Mechanical Behavior of Soil – Geotextile Composites: Effect of Soil Type
A.I. Droudakis and I.N. Markou
Department of Civil Engineering, Democritus University of Thrace, Greece
12 Vas. Sofias str., GR-67100 Xanthi, Greece
[email protected]
ABSTRACT
Triaxial compression tests were conducted on specimens of a granular and a cohesive soil both reinforced with 5 horizontal
layers of woven and non-woven geotextiles. The two soils used in this investigation were dry and dense Ottawa 20-30 sand
and a fine-grained soil of low plasticity compacted with water content close to optimum water content using energy comparable
to that of Standard Proctor compaction test. Stress – strain relationship, shear strength and failure mode of reinforced soil
specimens are used for the comparison of mechanical behavior of these reinforced soils. It is observed that reinforcement
effect on strength and axial strain at failure is greater in sand than in fine-grained soil. This observation can be attributed to the
better sand – geotextile cooperation in comparison to the insufficient cooperation between fine-grained soil and geotextiles.
Triaxial compression tests yielded bilinear failure envelopes for both reinforced soils attributed, however, to different behavior.
Introduction
Geotextile reinforced soil is used in a large number of applications because of its cost and engineering effectiveness. Free
draining granular materials, e.g. sands, are specified as backfill material for reinforced soil structures. The main reasons for
use of granular materials instead of cohesive soils are the volume change potential and inherent low strength of cohesive soils
which make them unsuitable. However, difficulties are encountered when the available quantity of granular materials is
insufficient. Since low-plasticity soils are generally not expansive, they could be used in reinforced soil structures provided the
reinforcement could increase the load-bearing capacity of the soil [1]. So, it is of merit to investigate the possibility of using
geotextiles as reinforcement of cohesive soils, since they are frequently used to improve the drainage characteristics of these
soils.
The mechanical behavior of sand – geotextile composites has been extensively investigated in the past and several research
efforts were based on the results obtained from triaxial compression tests [e.g. 2 – 7]. On the other hand, a number of
investigations of the mechanical behavior of geotextile reinforced cohesive soils was also performed by conducting triaxial
compression tests [8 – 12]. Although the results of the above mentioned investigations have provided valuable information, a
comparison between reinforced sand and reinforced cohesive soil has not been reported. Toward this end, triaxial
compression tests were conducted in order to compare the mechanical behavior of a granular and a cohesive soil both
reinforced with geotextiles and the observed results are presented herein.
Materials
For the purposes of this investigation, Ottawa 20-30 sand and a fine-grained soil were tested. The properties of soils are
presented in Table 1 and their grain size curves are shown in Figure 1. Ottawa 20-30 is a uniform, quartz sand consisting of
ο
rounded grains. This sand has angle of internal friction, φ, equal to 36 (Table 1) in dry condition and at an average relative
density of 84%. The fine-grained soil is a cohesive silty soil of low plasticity and is classified as CL according to Unified Soil
Classification System (U.S.C.S.). The compaction characteristics of fine-grained soil were obtained by conducting Standard
Proctor compaction test. The shear strength parameters of this soil (Table 1) were determined by conducting UnconsolidatedUndrained (UU) triaxial compression tests on specimens compacted with optimum moisture content and compaction energy
equal to that of Standard Proctor compaction test.
Three non-woven geotextiles and two woven geotextiles provided by five different manufacturers were used during this
investigation. These geotextiles were selected in order to have comparable mass per unit area and to cover a wide range of
types of commercially available products. More specifically, one thermally bonded (Typar SF 111), one needle-punched with
thermally treated surfaces (Fibertex F 500) and one needle-punched (Polyfelt TS 70) polypropylene non-woven geotextiles as
well as one high strength polyester (Bonar 150/60) and one standard grade polypropylene (Thrace Plastics 400) woven
geotextiles were tested. These geotextiles are designated as TB, TTS, NP, HS, and SG, respectively. Pertinent geotextile
properties, according to the manufacturers, are presented in Table 2.
