DOI: https://doi.org/10.62204/2336-498X-2023-4-13

ANALYSIS OF RESEARCH DETERMINATION OF

CHARACTERISTICS OF SOIL STRUCTURES REINFORCED

WITH GEOSYNTHETIC MATERIALS

Oksana Kushnirova,
Senior Lecturer,
National Transportation University, Ukraine,
kushnirovao@gmail.com; ORCID: 0000-0001-6011-5609

 

Annotation. Geosynthetic layers have a high flexibility index; therefore, the distribution of loads between geosynthetic layers will depend on the stress-strain response of the soil and geosynthetic layers. With this aim, the article investigates the development of deformations in geosynthetically reinforced structures and their performance considering the variation in reinforcement intervals. Various research methods for studying these issues are considered, such as laboratory tests in compact setups, large-scale constructions, and centrifuge-based investigations. It is expected that upon reaching a certain optimal reinforcement interval, the geosynthetically reinforced mass will act as a monolith. The findings of the review can be utilized for further research on the optimal geosynthetic reinforcement interval in structures and ensuring their quality performance in constructions.

Keywords: geosynthetic reinforcement, strength, stability, soil layer deformation, loading.

Introduction. Geosynthetic-reinforced soil (GRS) structures find wide applications in various engineering projects. Enhancing the performance of reinforced structures is achieved through effective separation and filtration of geosynthetic layers. These layers also mitigate soil deformation under loading, thereby increasing their bearing capacity by bolstering tensile strength and stiffness. It’s important to note the economically efficient utilization of such structures.

The most critical aspect of the structural behavior of GRS structures is how horizontal soil pressure is transmitted to the geosynthetic reinforcement. This necessitates ensuring the necessary strength of this reinforcement. There’s an assumption that soil and geosynthetic reinforcement deformation is significant enough to provide conditions for active earth pressure. It’s assumed that each layer of reinforcement resists the load applied by the horizontal pressure of the soil at each layer of reinforcement. However, this assumption overlooks the potential redistribution of the load among the layers of reinforcement. Load redistribution leads to increased bearing capacity of GRS structures, and understanding their mechanical behavior is crucial for assessing the potential limitations of the design method.

The nature of complex interactions that may develop between adjacent layers of geosynthetic reinforcement, potentially leading to “composite” behavior of the reinforced soil mass, requires further extensive study. It is expected that the degree of interaction between neighboring layers of reinforcement will influence the mechanical response of the reinforced soil mass. Additional benefits from the interaction between reinforcement layers will be particularly relevant for critical structures.

The interaction between the soil backfill and the geosynthetic reinforcement may be influenced by phenomena associated with the vertical spacing of the reinforcement. Such phenomena, which develop in the reinforced soil mass, may be related to soil arching. The soil arch forms during soil deformation and can take various shapes [2, 3, 4, 5]. This phenomenon can also occur in reinforced soil, especially in cases involving closely spaced reinforcement. It is expected that this phenomenon will depend on the soil density, particle size distribution, confining pressure, and interface characteristics.

Objective of the study. The aim of this study is to review the main testing methods for geosynthetic-reinforced structures to determine the factors influencing the performance parameters of these structures.

The main part. The consolidation of GRS structures occurs from the lower layers to the upper ones. At the initial stage of construction, the site is prepared by leveling the ground, where the first layer of reinforcement is then laid. If the GRS structure is intended to replace a slope, the slope is formed before construction begins. Then, the first layer of reinforcement is placed on the soil or foundation, and a layer of filler is applied on top of it, which is compacted before laying the next layer of reinforcement. This process is repeated until the desired height of the structure is achieved. A schematic view of the GRS structure can be seen in Figure 1 below.

                                                                                     ^ƵďƐŽŝů

Fig. 1. Schematic overview of GRS-wall

Possible failure modes of GRS structures include: base sliding, overturning, loss of bearing capacity due to settlement, pullout of the reinforcement layer, overstraining during tension, internal shear, failure of connection with facing, facing failure during sliding, and facing overturning. Depending on the geometric configuration of the structure, the strength of the reinforcement, facing, connection with reinforcement, backfill material, and soil, different failure scenarios may occur. Therefore, it is important to have a comprehensive understanding of all aspects of the structure during the design process.

