See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/265422132

Large-Scale Triaxial Compression Testing of Geocell-Reinforced Granular Soils

Article  in  Geotechnical Testing Journal · January 1993

DOI: 10.1520/GTJ10050J

CITATIONS READS
131 1,420

2 authors:

Richard J. Bathurst Rajagopal Karpurapu

Royal Military College of Canada Indian Institute of Technology Madras
284 PUBLICATIONS   7,766 CITATIONS    108 PUBLICATIONS   2,247 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Masters thesis View project

Peformance based design of reinforced soil structure View project

All content following this page was uploaded by Richard J. Bathurst on 09 September 2014.

The user has requested enhancement of the downloaded file.

Richard J. Bathurst 1 and Rajagopal Karpurapu 2

Large-Scale Triaxial Compression Testing of Geoceii-Reinforced

Granular Soils

REFERENCE: Bathurst, R. J. and Karpurapu, R., "Large-Scale filled with soil (typically a granular material) and compacted.
Triaxial Compression Testing of GeoceU-Reinforced Granular Soils,'' The composite geocell-soil mat exhibits greater load-bearing ca-
Geotechnical Testing Journal, GTJODJ, Vol. 16, No.3, September pacity and stiffness than the same infill constructed without the
1993. pp. 296-303.
polymeric materiaL The composite layers are often placed at or
ABSTRACT: The paper describes the results of a series of large-scale just below the surface of high-contact pressure pavements (e.g.,
triaxial tests carried out on 200-mm-high isolated geocell-soil com- intermodal terminal yard pavements) and below the tie soffit at
posite specimens and unreinforced soil specimens. Two different ag- railway track turnouts, transitions, and grade crossings (Pro-
gregate soils were used in the test program. The reinforced specimens gressive Railroading 1987). The cellular confinement mechanism
were tested with a heighHo-diameter ratio of unity, which matches
the dimensions of these systems in a typical base reinforcement ap- reduces the lateral spreading of aggregate materials that occurs
plication. The results illustrate the stiffening effect and strength in- under repeated loading in these structures. In some cases the
crease imparted to the soil by the enhanced confinement effect. Com- geocell materials have been used to create a stiffened mat to
parison of reinforced and unreinforced soil specimens shows that the support the base of narrow fills over compressible soils in pipe
frictional resistance described by the peak friction angle of the soil excavation and track applications (Bathurst and Jarrett 1989).
infill is applicable to the composite structure as well. A simple elastic
membrane model can be used to estimate the additional apparent Facia systems for gravity wall structures and for geosynthetic-
cohesion present in the composite structure. reinforced retaining walls have also been constructed using geocell-
soil composites (Crowe, Bathurst, and Alston 1989; Bathurst,
KEYWORDS: triaxial compression, gcocell, reinforcement, geosyn- Crowe, and Zehaluk 1993; Bathurst and Crowe 1992).
thetic reinforcement, base reinforcement

Experimental Method
This paper reports the results of a series of large-diameter
triaxial compression tests carried out on specimens of aggregate
Test Procedure
materials confined by a single flexible-walled geocell manufac-
tured from polymeric material. The geocell-aggregate composites A large-diamet~r triaxial test apparatus 700 mm high with an
were approximately 200 mm in diameter and 200 mm in height. inside diameter of 350 mm and an internal base plate diameter
The polymeric material comprised a single expanded cellular unit of 230 mm was used in the test program. Reinforced and un-
made of 1.15-mm-thick polyethylene. The aggregate materials reinforced specimens were encapsulated in a 230-mm-diameter
were a No. 40 silica sand placed at two different void ratios and rubber membrane 0.6 mm thick.
a well-compacted 20-mm-diameter 100% crushed limestone ag- Unreinforced specimens were prepared within a split cylinder
gregate. The purpose of the test program was to determine strength specimen former and were supported temporarily using a vacuum
properties of the composite geocell-soil systems under standard pressure. Infill soils for the reinforced specimens were prepared
triaxial compression conditions and to isolate the performance within the expanded geocelL All specimens were fully saturated
benefit due to the presence of the geocell membrane. under a back pressure of 15 kPa, which was maintained through-
out the test. Vertical axial stress was applied to the specimens
using a 890-kN capacity compression testing machine. The spec-
Geocell-Soil Confinement in Reinforcement Applications
imens were loaded axially using a rigid aluminum plate. For
The geocell materials that are the focus of this study are made reinforced tests the top platen was machined to just fit within
from strips of polyethylene polymer material welded together to the geocell walls and was inserted to an initial depth of 5 mm
form a three-dimensional cellular network 100 or 200 mm in below the top of the geocell. Reinforced and unreinforced spec-
depth. The cellular materials come in collapsed panels expanded imens were loaded at a constant vertical strain rate of 1% per
to cover an area of 2.44 x 6.1 m. After expansion, the cells are minute (based on original height of specimen), and free drainage
of pore water was permitted at the top and bottom of the spec-
1 Professor, Civil Engineering Department, Royal Military College of
imens. The change in volume of the specimens was measured
Canada, Kingston Ontario K7K 5LO. continuously during the test by weighing the amount of water
lAssistant professor, Civil Engineering Department, Indian Institute expelled or entering the specimen. The vertical deformation of
of Technology, Madras, India 6(XX)36. the specimen was measured at the loading piston using a linear
© 1993 by the American Society tor Testing and Materials
296
 BATHURST AND KARPURAPU ON GEOCELL SOILS 297

