Geosynthetic Reinforcement Applications

Abstract

Geosynthetic applications range from geotechnical to geo-environmental areas and include countless cases. In this paper, the aim was to introduce the reader to the reinforcement applications. A few examples of reinforcement were chosen to demonstrate the frequently used applications and to draw the attention of the reader to the vast variety of other reinforcement applications. First, a brief introduction to geosynthetics is given. As an example of a very frequently applied reinforcement application, information is given on geosynthetic reinforced soil retaining walls. Especially the benefits of geosynthetic-reinforced retaining walls under earthquake loading conditions are emphasized. Then, some information is given on using geosynthetic reinforcement below foundations to improve the bearing capacity. As a third and final example, the use of geosynthetic reinforcement as an encasement around stone columns is mentioned. The improvement of the geosynthetic encased columns compared to ordinary stone columns under earthquake loading conditions was shown.

1. Introduction

Geosynthetic is a generic term describing a product, at least one of whose components is made from a synthetic or natural polymer, in the form of a sheet, a strip, or a three-dimensional structure, used in contact with soil and/or other materials in geotechnical and civil engineering applications [1]. The concept of geosynthetics started with the introduction of geotextiles and has become a family that includes products, such as geogrid, geonet, geomembrane, geosynthetic clay liners (GCL), geofoam, geocell, prefabricated vertical drain (PVD), geotube, geosynthetic encased stone column, and geocomposites. The fact that the products are made of polymer material creates an important advantage because both the polymer properties and manufacturing procedures can be adapted to get the best properties for a required engineering solution. Especially in earthworks, polymers are very efficient, because soil protects them from UV light, which has a tendency to deteriorate the properties of the polymers. However, of course, they are also used successfully in applications open to the atmosphere under appropriate manufacturing and design conditions. Geosynthetic products are produced in a wide variety of forms and features and are produced in the most successful way to fulfill the intended function in the end use. Today, geosynthetics are used successfully in many civil engineering, geotechnical engineering, transportation engineering, environmental engineering, and hydraulic projects. Geosynthetics provide in many applications superior engineering solutions and at the same time reduce construction time and cost as well. Geosynthetics also make infrastructure more sustainable. They extend the service life of roads, reduce the use of aggregates, conserve and protect water, minimize land disturbance, and control soil erosion. Geosynthetics are used in a wide variety of applications, such as roads, railways, airports, embankments, retaining walls, reservoirs, canals, dams, erosion prevention projects, sediment control, solid and liquid waste landfill base and top liners, production facilities and tailing dams of mining activities, aquaculture, and agriculture.

Detailed descriptions of each of the various geosynthetic products mentioned above will not be provided within this article due to space limitations. However, the definitions of various geosynthetic products and applications can be found in Ref. [1]. Naturally, there are geosynthetics produced using a wide variety of production methods and materials within each main product group. For example, geotextiles can be classified as woven, nonwoven, or knitted. For woven geotextiles, there are various methods of weaving and also woven yarns can be for example in the form of filaments, tapes, etc. The nonwoven geotextiles can also be produced using different techniques like needle-punching or by thermal bonding, also called spun-bonding. In addition, the polymer used in the production of geosynthetics can also vary. Polypropylene and polyester are most commonly used in the production of geotextiles, but many different polymers can also be used. Again, for example, geogrids can be manufactured as woven, extruded, or welded, and can be made using a wide variety of polymers as needed. Naturally, although the production method and raw material remain the same, it is possible to produce products with varying properties. For example, tensile strength, permeability, etc., can vary significantly for the same production type and polymer. Likewise, there are various geomembranes produced with different raw materials and surface properties. This variability is a major advantage of geosynthetics. The scientific and technological level reached in polymer engineering today allows the production of polymer materials to be produced with the required features, so that the products can satisfy the needs of the civil engineer, who will design and manufacture facilities with these products. So, the combination of the proper raw material property and the production technique can help the geosynthetic to fulfill its required function in the most optimum way. While this wide variety of products is a very important advantage, it also creates a handicap, because civil engineers in general, but even geotechnical engineers, are not fully aware of the functions and design aspects of geosynthetic products, so it is difficult to determine what would be the most suitable geosynthetic for a project among products that are produced with a wide variety of raw materials and production techniques and have a wide range of mechanical and hydraulic properties. To overcome this complexity, it is important for geotechnical engineers to be well informed about geosynthetics. In fact, the International Geosynthetics Society IGS (International Geosynthetics Society) is a nongovernmental organization established just for this very purpose, namely to disseminate knowledge for the proper use of geosynthetics [2]. I would like to add that sustainability is at the heart of what IGS does. The publications, events, lectures and research highlight the many ways that geosynthetics contribute to the sustainable development goals. Although not included in this paper, because of space limitations, how the geosynthetics contribute to resolving the sustainability crises have been addressed in many papers and a synthesis paper that has been published recently [3].

In this paper only a sample of geosynthetic reinforcement applications will be highlighted to introduce the general reader about the versatile uses of geosynthetics.