Table 1. Soil properties
Soil
Sand
Fine-grained
2.67
2.70
Dmax: 0.85 mm
Sand: 32%
D50: 0.71 mm
Silt: 58%
Dmin: 0.60 mm
Clay: 10%
-
wL: 45%
Atterberg Limits
-
wP: 18%
-
IP: 27%
Compaction
Characteristics
emax: 0.77
wopt: 19.5%
emin: 0.46
γdmax: 1.70 g/cm3
Shear Strength
Parameters
c: 0 kPa
c: 72 kPa
Specific Gravity, Gs
Grain Size
Distribution
ο
φ: 17ο
φ: 36
100
Percent finer (by weight)rg
90
80
Fine-grained soil
70
60
50
40
30
20
Ottawa 20-30 sand
10
0
10.000
1.000
0.100
0.010
0.001
Grain size, (mm)
Figure 1. Grain size curves of soils
Table 2. Geotextile properties
Geotextile
Manufacturing
process *
Thickness
[mm]
Mass per
unit area
2
[g/m ]
TB
TTS
NP
HS
NW
NW
NW
W
0.85
2.20
1.30
1.30
375
370
325
375
SG
W
1.15
400
* NW: non-woven, W: woven
#
Machine direction / Cross machine direction
Tensile test results
Maximum tensile
Extension at
load
maximum load
[kN/m]
[%]
29
70
25
55
24
80 / 40 #
161 / 63 #
12 / 11 #
86 / 86 #
20 / 14 #
Experimental Procedures
Conventional laboratory triaxial compression equipment without modifications was used to conduct tests on geotextile
reinforced soil specimens, in order to investigate the mechanical behavior of composite materials. The cylindrical specimens
prepared, had a diameter of 50 mm and an overall height of 106 mm. Five geotextile discs having a diameter equal to the
diameter of the specimen and placed at equal distances perpendicular to the axis of the specimen, were used for soil
reinforcement. A schematic representation of the reinforced soil specimens is shown in Figure 2a. Specimen configurations
same as that of Figure 2a have been used previously [13, 14] in laboratory investigations using triaxial compression tests. The
reinforced sand specimens (Figure 2b) were prepared with compaction of dry sand using a special hand operated tamper and
extreme care was taken in order to produce sand layers with constant density. All tests were conducted at a relative density of
sand between 84% and 94%, with cell pressure, σ3, equal to 50, 100, 200, 400 and 600 kPa and at a constant axial
displacement rate of 0.57 %/min.
The fine-grained soil was compacted using energy comparable to that of Standard Proctor compaction test, with water content
ranging from 18.3% to 19.0%. As it can be seen in Figure 2c, all soil layers had the appropriate thickness at the end of the
compaction process, indicating a uniform distribution of the compaction energy. Reinforced fine-grained soil specimens had
3
saturation ratio, void ratio and dry unit weight values ranging from 75% to 90%, from 0.57 to 0.68 and from 1.61 g/cm to 1.72
3
g/cm , respectively. The observed differences in the parameter values mentioned above can be attributed to the
compressibility of geotextiles used, since it is believed that high geotextile compressibility leads to a decrease in compaction
energy and, as a result, to a decrease in compaction degree of the soil mass. All tests on reinforced fine-grained soil
specimens were unconsolidated-undrained (UU) and conducted with cell pressures, σ3, equal to 10, 25, 50, 100 and 200 kPa
and at a constant axial displacement rate of 0.57 %/min, which corresponds to undrained loading conditions.
(a)
(b)
(c)
Figure 2. Reinforced soil specimens: (a) schematic representation, (b) reinforced sand, (c) reinforced fine-grained soil
Stress – Strain Relationship
Typical stress – strain curves obtained by triaxial compression testing of sand and fine-grained soil reinforced with the same
geotextile, are presented in Figures 3a and 3b, respectively. It can be observed that reinforced sand presents a maximum
value of deviator stress (failure deviator stress) followed by a considerable decrease in deviator stress. On the contrary,
reinforced fine-grained soil presents either a peak deviator stress (failure deviator stress) followed by a negligible decrease in
deviator stress or a continuous increase in deviator stress as the axial strain increases. Accordingly, in tests not presenting a
maximum deviator stress, the failure of specimen and, as a result, the failure deviator stress were defined to correspond to a
value of axial strain equal to 20%. Stress – strain curve of reinforced fine-grained soil having similar shape to the ones
presented herein, were reported by Fabian and Fourie [12]. It is also observed (Figure 3a) that an increase in cell pressure, σ3,
causes an increase in failure deviator stress of reinforced sand. This behavior was also observed in fine-grained soil
specimens tested with cell pressures up to 50 kPa. On the other hand, reinforced fine-grained soil specimens tested with cell
pressures greater than 50 kPa, present either negligible increase or no increase or even a decrease in failure deviator stress
as cell pressure increases (Figure 3b). As an exception, fine-grained soil specimens reinforced with HS geotextile show the
same behavior with reinforced sand.