voltage differential transducer (LVDT). The load on the speci- 100

mens was measured using a 25-kN capacity load cell. All defor-
mation and load measurements were logged automatically using 80
a computer-controlled data acquisition system. ~ 60
Reinforced specimens were restricted in height by the 200 mm
depth of the geocell units and hence had an aSpect ratio of about
hid = I (height-to-diameter ratio). The sample height is ap-
.~
.s
:; 40
proximately one half that recommended in conventional practice ~
.!! 20 - - No. 40 silica sand
to minimize the influence of rough plate boundaries on the me- - - - limestone aggregate
chanical performance of specimens during triaxial compression 0
(e.g., Bishop and Green 1965; Rowe and Barden 1964; Goto 0.01 0.10 1.00 10.00 100.00
and Tatsuoka 1988). For comparison purposes, unreinforced tests Grain size (mm)
were carried out on specimens with aspect ratios of unity and FIG. 1-Grain-size distribution curves for test aggregates.
hid = 2.2 to quantify edge effects for one soil infill type.
using eight layers of aggregate and a tamping plate. The in situ
dry densities of silica sand specimens were measured using a
Test Program
water-balloon apparatus according to the ASTM Test Method
e A total of 23 triaxial compression tests was carried out using for Density and Unit Weight of Soil In-Place by the Rubber
r the specimen dimensions and materials described below. Balloon Method (D 2167-85). The density of the limestone ag·
gregate was determined from a mass-volume calculation based
Soils- A uniformly-graded silica sand with particle sizes falling
on measurements taken over the entire sample.
,. between No. 40 (0.425 mm) and No. 100 (0.15 mm) sieves was
used in the test program. The second material selected was a Tests-A summary of tests carried out in this study is given
n
100% crushed limestone aggregate with a top size of 19 mm. The in Table 1. A top view of a reinforced specimen prepared with
s the dense limestone aggregate infill is illustrated in Fig. 2. A
grain-size distribution curves are given in Fig. 1. The maximum
e failed specimen with silica sand infill is illustrated in Fig. 3.
and minimum void ratios of the silica sand are 0.76 and 0.56,
0
corresponding to dry unit weights of 14.8 and 16.7 kN/m 3 de-
e
termined according to ASTM Test Method for Minimum Index
I Test Results
Density and Unit Weight of Soils and Calculation of Relative
;.
Density (D 4254-83). The silica sand specimens were prepared Compacted Silica Sand- The stress-strain and volumetric strain
I- at two relative densities, one at its loosest state with zero relative response of the medium dense silica sand tests are plotted in Fig.
t,
density and the other at a medium dense condition with relative 4. The composite specimens showed greater shear strength and
densities ranging from 40 to 56%. The loosest state was achieved a strain-hardening response as compared to the unreinforced
by air pluviation and the denser state by compaction using eight specimens, which showed constant post-peak shearing resistance
layers of soil tamped in place at optimum moisture content. The or strain softening at large deformations. In addition, the com-
limestone specimens were compacted to a relatively dense state posite specimens exhibited greater axial stiffness than the un·

TABLE I-Summary of triaxial compre.uion tests.