2. Reinforced soil retaining structures

As it is known, retaining walls or slopes create level differences, which are often necessary for civil engineering projects. Historically, masonry walls were used as the only alternative. Later on, reinforced concrete retaining walls took over as the predominant retaining wall type. If there is enough space, an alternative is to fill or excavate with a slope. Today, it is a very common practice to keep excavations stable at steeper angles by equipping them with the so-called soil nails. In fill slopes, the most common method is to use geosynthetic reinforcement. The slope inclination can be varied infinitely with the help of the geosynthetic reinforcement up to geosynthetic-reinforced retaining walls with totally vertical facings (Figure 1). This brings flexibility to the design by optimizing the space available to reduce cost.

Figure 1.

(a) Reinforced concrete retaining structure, (b) geosynthetic-reinforced retaining structure, (c) geosynthetic-reinforced slope, and (d) slope.

Design details of geosynthetic-reinforced walls are not mentioned in this article, because these are too comprehensive to fit into the scope of this article, and also there are many specifications giving the details of the design [4, 5, 6].

Although the widespread use of geosynthetic-reinforced soil retaining walls is due to economic incentives, it is now known that they also exhibit a much better engineering performance, especially in earthquake-prone regions, and therefore they are preferred over rigid reinforced concrete retaining walls [7]. A case study [8] states that a total of approximately 3500 m long geosynthetic reinforced soil wall has been manufactured alone on the Hokkaido Shinkansen high-speed train line in Japan, and the maximum height of these walls is 11 m. In the same project, 29 bridge abutments with a maximum height of 13.4 m were formed as geosynthetic-reinforced walls. An integrated bridge with geosynthetic reinforcement was also constructed (Figure 2). This type of geosynthetic-reinforced integrated bridges is becoming increasingly common. The American highway administration has also published a design specification for the design of geosynthetic reinforced bridge abutments, which is called Geosynthetic Reinforced Soil-Integrated Bridge System (GRS-IBS) [9].

Figure 2.

Integrated bridge with geosynthetic reinforcement in Hokkaido Shinkansen line [8].

It is a well-accepted fact that geosynthetic-reinforced soil retaining structures are also much less demanding in terms of foundation soil bearing capacity because they do not exert an overturning moment to the foundation soil, which tends to create the maximum stress at the toe of the retaining structure. To highlight this advantage, in the comparative example given in Ref. [8], it can be clearly seen that for the reinforced concrete retaining wall system shown in Figure 3a, a pile foundation was necessary since the foundation soil was quite weak. However, in the solution given in Figure 3b, namely when the retaining walls are designed as geosynthetic-reinforced walls, a surficial foundation was found to be sufficient. This essentially indicates that the geosynthetic-reinforced wall technology provides an advantage, especially in retaining walls that need to be built on soft soils. It goes without saying how much of an advantage it would be to save both time and cost to be spent on pile foundation construction.

Figure 3.

Embankment wall alternative design comparison from Hokkaido Shinkansen line [8].

The information briefly summarized above shows that geosynthetic-reinforced retaining walls offer solutions that provide great advantages in terms of both engineering performance and economy. Another advantage of geosynthetic-reinforced retaining walls is that they can easily adapt to a wide variety of geometric conditions. Since they are not built-in panels like reinforced concrete retaining walls, they can easily adapt to the steep slopes existing along the length of the wall at the foundation level (Figure 4), and the top elevations of the walls to the desired geometry.

Figure 4.

Example of a geosynthetic-reinforced retaining wall constructed in a V-shaped valley.

As geosynthetic-reinforced retaining walls can easily adapt to geometric changes in the longitudinal direction, they can as easily adapt to geometric constraints on the cross-section as well. An interesting example to demonstrate this flexibility in cross-section is seen in Figure 5. In a reinforced concrete wall, if there is a need for stairs connecting the bottom and top elevations, the only alternative is to build a separate reinforced concrete ladder structure. However, as can be seen in Figure 5, the staircase has been created within the wall by systematically shifting the facing out of the wall at different elevations.

Figure 5.

Geosynthetic reinforced soil retaining structures with built-in stairs.

Because in reinforced concrete systems, it is not possible to place the foundation on the filled ground, the stem has to cover the elevation difference as one piece. However, as we can see in the example in Figure 3, retaining walls with geosynthetic reinforcement can easily be built on the embankment since they will exert nearly uniform vertical pressure on the foundation soil. In the figure, the distance between the two facings can vary between zero and any value (Figure 6). Undoubtedly, necessary analyses should be performed depending on this distance [4].

Figure 6.

Example of a tiered wall on a bridge approach fill. All the stair structure has also been constructed using the reinforced soil system.

Finite element analyses were used to investigate several parameters for a parametric study [10]. The results of the finite element analyses indicated that, under working loads, the maximum shear stresses occur along the classical expected potential failure plane. However, when in the finite element analysis the shear strength is gradually reduced up to failure (phi-c reduction), no internal stability problem within the reinforced soil zone is observed. The classical failure plane expected to occur in the unreinforced backfill, indicating an external stability problem (Figure 7).