Failure modes are often classified into internal and external, or global failures. Internal failures occur within the reinforced zone, while external failures are those that occur outside the reinforced soil.

Geosynthetic reinforcement is made from linear polymer elements that can be combined in various ways. Different types can be roughly categorized into the following categories:

  1. Geogrids: created by heating and stretching a polymer sheet or by combining strips or threads in two perpendicular directions with joints at intersections.
  2. Woven fabrics: composed of two perpendicular sets of parallel linear elements interlaced to form a flat fabric.
  3. Nonwovens: made from randomly arranged threads bonded together mechanically, thermally, or chemically.
  4. Strips: made from parallel threads secured and protected by a polymer coating.

Geosynthetic reinforcement can be made from various types of polymers, such as polyester, polypropylene, and polyethylene. One of the most significant differences between these materials is their deformation behavior under loading. Figure 2 illustrates the deformation behavior under loading for different polymer materials and other reinforcement materials. Polymers, compared to other reinforcement materials like steel, demonstrate significantly greater deformation under the same force. This is highly significant as it affects their interaction with the soil.

Fig. 2. Typical stress-strain relationship for different materials [6]

The purpose of using geosynthetics for soil reinforcement is to alter the force equilibrium of the soil by providing internal resistance to tension. This enhances the resistance to sliding of the structure. Resistance arises from the force mobilized when tensile deformation occurs in both the reinforcement and the soil. The greatest deformations in the soil occur in the sliding zone, hence the highest tensile forces in the reinforcement occur at the intersection of the sliding surface, as depicted in Figure 3. This implies that the structure has internal support, as the force in the reinforcement is mobilized locally in the sliding zone.

The mobilization of tensile force in the reinforcement occurs through the stress transfer between the soil and the reinforcement. This stress transfer arises from the frictional contact between soil particles and the surface of the reinforcement, as well as from bearing stresses on the transverse elements of the reinforcement in the case of geogrid reinforcement [8].

Fig. 3. Mobilized tensile force in the reinforcement and stress absorption on the sliding surface [8]

For a better understanding of the mobilization of tensile force in the reinforcement, let’s consider a mass of soil subjected to vertical loading. This loading leads to compression deformation in the vertical direction and tensile deformation in the horizontal direction. When horizontally oriented reinforcement is introduced into the soil, horizontal deformation creates shear forces at the interface between the reinforcement and the soil. These shear forces represent the interfacial stress between the soil and the reinforcement [7]. In the case of walls or steep slopes, the directions of the primary deformations do not coincide in the horizontal and vertical directions. However, the reinforcement needs to be installed horizontally [9].

To prevent soil from falling out onto the face of the wall, facing elements are attached to the end of the reinforcement or the reinforcement is wrapped around the soil layer. There are several types of facings, but mechanically they can be divided into three categories based on their bending stiffness [10]:

  1. Rigid – fully rigid facings without the possibility of accommodating differential lateral settlements, such as precast concrete facing covering the entire height.
  2. Flexible/partially deformable facings – rigid blocks capable of accommodating differential settlements between the blocks making up the facing, such as modular walls made of concrete blocks.
  3. Soft/deformable facing elements – facings without bending stiffness, where the soil is retained by geosynthetic reinforcement, i.e., gabions.

The stiffer the facing, the more load it carries, and the less load is applied to the reinforcement. This is because stiffer facing does not allow soil to freely move to the surface, so the facing bears more load. This general mechanical property means that stiffer elements in the construction carry more load than less stiff ones.