Reinforced/ Aspect Ratio, Dry Unit Weight Relative Density, Confining Pressure, Peak Stress, cr 1 - cr,,
Test Number Soil Unreinforced hid ("yd), kN/m' Dn o/o cr' 3 , kPa kPa
r I MDSS" U' 0.9H 15.7 51 10 105
2 MDSS u 0.98 15.8 56 25 195
3 MDSS u 0.98 15.7 51 50 295
"m 4 MDSS u 0.98 15.7 51 75 410
:d
5 MDSS u 0.98 15.7 51 150 850
6 LSS R 0.98 14.8 0 10 715
:d 7 LSS R 0.98 14.8 0 25 835
h- 8 LSS R 0.98 14.8 0 50 920
ns 9 LSS R 0.98 14.8 0 75 1095
C·
10 MDSS R 0.98 15.5 40 10 975
11 MDSS R 0.98 15.7 51 25 1025
or 12 MDSS R 0.98 15.7 51 50 1100
in 13 MDSS R 0.98 15.8 56 75 1270
m 14 MDSS R 0.98 15.7 51 100 1375
·c· 15 CLA R 0.98 16.0 25 1035
16 CLA R 0.98 16.0 50 1375
er 17 CLA R 0.98 16.0 1700
100
ge 18 CLA u 2.2 16.0 25 195
:c- 19 CLA u 2.2 16.1 50 345
ed 20 CLA u 2.2 16.1 100 535
er 21 CLA u 0.98 16.3 25 300
of
22 CLA u 0.98 15.9 50 420
23 CLA u 0.98 15.8 100 690
ar
liS aMDSS = medium dense silica sand; LSS = loose silica sand; CLA = crushed limestone aggregate.
hU = unreinforced; R = geocell reinforced; prime denotes effective stresses.
298 GEOTECHNICAL TESTING JOURNAL

FIG. 2-Test specimen of geocell-reinforced dense compacted crushed limesto11e aggregate (aspect
ratio h/d "" 1).

reinforced specimens at similar confining pressures. After initial tical deformation due to circumferential expansion of the geocell
elastic deformations, the reinforced specimens showed sup- wall. In some reinforced tests the specimen failed at large strain
pressed dilatancy, which is consistent with the notion of increased after rupture of the welded seam (Fig. 3).
confinement due to the flexible geocell wall. At large strains.
the effect of soil confinement by the geocell wall is to maintain Loose Silica Sand-A series of geocell reinforced tests with
the infill soil in a plastic state while increasing resistance to ver- silica sand in an initially loosest state was carried out. Strength
envelopes for silica sand tests are presented later in the paper.
Crushed Limestone Aggregate- Reinforced and umeinforced
tests with a well-compacted crushed limestone were carried out
with 200-mm-high specimens having a height-to-diameter aspect
ratio of unity. To investigate the influence of specimen geometry
on mechanical response, an additional set of unreinforced tests
was carried out that used a conventional aspect ratio of about
2. A summary of peak strengths from crushed limestone tests is
given in Table 1. Qualitative features of the stress-strain and
volumetric strain response are similar to the observations made
for medium dense silica sand tests carried out with and without
geocell confinement (Bathurst and Karpurapu 1992).
The results of triaxial compression tests on umeinforced spec-
imens of well-compacted crushed limestone with a height-to-
diameter ratio of hid = 2.2 and hid = 1 are given in Fig. 5. The
results confirm previous research that shows that nonlubricated
end platens for short samples with hid = 1 will result in a greater
specimen strength than that of specimens constructed with an
aspect ratio of hid ~ 2 or greater (Bishop and Green 1965).
Clearly, conclusions drawn from reinforced specimens in this test
program must be interpreted with caution.

Strength Parameters
FIG. 3-Geocell-silica sand specimen at failure showing seam rupture A summary of strength envelopes for reinforced and unrein-
at large axial strain (Ea = 0.15). forced soil specimens is given in Fig. 6. The strength envelopes
 BATHURST AND KARPURAPU ON GEOCELL SOILS 299

1500,.-------------------,
100
75

1000 25
----1o
gcocell-soil composite

5
a'3 (kPa)
- - - - - - - - - - 75
....... - - - - - - - - - - . 50 unreinforced
// - - - - - - - - - - - - - 25
1:--------- 10
{),1"'---~--~---~---~--........-··---,,---,-----J
0 0.05 0.10 0.15 0.20
vertical strain (ta)
a) principal stress difference versus axial strain

