Responses of single and group piles within MSE walls under static and cyclic lateral loads
Introduction
Due to space constraint, use of laterally loaded piles has become more common to support lateral loads as well as axial loads resulting from structures built with or behind mechanically stabilized earth (MSE) walls. Recently, several studies including full-scale tests (Pierson 2008; Pierson et al., 2009; Price 2012; Nelson 2013; Hatch 2014; Besendorfer 2015); reduced-scale model tests (Mohammed 2016; Mohammed and Han 2018), and numerical analyses (Huang et al. 2011, 2013) have been conducted to understand the interaction between piles and MSE walls when the piles are constructed close to the wall facing. These studies have mainly focused on some key influence factors (e.g., the pile location behind the wall facing, the length and the stiffness of reinforcement, and the height of the wall) on the behavior of piles and the performance of MSE walls under static lateral loading.
However, piles within MSE walls are subjected to not only static lateral loading but also cyclic lateral loading in practice. Integral abutment bridges, traffic signs, and sound barrier walls are common examples of structures generating cyclic lateral loading onto piles. Seasonal temperature changes in bridge girders generate contraction and expansion loads that are transferred by piles to their surrounding soil (Liu et al., 2020). The current literature review shows experimental tests have been rarely conducted on piles within MSE walls subjected to cyclic lateral loading. However, many previous studies investigated the responses of piles in soils subjected to cyclic lateral loading without the presence of MSE walls. For example, Poulos (1982) conducted a comprehensive study on the lateral response of cyclically loaded piles and found two phenomena for the response of piles under the increased number of load cycles: (1) structural “shakedown”, which takes place for a pile embedded in an elastoplastic soil mass after its permanent deformation becomes stable and (2) soil degradation, which increases the compressibility and reduces the strength of the soil due to the cyclic loading effect. Several studies were also carried out later to account for the responses of piles under different patterns of cyclic lateral loading (e.g., Brown et al., 1988; Rajashree and Sundaravadivelu 1996; Lin and Liao 1999; Allotey and El Naggar 2008; Tuladhar et al., 2008; Basack and Dey 2012; Giannakos et al., 2012; Ismael 2014; Byrne et al., 2015). These studies revealed that the responses of piles under static lateral loading were different from those under cyclic lateral loading. Giannakos et al. (2012) pointed out some general observations under cyclic lateral loading: (1) increased moment and deflection of the pile with the number of load cycles, (2) reduced ultimate lateral load capacity, (3) more cyclic degradation in stiff soils, and (4) larger cumulative deformations of the pile induced by one-way cyclic lateral loading than two-way cyclic lateral loading. Hence, it is important to investigate the effect of cyclic lateral loading on the response of piles constructed within the reinforced zone of MSE walls in addition to static lateral loading.
Arenas (2010) developed a three-dimensional numerical model to evaluate the behavior of the structural and geotechnical components (including piles within the MSE wall) during thermal changes for a bridge deck. The numerical model was first calibrated with the available data of a three-year monitored bridge, and then a parametric study was conducted. The numerical results from Arenas (2010) showed that the location of the pile within the MSE wall had a significant effect on the reinforcement tensile forces and the earth pressures behind the wall facing; moreover, the thermally-induced displacements affected the pressures and the reinforcement tensile forces within the upper quarter of the wall. Arenas (2010) also found that the use of expanded polystyrene styrofoam (EPS) behind the MSE wall reduced the lateral earth pressures and the reinforcement tensile forces by 75% and 50%, respectively.
Rahman et al. (2015) developed a numerical model to simulate one loading-unloading cycle on the drilled pile constructed within the reinforced zone of the MSE wall. In their model a stress-dependent Cap-Yield constitutive soil model was adopted for the backfill material and verified with three triaxial test results obtained by Pierson (2008). The pile loaded and then unloaded once resulted in a considerable amount of pile head plastic deformation as well as unrecoverable wall facing deflection especially within the lower portion of the MSE wall. Alam et al. (2019) conducted experimental and numerical investigations of the behavior of footings on geosynthetic-reinforced fill slopes under cyclic loading without any piles. Cardile et al. (2019) conducted cyclic pullout tests to investigate the effect of the cyclic loading history on the post-cyclic pullout resistance of uniaxial geogrids and found that the design pullout resistance parameters depended on the combination of load amplitude, number of cycles, and normal stress.
Based on the preceding studies, the effect of cyclic lateral loading on piles within MSE walls was only investigated through the numerical analysis at a few cycles of load or thermal displacement. Since the number of load cycles is expected to have an influence on the response of piles and the performance of MSE walls, an experimental study is necessary to evaluate the responses of the laterally-loaded piles within MSE walls during cyclic loading.
The objectives of this study were to investigate the effects of static and cyclic lateral loading on load capacities of single and group piles, displacements of wall facing, strains in geosynthetic reinforcement layers, and vertical distribution of lateral earth pressures behind the wall facing.
