near hill. Attempts to stabilise the slopes with45° re-profiling, plastic cover blankets, vegetation and gabionsfailed and only the use of permanent soil nailing solvedthe problem (Fernández and Guzman, 2005). Recently, semi-temporary soil nailing solutions have been used aroundConcepción, for example in the projects of Quinta Junge(Villalobos et al., 2013), Departamento Universitario ObreroCampesino, Lomas de San Andrés, Villuco, among others.They have in common the fact that they are located in residualsoil hills, which are prone to slide under the combination ofanthropic activity (excavations), seismic events and heavy rain.

Briefly, the soil-nailing technique consists of driving or drillingand then grouting under high pressures (for better penetrationof the grouting into the soil), along the whole length of inclinedbars as the excavation or re-profiling progresses downwards(Whitbread, 2012). Bars are normally inclined downwardsbetween 10° and 20° and spaced horizontally and verticallybetween 1 and 2·5 m. The slope facing is reinforced with steelmeshes and sprayed with shotcrete to create a flexible structuralfacing with a thickness of between 50 and 250 mm. A drainsystem is necessary for the soil behind the shotcreted wall to dis-sipate water pressures. Construction methods usually follow asequence of slope surface preparation before installing the nailbars with subsequent structural facing (Ball and Gavins, 2012;Jones, 1996). The reinforcing bars transmit tension loads,although they can resist some reduced amounts of shear andbending loads (e.g. Guilloux et al., 1983; Jewell and Pedley,1990, 1992); for practical purposes bending stiffness can be neg-lected even in cases with large displacements (e.g. Hajialilue-Bonab and Razavi, 2015; Wei and Cheng, 2010). From themeasurements of the nail axial stress distribution, it has beenfound that the tensile stress varies not only along the bars, butalso depending on the depth of the bar. The maximum tensilestress in the bar tends to be close to the wall facing at the toeand reduces from the one facing upwards. This creates a

parabolic distribution of maximum tensile stresses, which tracesthe potential failure surface (e.g. Guilloux et al., 1983; Shenet al., 1981; Stocker et al., 1979).

It is considered that the first soil-nailed wall was built in Francein 1972 (Whitbread, 2012). Later on, research programmes inGermany and France were dedicated to study soil nailing bymeans of large-scale tests. Results and recommendations ofthese research programmes are reported in the Clouterre (1991)project. Other field and laboratory studies of soil-nailed struc-tures have also shown that this technique can be an adequatesolution to retain excavations and stabilise steep cut slopesunder static loading (e.g. Davis et al., 1993; Ehrlich and Silva,2015; Gaessler and Gudehus, 1981) and also under strongseismic loadings with high ground accelerations of up to 0·7g(e.g. Felio et al., 1990; Tatsuoka et al., 1997; Tufenkjian andVucetic, 2000; Villalobos et al., 2013; Yazdandoust, 2017). Inpractice, soil-nailing analyses for different types of soils havebeen carried out using mainly the limit equilibrium method,where a failure surface is assumed before determining theglobal stability factor of safety FSG (e.g. FHWA, 1996, 2015;Stocker et al., 1979).

Research based on numerical analysis such as finite-elementand finite-difference methods does not have to assume a prede-fined failure surface shape since the failure surface is a resultof the calculations. Moreover, numerical results can be directlyinterpreted as stress and strain fields to obtain forces and dis-placements in two-dimensional models (e.g. Singh andSivakumar Babu, 2010) and three-dimensional (3D) models(e.g. Briaud and Lim, 1997; Razavi and Hajialilue-Bonab,2017; Smith and Su, 1997; Zhang et al., 1999). However, anumerical analysis does not directly allow the calculation of aglobal factor of safety, FSG, which is normally used by engin-eers. Although the above 3D works analysed soil displacementsand nail forces, they did not actually interpret their results interms of a global factor of safety. To obtain a global factor ofsafety, progressive failure procedures have been implemented,where two methods have been usually applied, namely thestrength reduction method and the method of uniform loadcollapse on top of the wall (e.g. Hajialilue-Bonab and Razavi,2015; Wei and Cheng, 2010). Furthermore, the above authorshave found that global factors of safety from the first methodare similar under most cases or are slightly larger than thatobtained by the limit equilibrium method (12% maximumdifference according to Wei and Cheng (2010)), whereas this isnot the case for the second method where large overburdensare needed. Due to its simplicity and efficiency, the limit equi-librium method is popular in engineering practice and for thatreason it is important to understand the results obtained fromthis method from a much wider analysis.

Within this context, this study deals with the limit equilibriumanalysis of soil-nailing solutions for the stabilisation of slopesand cuts in, but not restricted to, a residual soil known locally

Largest landslideDrinkable water tanks

Cracks due to landslides

Figure 1. Generalised residual soil landslides in Barrio Modelowhich have destroyed buildings and road accesses in 2000

113

Ground ImprovementVolume 171 Issue GI2

Observations from a parametric study ofthe seismic design of soil nailingVillalobos, Villalobos and Oróstegui

Downloaded by [ Universidad Católica de la Santísima Concepción] on [15/05/20]. Copyright © ICE Publishing, all rights reserved.