Figure 7.

Incremental shear strain contours: (a) at end of construction, and (b) at end of ϕ–c reduction in the same model [10].

As it is known, in practice often a suitable soil or rock excavated locally is used as a backfill in geosynthetic-reinforced retaining walls. Hence, it is important to understand how the cohesive backfill will affect the behavior of geosynthetic-reinforced walls. In this respect, Ref. [11] reports furthermore the behavior of geosynthetic-reinforced walls with clayey and granular backfills under earthquake loads. One of the most important results of the study is that if cohesive soil is used instead of granular soil as backfill, wall deformations are reduced by nearly 50%. This finding indicates that cohesive soils can be a suitable backfill material if proper drainage and compaction are provided. The parametric study further showed that lower displacements were obtained for higher-strength reinforcements, but the tensile forces on the reinforcements increased. Another finding mentioned by the paper is that doubling the earthquake acceleration magnitude caused the lateral displacements to increase more than twice, indicating that the acceleration and displacements do not have a linear relationship (Figure 8).

Figure 8.

Horizontal displacements for accelerations of 0.4 and 0.8 g [10].

This fact is also supported by the numerical study, where it is reported that less tensile forces occur in reinforcements in cohesive backfills than in granular backfill. As a result of the numerical model study, it has been shown that cohesive soils can also be a suitable filling material if the appropriate drainage measures are taken [11].

Ref. [12] tested 1.9 m high, 1/2 scaled reinforced soil walls with concrete modular block-facing elements on a shaking table. The models were exposed to the scaled El-Centro earthquake record. Two different reinforcement lengths, L/H = 0.6 and L/H = 0.9, were used in the experiments. As a result, it was concluded that the geosynthetic-reinforced walls performed well under earthquake loading and no permanent displacement was observed at the end of the seismic loading. The acceleration measurements indicated that for all models the accelerations at the facing increased with increasing height of the wall. Measured accelerations at different time intervals are seen in Figure 9. It can be seen that the accelerations vary at different times, but the general trend that accelerations increase with increasing height is seen clearly.

Figure 9.

Maximum accelerations measured on the front wall of various walls shaken by the El-Centro earthquake [12].

Ref. [12] further showed that the failure plane is in good agreement with the Rankine failure plane as can be seen in Figure 10.

Figure 10.

Internal failure surface determination and the regions where peak average tensile strains are observed during the El-Centro excitations [12].

Ref. [13] tested eight modular concrete blocks under earthquake loads again on a shaking table. A total of eight models were tested: four 2 meters tall and 1/2 scale, one 1-meter tall and 1/4 scale and three 2 meters tall and 1/4 scale. All the walls had concrete modular blocks as facing elements. Experimental results showed that reinforcement length and density did not affect the amplification factors much. Face displacements are also less affected as long as the reinforcement length meets the minimum requirements in the FHWA design criteria. No significant permanent displacement was observed at the end of the experiments. For short reinforcements, the measured tensile stress on reinforcements was higher than the tensile stresses calculated in the design. At the end of the study, the failure surfaces obtained by combining the locations of the maximum stresses on the geotextiles and the failure planes suggested by the Rankine theory were compared (Figure 11). It can be seen that the failure planes were in general in good agreement with the Rankine plane. The only exception was seen for the walls with longer reinforcements, where the potential failure plane developed closer to the facing.

Figure 11.

Potential failure surfaces determined from geotextile strains compared with the Rankine failure plane shown on the prototype walls represented by the tested model configurations (values beside illustrations are given for the prototype wall) [13].

In another series of shaking table tests, the model of a 3.8 m high prototype wall from a real design has been tested [14]. Because of the payload capacity of the shaking table, a scaling factor of 2 is used. So, four 1/2 scale and 1.9 meters high reinforced soil retaining wall models were tested under 1-g loading conditions. Their aim was to investigate the effects of various parameters on the seismic performance of the wall. The parameters investigated were the magnitude of horizontal acceleration, reinforcement length and strength, type of backfill soil (sand versus gravel), and face inclination (vertical versus 6-degree batter). The model walls were instrumented as follows: i) nine accelerometers, five of which were installed on the facing elements of the wall and three on top of the backfill and one accelerometer on the shaking table, ii) six laser displacement sensors to measure the displacement of the wall face and, iii) sixteen strain gages installed on three different geogrid reinforcement layers (bottom layer, at mid-height layer, and top layer). A schematic view of the model is seen in Figure 12. The model was shaken by scaled versions of El-Centro earthquake. When the amplitude was increased to 250% of the original record, the accelerations reached almost a level of 1 g. Even under these extreme accelerations, the walls manufactured in accordance with the current regulations showed a successful performance. Even at very extreme horizontal acceleration values, no stability problem occurred, and the deformations remained within acceptable limits. The maximum and residual (permanent) displacements at these loading values are given in Table 1.

Figure 12.

Schematic of the tested model wall (Guler and Basbug, 2012) [15].

