Journal of Contaminant Hydrology 107 (2009) 108113

Contents lists available at ScienceDirect

Journal of Contaminant Hydrology

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n h yd

Research article

Solubility enhancement of seven metal contaminants using

carboxymethyl--cyclodextrin (CMCD)
Magnus E. Skold ,a, Geoffrey D. Thyne b, John W. Drexler c, John E. McCray d
a
b
c
d

Department of Geology and Geological Engineering, Colorado School of Mines, Golden, CO 80401, United States
Hydrologic Science and Engineering Program, Department of Geology and Geological Engineering, Colorado School of Mines, Golden, CO 80401, United States
Geological Sciences, University of Colorado at Boulder, CO 80401, United States
Hydrologic Science and Engineering Program, Environmental Science and Engineering Division, Colorado School of Mines, Golden, CO 80401, United States

a r t i c l e

i n f o

Article history:
Received 5 December 2008
Received in revised form 25 February 2009
Accepted 16 April 2009
Available online 20 April 2009
Keywords:
Cyclodextrin
Solubility enhancement
Metals
Remediation

a b s t r a c t
Carboxymethyl--cyclodextrin (CMCD) has been suggested as a complexing agent for
remediation of sites co-contaminated with metals and organic pollutants. As part of an attempt
to construct a geochemical complexation model for metal-CMCD interactions, conditional
formation constants for the complexes between CMCD and 7 metal ions (Ba, Ca, Cd, Ni, Pb, Sr, and
Zn) are estimated from experimental data. Stable metal concentrations were reached after
approximately 1 day and estimated logarithmic conditional formation constants range from 3.2
to 5.1 with condence intervals within 0.08 log units. Experiments performed at 10 C and
25 C show that temperature affects the solubility of the metal salts but the strength of CMCDmetal complexes are not affected by this temperature variation. The conditional stability constants
and complexation model presented in this work can be used to screen CMCD as a potential
remediation agent for clean-up of contaminated soil and groundwater.
2009 Elsevier B.V. All rights reserved.

1. Introduction
Organic and inorganic contaminant mixtures present at
contaminated sites is one of the major challenges in
remediating soil and ground. Due to the fundamentally
different physiochemical properties of these contaminants,
research has to a large extent focused on remediating organic
and inorganic contaminants separately rather than with a
single treatment. Two basic approaches to remediation are to
wash (ex-situ) or ush (in-situ) the contaminated soil with
water (pump and treat) or with an agent to increase solubility
(enhanced ushing). If the complexing agent, or mixture of
complexing agents, is chosen appropriately both organic and
inorganic contaminants can be removed simultaneously. For
instance, biosurfactants (Mulligan et al., 2001) and solutions
of strong acids mixed with isopropyl alcohol (Semer and
Reddy, 1996) have successfully been used in bench scale
experiments. In both studies heavy metals were removed in
Corresponding author. Tel.: +1 303 471 3482; fax: +1 303 471 3482.
E-mail address: Magnus.Skold@Arcadis-us.com (M.E. Skold).
0169-7722/$ see front matter 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jconhyd.2009.04.006

the presence of organic compounds from a contaminated

sand and sandy silt, respectively.
One potential complexing agent to clean up mixed waste is
cyclodextrin (CD), a compound produced by microorganisms
consisting of 6, 7 or 8 glucose molecules in a ring structure. The
most common CD is the -cyclodextrin which is composed of 7
glucose molecules. The ring structure creates a non-polar cavity
enabling inclusion complexation of organic compounds (Wang
and Brusseau, 1993). Chemical properties of interest such as
aqueous solubility and metal complexation potential can be
altered by substituting functional groups to the outside of the
cyclodextrin. Laboratory and eld studies have demonstrated
that hydroxypropyl--cyclodextrin (HPCD) has high aqueous
solubility, strongly enhances solubility of organic compounds
and is not retained by soil (Boving and McCray, 2000).
A eld test at Hill Air Force Base, Utah demonstrated that
addition of 10 wt.% HPCD signicantly accelerated the removal
of organic contaminants via pump and treat (McCray et al.,
1999; McCray and Brusseau, 1999). The aquifer consisted of ne
to coarse sand interbedded with gravel and some silt. More TCE
was removed over 10 days than would have been removed in

