TECHNICAL NOTES

Impact of Soil Composition on Fall Cone Test Results

Bojana Dolinar1 and Ludvik Trauner2
Downloaded from ascelibrary.org by National Chung Hsing University on 04/11/14. Copyright ASCE. For personal use only; all rights reserved.

Abstract: The cone penetrometer method is being increasingly used for the determination of the liquid limit of cohesive soils. The same
method can be used to determine the plastic limit according to the known relationship between soil moisture content and depth of cone
penetration. This relationship is non-linear, yet becomes linear on a log–log scale resulting in a simple method of determination of liquid
limit and plastic limit values. This technical note determines the soil properties that define these index parameters. Experimentally
obtained results suggest that the index parameters primarily depend on the type, size, and quantity of the clay minerals in soil. Soil index
parameters such as the liquid limit and plastic limit can be determined from knowledge of the mineralogical properties of the soil because
this dependence appears well defined.
DOI: 10.1061/(ASCE)1090-0241(2005)131:1(126)
CE Database subject headings: Clays; Atterberg Limits; Cone penetration tests; Cohesive soils; Soil properties.

Introduction w = chm 共2兲

where c and m = soil-dependent parameters; c 共%兲 = water content
The cone penetrometer method is being used increasingly for the
at a penetration depth h = 1 mm; and m = slope of the linear func-
determination of liquid limit. The great advantages of this method
tion which represents the ratio between the logarithm of water
are its high reliability and simple operation. The same method can
content w 共%兲 and the logarithm of penetration depth h 共mm兲.
be used to estimate the plastic limit. The depth of cone penetra-
Parameters c and m can be obtained experimentally with a few
tion at the plastic limit can be computed by the following equa-
measurements of water content at cone penetration depths ranging
tion (Hansbo 1957):
from h = 3 to 25 mm. According to Eq. (2) the plastic limit
共PL100兲 is equal to the following equation for a British cone (30°,
W 80 g):
su = K 共1兲
h2
PL100 = c2m 共3兲
where su = undrained shear strength; K = constant, depending on This technical note explains the relationships between param-
the type of cone; and W = cone mass. For the British cone, the eters c and m and those soil properties that determine them. It is
theoretically determined value of the constant is K = 1.33 (Kou- shown that their values depend on the specific surface of the soil,
moto and Houlsby 2001) and the mass of the cone is W = 80 g. the portion of clay minerals in soil and the quantity of interlayer
The undrained shear strength is su = 2.66 kPa at the depth of cone water. As this dependence is well defined, the parameters c and m,
penetration hL = 20 mm, which corresponds to the water content at and thus the liquid limit and plastic limit, can be defined from
the liquid limit. The data of Skempton and Northey (1953) show knowledge of the mineralogical properties of the soil.
that the undrained shear strength at the plastic limit is about 100
times the undrained shear strength at the liquid limit. Conse-
quently the depth of cone penetration at the plastic limit can be Basic Assumptions
calculated by Eq. (1) and is about h P = 2 mm for the British cone.
Feng (2000) presents a relationship between the depth of cone Clay minerals, as well as water, are not chemically inert; therefore
penetration h and the water content w by the following equation: they are subject to interaction. Water is strongly attracted to clay
mineral surfaces, and results in plasticity; whereas nonclay par-
1
Senior Lecturer, Univ. of Maribor, Faculty of Civil Engineering, ticles have a much smaller specific surface and less affinity for
Smetanova ulica 17, 2000 Maribor, Slovenia. E-mail: bojana.dolinar@ water and do not develop significant plasticity even when in
uni-mb.si finely ground form. It is to be expected that almost all of the
2
Professor, Univ. of Maribor, Faculty of Civil Engineering, Smet- water in a soil is associated with the clay phase (Mitchell 1993).
anova ulica 17, 2000 Maribor, Slovenia. E-mail: [email protected] Water may be found only between the grains in soils with nonex-
Note. Discussion open until June 1, 2005. Separate discussions must panding clay minerals. The moisture content in soils with expand-
be submitted for individual papers. To extend the closing date by one
ing clay minerals represents the total quantity of water that is
month, a written request must be filed with the ASCE Managing Editor.
The manuscript for this technical note was submitted for review and bonded to both the external surface and the internal surface of the
possible publication on May 28, 2003; approved on April 23, 2004. This clay grains (Fig. 1). Using the standard method for measuring
technical note is part of the Journal of Geotechnical and Geoenviron- water content, the total quantity of intergrain water we and the
mental Engineering, Vol. 131, No. 1, January 1, 2005. ©ASCE, ISSN interlayer water wi can always be determined due to drying at a
1090-0241/2005/1-126–130/$25.00. temperature higher than 100° C.

