Determination of Cation Exchange Capacity

from Soil Water Retention Curve

Morteza Khorshidi, Aff.M.ASCE 1; and Ning Lu, F.ASCE 2

Abstract: Cation exchange capacity (CEC) has a significant influence on the physical and chemical behavior of soil. Quantification of the
Downloaded from ascelibrary.org by Colorado School of Mines on 02/21/17. Copyright ASCE. For personal use only; all rights reserved.

CEC is an essential yet challenging task. A new methodology for the determination of the CECs of soils by using the soil water retention
curve (SWRC) in the extremely high suction range is presented. The methodology is based on a theoretical SWRC model that explicitly
considers the contribution of the cation hydration on the matric potential. For a homoionic soil, the CEC governs the relationship among the
type of exchangeable cations, matric potential, and the corresponding equilibrated soil water. Thus, the CECs of homoionic soils can be
deduced from their SWRCs. For a natural soil with a multiple cation species, the CEC of bulk soil can be determined from the SWRC of the
Naþ -exchanged form of the soil. A suite of materials ranging from nonexpansive to expansive soils is used to demonstrate the validity of the
proposed methodology. The CECs of these soils are also measured independently by using the ammonium acetate method for comparison
purposes. The close match between the independently measured and the SWRC-based CEC values confirms the validity and illustrates the
potential promises of the SWRC-based methodology in determining the CEC of soils. DOI: 10.1061/(ASCE)EM.1943-7889.0001220.
© 2017 American Society of Civil Engineers.
Author keywords: Cation exchange capacity; Soil water retention curve; Cation hydration; Na+-exchanged soil.

Introduction
Soils containing clay minerals possess negatively charged particle
surfaces that hold exchangeable cations. These cations can be ex-
changed without a mineral structure change. The amount of ex-
changeable cations a soil possesses is referred to as the cation
exchange capacity (CEC). It is basically the sum of the permanent
CEC resulting from the isomorphous substitution within the soil
structure and the variable CEC, or pH-dependent CEC, resulting
from the deprotonation (e.g., Ellis and Foth 1996; Mitchell and
Soga 2005; Sposito 2008). CEC values are typically expressed
in units of milliequivalent electric charges per gram.
The CEC of soil reflects the layer charge and particle surface
properties. Thus, it has been recognized as a fundamental property
that influences soil behavior. For example, the commonly used geo-
technical engineering indices such as Atterberg limits as well as
specific surface area (SSA) have been closely correlated with
the CEC (e.g., Farrar and Coleman 1967; Smith et al. 1985;
Churchman and Burke 1991; Petersen et al. 1996; Yukselen and
Kaya 2006; Woodruff and Revil 2011). Numerous studies have also
shown that the swelling potential of soil may be related to the CEC;
the soil’s swelling potential is used for the expansive soil classifi-
cation (e.g., Rengasamy et al. 1986; Christidis 1998; Thomas et al.
2000; Dontsova and Norton 2002; Yilmaz 2004). In addition, the
CEC greatly influences water sorption, fluid flow, electric and
chemical flows, and coupled fluid in clayey soils (e.g., Shainberg
1
Postdoctoral Fellow, Dept. of Civil and Environmental Engineering,
Colorado School of Mines, 1500 Illinois St., Golden, CO 80401 (corre-
sponding author). E-mail: mkhorshi@alumni.mines.edu
2
Professor, Dept. of Civil and Environmental Engineering, Colorado
School of Mines, 1500 Illinois St., Golden, CO 80401. E-mail:
ninglu@mines.edu
Note. This manuscript was submitted on February 22, 2016; approved
on November 1, 2016; published online on February 15, 2017. Discussion
period open until July 15, 2017; separate discussions must be submitted
for individual papers. This paper is part of the Journal of Engineering
Mechanics, © ASCE, ISSN 0733-9399.
et al. 1980; Gaston and Selim 1990; Revil and Leroy 2004;
Dontsova et al. 2004; Khorshidi and Lu 2016a).
Numerous methods have been proposed for the CEC measure-
ment since the nineteenth century. The frequently used methods
include ammonium acetate method at pH ¼ 7.0 (e.g., Schollenberger
and Dreibelbis 1930), ammonium chloride method (e.g., Shuman
and Duncan 1990), sodium acetate method (e.g., Bower et al.
1952), or barium chloride (e.g., Mehlich 1938; Bascomb 1964)
at pH ¼ 8.2, and methylene blue absorption (Nevins and Weintritt
1967). The steps in these chemical methods involve the saturation
of the soil by index ions that replace exchangeable cations in the
soil and the subsequent displacement of the index ions with another
salt solution (i.e., the extracting solution). Finally, the amount of
the displaced index ions that represents the CEC of the soil is mea-
sured by using inductively couple plasma spectrometry or atomic
absorption. Ammonium acetate is the most commonly used
method, particularly for soil-classification applications (e.g., Sumner
and Miller 1996; Ross and Ketterings 2011). The index ion and
extracting solution in this method are ammonium (NHþ 4 ) and
KCl, respectively.
The CEC of soil may change depending on the type of saturat-
ing and extracting solutions used in the CEC measurement. This
change can only occur through the variable CEC, because it de-
pends on the soil pH, which might change when the saturating
or extracting solution is added to the natural soil. The less the
change in the soil pH after treatment with a saturation or extracting
solution is, the lower the error in the CEC measurement (Sumner
and Miller 1996; Laird and Fleming 2008). Therefore, each of the
chemical methods can be more appropriate for a particular soil sit-
uation. For example, the ammonium acetate method works well for
basic soils, while it overestimates the CEC of acidic soils (Sumner
and Miller 1996).
The methods described previously are cumbersome time-
consuming approaches that are subject to errors associated with salt
or sample loss during the decantation steps (e.g., Jaynes and
Bioham 1986; Sumner and Miller 1996). To overcome these prob-
lems, this present study investigates the potential application of the

© ASCE 04017023-1 J. Eng. Mech.