M.E. Skold et al. / Journal of Contaminant Hydrology 107 (2009) 108113

109

2. Materials and methods

purity. All other chemicals were + 98% pure unless otherwise noted. The background ionic strength was controlled
by KNO3 in the multi-day experiments, while NaNO3 was
used in the 2-h experiments. In all titrations, the solutions
were purged with nitrogen gas (ultra high purity) prior to
reaction and the pH was adjusted with sodium hydroxide
solution.
All metal salts (Ba, Ca, Cd, Ni, Pb, and Zn) were oxalates,
except for SrSO4. The purity of Cd oxalate was not reported. Zn
and Ni oxalates were formed via precipitation from solutions
of metal nitrates and oxalic acid as described in Donia (1997).
XRD analysis showed that crystalline Ni and Zn oxalates
formed (data not shown).
The acidbase behavior of CMCD was investigated by
potentiometric titrations in a CO2(g) free atmosphere. The
titrations were performed in a glass vessel. Temperature was
kept constant at 25 C and background ionic strength was
50 mM KNO3 while pH was increased by adding incremental
volumes of 0.1 M NaOH. The pH was measured with a
combination glass electrode. Before titrating cyclodextrin
with NaOH, the CMCD-Na salt was converted to its acidic form
by passing 50 ml of a 100 g/l CMCD solution twice through a
column containing an ion exchange resin. Ion-exchange resin
(Rexyn 101H) was purchased from Fisher Scientic. The
CMCD solution was bubbled with nitrogen gas for at least 1 h
at pH near 3.3 prior to each titration in order to remove any
carbon dioxide.
Batch experiments aimed at determining conditional
stability constants for complexation between CMCD and
metal ions were performed in sealed 15 ml centrifuge tubes
submerged in a water bath to control temperature at 10 0.5 or
25 0.5 C. Two separate sets of experiments were performed.
A preliminary investigation at 10 and 25 C lasting 2 h was
performed prior to a second set of experiment at 25 C lasting
7 days. A small amount (2050 mg) metal salt was added to
10 ml solutions containing up to 20 g/l CMCD in 50 mM
background ionic strength. In the preliminary experiments,
background ionic strength was varied from 0 for 20 g/l CMCD
solutions, to 50 mM at the lower CMCD concentrations. In
the longer time experiments, the CMCD concentration ranged
from 0 to 100 g/l and background ionic strength was held
constant at 50 mM KNO3. Hydrogen ion activity was adjusted to
pH 6.00 0.05 in all CMCD solutions before adding the metal
salts. The pH was not controlled during the experiment to avoid
interference with buffering compounds; the nal pH varied
from 5.3 to 7.3. Effects of pH variations on the metal speciation,
such as formation of metal carbonates and hydroxides
was taken into account by geochemical modeling as explained
below. The centrifuge tubes were slowly rotated top-overbottom throughout the experiment to ensure constant and
gentle mixing.
The solid metal salt was separated from the aqueous
solution by 0.45 m ltering and the pH was measured
immediately. The aqueous metal concentration was subsequently analyzed with inductively couple plasma-optical
emission spectrometry (ICP-OES).

2.1. Experimental techniques

2.2. Modeling techniques

Industrial grade CMCD was provided by Wacker Company.