0.06~-------------------~
geocell soil composite
0.05 unreinforced

c
0.0 a'3 (kPa)
II ·;;;
n !:: 0.03 10
50~/y
~

h
·"E
!::
0 0.02 75
/
i silica
h ::>
0 h sand
> 0.01
d
Jt 0.00
100 l
ct geocell
·y -0.01
ts 0 0.05 0.10 0.15 0.20
Jt
vertical strain (Ea)
is
td b) volumetric strain versus axial strain
le FIG. 4- Triaxial behavior of compacted silica sand with and without geocell confinement
(aspect ratio hld = 1).
"'
c- have been based on peak shearing resistance and plotted with from 156 to 190 kPa. The influence of specimen aspect ratio on
J- respect to shear and hydrostatic stress invariants (p,q) defined strength parameters can be seen from the data in Fig. 6b and
Je on the figure. The data from each series of tests plot as straight Table 2 for the unreinforced limestone tests. Reducing the height
:d lines. Unreinforced specimens can be seen to have negligible of the specimens by approximately half increases the peak friction
er apparent cohesion over the range of average normal stress ap- angle from 46.7 to 52.0°. If the relative strength of reinforced
m plied. The extrapolation of a first order linear regression on the and unreinforced limestone specimens with an aspect ratio hid
). reinforced specimen data points shows that these composite ma- = 1 is also valid for specimens constructed with hid = 2.2, then
st terials have essentially the same friction angle as the unreinforced the reinforced coarse aggregate soil specimens in this study may
specimens placed at the same density but that they exhibit a large be expected to overestimate the composite peak friction angle
apparent cohesion due to the geocell confinement effect. Strength by about 5'. The issue of specimen height on geocell composite
parameters in conventional Mohr-Coulomb terminology are strength is further complicated by the observation that the max-
n- summarized in Table 2. The apparent cohesion for geocell com- imum height of the geocell panels (of the generic type used in
os posites constructed with two different aggregate infills varied this study) is 200 mm. The aspect ratio of hid = 1 has been
300 GEOTECHNICAL TESTING JOURNAL

7'~-----------------=-=-~-=-~------------,

/
/
/,../ - 100
/
/
100
o'3 (kPa)
--------
/
,..,....---- ----------- 50
50

-- -- --- ------------ --- -- 25

long sample, h/d=2.2

short sample, h/d=1

0 0.05 0.10 0.15 0.20

vertical strain ( ea)

a) principal stress difference versus axial strain

0.06-r-----------------------,
25

0.
o'3 (kPa)
25 50

0.02 !
j o'3 crushed
f•
h limestone

100
------ --- f--d--
-0.02~+----~--~-~----~-~--~---<
0 0.05 0.10 0.15 0.20
vertical strain ( ea)
b) volumetric strain versus axial strain
FIG. 5~ Triaxial behavior of unreinforced crushed limestone specimens with different aspect
ratios h/d.

purposely selected in the manufacturing process to allow for the (1952) is used as a model for the geocell wall confining effect.
efficient compaction of granular infill materials routinely used The increased confining pressure (6.a3 ) due to the membrane
to construct composite mats for the support of pavement and stresses in the geocell wall can be described by the following
track formations (e.g., Webster 1979, 1981). Hence, geocell units equations
in the field typically have the geometry and the boundary con-
ditions simulated in the large-scale triaxial compression tests. 2M E< _ _I:_~
(I)
d (1 E.)
Calculation of Geocell Confinement Effect from Elastic
Membrane Theory
(2)
A number of theories have been advanced to estimate the
additional confining pressure due to membrane action in con-
ventional triaxial testing (e.g., Henkel and Gilbert 1952; Duncan where M is the modulus of the membrane (kN/m), Ec is the
and Seed 1967). In this paper the method of Henkel and Gilbert circumferential strain, Ea is the axial strain, and dis the original
 BATHURST AND KARPURAPU ON GEOCELL SOILS 301

1~~--------------------------------------,
... .a.
geocell/dense silica sand composite (h/d=1)
• • geocell/loose silica sand composite (hid= 1)
• • unreinforced dense silica sand (h/d=1)

OJ-OJ

j o'l
h

geoccll
1 t---d--

0 200 400 600 800 1000

p (kPa)
a) No. 40 silica sand

.a. geocell/limestone aggregate composite (h/d=1)

-----<•• unreinforced limestone aggregate (h/d=1)
+----<•• unreinforced limestone aggregate (h/d=2.2)

OJ -OJ
l l l

i
h
0 '3 1-
r-
crushed
limestone

1 1-

--
·.

geocell t---d

0 200 400 600 800 1000

p (kPa)
b) crushed limestone aggregate
FIG. 6-Failure strength envelopes for geocell-soil composites and unreinforced triaxialtesl
specimens.

e
g

TABLE 2-Mohr-Coulomb strength properties from triaxial te.~t data.