J. Eng. Mech., -1--1

 soil water retention curve (SWRC) in the determination of the CEC.
The SWRC is an important constitutive function of the soil that can
be used in the prediction of numerous soil properties and processes
in the soil such as water flow, solute and contaminant transport,
coupled heat and mass transfer, stress and deformation, specific
surface area, and swell potential (e.g., Beven and Germann
1982; Padilla et al. 1999; Hansson et al. 2004; Lu and Likos
2004; Lu et al. 2010, 2014; Leao and Tuller 2014; Akin and Likos
2014; Moradi et al. 2015; Khorshidi and Lu 2016b).
Following the SWRC approach to obtain soil properties, a new
SWRC-based methodology for determining the CEC of soil is pro-
posed. This methodology is based on the relationship between the
cation hydration and matric potential of soil at the very dry end of
Downloaded from ascelibrary.org by Colorado School of Mines on 02/21/17. Copyright ASCE. For personal use only; all rights reserved.

the SWRC. More specifically, the CEC of the soil can be estimated
given the SWRC at high suction values (soil water sorption iso-
therm) of any bulk soil saturated with Naþ cations. Furthermore,
independently measured CEC values by the ammonium acetate
method are also used to verify the proposed method for the
CEC determination.

Methodology

Relationship between Matric Potential

and Cation Hydration
The SWRC is the relationship between the total potential and cor-
responding equilibrated soil water (e.g., Brooks and Corey 1964; van
Genuchten 1980; Lu and Likos 2004; Khorshidi 2015). The total
potential consists of three forms of potential: matric, gravitational,
and osmotic. Due to the negligible effects of gravitational and
osmotic potentials compared to matric potential for problems con-
cerning total potentials less than −100 MPa, matric potential can
be effectively considered as total potential (Khorshidi et al. 2016d).
There exist three water retention regimes. Starting from high
matric potentials to low matric potentials, they are capillary water,
adsorbed water, and tightly adsorbed water (McQueen and Miller
1974; Lu and Likos 2004). The pertinent range to these regimes
depend on type of soil; but in general, matric potential ranges
between 0 to −400 kPa, −400 kPa to −10 MPa, and −10 to
−1,200 MPa were proposed for these regimes, respectively
(e.g., Tuller et al. 1999; Frydman and Baker 2009; Lu and
Khorshidi 2015).
Since the water sorption is controlled solely by cation and sur-
face interactions in tightly adsorbed water regime, two main fun-
damental soil properties determine the amount of equilibrated soil
water at low matric potentials: CEC and SSA (e.g., Tuller et al.
1999; Dontsova et al. 2004; Srodon et al. 2008; Khorshidi and
Lu 2016a; Khorshidi et al. 2016c). The former property is impor-
tant at very low water contents, where cation hydration is the
governing mechanism to retain water in the soil (e.g., Berend
et al. 1995; Cases et al. 1997; Prost et al. 1998; Laird 1999;
Xu et al. 2000; Sposito 2008; Woodruff and Revil 2011; Khorshidi
et al. 2016d). Water molecules at these water contents are located
in the first hydration shell of the exchangeable cations, as illustrated
in Fig. 1. The exchangeable cation in the homoionic soils
shown in Fig. 1 is the Naþ with a hydration shell distance of
0.24 nm. The cation hydration is followed by surface hydration
until the monolayer surface coverage on soil particles becomes
complete. The boundary between these two mechanisms depends
on the type of exchangeable cations in soils and it can occur at rel-
ative humidity values between 5 and 30% (Khorshidi and Lu 2016a).
The contribution of the cation hydration on the matric po-
tential has been recently identified by Khorshidi et al. (2016d)
Fig. 1. Arrangement of water molecules around exchangeable cations
at very low water contents in: (a) typical nonexpansive soil; (b) typical
expansive soil (adapted from Khorshidi et al. 2016d)
by considering all different types of works (such as pressure work,
temperature work, surface tension work, and surface hydration
work) acting on water molecules during sorption of water on soil.
Using Gibbs free energy in the range of very low water contents,
the matric potential ψ (Pa) in homoionic soils due to all water sorp-
tion mechanisms, particularly the exchangeable cation-water mol-
ecule (charge-dipole) interaction, has been captured by (Khorshidi
et al. 2016d)
V ci
ψ ¼ −3 × 10−10 ð1Þ
κi r21i
where V ci = cation type i valence; r1i (m) = distance between the
cation and water molecules in its first hydration shell (as shown in
Fig. 1); and κi = parameter used to include the works due to the
electric force and cation-water interaction in the matric potential
expression. Due to the hydration hysteresis, Eq. (1) was developed
for the drying path since this path is more representative to infer
surface and exchangeable cation-related properties of soils
(e.g., Lu and Khorshidi 2015). The value of κi is directly related
to the number of water molecules around each cation N i in the soil,
regardless of the type of the exchangeable cation i. The relationship
between κi and N i can be obtained by back-calculation of κi values
by fitting Eq. (1) to the measured water sorption isotherms of

© ASCE 04017023-2 J. Eng. Mech.