The compound was delivered as a sodium salt with unspecied

In our batch experiments, the free metal ion activity, Me2+

for all divalent cations investigated in this study, is controlled by

1 year with regular pump and treat. In addition, biotracer

studies indicated that ushing with HPCD enhanced the
biodegradation potential for ethanol and hexanol (Alter et al.,
2003). Vertical ushing with HPCD of a contaminant source
zone was attempted at another superfund site (Air Force Plant44, Tucson, Arizona) where the subsurface consisted of silt
and clay layers. Despite rate limited transfer of TCE to the
aqueous phase in this setting, 20 wt.% HPCD increased the TCE
concentration in the extracted groundwater threefold (Blanford
et al., 2001). In a third eld trial, enhanced ushing with HPCD
increased efuent TCE concentration up to 22 times compared
to water ush and the HPCD solution was recovered and reused
after removing 99.99% of TCE with an air stripper (Tick et al.,
2003). The study was conducted in an unconned aquifer
consisting of medium to ne sand with interbedded strings of
gravel and silt and clay lenses at the Dover Air Force Base, Dover,
Delaware. These studies show that HPCD is a potential agent for
remediation of organic contaminants.
Laboratory studies have shown that carboxymethyl-cyclodextrin (CMCD) has the ability to complex heavy metals
such as cadmium, nickel, strontium and mercury in the presence
of various organic contaminants (Wang and Brusseau, 1995;
Brusseau et al., 1997; Wang et al., 2004). A comparison between
HPCD and CMCD showed that both CDs enhanced the solubility
of organic compounds but that HPCD was the more effective
(Brusseau et al., 1997). Because HPCD does not complex metal
ions but CMCD does, the ability of the latter to simultaneously
enhance the solubility of organic and inorganic contaminants
is evaluated.
A literature review revealed only one study where the
stability constant for a complex between a single metal (Cd)
and CMCD was investigated (Wang and Brusseau, 1995),
making quantitative evaluation of potential remediation
performance and comparison to other complexing agents
difcult. If formation constants would be available for the
formation of aqueous complexes and metal complexation
agents such as CMCD, the solubility enhancement of metal
contaminants could be estimated and the use of each
complexation agent for in-situ remediation could be evaluated. This study conducted a series of batch equilibrium
experiments using industrial grade CMCD to derive conditional stability constants for aqueous complexes between
CMCD and Ba, Ca, Cd, Ni, Pb, Sr, and Zn. These elements were
selected to represent a suite of common metal contaminants
(Ba, Cd, Ni, Pb, Sr, and Zn) as wells as provide insight into
potential competition of elements present in most aquifers
(group 2 elements). In addition, Sr has a radioactive isotope
and is of concern at some sites. These constants can be
incorporated into a geochemical equilibrium model to simulate
the competition for complexation sites on CMCD between
common cations and metal contaminants and investigate
the effect of aqueous metal speciation and solid composition,
i.e. the composition of the mineral phases hosting the metal
contaminant.

110

M.E. Skold et al. / Journal of Contaminant Hydrology 107 (2009) 108113

the presence of a sparingly soluble metal salt. CMCD complexes

the free metal ion which causes more of the salt to dissolve
(Eqs. (1) and (2)). Previous systematic evaluation of experimental data for lead complexation by CMCD indicated the data
is best modeled by representing 3 sets of mass action equations
(Skold et al., 2007b). The CMCD has two protonation reactions
and a complexation reaction with a 1:1 (metal ion:CMCD)
stoichiometry (Eqs. (1)(4)). The previous evaluation investigated conceptual models based on various Pb:CMCD ratio,
different acid strength (pKa) for the individual carboxyl groups
that are part of the CMCD molecule, different methods for
calculating the activity coefcients of these carboxyl groups,
and methods of estimating the effect on ionic strength of high
CMCD concentrations. Titrations indicated that each CMCD
molecule has on average 4.27 titrable groups. Additional CMCD
acidity (2.27 carboxyl groups per CMCD) is modeled as separate
by carboxylic acid reactions (Eq. (5)). It is important to include
the acid/base behavior of CMCD in the model because hydrogen
ions and metal ions will compete for sites on CMCD at low pH.
Also, the buffering capacity of CMCD will affect pH dependent
dissolution of minerals in both lab experiments and potential
eld applications.
MeC2 O4 s = Me
2

CMCD

2+

2+

+ Me

= CMCDMe

CMCDH2 = CMCDH
CMCDH

+ C2 O4

= CMCD

COOH = COO

+H

+H

+H

The activity coefcient for the species CMCD 2 is

calculated using the WATEQ DebyeHckel equation with
a0,CMCD2- = 3.50 and bCMCD2- = 0.65 (Eq. (6)). The Davies
equation was used to calculate the activity coefcients of all
other species for specic ionic strength of the solution ()
(Stumm and Morgan, 1995).
2

log CMCD2 = A zCMCD2

+ bCMCD2

1 + B a0;CMCD2

!
p

6

The conditional stability constant for metal-CMCD complexation is calculated according to:
fCMCDMeg


KCMCD Me = 
CMCD2 Me2+

Values within curved brackets represent activities. The

geochemical modeling program PHREEQC (Parkhurst and
Appelo, 1999) was used to model metal complexation by
CMCD and the conditional stability constants were derived
with the parameter optimization program UCODE_2005
(Poeter and Hill, 1998; Poeter et al., 2005). The optimization
process is described in Skold et al. (2007a). Eqs. (1)(7)
were included in the PHREEQC model and the standard
PHREEQC database was amended with thermodynamic data

Table 1
Logarithmic constants for aqueous metal-oxalate complexes used in
modeling (Martell and Smith, 1977).