In order to quantify and compare the deformability of reinforced soils, the axial strain ratio, εfr/εfu, defined as the ratio of the
axial strains at failure of reinforced and unreinforced soil for the same cell pressure, is used. The εfu, εfr and εfr/εfu values
obtained, are presented in Table 3. It can be seen that unreinforced fine-grained soil presents higher values of axial strain at
failure than unreinforced sand. For this reason, the values of εfr/εfu ratio of fine-grained soil are lower than the ones of sand,
although reinforced fine-grained soil presents higher values of axial strain at failure than reinforced sand. Therefore, it is
concluded that the effect of reinforcement on the axial strain at failure is greater in sand than in fine-grained soil. As regards
the effect of geotextile type and properties on the deformability of reinforced soil, it is evident that it depends on soil type, as
well. More specifically, reinforced sand with geotextiles of higher compressibility (NP and TTS) presents higher εfr values than
sand reinforced with the other geotextiles (Table 3). This influence is not observed in reinforced fine-grained soil, where the
axial strain at failure is generally independent of geotextile type.
2500
500
(a)
400 σ3, (kPa)
2000
Deviator stress, σ1-σ3 (kPa)
Deviator stress, σ1-σ3 (kPa)
100
1500
200
1000
100
500
50
(b)
50
400
200
25
300
σ3, (kPa)
10
200
100
Ottawa 20-30 sand
TB geotextile
Fine-grained soil
TB geotextile
0
0
0
5
10
15
20
0
5
Axial strain, ε (%)
10
15
20
Axial strain, ε (%)
Figure 3. Typical stress – strain curves of (a) reinforced sand and (b) reinforced fine-grained soil
Table 3. Values of axial strain at failure and axial strain ratio
Fine-grained
Sand
Soil
σ3
Unreinforced
TB
NP
TTS
HS
SG
[kPa]
εfu [%]
εfr [%]
εfr/εfu
εfr [%]
εfr/εfu
εfr [%]
εfr/εfu
εfr [%]
εfr/εfu
εfr [%]
εfr/εfu
50
2.36
2.35
0.99
5.75
2.44
5.40
2.29
2.36
1.00
3.54
1.50
100
3.27
4.25
1.30
8.73
2.67
6.34
1.94
3.30
1.01
4.01
1.23
200
1.67
4.69
2.81
8.02
4.80
11.03
6.60
3.77
2.26
4.46
2.67
25
8.12
13.21
1.63
19.28
2.37
20.00
2.46
9.29
1.14
10.89
1.34
50
15.02
11.79
0.78
19.73
1.31
19.71
1.31
19.76
1.32
19.56
1.30
100
19.29
16.27
0.84
19.51
1.01
19.71
1.02
20.00
1.04
19.33
1.00
200
17.72
18.66
1.05
19.06
1.08
17.07
0.96
15.95
0.90
18.00
1.02
Shear Strength
Shown in Figure 4 are the failure envelopes of unreinforced and reinforced sand and fine-grained soil. It can be observed that
triaxial compression tests yielded bilinear failure envelopes for reinforced sand which is in agreement with the observations of
other investigators [2, 3]. This bilinear form is attributed to a change of the interaction behavior at the sand – geotextile
interface. More specifically, the part of failure envelope before the break point corresponds to failure of the composite
600
600
TB Geotextile
Unreinforced
Reinforced
400
200
Fine-grained
Soil
(a)
0
0
TTS Geotextile
Sand
Shear stress, τf (kPa)
Shear stress, τf (kPa)
Reinforced
200
400
200
Fine-grained
Soil
(b)
0
0
600
400
200
600
NP Geotextile
Reinforced
400
200
Fine-grained
Soil
(c)
0
200
400
200
Fine-grained
Soil
(d)
0
200
Normal stress, σnf (kPa)
600
400
Normal stress, σnf (kPa)
HS Geotextile
Reinforced
Sand
Unreinforced
400
200
Fine-grained
Soil
(e)
0
0
Sand
Unreinforced
0
600
400
Shear stress, τf (kPa)
Unreinforced
0
SG Geotextile
Sand
Shear stress, τf (kPa)
Shear stress, τf (kPa)
Reinforced
600
400
Normal stress, σnf (kPa)
Normal stress, σnf (kPa)
600
Sand
Unreinforced
200
400
600
Normal stress, σnf (kPa)
Figure 4. Failure envelopes of reinforced and unreinforced soils
600
material by slippage of the geotextile with regard to the surrounding sand. The part of failure envelope after the break point
corresponds to failure caused by excessive deformation during which the geotextile is stretched in unison with the surrounding
sand. The break point of failure envelopes of reinforced sand corresponds to critical values of normal stress, σvcr, ranging from
257 kPa to 290 kPa. It can be clearly seen (Figure 4) that reinforced sand presents higher shear strength than unreinforced
sand, for all the geotextiles tested.