Peak Friction Angle.
Materials Apparent Cohesion, kPa degrees

Unreinforced dense silica sand (hid = 1) 5.80 45.4

) Geocell/dense silica sand composite (hid= 1) 190 44.4
Geocell/loose silica sand composite (hid ~ l) 156 42.1
Unreinforced limestone aggregate (hid = 1) 11.0 52.0
Unreinforced limestone aggregate (hid = 2.2) 6.80 46.7
' Geocell/limestone aggregate composite (hid = 1) 169 51.9
302 GEOTECHNICAL TESTING JOURNAL

diameter of the specimen. The triaxial tests were performed at tests. The modulus Min Eq 1 was estimated using a secant slope
1% axial strain/min. The corresponding circumferential strain intersecting the load-strain data at a circumferential strain pre-
rate can be computed from Eq 3. dicted at peak shearing resistance for the reinforced specimens.

Prediction of Geocell-Soil Composite Strength

A simple model to relate the composite geocell-soil Mohr-
(3)
Coulomb strength envelope to that of a cohesionless soil infill is
given in Fig. 8. Using the construction in the figure, the enhanced
The circumferential strain rate was estimated to vary from 0.5 strength of the soil infill due to geocell confinement can be de-
to 0. 7% strain/min over the course of triaxial compression. In- scribed by an equivalent cohesion term c~ using Eq 4.
isolation tension tests were performed on specimens of geocell
polyethylene sheet (Fig. 7). One test was carried out at a reduced
rate of 0.5% axial strain/min and another at 1% axial strain/min.
The load-deformation response was essentially the same for both
c
'
= .:l.rr, tan
2
(:": +
4 2
!) (4)

10% per minute

15
1% per minute

0
0.5% per minute

:;;"//////////./
T
L=IOOmm
1 :;;"///////////

W=200mm

0
s
s
r
0 0.05 0.10 0.15 0.20 0.25 0.30 c
axial strain r
t
FIG. 7-ln-isolation tension tesrs on geocell polyethylene sheet.
I!
g
geocell tl
reinforced b
p
unreinforced g
a
b

c
c
effect of il.OJ on unreinforced soil c
c, I ~-...__,L__
OJ OJ + il.OJ
_ ____.__ __.,t_

normal stress
F1G. 8- Mohr circle construction for calculation of equivalent cohesion for geocell-soil composites.
 BATHURST AND KARPURAPU ON GEOCELL SOILS 303