Ba
Ca
Cd
Ni
Pb
Sr
Zn

MeC2O4

Me(C2O4)22

Me(C2O4)43

MeHC2O+
4

Me(HC2O4)2

2.31
2.54
3.89
5.16
4.91
NA
4.87

NA
2.69
5.66
NA
6.76
NA
7.65

NA
NA
5.06
NA
NA
NA
NA

NA
1.84
NA
NA
NA
NA
1.72

NA
NA
NA
NA
NA
NA
3.12

for aqueous complexes between oxalate and metal ions as

shown in Table 1 (Martell and Smith, 1977).

3. Results and discussion

3.1. Potentiometric titrations
Titration end-points and pKas were evaluated with halfpoint analysis; the pKa of the hydroxyl groups on the CMCD
was determined as the pH at a point in the titration where half
of the groups were titrated and the total concentration of
carboxylic acid groups were determined at the inection
point when the change in pH was the greatest per addition of
acid or base (Stumm and Morgan, 1995). The analysis
indicated an average carboxyl concentration of 4.27 functional groups/CMCD molecule. This is equivalent to 0.610
substituted carboxymethyl groups per glucose ring. The 95%
condence interval ranges from 0.601 to 0.619. The molecular
weight of CMCD was calculated to 1390 g/mol based on
substitution ratio and its molecular structure. The average pKa
from ve titrations was 3.80 with 95% condence interval
ranging from 3.77 to 3.84. This value compares well with
published pKas of metoxyacetic acid and etoxyacetic acid,
which have similar structure to functional groups on CMCD:
3.57 and 3.67, respectively (CRC, 2004). The calculations were
performed assuming pure CMCD. If impurities containing less
carboxyl groups than CMCD were present in the experiments
the calculated concentration of functional groups would be
underestimated. However, it is difcult to know the composition of impurities.
The carboxymethyl groups are separated from each other
by several single carbon\carbon bonds. Very little electron
sharing is expected and we assume above that all carboxyl
groups have identical strength. This assumption was investigated by comparing experimental data to a model based on
experimentally derived acidity constants and assuming that
all carboxyl groups have the same strength. The model ts
experimental data well. The correlation coefcient (R2)
between observed and simulated pH for all 5 titrations
modeled simultaneously was 0.996. Experimental data show
a slightly smoother pH transition than do calculated data.
Subsequently, a model using two different pKa values for the
carboxyl groups was used to evaluate the data. This two-site
model enhanced the t somewhat but the improvement was
not considered great enough to warrant the use of a more
complicated model (Fig. 1).

M.E. Skold et al. / Journal of Contaminant Hydrology 107 (2009) 108113

Fig. 1. Base titrations at 0.518, 0.698, 0.977, 1.22, and 1.49 g/L of CMCD. Solid
lines represent single-site model and dotted lines represent two-site model.

3.2. Transient metal solubilization

In all experimental solutions, metal salts reached stable
concentrations within approximately 1 day (Fig. 2). IR
spectroscopy of CMCD solutions indicate that the large
molecule does not decompose during the remaining experimental duration (data not shown). Ca, Sr, Ba, Pb, Ni, and Zn
trends are stable over 7 days duration. In contrast, Cd shows
concentrations that decline with time in the presence of
CMCD. The dissolution of all the salt was at equilibrium in less
than a day except Ca-oxalate dissolution, which reached
equilibrium after 2 days.