In many studies, the long-term behavior of GRS structures has been investigated. In laboratory tests to assess the creep behavior of geosynthetic reinforcement due to interaction with soil, the geosynthetic reinforcement and the restraining soil were loaded for a long period and allowed to deform interactively. Two different tests used clean sand and kaolin clay as backfill. The results showed that reinforced soils have greater stiffness and strength than unreinforced ones. However, some vertical and horizontal deformations must be present to mobilize reinforcement effects. Additionally, it was concluded that the time-dependent deformation of the restraining soil plays an important role in the long-term creep potential of GRS structures. That is, the time-dependent deformation significantly affects the creep deformation of the reinforcement and depends on the time deformation characteristics of the confined soils. Sand and clay deformations differ significantly over time. Sands demonstrate less deformation rate than clays. Conversely, clays tend to deform more rapidly than sands. It was found that the assessment of creep potential of soil-geosynthetic composites can be misleading if it is based solely on the results of geosynthetic creep tests. Creep deformation decreases over time. In sands, vertical and lateral deformations are limited and similar. In clay, lateral deformation is insignificant, but significant vertical deformation is observed. Moreover, higher reinforcement strength further reduces deformation.

Research on the behavior of structures under constrained stress-strain conditions and various loadings has led to the following conclusions:

  • Preloading increases the stiffness of the soil but does not affect the shear strength. The stiffness of reloading depends on the normal pressure and the level of unloading.
  • Preloading enhances the geosynthetic stiffness and marginally reduces the tensile strength. Conversely, the stiffness of reloading decreases with an increase in the load level before loading.
  • Preloading has no effect on the interface shear force. The stiffness of reloading increases with preloading and normal loading on the interface.
  • Preloading does not impact the bearing capacity of GRS.
  • Creep of geosynthetic reinforcement is negligible when using well-compacted granular backfill, and stress relaxation occurs immediately after construction.
  • Unloading and reloading stress behavior almost coincide.

Large-scale general testing of the soil geosynthetic composite enabled the development of an analytical model to describe the relative contribution of reinforcement strength and spacing between reinforcements. Additionally, equations were derived based on the analytical model. These equations were formulated to calculate the apparent cohesion of the GRS composite (acquired cohesion due to the inclusion of reinforcement), ultimate load-bearing capacity of the GRS mass, and required strength of the reinforcement for a specified reinforcement interval. The adequacy of the developed equations was evaluated by comparing predictions with the results of tests, large-scale experiments conducted by other researchers, and finite element modeling [11,12, 13, 14].

The overall bearing capacity of the GRS structure was assessed using an analytical formula [11], as depicted in Equation 1. This formula is applicable to GRS structures subjected to vertical loading.

𝑆𝑆

(

𝑞𝑞𝑢𝑢𝑢𝑢𝑢𝑢 = [𝜎𝜎𝑐𝑐 + 0,7 𝑆𝑆𝑣𝑣 𝐾𝐾𝑝𝑝𝑝𝑝 + 2𝑐𝑐√𝐾𝐾𝑝𝑝𝑝𝑝                 (1)

where: σc –  lateral normal pressure; Sv – vertical spacing between reinforcements; dmax– maximum particle size of the backfill material; Tf –  ultimate strength of the reinforcement; Kpr –  coefficient of passive earth pressure.

The lateral pressure exerted on the facing from the soil is relatively insignificant compared to the lateral pressure predicted by soil pressure theory. It has also been reported that the pressure on the soil near the reinforcement layers is almost negligible. However, under certain conditions, some pressure may occur due to support stress. A method for estimating the lateral pressure on the soil acting on the facing of GRS structures has been proposed. It is noted that the primary function of the facing is to prevent soil spalling for GRS structures reinforced at small intervals (less than 0.3 m). A comparison was made between certain measured values of lateral pressure on the soil and two prediction methods. However, this comparison was not very reliable due to the inclusion of several measured values, most of which were negative. To enhance the assessment of prediction models, more reliable data were found and utilized. However, the effect of superstructure loading was ignored, meaning the predictions were based solely on the self-weight of the fill. It is worth noting that the investigated models mainly do not account for the effect of additional loading on the superstructure.