	Gravel backfill		Sand backfill
	Peak displacement (mm)	Residual displacement (mm)	Peak displacement (mm)	Residual displacement (mm)
100% El-Centro	1.9	0.5	2.2	0.6
250% El-Centro	28.1	4.7	30.6	4.3

Table 1.

Peak and permanent displacements under various earthquake magnitudes [15].

Ref. [16] studied the components of the Geosynthetic Reinforced Soil-Integrated Bridge Abutment (GRS-IBS) method suggested by Ref. [9] with a finite element analysis. For the finite element model, a fully instrumented real prototype bridge abutment was used to calibrate and validate the finite element model (Figure 13).

Figure 13.

Field monitored GRS-IBS bridge abutments built in Virginia [16].

Effects of the presence of a bearing bed and reinforced soil foundation (RSF), the vertical reinforcement spacing and reinforcement stiffness, and subgrade compressibility on design components, including the lateral displacement of facing, maximum tension in reinforcement (T_max), connection strength (T_o), and differential settlement in reinforced soil structures was investigated [16]. It was determined that the inclusion of a bearing bed, which is a densely reinforced zone below the superstructure, contributes to the reduction in the lateral displacement of facing. The bearing bed also reduces the maximum tension in reinforcements by 20%. The presence of the bearing bed decreased the settlement difference between the superstructure and integrated approach zones for the reinforcement spacing for the wall, which had a reinforcement spacing of S_v = 0.4 m. The vertical reinforcement spacing is an important variable that controls the behavior of the geosynthetic-reinforced soil walls. Decreasing S_v from 0.4 m to 0.2 m decreases the lateral displacement of the facing by up to 1.8 times, the maximum tension in the reinforcement by up to 2.2 times, and the connection strength by up to 4 times. It was also determined that the magnitude of the pressure applied from the superstructure affects the distribution of T_max with depth. When the GRS was modeled with S_v = 0.2 m and nonuniform reinforcement distribution (a typical GRS-IBS structure), the results showed that the distribution of T_max under typical working load conditions (surcharge of approximately 40 kPa) shows a uniform trend with depth but with a high bridge superstructure load (approximately 200 kPa) the trend shifts to become a bilinear distribution. However, when the GRS was modeled with Sv = 0.4 m and uniform reinforcement distribution (a typical MSE structure) the distribution, regardless of the magnitude of the superstructure loads, shows a linear distribution with depth. This behavior difference can be most likely attributed to the stiffer behavior due to closely spaced reinforcement. Figure 14 shows changes in lateral displacement of facing with depth. A comparison of the results from the model with reinforced soil foundation (RSF) and ordinary foundation (NRSF) shows that the presence of RSF reduces the lateral displacement of the facing by up to 60%. It was further recommended based on the parametric study that the connection strength must be considered in design even in closely spaced reinforced soil structures. As summary, the inclusion of the bearing bed and reinforced soil foundation was found to have beneficial effects in reducing the lateral displacement of the facing, the maximum tension in the reinforcement, the connection strength, the foundation settlement, and the settlement difference within the abutment.

Figure 14.

Effect of foundation soil reinforcement suggested by GRS-IBS on horizontal deformation of the wall [16].

As mentioned above, in many projects, cohesive soils or weathered rock is used as a backfill because it is locally available and therefore a great saving both in terms of money, and probably more important in terms of CO2 footprint can be achieved. Having seen that the cohesive soil can be efficiently implemented as a backfill in geosynthetic reinforced walls, [17] investigates the seismic performance of geosynthetic-reinforced modular block retaining walls backfilled with cohesive, fine-grained clay-sand soil mixture on a shaking table. Tests were performed for three 1/2 scaled (wall height 1.90 m) and 1/4 scaled model walls to investigate the effects of backfill type, the influence of reinforcement length, and reinforcement stiffness effects. The El-Centro and Kobe earthquake records of varying amplitudes were used as base acceleration. In conclusion, when relative displacement values were compared, model walls with cohesive backfill exhibited superior seismic performance compared to the model walls with granular backfill. As seen in Table 2, the maximum displacements observed even under 125% Kobe earthquake were lower than the maximum deformation, which occurred under 100% Kobe earthquake in sand backfill. When the results of reinforcement stress measurements were compared, it was seen that seismic loading-induced strain demand on geogrids in cohesive backfill was as high as their counterparts embedded in granular backfill. No remarkable impact of reinforcement length was observed during the tests for cohesive backfills. For all tested model walls, no failure and extreme deformation were observed even under extreme seismic loads. The model walls exhibited minimal residual deformations on the facing as a result of the applied base excitations for both cohesive and granular backfill. Seismic loading-induced settlements were observed, however only at the back of the model. Some minimal tension cracks were detected in the unreinforced backfill zone for cohesive backfills. The seismic performance of all setups could be acknowledged as successful when the results were compared with the limit values reported in the literature. The authors provide the following important disclaimer: “It must be stated that these favorable results were observed in the protected laboratory environment, where atmospheric conditions are controlled. In order to transfer the advantage of cohesive backfill into practice, proper drainage measures must be taken to prevent the wetting of the cohesive backfill.”