A comparison of the measured cohesion for the composite spec- alent strength parameters of geocell-granular soil composite lay-
imens and the predicted values using Eqs 1-4 is given in Table ers without carrying out the large-scale triaxial compression tests
3. The data show that predicted values vary by 5 to 18% from described in the paper.
the measured values. Consequently, the elastic membrane theory
and its implementation within the enhanced confinement model
Acknowledgments
gives a reasonably good estimate of the apparent cohesion im-
parted to a granular soil by geocell confinement. The underpre- Financial support for this study was provided by the Depart-
diction of apparent cohesion for the dense compacted soil spec- ment of National Defence (Canada) in the form a research grant
imens may be due to frictional resistance between the soil and awarded to the first author. The authors also thank Presto Prod-
geoceH wall materials which is not accounted for in the membrane ucts Company, Geosystems Division for provision of the poly-
model. In practice, there will be interactions between connected meric geocell material.
geocell units in the field that may further increase the stiffness
and strength of the composite. In particular, shear and moment
References
transfer between cells in a road base layer application can be
expected. Hence, the strength of the in situ structure may be Bathurst, R. J., Crowe, R. E., and Zehaluk, A. C., ''Geosynthetic
underestimated by the single cell ela:stic membrane model de- Cellular Confinement Cells for Gravity Retaining Wall, Richmond
veloped here, but the error is likely to be on the safe side for Hill, Ontario. Canada," Geosynthetics Case Histories, G. P. Raymond
and J.P. Giraud, Eds., BiTech Publishers, Vancouver, Canada, 1993,
design. pp. 266-267.
Bathurst, R. J. and Crowe, R. E., "Recent Case Histories of Flexible
Geocell Retaining Walls in North America," Proceedings Symposium
Conclusions on Recent Case Histories of Permanent Geosynthetic-Reinforced Soil
Thin layers of polyethylene geocell-soil composites are used Retaining WaltS, Tokyo, Japan, held 6-7 Nov. 1992, A. A. Balkcma,
pp. 17 (in press).
to reinforce the near surface of granular base layers in pavement Bathurst, R. J., and Jarrett, P.M., '·Large-Scale Model Tests of Geo-
and track support structures. Multiple layers of geocell-soil com- composite Mattresses Over Peat Subgrades," Transportation Research
posites have also been used to construct gravity wall structures Record 1188, Transportation Research Board, Washington. D.C., 1989,
and to form the facia of geosynthetic-reinforced soil walls and pp. 28-36.
Bathurst, R. J. and Karpurapu, R., "Stability of Base Layers in Highway
steepened slopes. The results of large-scale triaxial tests carried and Airfield Pavements," Military Engineering Research Group Re-
out with isolated geocell-soil specimens matching the geometry port, Royal Military College of Canada, Kingston, Ontario, January
in the field have shown the stiffening effect and strength increase 1992.
imparted to the soil by the enhanced confinement effect. Com- Bishop, A. W. and Green, G. E., "The Influence of End Restraint on
parison of reinforced and unreinforced soil specimens shows that the Compression Strength of a Cohesionless Soil," Geotechnique, Vol.
15, No. 3. 1965. pp. 243-266.
the frictional resistance described by the peak friction angle of Crowe, R. E., Bathurst, R J. and Alston, C., "Design and Construction
the soil infill is applicable to the composite structure as well. A of a Road Embankment Using Geosynthetics," Proceedings, 42'nd
simple elastic membrane model has been shown to give a rea- Canadian Geotechnical Conference, The Canadian Geotechnical So-
sonably accurate estimate of the additional apparent cohesion ciety, Winnipeg, Manitoba, October 1989, pp. 266-271.
Duncan. J. M. and Seed. H. B., "Corrections for Strength Test Data,"
present in the composite structure as a result of the enhanced ASCE Journal of Soil Mechanics and Foundations Division, Vol. 93,
confinement effect of the flexible wall geocell material. The im- No.5. 1967, pp. 121-137.
portant technical contribution of the paper is that it illustrates Goto, S. and Tatsuoka, F., "Effects of End Conditions on Triaxial Com-
that large-scale triaxial tests may not be required to estimate the pressive Strength for Cohesionless Soil," Ad~·anced Triaxial Testing of
strength properties of geocell-granular soil composites. The elas- Soil and Rock, ASTM STP 977, R. T. Donaghe, R. C. Chaney, and
M. L. Silver, Eds .. American Society for Testing and Materials, Phil-
tic membrane model proposed here has been demonstrated to adelphia, 1988, pp. 692-705.
give a reasonably accurate but slightly conservative estimate of Henkel, D. J. and Gilbert, G. C., "'The Effect of Rubber Membranes
the apparent cohesion imparted to a compacted granular fill on the Measured Triaxial Compression Strength of Clay Samples,"
based on unreinforced strength data and the modulus of the Geotechnique, Vol. 3, No. 1. 1952, pp. 20-29.
Progressive Railroading, "ATSF Tests Grid System for Track Stabili-
polymeric cell material. Foundation design in geotechnical en- zation," March 1987, p. 54.
gineering practice often involves limit-equilibrium approaches Rowe, P. W. and Barden, L., "Importance of Free Ends in Triaxial
and the assumption of Coulomb-type soils. The model proposed Testing,'' Journal of Soil Mechanics and Foundation Division, ASCE,
by the authors may be used by engineers to estimate the equiv- Vol. 90, No. 1, 1%4. pp. 1-27.
Webster, S. L., "Investigation of Beach Sand Trafficability Enhancement
Using Sand-Grid Confinement and Membrane Reinforcement Con-
TABLE 3-Comparison of measured and predicted cohesion for cepts," Report GL-79-20(1), Geotechnical Laboratory, U.S. Army
geocell-soil composite samples. Engineer Waterways Experiment Station, Vicksburg, MS, November
Materials Measured, kPa Predicted, kPa 1979.
Webster, S. L. "Investigation of Beach Sand Trafficability Enhancement
Geocell/dense silica sand composite 190 156 Using Sand-Grid Confinement and Membrane Reinforcement Con-
GeocelUloose silica sand composite 156 174 cepts," Report GL-79-20(2), Geotechnical Laboratory, U.S. Army
Geocell/limestone aggregate composite 169 160 Engineer Waterways Experiment Station, Vicksburg, MS, February
1981.

View publication stats