One of the methods for predicting the maximum lateral displacement in GRS supports assumes no volume change. This means that vertical deformation is compensated by an equivalent lateral deformation while maintaining the same volume of GRS mass. This method is considered conservative. Maximum lateral displacement and deformation can be estimated using appropriate equations [15].

𝐷𝐷𝑣𝑣

𝐷𝐷𝐿𝐿 = 2 ∙ 𝐵𝐵 ∙ 𝐻𝐻                                                       (2)

𝜀𝜀𝐿𝐿 = 2𝜀𝜀𝑣𝑣                                                                (3)

where: Dl.- maximum lateral displacement;

B –  width of the load along the top of the wall, including setbacks;

Dv –  vertical load calculation; H –  height of the structure; εl – maximum lateral deformation; εv –  vertical deformation.

This method is based on simplified deformation geometry, assuming that vertical deformation under loading is uniform, while lateral deformation takes the form of a triangular prism. Regardless of the location of the maximum ordinate, the equation would still provide an approximate estimate of the maximum lateral displacement (the peak of the triangle). To properly assess the effectiveness of this method [15], a modification was made to account for the fact that supports allow deformation only on one side, whereas square supports deform on four sides, and supports with a fixed wall deform on two sides. This modification allows for the use of this method with structures of different geometric configurations.

Equations 4 and 5 modify the maximum lateral displacement and deformation, respectively.

𝐷𝐷𝐿𝐿 = 2 ∙ 𝐵𝐵 ∙ 𝐷𝐷𝐻𝐻𝑣𝑣 ∙ 𝑛𝑛1                                                   (4)

2𝜀𝜀𝑣𝑣

𝜀𝜀𝐿𝐿 = 𝑛𝑛                                                                         (3)

 

Where n is the number of deforming sides of the structure (n = 4 for square abutments; n = 2 for abutments with fixed walls; and n = 1 for abutments).

Large parametric studies were conducted using finite difference software, which employs a finite difference approach. The results indicate that the effect of closely spaced reinforcement increases with the increasing shear strength of the fill material. This trend becomes more pronounced when the foundation soil is stiff. For distances between reinforcements of less than 200 mm, it was found that reinforced soil masses behave as integral masses and do not exhibit internal plastic zones. However, at large intervals (over 600 mm), connection failure is observed. The research has shown that the distance between reinforcement layers plays a key role in the behavior of the structures and significantly influences the predominant failure mode, which may not align with modern design approaches. Generalizing the numerical results, it was found that the interaction of all construction components (i.e., facing, foundation, retained soil, reinforced soil, and reinforcement properties) can significantly affect its performance. Additionally, it was established that for high-quality backfills, the “close interval” corresponds to values less than 400 mm, but this value is highly dependent on various factors. [16, 17].

Full-scale field investigations, while important, are often laborious and expensive, which limits their use in creating a comprehensive database. Therefore, many researchers turn to small-scale laboratory methods. For example, to study the behavior of GRS structures under vertical loading, small-scale models are used under normal gravity conditions (1 g) [18].

However, such scaled-down models are not always capable of reproducing the same stress levels that occur in real conditions. Therefore, a methodology for modeling engineering-geological centrifuges has been developed to analyze the performance of soil structures at representative stress levels. The use of these centrifuges allows for obtaining more realistic indicators and facilitates understanding of mechanical behavior by observing real conditions.

The engineering-geological centrifuge creates an environment with inertial acceleration higher than free-fall acceleration, allowing for the reproduction of real conditions for test models. This enables conducting parametric studies on small-scale structures with the generation of highly realistic results reflecting the behavior of the structures.

The model testing was conducted under normal Earth gravity acceleration (1 g) using various reinforcement schemes, including hybrid vertical spacing and lengths of reinforcement bars. It was found that the reinforcement arrangement significantly influences the external and internal deformation of the reinforced soil mass. Specifically, it was concluded that short secondary layers of reinforcement with predominantly distributed primary reinforcement facilitate construction but provide limited benefit in restraining deformation. It was also reported that the maximum peak deformation of the reinforcement layers occurs at the mid-height of the GRS structures, contrary to the commonly accepted triangular stress distribution of the reinforcement, and occurs below the crest of the structure.