J. Eng. Mech., -1--1

 40
produce the homoionic soils. More specifically, NaCl can be used
35 as the saturating solution to replace all exchangeable cations in the
= 4N + 9.3
soil with Naþ . This solution is sometimes used as the saturating
30
solution for the quantification of the CEC (Pearcy et al. 2012).
25
R 2 = 0.96
The basic assumption in the SWRC-based CEC measurement is
20
that the CEC of the bulk soil is equal to the CEC of Naþ -exchanged
soil.
15 Li -bentonite
The reason for the use of NaCl as the saturating solution, instead
Na -bentonite
10 of other solutions (e.g., KCl and MgCl2 ), can be explained by em-
K -bentonite
Mg -bentonite
ploying Fig. 2. Based on this figure, the obtained experimental κi
5
Ca -bentonite values for the Naþ cation (in Naþ -bentonite) are found to fit much
0 better than the other cations to the average line indicated by Eq. (2)
0 2 4 6 8 [Eq. (2) was ultimately used in Eq. (5)]. The good fitting also cov-
Downloaded from ascelibrary.org by Colorado School of Mines on 02/21/17. Copyright ASCE. For personal use only; all rights reserved.

Number of water molecules around cation, N

Fig. 2. Relationship between parameter κ and number of water mole-
cules around each cation (N) for five homoionic cation-exchanged ben-
tonites (reprinted from Khorshidi et al. 2016d, with permission)
cation-exchanged bentonites. Resultantly, Fig. 2 demonstrates the
relationship between κi and N i , which can be approximated as
(Khorshidi et al. 2016d)
κi ¼ 4.0N i þ 9.3
Substituting Eq. (2) into Eq. (1) yields
V ci
ψ ¼ −3 × 10−10
ð4N i þ 9.3Þr21i
Assuming the cation hydration as the only mechanism for hold-
ing water, N i for homoionic soils can be obtained by dividing the
number of water molecules by the number of exchangeable cations.
Regardless of the location of exchangeable cations in a diffuse dou-
ble layer (either in inner sphere, outer sphere, or in diffuse double
layer), all ions in water are hydrated as the water dipole is attracted
to the charged ion and forms a structure with it (Appelo and Postma
2005). Accordingly, N i is yielded from (Khorshidi et al. 2016d)
wV ci
Ni ¼
ww CEC
where ww = molar mass of water; and w = gravimetric water content
of the soil. Substituting Eq. (4) into Eq. (3) yields the SWRC model
at very low water contents
ers a wider range of N i values for the Naþ cation, as compared with
the other cations. Thus, using Naþ as the saturating cation by em-
ploying NaCl, instead of the other solutions, reduces the errors in-
troduced by Eq. (5) for the determination of the CEC. In addition,
NaCl has a negligible effect on the soil pH (Pearcy et al. 2012).
Therefore, there would be a negligible difference between the var-
iable CEC of the bulk soil and that of the Naþ -exchanged form of
the soil.
The SWRC model for the Naþ -exchanged soils can be obtained
by using Eq. (5). The V ci and r1i values for Naþ are 1 and 0.24 nm
(Chang et al. 1998; Măhler and Persson 2012), respectively; there-
ð2Þ fore, Eq. (5) reduces to
5.40 × 109 CEC
ψ¼− ð6Þ
0.23w þ 9.66CEC
ð3Þ
In this equation, the CEC can be used as the fitting parameter
that will be determined. The SWRC model of Eq. (6) should be
fitted to the measured isotherm in the range of −350,000 kPa < ψ <
−300,000 kPa by using a least-squares regression to determine the
best fit value of the CEC. A fitting in this range of matric potentials,
which corresponds to 1 < N Na < 2, provides the lowest error in the
CEC estimation. This is because the predicted κi values obtained
by Eq. (2) match the experimental κi values for the Naþ cation very
well, as depicted in Fig. 2.
Overall, the CEC of each soil can be deduced by using
Eq. (6) and the SWRC of the Naþ -exchanged form of the
ð4Þ
soil in the range of −350,000 kPa < ψ < −300,000 kPa [≈7% <
relative humidityðRHÞ < 11%]. The experimental program to vali-
date the proposed SWRC-based methodology is addressed in the
next section.

3 × 10−10 V ci CEC
ψ¼−
r21i ð4V ci wþ 9.3CECÞ
Eq. (5) is valid for water contents up to the monolayer cation
hydration water content at which all exchangeable cations in the
soil are covered with monolayer water. At this water content, N i
is equal to the cation hydration number (Khorshidi and Lu
2016a). The SWRC model of Eq. (5) can be used to determine
the CEC of natural soils with a multiple cation species.
Determination of the CEC from Soil Water Sorption
Isotherm
The SWRC model in Eq. (5) was established for homoionic soils.
In these soils, V ci and r1i can be replaced by their associated values,
yielding the SWRC model as a function of the CEC. In order to use
Eq. (5) for natural soils containing multiple cation species, the bulk
natural soils should first be treated with saturating solutions to
ð5Þ Experimental Program
A suite of natural silty and clayey materials ranging from expansive
to nonexpansive soils was used to demonstrate the validity of the
SWRC-based methodology for the CEC determination. All but one
soil sample was collected from different places in the United States;
the last sample was collected in China. The engineering properties
of the soils are summarized in Table 1.
The soil samples were treated to produce Naþ -exchanged soils
for the purpose of measuring their water sorption isotherms.
Naþ -exchanged soils were prepared by saturating a bulk soil with
1 M NaCl for 24 h. The particles of the soil samples were separated
from the salt solution in a high-speed centrifuge, and the soils were
suspended again in fresh NaCl two more times. The samples were
then repeatedly rinsed with deionized water to remove the chlorides
detected in the supernatant. In the absence of chloride in the super-
natant, the supernatant tested negative with AgNO3 (no salt forms;
Moore and Reynolds 1997).