126 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JANUARY 2005

J. Geotech. Geoenviron. Eng. 2005.131:126-130.

Downloaded from ascelibrary.org by National Chung Hsing University on 04/11/14. Copyright ASCE. For personal use only; all rights reserved.

Fig. 1. Graphical presentation of water distribution in fully saturated clays; intergrain water we = wef + wea

The liquid limit represents the water content at which different Taking into consideration the above assumptions, the soil-
fine-grained soils have approximately equal undrained shear dependent parameters c and m, determined by the fall cone test,
strength (Casagrande 1932). It can be expected that undrained must depend on the specific surface and the portion of clay grains
shear strength depends on the quantity of intergrain water only in the nonexpanding soils, and on the specific surface, portion of
because the interlayer water is tightly tied between the layers and the clay minerals, and the quantity of interlayer water in the ex-
thus cannot influence it. The clays have almost equal hydraulic panding soils. The experimental results completely confirm these
conductivity at the liquid limit (Nagaraj et al. 1991), which means presumptions.
that the effective pore sizes controlling fluid flow must be about
the same for all clays (Mitchell 1993). It was also found that the
clays have similar pore water suction at the liquid limit (Russel Materials
and Mickle 1970). It can be expected that the interaction forces
between surface grains and the adsorbed water are the same for
Three monomineral samples (Samples 1, 2, 3) and four samples
all clay minerals due to the equality of surface structure. This also
containing 20 and 50% of nonclay particles (Samples 5, 6, 7, and
means that the quantity of the adsorbed water per clay surface
8) were used to determine dependence between the quantity of
unit should be about the same (Mitchell 1993). Considering these
water and the external surfaces of clay grains at different cone
facts, the quantity of intergrain water we for soils at the liquid
penetration depths. Investigated clays belong to well and poorly
limit (LL) can be expressed in the form
crystallized kaolinite and Ca-montmorillonite. They originate
we兩LL = p共wef兩LL + wea兩LL兲 = p共wef兩LL + ta兩LLASC兲 共4兲 from regions of the United States and are intended for different
fundamental studies, therefore their mineral composition, struc-
where wef兩LL= quantity of free pore water; and wef兩LL = constant for tural formula of unit cell and specific surface are known. The data
different clays at the liquid limit. The quantity of adsorbed water referring to these properties of the tested samples has been taken
on the clay surface is wea兩LL. This quantity of water depends only from literature (Van Olphen and Fripiat 1979). The specific sur-
on the external surface sizes of the clay grains ASC (m2 / g = the faces for Samples 5, 6, 7, and 8, where quartz sand of grain size
external surface of the clay minerals per 1 g of clay minerals) 0.08– 0.40 mm was added to kaolinite and Ca-montmorillonite,
because the thickness ta兩LL of the firmly adsorbed water on the was calculated from the known weight ratio between individual
external clay surface is constant for all clays at the liquid limit. components. The data for Sample 4 (kaolinite) were selected from
The intergrain water content depends on the quantity of clay min- the literature (Sudhakar et al. 1985) for comparison purposes
erals p 共0 ⬍ p 艋 1兲 in soil supposing that all of the water in the (Table 1).
soil is associated with the clay minerals. Parameters c and m depend mainly on soil composition, how-
Considering the assumption that the penetration depth depends ever the temperature, the structure of the soil, organic additions,
only on the quantity of intergrain water we, the expression Eq. (2) and the chemical composition of the pore water can also influence
can be written as them. Laboratory tests were performed at a temperature 20° C
using distilled water and samples without organic matter with a
w e = c eh me 共5兲 parallel arrangement of clay particles to adjust test conditions as
where ce and me = parameters determined by the quantity of inter- much as possible.
grain water. According to Eq. (4) the quantity of intergrain water
at the penetration depth h = 1 mm, where we兩h=1 = ce can be written
as Test Results
we兩h=1 = ce = p共wef兩h=1 + wea兩h=1兲 = p共wef兩h=1 + ta兩h=1ASC兲 共6兲
A cone penetrometer with a 30° cone and mass of 80 g was used
Eq. (6) shows that the parameter ce depends on the specific sur- for determining the liquid limit (LL) and the parameters c and m
face of clay minerals ASC and the portion of clay grains p in the (except for Sample 4). Parameters c and m were obtained with
soil, because wef兩h=1 and ta兩h=1 = constants. It follows from Eqs. measurements of water content at cone penetration depths from
(4)–(6) that the parameter me depends on the specific surface of h = 10 to 25 mm (Table 1). The plastic limit PL100 was calculated
clays ASC only. by Eq. (3). Liquid limit and plastic limit (PL) for Sample 4 were