Backfill Soil	L/H	Base excitation	Peak displacement (mm)
Sand	0.8	100% Kobe Earthquake	34.6
30% Clay and 70% sand	0.8	125% Kobe Earthquake	32.4
30% Clay and 70% sand	0.6	125% Kobe Earthquake	34.3

Table 2.

Properties of model walls and measured maximum displacements [17].

Although retaining structures are commonly analyzed in two dimensions because they are relatively long structures, in some cases, due to property boundaries, it may be necessary to include corner turns in geosynthetic-reinforced retaining walls. This corner turns can be induced at a right angle or greater/smaller angles. Ref. [18] reports about some cracking and separation of facing blocks in the near vicinity of a right-angle corner of a 13 m high real wall as seen in Figure 15.

Figure 15.

Geosynthetic-reinforced soil wall with concrete block facing elements [18].

In order to analyze this problem, a 3D finite element model was built using 467,221 elements. However, this model could not be run on a windows-based computer. So, a scaled-down model was constituted to understand the mechanisms involved (Figure 16). In conclusion, it was seen that the soil modulus and reinforcement stiffness both play important roles in the cracking of facing blocks.

Figure 16.

3D finite element model of the corner of a geosynthetic-reinforced wall with block-facing elements [18].

As seen in Figure 17, as the modulus of the backfill soil increases, the deviator stress, which can be influential on the cracking of the facing elements, reduces significantly. Similarly, the relationship between reinforcement stiffness and deviator stress acting on the facing blocks as shown in Figure 18 indicates that the reinforcement stiffness is equally effective in reducing the deviator stress acting on the facing block. So, it is recommended that in order to reduce the cracking of the facing blocks at corner turns, the backfill should be compacted as much as possible and a stiffer geosynthetic reinforcement should be used. Furthermore, the results of the 3D finite element model indicated that the wall undergoes a curved form in the plan and this causes the separation of the blocks.

Figure 17.

Maximum deviator stress acting on facing blocks versus soil modulus [18].

Figure 18.

Maximum deviator stress acting on facing blocks versus reinforcement stiffness [18].

It was determined that the abrupt change in the slope, which is also responsible for the cracking of the blocks, occurs only when the reinforcement stiffness is relatively low. When the stiffness was increased, the abrupt change in the slope ratio suddenly diminished (Figure 19). This indicates that there should be a minimum reinforcement stiffness that is required for the integrity of the blocks on the wall corner.

Figure 19.

Horizontal stress concentration was observed close to the corner in the FE model, where block cracking was observed in the real wall [18].

3. Reinforced embankments

Another reinforcement application of geosynthetics is reinforced foundations. This can be a very economical solution if the bearing capacity of the soil is not sufficient to carry the foundation loads safely, because the other alternatives are improving the bearing capacity with solid inclusions, which are relatively costly compared to just using one or several layers of geosynthetic below the foundation. There are many scientific publications reporting on theoretical and experimental studies to understand the role of reinforcement materials in improving the bearing capacity of foundation soils. Experimental studies to evaluate the bearing capacity of footings on reinforced sandy soil were reported by Refs. [18, 19, 20, 21, 22, 23, 24, 25]. These studies show that the reinforcement configuration values that give the maximum bearing capacity value depend on soil and footing types. Ref. [20] investigated the influence of the model footing diameter and embedment depth on the bearing capacity of circular shallow footings by centrifugal model testing. Ref. [19] performed laboratory tests to determine the behavior of geosynthetic-reinforced sandy soil foundations and studied the effect of different parameters contributing to their performance. The ultimate bearing capacity of a circular footing, placed over a soil mass reinforced with horizontal layers of circular reinforcement sheets, was determined with the limit analysis in conjunction with finite elements and linear optimization by Ref. [21]. The critical positions and corresponding optimum diameter of the reinforcements to achieve maximum bearing capacity were established, and a marked improvement in the bearing capacity is evident in the case of two layers of reinforcements rather than a single layer of reinforcement. All these studies show the importance of reinforcement for the bearing capacity of foundations. Several researchers have investigated the degree of improvement achieved with different reinforcement lengths. Ref. [22] determined the optimum length of the reinforcement layer as L = 2.5B for a square footing on geogrid-reinforced sand. Ref. [23] found the optimum reinforced length for strip footing for reinforced sand and reinforced clay to be L = 8B and L = 5B, respectively. Ref. [24] stated that the optimum reinforcement length is between L = 5B and L = 7B. Ref. [25] found the optimum reinforcement length to be L = 5B. As seen from the studies in the literature, there is no unique value proposed for the reinforcement length. If one was to accept the conclusions of the previous studies, the optimum length of the reinforcement layer for maximum bearing capacity lies anywhere between L = 2B and L = 8B.

Only a few studies have compared the behavior of different reinforcement types [22, 26, 27]. Thus, the effect of reinforcement type has not been investigated thoroughly. However, the performance of reinforced soil foundation depends on the interaction between the soil and the reinforcement and, therefore, the properties of the geogrid play a major role.