A significant amount of research has been conducted on scaling centrifuge models to real structures. Overall, practically increasing the gravitational force in centrifuge models can lead to a proportional increase in stresses. This can be achieved by rotating the structural model in the centrifuge to create a large centrifugal acceleration acting as an amplified virtual gravitational acceleration for the model. It is important for the model to be positioned in the centrifuge in such a way that its initial direction of gravitational force aligns with the centrifugal acceleration during its flight in the centrifuge. Elevated stresses in centrifuge models reflect the stresses observed in real structures. Such modeling provides analogous stress and deformation conditions as in real structures.

Geotechnical modeling using centrifuges has its limitations that may affect the accuracy of the results. These limitations can be classified into four main sources of errors:

  1. Acceleration field variability within the centrifuge model: The acceleration field inside the model is proportional to the centrifuge arm radius. However, the variation in acceleration level within the model is proportional to the size of the model compared to the centrifuge arm.
  2. Mismatch between stress trajectories of the prototype and the model: Stress paths in the model may differ from those actually observed in the prototype during construction.
  3. Model boundary effects: Boundary effects can arise from the walls of the container in which the model is placed. These effects can be mitigated by using a material with a low friction coefficient to smooth the inner surfaces of the container to determine the state of plane deformation during testing.
  4. Model scale effects: Scale effects arise due to the relative size of the fill particles between the model and the prototype. These effects can be reduced by using fill and reinforcing materials that can behave as a continuum. Additionally, the width of the contact zone should be greater than approximately 15 particle diameters to ensure adequate modeling of scale effects.

Conclusions. A review of GRS (Geosynthetic Reinforced Soil) construction testing methods has shown that the interaction between soil layers and geosynthetic reinforcement layers can be significant and can provide the construction with composite behavior. Testing of dual geosynthetic reinforcement systems has shown that the soil mass between geosynthetic reinforcements mobilizes as a monolithic system.

The results of experimental and field research components collectively indicate a favorable effect of closely spaced reinforcement on the characteristics of reinforced soil structures, particularly on the influence of closely spaced reinforcement on the stresses acting on wall components. Although the value of the vertical reinforcement interval below which composite behavior is expected has not been established, the following practical recommendations can be made: (1) composite behavior is not expected for vertical reinforcement intervals above 0.6 m, although this value is expected to correspond to the minimum stiffness value of the geosynthetic reinforcement; (2) the length of the geosynthetic reinforcement is expected to be governed by external stability considerations; and (3) the influence of closely spaced reinforcement on reducing stresses acting on wall components is significant.

The technology of geotechnical centrifuges is highly powerful in modeling GRS constructions, allowing for the modeling of real conditions in scaled-down models, enabling the investigation of deformation and stress states scaled to real structures. This provides important practical insights into the type of behavior expected from these structures and optimizes design accordingly.

The resulting conclusions can be summarized as follows:

  • Reducing the vertical spacing of reinforcement increases the stability of wall constructions even at small L/H ratios.
  • Reviewing the research results has shown that reducing the distance between reinforcements mitigates lateral pressure on the soil at the facing, resulting in reduced lateral deformation and, consequently, reduced vertical deformation.
  • GRS constructions, built with the same total tensile capacity and total stiffness of reinforcement, behaved differently depending on the vertical spacing of the reinforcement.
  • Using a large number of low strength-to-break and stiffness layers of reinforcement, but placed at a small vertical distance, may result in better overall structural characteristics than a comparable construction with high strength-to-break and stiffness reinforcement, but placed at a large vertical interval. This difference in operational characteristics is explained by the higher number of soil-reinforcement connections in constructions with closely spaced reinforcement.