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JANUARY 2005 / 127

J. Geotech. Geoenviron. Eng. 2005.131:126-130.

 Table 1. Mineral Composition, Specific Surface AS, Parameters ce and me, Liquid Limit (LL), Intergrain Water weL, and Interlayer Water wiL at Liquid
Limit, Plastic Limit 共PL100兲, Intergrain Water weP, and Interlayer Water wiP at Plastic Limit
Mineral composition/ AS ce LL weL wiL PL100 weP wiP
portion of clay in sample 共m2 / g兲 (%) me (%) (%) (%) (%) (%) (%)
1-kaolinite/ p = 1 10.05 22.83 0.189 40.1 40.1 0.0 25.9 25.9 0.0
2-kaolinite/ p = 1 23.50 25.22 0.235 51.0 51.0 0.0 29.7 29.7 0.0
3-Ca-montmorillonite/ p = 1 97.42 39.42 0.345 129.0 110.8 18.2 68.2 50.0 18.2
4-kaolinitea / p = 1 21.24 24.76 0.228 49.0 49.0 0.0 29.0 29.0 0.0
5-80% 共1兲 + 20% sandb / p = 0.8 8.04 18.32 0.184 31.8 31.8 0.0 20.8 20.8 0.0
6-50% 共1兲 + 50% sandb / p = 0.5 5.02 11.33 0.188 19.9 19.9 0.0 12.9 2.9 0.0
7-80% 共3兲 + 20% sandb / p = 0.8 77.93 31.52 0.345 103.2 88.6 14.6 54.6 40.0 14.6
Downloaded from ascelibrary.org by National Chung Hsing University on 04/11/14. Copyright ASCE. For personal use only; all rights reserved.

8-50% 共3兲 + 50% sandb / p = 0.5 48.71 19.71 0.345 64.5 55.4 9.1 34.1 25.0 9.1
a
Liquid limit was determined by Casagrande method and plastic limit by rolling thread test.
b
Quartz sand of grain size 0.08– 0.40 mm.