© ASCE 04017023-3 J. Eng. Mech.

J. Eng. Mech., -1--1

 Table 1. Fundamental Engineering Properties of the Tested Soils
Soil Mineralogy Clay (%) LL (%) PL (%) PI (%) USCS classification Origin
Georgia kaolinite K 35 44 26 18 CL Georgia
Wyoming montmorillonite Discrete S, trace quartz 100 485 353 132 CH Wyoming
Naþ =Caþþ bentonite — — — — — — —
Denver bentonite Discrete S, trace K, quartz, K-Feld 90 118 45 73 CH Colorado
Denver claystone Mixed layer I/S, trace K, quartz, gypsum 55 44 23 21 CL Colorado
Bonny silt — 14 25 21 4 ML Colorado
BALT silt I — 14 26 19 7 CL-ML California
BALT silt II — 5 28 23 5 SM California
Hopi silt I — 3 26 19 7 SC-SM Arizona
Hopi silt II — 2 — — — ML Arizona
Iowa silt — 10 33 24 9 ML Iowa
— —
Downloaded from ascelibrary.org by Colorado School of Mines on 02/21/17. Copyright ASCE. For personal use only; all rights reserved.

Heifangtai silt 26 21 5 ML Gansu, China

Note: I ¼ illite; K ¼ kaolinite; K-Feld ¼ K-Feldspar; LL = liquid limit; PI = plastic index; PL = plastic limit; S ¼ smectite; USCS ¼ unified soil
classification system.

One of the common issues that occurs during rinsing Naþ - Table 2. Comparison between the Measured CEC Values of Soils
saturated clays (e.g., montmorillonite) with water is colloidal Obtained by Ammonium Acetate Method and Those Obtained from the
dispersion, which makes soil settlement by centrifuge almost SWRC Model (Data from Akin 2014)
impossible. In the case of a colloidal dispersion, a 50:50 etha-
Ammonium acetate- SWRC-based
nol-deionized water mixture, instead of deionized water, was used Soil based CEC (meq=g) CEC (meq=g)
to rinse the soils with chlorides (Moore and Reynolds 1997).
The CECs of the tested soils were independently measured by Heifangtai silt 0.08 0.07
using the ammonium acetate method to validate the new method- Georgia kaolinite 0.09 0.09
Hopi silt II 0.18 0.22
ology. The detailed procedure for the ammonium acetate method is
Iowa silt 0.19 0.22
described in ASTM D7503 (ASTM 2010). Hopi silt I 0.20 0.20
The water sorption isotherms of Naþ -exchanged soils were Bonny silt 0.21 0.21
measured by using a vapor sorption analyzer (VSA), an automated BALT silt II 0.23 0.24
device that has recently become widely used for measuring iso- BALT silt I 0.23 0.20
therms in the drying (desorption) and wetting (adsorption) cycles Denver claystone 0.32 0.35
(e.g., Arthur et al. 2013; Lu and Khorshidi 2015). VSA works based Wyoming montmorillonite 0.71 0.75
on the chilled-mirror dewpoint technique, continuously measuring Naþ =Caþþ bentonite 0.76 0.74
RH (or ψ) and w (e.g., Lu and Khorshidi 2015). Soil samples were Denver bentonite 1.02 1.69
oven-dried at 105°C and cooled down to ambient room tempera-
ture. Soil water sorption isotherms were then measured at 25 
0.2°C with the VSA. The measurement for the drying and wetting
cycles typically lasts between 24 and 48 h. However, the time re- hydration) pronounced in the soils with a crystalline structure
quired to measure the isotherms in the range of matric potentials (Lu and Khorshidi 2015; Khorshidi and Lu 2016a).
recommended for the determination of the CEC favorably reduces The direct relationship between the CEC and the amount of
to 1–3 h, depending on the type of the soil. equilibrated soil water content at low RH values can be deduced
from Fig. 3. To illustrate this, the gravimetric water content
values of Naþ -Georgia kaolinite, Naþ -Denver claystone, and
Validation of the Proposed SWRC-Based Naþ -Wyoming montmorillonite at 10% RH were obtained. The
Method for the CEC Determination corresponding water contents were 0.33, 1.12, and 2.57%, respec-
tively. The CECs of these soils measured by the ammonium acetate
To examine the performance of the proposed SWRC-based meth- method were 0.09, 0.32, and 0.71 meq=g (Table 2), respectively.
odology, the CECs of the tested soils were measured independently The ratio of the CEC values of the soils compared with each other
using the ammonium acetate method. The CEC values, ranging are almost the same as the ratios of their corresponding water con-
from 0.08 meq=g for Heifangtai silt to 1.02 meq=g for Denver tents. This implies that the CEC of the soil is the governing factor
bentonite, are listed in Table 2 (data from Akin 2014). for the amount of water in the soil at low RH values, and can be
The measured water desorption isotherms of three Naþ -ex- correlated to the soil water sorption isotherm. On the other hand,
changed soils (Naþ -Georgia kaolinite, Naþ -Denver claystone, the SSA of the soil is the governing factor at higher RH values. The
and Naþ -Wyoming montmorillonite) are plotted in Fig. 3. At a methodology to calculate SSA by using isotherm at higher RH
given RH, the values of equilibrated soil water content for these values is described in Khorshidi et al. (2016c).
three soils differ significantly. These differences depend on the The water desorption isotherms of Naþ -exchanged soils are
mineral type, CEC, and SSA of the soils. At approximately 94% used for the determination of their CECs. More specifically,
RH, kaolinite was found to adsorb only 4.5% gravimetric water Eq. (6) was fitted to the isotherms of Naþ -exchanged soils in
content, claystone (illite/smectite mixture) can adsorb 8.2% gravi- the range of 7% < RH < 11%. Consequently, the CECs of the soils
metric water content, while montmorillonite can adsorb as much as were calculated as the fitting parameter in this equation. Overall,
30.5% gravimetric water content. In addition, noticeable wavy or the model fits the experimental data well, as shown in Fig. 4 for the
stepwise behavior is observed in montmorillonite, but is less promi- three soils considered in Fig. 3. The SWRC-based CEC values ob-
nent in claystone. This behavior is associated with two different tained as the fitting parameter are listed in Table 2, along with the
hydration mechanisms (cation hydration and particle surface values measured by the ammonium acetate method. The difference