Ref. [28] conducted laboratory model tests using four different geosynthetic reinforcements in order to understand the effect of reinforcement type along with the effects of the depth of the first reinforcement layer, the vertical spacing between consecutive layers of reinforcement, the total number of reinforcement layers, and the width of the geosynthetic reinforcement. The test setup is seen in Figure 20. The soil used in the tests was a dry poorly graded sand (C_u = 2.5) and was placed into the model box using a raining technique, at a relative density of 46%. As reinforcement, four different geosynthetic products were used. One of them was a polypropylene woven geotextile with an ultimate tensile strength of 60 kN/m. The remaining three were geogrids. Geogrid 1 and Geogrid 2 were woven polyester geogrids with an ultimate tensile capacity of 35 and 55 kN/m respectively. Geogrid 3 was an extruded polypropylene geogrid with an ultimate tensile capacity of 45 kN/m.

Figure 20.

Test setup (a) schematic diagram, and (b) photo of test setup [28].

Ref. [28] concluded that the improvement in the bearing capacity due to reinforced foundation soil is typically reported for the ultimate condition. However, from a practical point of view, we have also to limit the amount of vertical settlement. As such, it is important to note the vertical settlement, which occurs with increases in the bearing capacity. The results of the test indicated that the amount of bearing capacity at small settlements and large settlements varies significantly and that one product showing a very good improvement under large settlements may give poor results under small settlements and vice versa.

Figure 21 shows that for all reinforcement types, for all number of reinforcement layers, the bearing capacity of the reinforced soil foundation increased with increasing settlement ratios. Especially at smaller settlement ratios (s/B = 0.05–0.2) and for one to four reinforcement layers, the woven geogrid gave a higher improvement in the bearing capacity compared to extruded geogrid. The Bearing Ratio values of geotextile reinforced soils were smaller than those of the other reinforced types at all settlement ratios. The authors further highlighted that as the number of reinforcement layers increased, the increment of the Bearing Ratio values rose and it had a linear effect. For N = 5, the maximum BR occurred for Geogrid 1 reinforced sand, but the minimum BR value occurred for Geogrid 3.

Figure 21.

Comparison of different reinforcement types to bearing ratio for reinforcement length of L/B = 3 and different settlement ratios (a) s/B = 0.05, (b) s/B = 0.2, (c) s/B = 0.4, and (d) s/B = 0.6 [28].

Based on the results of many experimental data, Ref. [29] suggested a limit equilibrium method to determine the bearing capacity of strip foundations on geosynthetic-reinforced sand soils. When a reinforcement of the foundation soil is considered, the geosynthetic reinforcement will be embedded into the soil. Therefore, in general, there are two soil layers below the foundation, namely the natural soil and the fill soil. Therefore, the limit equilibrium approach that allows the calculation of the bearing capacity of a two-layered soil was adapted to analyze the reinforced foundation problem. Since the soil above the reinforcement will be compacted properly, it was considered that the natural soil is less dense than the fill soil. The geometry and failure concept adapted is shown in Figure 22.

Figure 22.

Simplified slip surface with coulomb wedges to analyze a geosynthetic-reinforced foundation [29].

Although the literature suggests that it is beneficial to extend the reinforcement outside the footprint of the foundation, construction restraints may not allow this and this will lead to the condition that the reinforcement can be placed only as wide as the foundation itself. Therefore, the formula developed also considers the reinforcement length, which is equal to the foundation width (L/B = 1). The most important unknown in the bearing capacity calculation is the activated reinforcement stress. Ref. [29] suggested a relation obtained by the multiple regression as shown in Eq. (1).

$$\mathrm{T}=110\left(\mathrm{h}/\mathrm{B}\right)+68.8\mathrm{B}+5.07\mathrm{\varphi }+8.05\left(\mathrm{L}/\mathrm{B}\right)-267$$E1

where T is the tensile stress in the reinforcement, h is the distance of the reinforcement from the base of footing, B is the footing width, ϕ is the internal friction angle of the upper strata, and L is the reinforcement length.

Along with the bearing capacity, the amount of allowable settlement is also very important in determining the allowable bearing capacity of soil. In this respect, the pressure distribution below a foundation resting on reinforced soil must be known. However, the stress distribution below foundations resting on reinforced soil foundations was investigated very little. In the literature related to reinforcing unpaved roads, the general assumption is that the reinforcement helps to distribute the load to a larger area and hence reduces the settlement. However, these studies are in general related to pavement analyses, and hence consider well-compacted granular materials, rather than loose sands, which necessitate a soil improvement. Consequently, the stress distribution in medium dense and loose sand is not considered. Ref. [30] provides important information on the stress distribution below the foundation based on experimental data and finite element verifications. Although stress values measured in the laboratory test are slightly smaller than FEM in some conditions, the authors conclude that there is a good correlation between the experimental measurements and the FE analysis results. The most important finding of this research was that the presence of reinforcements transfers vertical stress to deeper levels. However, it must be noted that a difference in the behavior exists depending on the reinforcement type and the reinforced zone depth.