Reinforced fine-grained soil also presents bilinear failure envelopes (Figure 4). With the exception of soil reinforced with HS
geotextile, the part of failure envelopes after the break point is horizontal showing that shear strength of reinforced fine-grained
soil is independent of the applied normal stress. This behavior is similar to that of fully saturated fine-grained soils loaded
under undrained conditions. Therefore, the bilinear form of these failure envelopes is possibly attributed to a change of
draining conditions in reinforced soil as the applied normal stress increases. Curved and bilinear failure envelopes were also
reported by Athanasopoulos [15] for reinforced fine-grained soil, indicating a continuous transition from drained to undrained
behavior as normal stress is increased. The break point of failure envelopes of reinforced fine-grained soil corresponds to
normal stress values ranging from 133 kPa to 208 kPa. However, these critical values of normal stress should not be taken to
represent the transition from slippage failure to stretching failure of the reinforcement but appear to represent transition from
drained to undrained behavior. It is also observed (Figure 4) that reinforced fine-grained soil does not always present higher
shear strength than unreinforced soil.
The strength ratio, SR, defined as the ratio of failure deviator stress of reinforced soil to the failure deviator stress of
unreinforced soil for the same cell pressure, is used for the quantification of the strength increase due to reinforcement of soils.
As it can be seen in Figure 5, the SR values range from 2.17 to 4.62 and from 0.94 to 1.84 for reinforced sand and fine-grained
soil, respectively. It is evident that the reinforcement effect on strength is greater in sand than in fine-grained soil. Reported SR
values by Ingold [10] and Fabian and Fourie [12] for clays reinforced with geotextiles, are generally lower than 2 and,
therefore, are in good agreement with the values presented herein. Strength ratios of sand decrease as cell pressure
increases and strength ratios of fine-grained soil attain to maximum values for cell pressures between 50 and 100 kPa (Figure
5). This difference in behavior is justified by the positions of failure envelopes of reinforced soils relative to those of
unreinforced soils (Figure 4).
5.0
Strength Ratio, SR
4.0
Ottawa 20-30 Sand
3.0
2.0
1.0
Fine-grained Soil
0.0
0
100
200
300
Cell Pressure, σ3 (kPa)
Figure 5. Strength ratio values
Failure Mode of Reinforced Soil Specimens
Shown in Figure 6 are typical forms of specimens of reinforced sand and fine-grained soil, respectively, after triaxial
compression testing. The failure mode of reinforced soil specimens generally consisted of bulging of soil between geotextile
layers. In reinforced sand (Figure 6a), bulges and geotextile discs closer to the mid-height of the specimen were greater in
diameter than the ones closer to the loading surfaces of the specimen. The increase in diameter of geotextile layers indicates
good sand – geotextile cooperation. On the other hand, reinforced fine-grained soil specimens (Figure 6b) presented a nearly
uniform diameter increase in all soil layers and no remarkable diameter change in geotextile layers, indicating an insufficient
cooperation between fine-grained soil and geotextiles. The better sand – geotextile cooperation possibly justifies the higher
strength increases observed in reinforced sand compared to those obtained from reinforced fine-grained soil (Figure 5).
(a)
(b)
Figure 6. Typical specimens of reinforced (a) sand and (b) fine-grained soil after triaxial compression testing
Conclusions
Based on the results of this investigation and within the limitations posed by the number of tests conducted and the materials
used, the following conclusions may be advanced:
•
•
•
•
•
The triaxial compression tests yielded bilinear failure envelopes for geotextile reinforced sand and fine-grained soil. In
reinforced sand, the bilinear form is attributed to a change of the interaction behavior at the sand – geotextile interface,
while in reinforced fine-grained soil, is possibly attributed to a transition from drained to undrained behavior.
Values of strength ratio range from 2.17 to 4.62 and from 0.94 to 1.84 for reinforced sand and fine-grained soil,
respectively, indicating that the reinforcement effect on strength is greater in sand than in fine-grained soil.