Section snippets
Experimental tests
Six reduced-scale models in a 1/6 scale as typical MSE walls in field (e.g., Pierson et al., 2009) were designed and constructed inside a test box with interior dimensions of 1760 mm long × 1000 mm wide × 1480 mm high. All these model walls did not have any embedment to minimize the effect of toe restraints on wall performance as demonstrated by Zhang et al. (2019). Four of the model tests (referred to as Tests 1, 2, 4, and 5) contained single piles within the MSE wall and the other two model
Materials
A poorly-graded dry Kansas River Sand was used as a granular backfill material in this study. This sand was used as the backfill material in some previous studies related to MSE walls (e.g., Kakarasul et al., 2018; Kakrasul et al., 2020). The coefficients of uniformity (Cu) and curvature (Cc) were 3.18 and 0.99, respectively. The minimum and maximum dry unit weights were determined according to the ASTM D4254-14 and ASTM D4253-14 as 16.02 and 18.85 kN/m3, respectively. The dry unit weight of
Instrumentation
Earth pressure cells of 50 mm in diameter and 200 kPa in capacity were used to monitor lateral earth pressures behind the MSE wall facing induced by pile movement. In each single pile test, eight pressure cells were placed in front of the pile and directly behind the MSE wall facing above each geogrid layer to capture the vertical distribution of the lateral earth pressures along the centerline of the wall height. In the group pile test, the lateral pressures behind the wall facing were
Test procedure
After the completion of the model test construction, all sensors were connected to a data logger prior to the loading stage. In the static loading test, lateral loading was applied in increments on the pile for 10 min for each load increment. The test was terminated after the pile moved laterally to a significant displacement (i.e., larger than 20% of the pile diameter). From the static loading test, the ultimate lateral load capacity of a pile was determined based on Broms's recommended
Load-displacement curves
Fig. 3 presents the load-displacement curves of the pile in three model tests (i.e., two single pile tests with 2D or 4D offsets and one group pile test with a 2D offset) under static loading. All these curves have small initial linear portions (i.e., elastic behavior) and then non-linear portions due to the plastic deformations of the soil. As expected, the pile with a farther distance behind the wall facing had a higher lateral load capacity. For example, the lateral load capacity of the pile
Strains in geogrid layers induced by static and cyclic lateral loading
Fig. 11 presents the measured strains in the geogrid layers at different elevations in three model tests (i.e., Tests 1, 2, and 3). The measured strains were obtained from the strain gauges attached on the ribs beside the pile and also on the ribs located in the middle distance between the piles in the group pile test. Same as the measured wall facing displacements, the measured strains are reported at the lateral displacement of the pile head equal to 20% its diameter. The strains in the
Additional lateral earth pressure behind wall facing
Fig. 14 presents the vertical distribution of additional lateral earth pressures at the ultimate lateral load capacities of the piles in Tests 1, 2, and 3. The dashed lines with the legends Ka and Kp represent the Rankine active and passive earth pressures, which were calculated based on the soil friction angle of 38°. This figure shows that in all the model tests the additional lateral earth pressure distributions with the wall elevation were non-linear. The maximum lateral pressures were
Conclusions
This paper presents six reduced-scale model tests of single and group piles seated on a rigid foundation within the MSE wall subjected to lateral loading. Three influence factors were investigated in this study: (1) the offset distance of the pile behind the MSE wall facing, (2) the effect of group piles, and (3) the effect of loading type. During each test, the pile displacements, the wall facing displacements, the strains in geogrid layers, and the pressures behind the wall facing were
Acknowledgment
The first author would like to express his gratitude to his sponsor, the Higher Committee for Education Development in Iraq (HCED), and the Iraqi government, for providing him the opportunity to conduct his graduate study at the University of Kansas.
References (38)
- et al.
The influence of a cyclic loading history on soil-geogrid interaction under pullout condition
Geotext. Geomembranes
(2019) - et al.
Cyclic lateral response of piles in dry sand: finite element modeling and validation
Comput. Geotech.
(2012) - et al.
Refined numerical modeling of a laterally-loaded drilled shaft in an MSE wall
Geotext. Geomembranes
(2013) - et al.
Numerical analysis of a laterally loaded shaft constructed within an MSE wall
Geotext. Geomembranes
(2011) - et al.
A new generation of soil-geosynthetic interaction experimentation
Geotext. Geomembranes
(2019) - et al.
Degradation model for one-way cyclic lateral load on piles in soft clay
Comput. Geotech.
(1996) - et al.
Experimental study on performance of geosynthetic-reinforced soil model walls on rigid foundations subjected to static footing loading
Geotext. Geomembranes
(2016) - et al.
Influence of toe restraint conditions on performance of geosynthetic reinforced soil retaining walls using centrifuge model tests
Geotext. Geomembranes
(2019) - et al.
Experimental and numerical investigations of the behaviour of footing on geosynthetic reinforced fill slope under cyclic loading
Geotext. Geomembranes
(2019) - et al.
Investigation of stability of soil arching under surface loading using trapdoor model tests