determined in accordance with the ASTM standard (data from Relationships between the quantity of intergrain water we and
literature) thus the parameters c and m were calculated from these the depth of cone penetration h for tested samples are shown in
values by the following equations: Fig. 2. It is evident from the Fig. 2 and Table 1 that the value of
parameter me increases with any increase in the specific surface of
log LL − log PL
m= 共7兲 clays (solid lines) but remains unchanged when the portion p of
log hL − log h P the clay minerals in samples changes (dashed and dotted lines).
The change of p is reflected in the change of ce.
c=
LL
=
PL
共8兲 Measured values of parameter ce 共%兲 for corresponding spe-
h Lm h Pm cific surface AS 共m2 / g兲 are shown in Fig. 3. The relationship be-
tween ce and AS for p = 1 is aproximated by the following function
To determine dependence between the depth of cone penetra- (solid line in Fig. 3):
tion and soil composition, only intergrain water we, which deter-
mines these values, must be considered. Thus, parameters c and m
are replaced by parameters ce and me, where suffix e shows that ce = ␣ + ␤AS 共10兲
they are determined only on the basis of intergrain water quantity.
In this case, c = ce and m = me for nonexpanding soils, while both
parameters change for expanding soils. For montmorillonite with Parameters ␣ = 20.72 and ␤ = 0.19 were determined by approxima-
divalent or multivalent exchangeable cations, c ⬎ ce and m ⬍ me tion using the least squares method.
while for montmorillonite with monovalent exchangeable cations Laboratory tests on samples with nonclay additives (Samples
c ⬎ ce and m ⬎ me. 5, 6, 7, and 8) has shown that measured values of ce for the same
An example of interlayer water content wi calculation is given portion of clay minerals are placed parallel to one with p = 1
for the sample of pure montmorillonite with the Ca-exchangeable (dashed lines in Fig. 3), therefore they can be expressed as
cation (Sample 3) having the liquid limit LL= 129.0% and plastic
limit PL100 = 68.2%. The adsorption of water between layers is
already completed at the plastic limit in case a Ca-exchangeable
cation is present. This means that the initial distance between
layers d1 = 0.96 nm for a completely dry montmorillonite (dried at
105° C) has increased to the distance d2 = 1.54 nm after the ad-
sorption of water. The volume of interlayer water Vi can be cal-
culated by the following equation (Fink and Nakayama 1972):
ASCi共d2 − d1兲
Vi = 共9兲
2
where Vi = 0.1817 cm3 = volume of water between layers per 1 g
of clay; wi = 18.17%; d1 = 0.96 nm= distance between layers for a
totally dry Ca-montmorillonite; d2 = 1.54 nm= distance between
layers at the liquid limit and plastic limit for Ca-montmorillonite;
and ASCi = 626.80 m2 / g = internal specific surface for Ca-
montmorillonite.
The internal specific surface of Ca-montmorillonite ASCi can
be determined from the total specific surface of 1 g of clay ASCt
reduced for the value of the measured external specific surface
ASC. The calculated total specific surface of Ca-montmorillonite is
ASCt = 724.22 m2 / g (Van Olphen 1977). In Samples 7 and 8 where
quartz sand is added to Ca-montmorillonite, the quantity of inter-
layer water wi is proportional to the portion of clay mineral in Fig. 2. Quantity of intergrain water we as function of penetration
samples (Table 1). depth h

128 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JANUARY 2005

J. Geotech. Geoenviron. Eng. 2005.131:126-130.

 me = ␥ · ASC␭ 共12兲
The parameter ␥ = 0.10 represents the value me at ASC = 1 m2 / g;
and ␭ = 0.27= slope of the function because the dependence be-
tween me and ASC is a linear function in logarithmic scale. It is
evident from Table 1, Fig. 2, and expression (12) that the portion
of nonclay particles does not affect the value of parameter me.
Combining Eqs. (5), (11), and (12) yields
we = p共20.72 + 0.19ASC兲h0.10ASC0.27 共13兲
In soils that contain nonexpanding minerals, the above expression
allows calculation of the soils’ moisture at different depths of
Downloaded from ascelibrary.org by National Chung Hsing University on 04/11/14. Copyright ASCE. For personal use only; all rights reserved.

cone penetration if we know the specific surface of soil and the

portion of clay minerals. This enables also to obtain the liquid
limit and plastic limit of the soils. In cases where expanding min-
erals are present, Eq. (13) determines the quantity of intergrain
water at both the liquid and plastic limits, while the actual values
of the consistency limits are higher due to the quantity of inter-
layer water.