111

Cadmium oxalate dissolved in a different manner than

the other metal salts. For all salts equilibrium was approached
from undersaturation by adding the metal salt to a solution
without the metal. The metal concentration was therefore
expected to increase until stabilizing at equilibrium. This
pattern was observed for all metals but Cd. The concentration
of Cd was 60.7 and 60.5 mg/l after 1 and 2.5 h, respectively,
in the blank solutions. The concentration then started to
decrease (58.6 and 58.7 mg/l after 4.5 and 19.5 h) and reached
44.9 mg/l after 7 days. The most plausible explanation for this
behavior is a change in the metal salt structure resulting in a
change in solubility. Cadmium oxalate exists as a hydrous salt
(CdC2O43H2O) and as an anhydrous salt (CdC2O4) (CRC, 2004).
It is postulated that the salt used in the experiments transitioned from one phase to a thermodynamically more stable
phase. A literature did not reveal any studies investigating the
solubility of the anhydrous form of cadmium oxalate.
3.3. Measured conditional stability constants
Results from batch experiments show that CMCD
enhances the solubility of all the investigated metal salts.
The conditional formation constants were optimized by
systematically varying their values such that the model
calculated aqueous metal concentration matched the
Table 2
Measured conditional formation constants and solubility products.
Metal salt Temp. texperimental logKsp a
BaC2O4

CaC2O4

CdC2O4

NiC2O4

PbC2O4

SrSO4

ZnC2O4

Fig. 2. Aqueous metal concentrations as a function of time. Each symbol

represents a single measurement.

10 C
25 C
25 C
10 C
25 C
25 C
10 C
25 C
25 C
25 C
25 C
10 C
25 C
25 C
10 C
25 C
25 C
10 C
25 C
25 C
10 C
25 C
25 C

2h
2h
7 days
2h
2h
7 days
2h
2h
4.5 h
19 h
7 days
2h
2h
7 days
2h
2h
7 days
2h
2h
7 days
2h
2h
7 days

6.71
6.55
6.59
8.76
8.49
8.03
7.99
7.80
7.80
7.80
7.80
NA
NA
9.73
10.79
10.46
10.46
6.81
6.69
6.69
NA
NA
8.86

logKsp b

logKCMCDMe 95% C.I. c

6.79 d 3.60
3.60
3.58
8.63 e 3.64
3.67
3.56
f
7.85
5.14
4.78
4.95
4.40
4.04
g
9.4
NA
NA
3.48
9.32 h 5.21
5.27
5.18
6.46 i 3.49
3.53
3.55
j
8.86
NA
NA
3.64

3.543.66
3.583.64
3.553.60
3.583.70
3.593.75
3.473.64
4.765.52
4.495.07
4.755.16
4.304.50
3.914.18
NA
NA
3.443.51
5.135.30
5.215.34
5.145.23
3.293.69
3.283.78
3.473.63
NA
NA
3.613.67

NA, not analyzed.

a
Experimentally determined solubility product, Ksp, was calculated from
aqueous metal concentration measured in experiments with metal salt and
deionized water.
b
Ksp values reported in Dean (1999).
c
95% condence interval for experimentally determined complexation
constant (KCMCD-Me) for complexation reaction between the free metal ions
and CMCD.
d
Value for BaC2O4H2O.
e
Value for CaC2O4H2O.
f
Value for CdC2O43H2O.
g
Value for NiC2O4.
h
Value for PbC2O4.
i
Value for SrSO4.
j
Value for ZnC2O42H2O.

112

M.E. Skold et al. / Journal of Contaminant Hydrology 107 (2009) 108113

laboratory analyzed metal concentration. The procedure is

automated in UCODE_2005 and statistics including 95%
condence intervals are calculated during the optimization
process (Poeter et al., 2005; Poeter and Hill, 1998). The
calculated conditional stability constants for the CMCD-metal
complexes were based on 7-day data at 25 C and range from
10 3.2 to 10 5.1 (Table 2). Estimated 95% condence intervals
are within 0.08 log units for all elements but Cd.
Fig. 3 shows that in most cases the agreement between
experimental data and calculated solubilization based on the
tted conditional stability constants models is very good. The
condence intervals are estimated based on assumption of a
correctly formulated conceptual model, which includes
assumptions regarding number of metal complexation
sites per CMCD molecule, no difference in strength between
the different complexation sites, and model used to calculate
non-ideal interactions of functional groups on the large organic
molecule. These assumptions have been investigated previously
(Skold et al., 2007a,b) and the optimized conceptual model was
used in this work. Given the difculty of modeling complexation
reactions for multiple functional groups on organic molecules,
the conceptual model may not be entirely correct. Although the
difference between estimated and observed data is small, the
true uncertainty is likely somewhat larger than the reported
95% condence intervals because of uncertainties in the
conceptual model. However, any uncertainties related to, for
instance, ionic strength model cannot be quantied.