© ASCE 04017023-4 J. Eng. Mech.

J. Eng. Mech., -1--1

 35 1.8
Heifangtai silt
Na -Wyoming montmorillonite
Georgia kaolinite
30 1.6

Gravimetric water content, w (%)

Na -Denver claystone Hopi silt II
Iowa silt

SWRC-based CEC (meq/g)

Na -Georgia kaolinite 1.4 Hopi silt I
25 Bonny silt
1.2 BALT silt II
BALT silt I
20 Denver claystone
1.0 Wyoming montmorillonite
1
Na / Ca bentonite
15 0.8 Denver bentonite 1

10 0.6

5
0.4
R2 = 0.99
0.2
Downloaded from ascelibrary.org by Colorado School of Mines on 02/21/17. Copyright ASCE. For personal use only; all rights reserved.

0
0 20 40 60 80 100 0.0
Relative humidity, RH (%) 0.0 0.2 0.4 0.6 0.8 1.0 1.2
Ammonium acetate-based CEC (meq/g)
Fig. 3. Soil water desorption isotherms of three Naþ -exchanged soils
with low, medium, and high CEC values Fig. 5. Comparison between cation exchange capacity measured
by ammonium acetate method and those obtained by soil water
sorption isotherm

5
Na -Wyoming montmorillonite (Exp)
sites will be occupied by the saturating cation when using the am-
Gravimetric water content, w (%)

Na -Wyoming montmorillonite (Pred)

4
Na -Denver claystone (Exp) monium acetate method for the CEC measurement (e.g., Sumner
Na -Denver claystone (Pred)
and Miller 1996). This results in the underestimation of the
Na -Georgia kaolinite (Exp)
Na -Georgia kaolinite (Pred) CEC. For these soils, the method of Amrhein and Suarez (1990)
3 is recommended because it takes into account the dissolution of
R2 = 0.99 calcite and gypsum during the saturation and extraction steps
CEC = 0.75 meq/g (Sumner and Miller 1996). In addition, fixation of ions in some
2
clay minerals (e.g., vermiculites and smectite) is a well-known
R2 = 0.98 problem that results in the underestimation of CEC (e.g., Bower
CEC = 0.35 meq/g
1 1950; Sumner and Miller 1996). The fixation of ions occurs when
R2 = 0.95 cations such as NHþ þ
4 and K are trapped in the interlayers and are
CEC = 0.09 meq/g
not readily available for cation exchange during the saturation or
0 extracting processes.
5 7 9 11 13
Relative humidity, RH (%)
Overall, it is anticipated that the CEC measurement with the
ammonium acetate method for Denver bentonite is underestimated.
Fig. 4. SWRC-based CEC estimated for three clayey soils On the other hand, it is possible that all exchange sites in Denver
bentonite with Ca2þ and Mg2þ cations were not replaced by the
Naþ cation when using NaCl as the saturating solution. Even a
small amount of Ca2þ and Mg2þ remaining in the interlayers of
between the SWRC-based and ammonium acetate-based CEC val- Denver bentonite after saturation can substantially increase the
ues is less than 15%, except for Denver bentonite. Fig. 5 is also amount of retained water because the hydration enthalpy of Ca2þ
plotted to compare these two sets of data. The high value of the and Mg2þ are much greater than Naþ . Hence, the SWRC-based
coefficient of determination (R2 ) in this figure indicates a very good CEC value is possibly overestimated.
match between the CEC values deduced from the SWRC model Several problems may arise when the chemical methods are
and those measured by the ammonium acetate method. Hence, it used for the CEC measurement (Laird and Fleming 2008). These
can be concluded that the SWRC can readily be used to determine methods might not work properly when used for clay samples.
the CEC of soils. Fixation of index ions in the interlayer positions of minerals,
hydrolysis of water molecules during the process of rinsing soil
with deionized water, change in soil pH by some of the saturating
Discussion solutions, and dissolution of soluble salts or soluble minerals in
soils containing them are among the problems that make the exist-
The R2 value in Fig. 5, calculated for all soils excluding Denver ence of a perfect method for CEC determination almost impossible
bentonite, is 0.99; this implies a very good match between the (e.g., Sumner and Miller 1996; Laird and Fleming 2008).
SWRC-based and ammonium acetate-based CEC values of the Using the SWRC at very high suction for the CEC calculation
tested soils. In general, the former values are slightly larger than resolves the problems associated with the fixation of index ions
the latter values. However, the SWRC-based value for Denver during the saturation and extraction processes, as well as the change
bentonite is much larger than the ammonium acetate-based value. in soil pH, because it is not required to extract the saturating sol-
A couple of reasons can be discussed as the possible sources of ution in the new methodology anymore. Furthermore, the NaCl
the large discrepancy for Denver bentonite. First, Denver bentonite does not change the soil pH considerably. The new SWRC-based
is composed of smectite and calcite minerals, containing Ca2þ and method is also much less labor intensive and time consuming, with
Mg2þ as the main exchangeable cations in the soil (Likos et al. less of a sample loss that typically occurs during the many decant-
2011). In soils with calcite and soluble salts, not all the exchange ation steps of the chemical methods. Therefore, it can be used as an

© ASCE 04017023-5 J. Eng. Mech.