Fig. 3. Parameter ce as function of external specific surface AS for Conclusion

different portions of clay minerals p in the soils
The relationship between water content w and penetration depth h
at the fall cone test can be described by the equation w = chm,
where c and m are parameters which depend mostly on the soils’
ce = p␣ + ␤ · AS = p共␣ + ␤ASC兲 共11兲 composition.
This technical note explains how the clay minerals in soils
where AS = surface of clay minerals per 1 g of soil and ASC influence the values of both these parameters. The findings are
= surface of clay minerals per 1 g of clay minerals in the soil. based on the following assumptions:
Note that AS = pASC and AS = ASC for p = 1. 1. All of the water in a soil is associated with the clay phase.
Eqs. (6) and (11) have the same structure, therefore, the first 2. The depth of cone penetration depends on the quantity of
and the second term in both of them have to be equal. Conse- intergrain water in the soil, while the interlayer water in
quently the quantity of free pore water wef兩h=1 = ␣, the average swelling minerals does not affect it. The intergrain water
thickness of the firmly adsorbed water around the clay grains consists of free pore water and adsorbed water on external
ta兩h=1 = ␤, and the quantity of the adsorbed water around the clay surface of clay grains.
grains wef兩h=1 = ta兩h=1ASC = ␤ASC. 3. The free water quantity is the same for different clays at the
The relationship between parameter me and the specific sur- same penetration depth.
face ASC of the clay minerals (Fig. 4) is aproximated by the fol- 4. The water film thickness around the clay grains is equal for
lowing function: different clays at equal penetration depth. The total quantity
of the adsorbed water in clays depends on the size of clay
grains.The results of experimental tests confirm the above
assumptions. The parameters c and m are dependent on the spe-
cific surface and the quantity of clay minerals in the soil for
nonexpanding soils. For expanding soils the parameters ce and me
are calculated from the portion of intergrain water only.
The values for the parameters ce and me given in this technical
note are valid for the w – h relationship determined by the British
cone 共30° , 80 g兲. If a 60°, 60 g cone is used, parameter me
would remain unchanged, but a different value for ce would be
required, and Eq. (13) should be revised accordingly.

References
Casagrande, A. (1932). “Research on the Atterberg limits of soils.” Public
roads.
Feng, T. W. (2000). “Fall-cone penetration and water content relationship
of clays.” Geotechnique, 50(2), 181–187.
Fink, D. H., and Nakayama, F. S. (1972). “Equation for describing the
free-swelling of montmorillonite in water.” Soil Sci., 114(5), 355–
358.
Hansbo, S. (1957). “A new approach to the determination of the shear
Fig. 4. Parameter me as function of external specific surface of clays
strength of clay by the fall-cone test.” R. Swedish Geotech. Inst. Proc.,
ASC 14, 7–47.

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JANUARY 2005 / 129

J. Geotech. Geoenviron. Eng. 2005.131:126-130.

 Koumoto, T., and Houlsby, G. T. (2001). “Theory and practice of the fall Skempton, A., and Northey, R. D. (1953). “The sensitivity of clays.”
cone test.” Geotechnique, 51(8), 701–712. Geotechnique, 3(1), 30–53.
Mitchell, J. K. (1993). Fundamentals of soil behaviour, 2nd Ed., Wiley, Sudhakar, A., Rao, M., and Sridharan, A. (1985). “Mechanism control-
New York. ling the volume change of kaolinite.” Clays Clay Miner., 33(4), 323–
Nagaraj, T. S., Pandian, N. S., and Narasimha Raju, P. S. R. (1991). “An 328.
approach for prediction of compressibility and permeability behaviour Van Olphen, H. (1977). An introduction to clay colloid chemistry, 2nd
of sand–bentonite mixes.” Indian Geotech. J., 21(3), 271–282. Ed., Wiley Interscience, New York.
Russell, E. R., and Mickle, J. L. (1970). “Liquid limit values of soil Van Olphen, H., and Fripiat, J. J. (1979). Data handbook for clay miner-
moisture tension.” J. Soil Mech. Found. Div., 96(3), 967–989. als and other non-metallic materials, Pergamon, New York.
Downloaded from ascelibrary.org by National Chung Hsing University on 04/11/14. Copyright ASCE. For personal use only; all rights reserved.

130 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / JANUARY 2005

J. Geotech. Geoenviron. Eng. 2005.131:126-130.