4. Geosynthetic encased columns (GECs)

Ground improvement using ordinary stone columns (OSCs) in soft clayey soils is a cost and time-efficient soil remediation technique that has been practiced for decades. OSCs have been used in a wide spectrum of applications for enhancing the foundation soil properties of rigid and flexible structures, such as buildings, embankments, and oil storage tanks, that are founded on weak clays [31]. OSCs are also used efficiently used in mitigating the liquefaction risk of loose saturated sands [32]. While OSCs have proven themselves to be viable method, there are inherent shortcomings associated with the use of OSCs in soft soils. Early studies have pointed out that their stability is predominantly based on the available lateral support that is provided by the surrounding soil. When implemented in extremely soft soils (s_u < 15 kPa), the columns usually fail in bulging due to the lack of lateral support that the weak soil can offer. One way to overcome bulging failure is to encase the granular column materials with a reinforcing geosynthetic and thereby forming a geosynthetic encased column (GEC), which increases column performance by providing lateral confinement [33, 34]. The engineering behaviors of OSCs and GECs have been studied by means of both laboratory and field experiments, finite element methods, and analytical models. Most of the laboratory tests reported in the literature deal with the physical modeling of columns with small-scale models [35]. Also, field tests of encased columns have been conducted and it was demonstrated that the presence of encasement around the column enabled the column to support 2.3 times the total applied vertical stress [36].

Although there is plenty of literature on the vertical stress-strain behavior of GECs and OSCs, there are not many studies in the literature on the behavior of GECs and OSCs under the action of dynamic loads. Ref. [37] conducted a finite element analysis to model GECs under the action of seismic input motions and determined that implementation of GECs to support embankments underlain by weak soils greatly reduces the seismically induced settlements. The aim of this study is to shed light on the seismic behavior of GECs and OSCs supporting earthen embankments underlain by weak clay.

Because of the limitations of finite element analysis, it is not possible to simulate the intrusion of the stones of the stone column into the soft clay and thereby expand in volume and hence reduce in stiffness. Therefore, it is obvious that the improvements determined with the help of finite element analysis show only a partial contribution of the geosynthetic encapsulation limited to the increased stiffness due to the confining it applies to the stone column. However, with the use of three-dimensional finite element analysis, it has been determined that the total settlement under earthquake loading conditions can be reduced with the help of geosynthetic encapsulation. It was also shown that the GECs undergo less lateral displacements than the ordinary stone columns. Finite element analyses also show that significant tensile stress develops in the geosynthetic encapsulation, indicating the prevention of further bulging under earthquake loading conditions. In summary, the encapsulation of stone columns by a geotextile reinforcement (instead of installing ordinary stone columns), helps with the integrity of the column in very soft soils and reduces the system displacements under seismic impact.

Ref. [38] investigated the seismic behavior of GECs and OSCs installed in soft clay under embankments. The clay had an undrained cohesion of 4.25 kN/m² and the geosynthetic encasements GT1, GT2, and GT3 had a secant modulus of 18, 440, and 1150 kN/m, respectively at a strain of 10%. 1-g shaking table tests are conducted on OSCs and GECs. Figure 23 shows the results from stress-controlled vertical load tests of all columns and the clay beds after the seismic excitations phase. The difference in pressure-settlement behavior of ordinary columns and columns reinforced with GT1 is not significant. This is because GT1 offered very little confinement therefore including GT1 as a media to attach the strain rosettes did not introduce significant error to the assumption that columns reinforced with GT1 in fact behave like ordinary stone columns. GEC-3 with gravel infill experienced a minimum settlement value of about 55 mm at the end of the stress-controlled loading phase. GEC-2 experienced larger maximum settlements of 73 and 68 mm, for sand and gravel infill, respectively.

Figure 23.

Load settlement curves of clay, OSC, and GECs after the system has been exposed to earthquake loading conditions [38].

For the testing of inclusions into soft clay under earthquake loading conditions, the boundaries of the model must satisfy the free field condition. Therefore, flexible containers or more specifically laminar boxes constitute the vast majority of the cases reported in the literature for models tested with shaking tables. Such laminar boxes are able to efficiently model soil’s free field response to input motions. The laminar box reported in Ref. [39] is square in plan with inner clearances of 900 × 900 mm. The height of the samples that can be accommodated in the laminar box is 1932 mm. A 300 mm deep rigid base cavity underlay the laminates (Figure 24). The laminar box consists of 16 individually supported laminates, which are made up of aluminum sigma profiles. A vertical clearance of 2 mm is provided between the laminates. The sigma profiles used are rectangular in cross-section with a width and height of 50 and 100 mm, respectively. The laminar box design makes use of the hollow space at the center of the sigma profiles and smooth guide rods are placed throughout the sigma profiles in the direction of shaking. Teflon riders are fitted in the ends of the aluminum profiles so that there is minimum resistance to horizontal movement. A laminar box commissioned as such, enables purely horizontal laminate movements without causing any rocking or twisting with respect to the vertical axis and diminishes the possibility of cantilever deformations and the toppling of the laminar box. Since the sigma profiles have hollow cross-sections, a significant reduction in mass is achieved while retaining the flexural rigidity of the laminates, which provides unyielding boundaries for the housed soil specimen. The resulting laminates have an assembly-to-soil mass ratio of about 7.5%, which ensures that the inertia effects of the laminates are within tolerable limits to study the 1D response of the soil.