J. Eng. Mech., -1--1

 alternative method for the CEC measurement, as it will result in
fewer errors, as compared to the existing chemical methods.
Along with all the advantages of SWRC-based methodology, it
still might not work well for soils containing calcite, carbonate, and
soluble salts. Indeed, some discrepancies might arise when using
this method for soils with considerable amount of Alþ . It is also
important to measure SWRC accurately because small change in
water content can greatly change the obtained CEC value.
The new methodology for CEC measurement can become even
simpler and more inexpensive if just a single-point soil water re-
tention is used instead of measuring the whole range of isotherm
from RH 7 to 11%; of course, the accuracy of the estimate will be
sacrificed a little bit. This can be achieved by equilibration of soil
Downloaded from ascelibrary.org by Colorado School of Mines on 02/21/17. Copyright ASCE. For personal use only; all rights reserved.
specimens in closed chambers containing saturated solution of so-
dium hydroxide or potassium hydroxide that produces RH of 8.2%.
The CEC can then be estimated by using Eq. (6) for a single pair of
ðw; ψÞ corresponding to the RH of 8.2%.
Summary and Conclusions
The applications of SWRC for determination of fundamental soil
properties have recently been explored. CEC is one of these proper-
ties that has great impacts on soil’s physical and chemical behavior.
The CEC can be directly linked to the water content of soil at low
relative humidity values. The linkage is provided by a theoretical
SWRC model at high suction values that explicitly considers the
effect of cation hydration on the soil water content.
A new experimental method is proposed to estimate CEC of soil
based on the SWRC at the high suction range. In the experimental
procedure, the bulk soil is first saturated with 1 M NaCl to produce
the Naþ -exchanged form of the soil. The soil water desorption
isotherm of the Naþ -exchanged soil in the range of the RH from
7 to 11% is then measured. Finally, the theoretical SWRC model is
fitted to the measured isotherm to obtain the CEC.
The results reveal that the CEC values calculated by the SWRC
model are comparable to those measured by the ammonium acetate
method. One of the main advantages of the SWRC-based method
as compared to the existing methods for CEC measurement is that
the extraction process is no longer needed here. This indeed elim-
inates the errors that typically occur during the extraction process,
leading to more accurate CEC measurement of soil. Moreover, the
new method is much less cumbersome and time-consuming.
Overall, the proposed methodology presents new opportunities
and potential promises in the fields of geotechnology, geoenviron-
mental, and soil sciences for the practical use of SWRC in deter-
mining the CEC of soils.
Acknowledgments
This research is funded by a grant from the National Science
Foundation (NSF CMMI 1233063).
References
Akin, I. D. (2014). “Clay surface properties by water vapor sorption
methods.” M.S. thesis, Univ. of Wisconsin-Madison, Madison, WI.
Akin, I. D., and Likos, W. J. (2014). “Specific surface area of clay using
water vapor and EGME sorption methods.” Geotech. Test. J., 37(6),
1–12.
Amrhein, C., and Suarez, D. L. (1990). “Procedure for determining
sodium-calcium selectivity in calcareous and gypsiferous soils.” Soil
Sci. Soc. Am. J., 54(4), 999–1007.
© ASCE 04017023-6
J. Eng. Mech., -1--1
Appelo, C. A. J., and Postma, D. (2005). Geochemistry, groundwater, and
pollution, 2nd Ed., CRC Press, Boca Raton, FL.
Arthur, E., Tuller, M., Moldrup, P., and de Jonge, L. W. (2013). “Rapid and
fully automated measurement of water vapor sorption isotherms: New
opportunities for vadose zone research.” Vadose Zone J., 13(1), in press.
ASTM. (2010). “Standard test method for measuring exchange complex
and cation exchange capacity of inorganic fine-grained soils.” ASTM
D7503, West Conshohocken, PA.
Bascomb, C. L. (1964). “Rapid method for the determination of cation-
exchange capacity of calcareous and non-calcareous soils.” J. Sci. Food
Agric., 15(12), 821–823.
Berend, I., et al. (1995). “Mechanism of adsorption and desorption of water
vapor by homoionic montmorillonites: 2. The Liþ , Naþ , Kþ , Rbþ and
Csþ -exchanged forms.” Clays Clay Miner., 43(3), 324–336.
Beven, K., and Germann, P. (1982). “Macropores and water flow in soils.”
Water Resour. Res., 18(5), 1311–1325.
Bower, C. A. (1950). “Fixation of ammonium in difficulty exchangeable
form under moist conditions by some soils of semi-arid regions.”
Soil Sci., 70(5), 375–384.
Bower, C. A., Reitemeier, R. F., and Fireman, M. (1952). “Exchangeable
cation analysis of saline and alkali soils.” Soil Sci., 73(4), 251–262.