Values of axial strain ratio at failure range from 0.99 to 6.60 and from 0.78 to 2.46 for reinforced sand and fine-grained
soil, respectively, indicating that the reinforcement effect on the axial strain at failure is greater in sand than in finegrained soil.
Failure modes of reinforced soil specimens indicate that geotextiles cooperate more effectively with sand than with finegrained soil. The better sand – geotextile cooperation possibly justifies the higher strength increases observed in
reinforced sand compared to those obtained from reinforced fine-grained soil.
The suitability of sands for use as backfill materials in reinforced soil structures is confirmed, while the use of finegrained soils of low plasticity in reinforced soil structures requires very careful consideration.
Acknowledgments
The Research Committee of Democritus University of Thrace provided financing for the research effort reported herein. This
financial support is gratefully acknowledged. Thanks are expressed to Mr. G. Maroulas for his contribution to the construction
of the fine-grained soil specimen preparation equipment. Part of the triaxial compression tests described in this paper, were
conducted by Mr. G. Sirkelis and Mr. Ch. Ioannou whose careful work is acknowledged.
References
1.
2.
3.
4.
5.
Fourie, A.B. and Fabian, A.G., “Laboratory Determination of Clay-Geotextile Interaction”, Geotextiles and Geomembranes,
6, 275-294 (1987).
Gray, D.H., Athanasopoulos, G.A. and Ohashi, H., “Internal / External Fabric – Reinforcement of Sand”, Proceedings, 2nd
International Conference on Geotextiles, Las Vegas, U.S.A., 3, 611-616 (1982).
Gray, D.H. and Al-Refeai, T., “Behavior of Fabric- vs. Fiber- Reinforced Sand”, Journal of Geotechnical Engineering, 112,
804-820 (1986).
Baykal, G., Guler E. and Akkol, O., “Comparison of Woven and Nonwoven Geotextile Reinforcement Using Stress Path
Tests”, Proceedings, International Symposium on Earth Reinforcement Practice, Fukuoka, Japan, 1, 23-28 (1992).
Ashmawy, A.K. and Bourdeau P.L., “Effect of Geotextile Reinforcement on the Stress-Strain and Volumetric Behavior of
th
Sand”, Proceedings, 6 International Conference on Geosynthetics, Atlanta, U.S.A., 2, 1079-1082 (1998).
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Haeri, S.M., Noorzad, R. and Oskoorouchi, A.M., “Effect of Geotextile Reinforcement on the Mechanical Behavior of
Sand”, Geotextiles and Geomembranes, 18, 385-402 (2000).
Wu, J.H., Wang, D.Q. and Wang, L.J., “Experimental Study on Geosynthetic Reinforcement”, Proceedings, 7th
International Conference on Geosynthetics, Nice, France, 4, 1285-1288 (2002).
nd
Ingold T.S. and Miller, K.S., “The Behavior of Geotextile Reinforced Clay Subject to Undrained Loading”, Proceedings, 2
International Conference on Geotextiles, Las Vegas, U.S.A., 3, 593-597 (1982).
nd
Christie, I.F., “Economic and Technical Aspects of Embankments Reinforced with Fabric”, Proceedings, 2 International
Conference on Geotextiles, Las Vegas, U.S.A, 3, 659-664 (1982).
Ingold, T.S., “Reinforced Clay Subject to Undrained Triaxial Loading”, Journal of Geotechnical Engineering, 109, 738-744
(1983).
Ingold, T.S., “Fully and Partially Saturated Reinforced Clay under Undrained Axisymmetric Loading”, Ground Engineering,
18, 27-31 (1985).
Fabian, A.G. and Fourie, A.B., “Performance of Geotextile – Reinforced Clay Samples in Undrained Triaxial Tests”
Geotextiles and Geomembranes, 4, 53-63 (1986).
Atmatzidis, D.K and Athanasopoulos, G.A., “Sand – Geotextile Friction Angle by Conventional Shear Testing”, Proceeth
dings, 13 International Conference on Soil Mechanics and Foundation Engineering, New Delhi, India, 3, 1273-1278
(1994).
Markou, I. and Droudakis, A., “Effect of Triaxial Compression Testing Factors on Sand - Geotextile Interface Friction,”
Proceedings, 8th International Conference on Geosynthetics, Yokohama, Japan, 4, 1401-1404 (2006).
Athanasopoulos, G.A., “Results of Direct Shear Tests on Geotextile Reinforced Cohesive Soil”, Geotextiles and
Geomembranes, 14, 619-644 (1996).