References:

  1. Koerner, R. M. & Song, T. Y., 2001. Geosynthetic reinforced segmental retaining walls. Geotextiles and Geomembranes, 24 May, Issue 19, pp. 359-386.
  2. Chen, Y.M., Cao, W.P., & Chen, R.P. (2008). An experimental investigation of soil arching within basal reinforced and unreinforced piled embankments. J. G&G, 26(2):16474.
  3. Costa, Y.D., Zornberg, J.G., Bueno, B.S., & Costa, C.L. (2009). Failure Mechanisms in Sand over a Deep Active Trapdoor. JGGE, ASCE, 135(11):1741-1753.
  4. Iglesias G.R., Einstein H.H., & Whitman R.V. (2013). Investigation of soil arching with centrifuge tests. JGGE, 140(2):248-56.
  5. Rui, R., van Tol, F., Xia, X. L., van Eekelen, S., Hu, G., & Xia, Y. Y. (2016). Evolution of soil arching; 2D DEM simulations. Journal of Computers & Geotechnics, 73, 199-209.
  6. Carlsson, B., 1987. Armerad jord berakningsprinciper for vertikala vaggar, branta slanter, bankar pa los undergrund, bankar pa palar, s.l.: Terrateam AB.
  7. Hoffman, P., 2015. Plasticity and the mechanics of reinforced soil. 1st ed. Denver: Preservation Engineering.
  8. SGF, 2004. Armerad jord och fyllning, Nordisk vagledning. Linkoping: rapport 2:2004, Unitryck 2004.
  9. Shukla, S., Sivakugan, N. & Das, B., 2011. A state-of-the-art review of geosyntheticreinforced slopes.International Journal of Geotechnical Engineering, 5(1), pp. 17-23.
  10. Nordic Geosynthetic Group, 2005. Nordic guidelines for reinforced soil. 2nd ed.

s.l.:The Nordic Geotechnical Socites Nordic Industrial Fund.

  1. Wu, J., Pham, T.Q., and Adams, M. (2013). Composite Behavior of Geosynthetic- Reinforced Soil (GRS) Mass. Report No. FHWA-HRT-10-077, Federal Highway Administration, McLean, VA.
  2. Elton, D., and Patawaran, M.A. (2004). Mechanically Stabilized Earth Reinforcement Tensile Strength from Tests of Geotextile-Reinforced Soil. Transportation Research Record: Journal of the Transportation Research Board, Vol. 1868, pp. 81-88.
  3. Elton, D., and Patawaran, M.A. (2005). Mechanically Stabilized Earth (MSE) Reinforcement Tensile Strength from Tests of Geotextile-Reinforced Soil. A report to the Alabama Highway Research Center, June 2005, 77p.
  4. Ruiken, A. and Ziegler, M. (2009). Large Scale Laboratory Element Testing of Geogrid Reinforced Soil. Proceedings of GIGSA GeoAfrica 2009 Conference, Cape Town, September.
  5. Adams, M.T., Nicks, J.E., Stabile, T., Wu, J.T.H., Schlatter, W., and Hartmann, J. (2012). Geosynthetic Reinforced Soil Integrated Bridge System Interim Implementation Guide. Report No. FHWA-HRT-11-026, Federal Highway Administration, McLean, VA.
  6. Leshchinsky, D., Kaliakin, V., Bose, P., & Collin, J. (1994). Failure Mechanism in Geogrid-Reinforced Segmental walls: Experimental Implications. Soils & Foundations, Journal of the Japanese Society of Soil Mechanics & Foundation Engineering, 34(4):3341.
  7. Leshchinsky, D. & Vulova, C. (2001). Numerical investigation of the effects of geosynthetic spacing on failure mechanisms in MSE block walls. Geosynthetics Int., 8(4):343-365
  8. Vafaeian, M. and Abbaszadeh, R. (2006). Laboratory Small Scale Tests to Study the Behaviour of Reinforced Soil Wall. Proceedings of 8th International Conference on Geosynthetics, Kuwano, J., Kuseki, J. (eds.), Millpress Science, Rotterdam, Vol. 4, pp. 1409-1412.