Brooks, R. H., and Corey, T. (1964). “Hydraulic properties of porous
media.” Hydrology Papers, Colorado State Univ., Fort Collins, CO.
Cases, J. M., Berend, I., Francois, M., Uriot, J. P., Michot, L. J., and
Thomas, F. (1997). “Mechanism of adsorption and desorption of water
vapor by homoionic montmorillonite. 3: The Mg2þ , Ca2þ , Sr2þ and
Ba2þ exchanged forms.” Clays Clay Miner., 45(1), 8–22.
Chang, F.-R. C., Skipper, N. T., Refson, K., Greathouse, J. A., and Sposito,
G. (1998). “Interlayer molecular structure and dynamics in Li-, Na-, and
K-montmorillonite-water systems.” ACS Symp. Ser., 715(1), 88–106.
Christidis, G. E. (1998). “Physical and chemical properties of some
bentonite deposits of Kimolos Island, Greece.” Appl. Clay Sci., 13(2),
79–98.
Churchman, G. J., and Burke, C. M. (1991). “Properties of subsoils in re-
lation to various measures of surface area and water content.” J. Soil
Sci., 42(3), 463–478.
Dontsova, K. M., and Norton, L. D. (2002). “Clay dispersion, infiltration,
and erosion as influenced by exchangeable Ca and Mg.” Soil Sci.,
167(3), 184–193.
Dontsova, K. M., Norton, L. D., Johnston, C. T., and Bigham, J. M. (2004).
“Influence of exchangeable cations on water adsorption by soil clays.”
Soil Sci. Soc. Am. J., 68(4), 1218–1227.
Ellis, B., and Foth, H. (1996). Soil fertility, 2nd Ed., CRC Press, Boca
Raton, FL.
Farrar, D., and Coleman, J. (1967). “The correlation of surface area
with other properties of nineteen British clay soils.” J. Soil Sci.,
18(1), 118–124.
Frydman, S., and Baker, R. (2009). “Theoretical soil-water characteristic
curves based on adsorption, cavitation, and a double porosity model.”
Int. J. Geomech., 10.1061/(ASCE)1532-3641(2009)9:6(250), 250–257.
Gaston, L. A., and Selim, H. M. (1990). “Predicting cation mobility
in montmorillonite media based on exchange selectivity of
montmorillonite.” Soil Sci. Soc. Am. J., 54(6), 1525–1530.
Hansson, K., Šimůnek, J., Mizoguchi, M., Lundin, L. C., and Van
Genuchten, M. T. (2004). “Water flow and heat transport in frozen
soil.” Vadose Zone J., 3(2), 693–704.
Jaynes, W. F., and Bioham, J. M. (1986). “Multiple cation-exchange capac-
ity measurements on standard clays using a commercial mechanical
extractor.” Clays Clay Miner., 34(1), 93–98.
Khorshidi, M. (2015). “Soil-water interaction at high soil suction.” Ph.D.
thesis, Colorado School of Mines, Golden, CO.
Khorshidi, M., and Lu, N. (2016a). “Intrinsic relation between soil water
retention and cation exchange capacity.” J. Geotech. Geoenviron. Eng.,
10.1061/(ASCE)GT.1943-5606.0001633, 04016119.
Khorshidi, M., and Lu, N. (2016b). “Soil water retention-based methodol-
ogy for classification of expansive soils.” Proc., Geo-Chicago 2016,
ASCE, Reston, VA, 865–873.
Khorshidi, M., Lu, N., Akin, I. D., and Likos, W. J. (2016c). “Intrinsic
relationship between specific surface area and soil water retention.”
J. Eng. Mech.
 J. Geotech. Geoenviron. Eng., 10.1061/(ASCE)GT.19435606.0001572,
04016078.
Khorshidi, M., Lu, N., and Khorshidi, A. (2016d). “Intrinsic relationship
between matric potential and cation hydration.” Vadose Zone J., 15(11),
in press.
Laird, D., and Fleming, P. (2008). “Analysis of layer charge, cation and
anion exchange capacities and synthesis of reduced charge clays.”
Methods of soil analysis. Part 5: Mineralogical methods, A. L. Ulery
and R. L. Drees, eds., Soil Science Society of America, Madison,
WI.
Laird, D. A. (1999). “Layer charge influences on the hydration of
expandable 2:1 phyllosilicates.” Clays Clay Miner., 47(5), 630–636.
Leao, T. P., and Tuller, M. (2014). “Relating soil specific surface area, water
film thickness, and water vapor adsorption.” Water Resour. Res.,
Downloaded from ascelibrary.org by Colorado School of Mines on 02/21/17. Copyright ASCE. For personal use only; all rights reserved.
50(10), 7873–7885.
Likos, W. J., Lu, N., and Wenszel, W. (2011). “Performance of a dynamic
dew point method for moisture isotherms of clays.” Geotech. Test. J.,
34(4), 1–10.
Lu, N., Godt, J. W., and Wu, D. T. (2010). “A closed-form equation for
effective stress in unsaturated soil.” Water Resour. Res., 46(5),
W05515.
Lu, N., Kaya, M., and Godt, J. W. (2014). “Interrelations among the soil-
water retention, hydraulic conductivity, and suction-stress characteristic
curves.” J. Geotech. Geoenviron. Eng., 10.1061/(ASCE)GT.1943-5606
.0001085, 04014007.
Lu, N., and Khorshidi, M. (2015). “Mechanisms for soil-water retention
and hysteresis at high suction range.” J. Geotech. Geoenviron. Eng.,
141(8), 04015032.
Lu, N., and Likos, W. J. (2004). Unsaturated soil mechanics. Wiley,
New York.
Măhler, J., and Persson, I. (2012). “A study of the hydration of the alkali
metal ions in aqueous solution.” Inorg. Chem., 51(1), 425–438.