Figure 24.

(a) Dimensions of the laminar box, (b) close-up image of the guide rods and laminates, and (c) top view of the laminar box [39].

When an impact-type input acceleration is applied to the table, it reaches a peak value at each laminate at a given time. Using this information, it was possible to determine the delay of peak value for all laminate levels. This information made it possible to determine the speed of the upward traveling shear wave or, in other words, the shear wave velocity.

Using the measured shear wave velocities, the small strain shear modulus (G_max) of the soft soil composite inside the laminar box could be calculated by making use of Eq. (2).

$$\text{Gmax}={{\mathrm{\rho V}}_{\mathrm{s}}}^2$$E2

where ρ is the mass density of soil and V_s is the shear wave velocity. Using an average bulk mass density of 1630 kg/m³ for all models tested, the small strain shear modulus for models of clay, clay with OSC, and clay with two different geosynthetic encasements were calculated to be 115, 259, 595, and 978 kPa, respectively. These results indicate that the inclusion of high modulus GECs at an area replacement ratio of 7% increased the small-stiffness shear modulus of a soft clay bed by almost an order of magnitude.

The shear modulus of the clay and the OSC or GEC-enhanced clay were also determined by plotting the shear strain and shear stress loops given in Figure 25. Figure 25 illustrates the hysteresis loops extracted for the 10th cycle of 0.5 Hz sinusoidal excitation. The secant shear stiffnesses values for models with clay, clay improved with OSC and clay improved with a less stiff encasement and clay improved with a stiffer encasement are 14.5, 21.1, 39.6, and 90.6 kPa, respectively.

Figure 25.

Hysteresis loops for models consisting of clay alone and clay bed with OSC and GECs [39].

Most of the available experimental data pertain to the engineering behavior of GECs under vertical loading as briefly summarized above. However, especially the rigid inclusions, such as piles, OSCs, or GECs, close to the edge experience shear stresses. This is a well-known fact and studied for all embankment designs under the concept of edge failure. Although this constitutes a real threat to the stability of both OSCs and GECs, there are only limited studies on the behavior of such columns under shear loading [40].

Ref. [41] developed a unit cell shear device to study the behavior of OSCs and GECs behavior under shear forces. This device has a maximum height of 1850 mm and a diameter of 460 mm. The device allows for the installation of a single granular column with a diameter of 113 mm at the center of the unit cell. This corresponds to an area replacement ratio (the ratio of column area to unit cell area) of 6%, which is a lower-bound value in field applications. The device can be operated to shear the specimen under static loads or under cyclic loads. In the tests along with the pure loose sand also soil improved with OSCs and GECs were tested. For GECs, three types of geosynthetics have been used, namely with secant modulus values of 36, 380, and 1050 kN/m at 2% strain and are denoted as J35, J400, and J1000, respectively. The Mohr-Coulomb failure envelopes obtained are shown in Figure 26. From Figure 26, it is clear that the shear resistance of stone columns is significantly improved with the help of the encasement. It can be seen, that the internal friction angle of the loose sand (23.3°) could only be increased to 24.7° with stone columns alone. However, the overall internal friction angle for the unit cell was increased to 28° and 35° for geosynthetic encasements of J400 and J1000, respectively. The cohesion intercepts varied between 2 and 8 kPa.

Figure 26.

Mohr-coulomb failure envelopes for the static shear tests [41].

The encasements of OSCs were equipped with strain gages to determine how the shear plane affects the columns. From Figure 27 it can be seen that the shear applied on a single plane affects the column along a height approximately four times the diameter of the OSCs and GECs. It is also seen from the same figure that the GEC encasements are both strained in the vertical and horizontal directions. Especially the significant strain in the vertical direction hints at the tensile stress developed, which shows that the tensile strength in the encasement significantly increases the shear resistance of the column.

Figure 27.

Maximum values of (a) vertical and (b) horizontal strains were measured in static shearing of unit cells over the height of the columns [41].

5. Conclusion

Reinforcement applications using geosynthetics cover a very wide area. In this synthesis paper, it was tried to highlight some of the more common and frequent reinforcement applications as well as some less frequently used reinforcement applications.

In general, the use of geosynthetics allows the engineer to come up with an alternative design, which is generally more economical, takes a shorter time to construct, and reduces the CO2 footprint.

The reader is encouraged to learn more about the advantages geosynthetics provide to geotechnical construction to be able to optimize their design.

Obviously, being a construction material, the properties of each geosynthetic product should be well understood and demonstrated with help of tests. Furthermore, appropriate QA/QC procedures have to be followed during the installation.

Last but not least, since geosynthetics allows the engineer to come up with innovative solutions, it also provides joy and professional satisfaction to the engineer.