McQueen, I. S., and Miller, R. F. (1974). “Approximating soil moisture
characteristics from limited data: Empirical evidence and tentative
model.” Water Resour. Res., 10(3), 521–527.
Mehlich, A. (1938). “Use of triethanolamine acetate-barium hydroxide
buffer for the determination of some base exchange properties and lime
requirement of soil.” Soil Sci. Soc. Am. Proc., 3, 162–166.
Mitchell, J. K., and Soga, K. (2005). Fundamentals of soil behavior, Wiley,
New York.
Moore, D. M., and Reynolds, R. C. (1997). X-Ray diffraction and the iden-
tification and analysis of clay minerals. 2nd Ed., Oxford University
Press, London.
Moradi, A., Smits, K. M., Massey, J., Cihan, A., and McCartney, J. (2015).
“Impact of coupled heat transfer and water flow on soil borehole ther-
mal energy storage (SBTES) systems: Experimental and modeling
investigation.” Geothermics, 57(1), 56–72.
Nevins, M. J., and Weinttitt, D. J. (1967). “Determination of cation ex-
change capacity by methylene blue adsorption.” Am. Ceram. Soc. Bull.,
46(1), 587–592.
Padilla, I. Y., Yeh, T. C. J., and Conklin, M. H. (1999). “The effect of water
content on solute transport in unsaturated porous media.” Water Resour.
Res., 35(11), 3303–3313.
Pearcy, R. W., Ehleringer, J., Mooney, H. A., and Rundel, P. W. (2012).
Plant physiological ecology: Field methods and instrumentation,
Springer, Amsterdam, Netherlands.
© ASCE 04017023-7
J. Eng. Mech., -1--1
Petersen, L. W., Moldrup, P., Jacobsen, O. H., and Rolston, D. E. (1996).
“Relations between specific surface area and soil physical and chemical
properties.” Soil Sci., 161(1), 9–21.
Prost, R., Benchara, A., and Huard, E. (1998). “State and location of water
adsorbed on clay minerals: consequences of the hydration and swelling-
shrinkage phenomena.” Clays Clay Miner., 46(2), 117–131.
Rengasamy, P., Greene, R. S. B., and Ford, G. W. (1986). “Influence of
magnesium on aggregate stability in sodic red-brown earths.” Aust.
J. Soil Res., 24(2), 229–237.
Revil, A., and Leroy, P. (2004). “Governing equations for ionic transport in
porous shales.” J. Geophys. Res., 109.
Ross, D. S., and Ketterings, Q. (2011). “Recommended methods for
determining soil cation exchange capacity.” Recommended Soil Testing
Procedures for the Northeastern United States, 3rd Ed., A. Wolf and J.
McGrath, eds., Northeastern Regional Publications, College Park, MD.
Schollenberger, C. J., and Dreibelbis, E. R. (1930). “Analytical methods in
base-exchange investigations on soils.” Soil Sci., 30(3), 161–174.
Shainberg, I., Rhoades, J. D., and Prather, R. J. (1980). “Effect of
exchangeable sodium percentage, cation exchange capacity, and soil
solution concentration on soil electrical conductivity.” Soil Sci. Soc.
Am. J., 44(3), 469–473.
Shuman, L. M., and Duncan, R. R. (1990). “Soil exchangeable cations and
aluminum measured by ammonium chloride, potassium chloride and
ammonium acetate.” Commun. Soil Sci. Plant Anal., 21(13–16),
1217–1228.
Smith, C. W., Hadas, A., Dan, J., and Koyumdjisky, H. (1985). “Shrinkage
and Atterberg limits relation to other properties of principle soil types in
Israel.” Geoderma, 35(1), 47–65.
Sposito, G. (2008). The chemistry of soils. 2nd Ed., Oxford University
Press, New York.
Środoń, J., and McCarty, D. K. (2008). “Surface area and layer charge
of smectite from CEC and EGME=H2 O-retention measurements.”
Clays Clay Miner., 56(2), 155–174.
Sumner, M. E., and Miller, W. P. (1996). “Cation exchange capacity and
exchange coefficients.” Methods of soil analysis. Part 3: Chemical
methods, D. L. Sparks, ed., Soil Science Society of America, Madison,
WI.
Thomas, P. J., Baker, J. C., and Zelazny, L. W. (2000). “An expansive soil
index for predicting shrink-swell potential.” Soil Sci. Soc. Am. J., 64(1),
268–274.
Tuller, M., Or, D., and Dudley, L. M. (1999). “Adsorption and capillary
condensation in porous media: Liquid retention and interfacial configu-
rations in angular pores.” Water Resour. Res., 35(7), 1949–1964.
van Genuchten, M. T. (1980). “A closed-form equation for predicting the
hydraulic conductivity of unsaturated soils.” Soil Sci. Soc. Am. J., 44(5),
892–898.
Woodruff, W. F., and Revil, A. (2011). “CEC-normalized clay-water
sorption isotherm.” Water Resour. Res., 47(11), W11502.
Xu, W., Johnston, C., Parker, P., and Agnew, S. (2000). “Infrared study of
water sorption on Na-, Li-, Ca- and Mg-exchanged (SWy-1 and SAz-1)
montmorillonite.” Clays Clay Miner., 48(1), 120–131.
Yilmaz, I. (2004). “Relationships between liquid limit, cation exchange
capacity, and swelling potentials of clayey soils.” Eurasian Soil Sci.,
37(5), 506–512.
Yukselen, Y., and Kaya, A. (2006). “Prediction of cation exchange capacity
from soil index properties.” Clay Miner., 41(4), 827–837.
J. Eng. Mech.