Fig. 3. Enhanced metal solubility as a function of CMCD concentration.

Experimental data (symbols) represents results of single measurements are
compared to models (lines) based on 1:1 metal:CMCD complexation.

Fig. 4. Enhanced cadmium oxalate solubility as a function of CMCD

concentration at various contact times.

The Cd behavior was investigated by measuring the

solubility of cadmium after 1, 2, 4.5, 19.5 h and after 7 days at
25 C, Fig. 4. The aqueous Cd concentration increases during the
rst 4.5 h and then decreases until the end of the experiment.
The decrease in cadmium concentration is more pronounced in
the presence of CMCD than in its absence. In 1 mg/l CMCD the
cadmium concentration decreases from a maximum of 165 mg/
l after 2.5 h to 54.8 mg/l after 7 days contact time. This supports
the hypothesis of a phase transfer to a thermodynamically more
stable metal salt structure. This potential phase transfer appears
to be occurring more rapidly when the total aqueous metal
concentration increases due to CMCD-Cd complexation. When
calculating the conditional formation constant for Cd complexation the value of the solubility product was not varied
between the different data sets. If the solubility product would
have been recalculated for each time series a smaller solubility
product for the 7-day series would have resulted in a larger
conditional formation constant for Cd-CMCD complexation.
The strength of the Cd-CMCD complexes are not assumed to
change with time and the most appropriate conditional
formation constant for complexation with Cd is the suggested
to be the one determined after 4.5 h (4.95). The experimental
data for enhanced cadmium dissolution is, however, less
accurate and precise than the other metals and further
investigations of his behavior would be benecial.
Temperature affects the solubility of the salts. The concentrations of all metals, except Cd, increase with increasing
temperature. The solubility product used to calculate
the complexation was derived from the metal salt solutions.
Table 2 shows the calculated solubility products are higher at
25 C than at 10 C but the estimated conditional formation
constants do not change. Thus, it appears that the greater metal
solubility in the warmer solutions is due to greater salt
solubility rather than increased metal complexation by CMCD.
The conditional formation constants of the transition
elements increase with increasing period and group in the
periodic table. The heavier elements Cd and Pb have greater
conditional formation constants than the lighter transition
elements and the group 2 elements. Based on this trend, we
hypothesize that other common metal contaminants such as
Cu and Hg will also form relatively strong complexes with
CMCD. While the solubility enhancement of some important
metals not used in these experiments can be interpolated,
such is not the case for the metalloids. For example, As and Se
are normally oxy-anions in solution (Smedley and Kinniburg,

M.E. Skold et al. / Journal of Contaminant Hydrology 107 (2009) 108113

2002). Because anions do not form strong complexes with

carboxyl groups, the behavior of these complexes cannot be
interpolated based on other elements but needs to be
investigated separately. Indeed, Chatain et al. (2004) found
that Cu and Fe were removed more readily from a mining
contaminated soil with CMCD than was As.
The effectiveness of metal remediation will be inuenced
by the presence of non-target cations, such as Ca and Mg. The
extent of competition will depend on their concentrations
and strengths of CMCD complexes. The conditional formation
constants for the group two elements Ca, Sr, and Ba, range
from 3.55 to 3.58. There is no trend in complexation strength
with their period in the periodic table because the 95%
condence intervals overlap. For the transition elements,
however, there does appear to be a trend. The magnitude
of the formation constants increases with increasing group
and period. The period 4 elements Ni and Zn have similar
conditional formation constants as the group 2 elements: 3.48
and 3.64, respectively. The measured constant for Cd, a period
5 transition element, ranged between 4.95 (4.5 h) and 4.04
(7-day). The strongest complexes were formed between
CMCD and Pb, a period 6 element, the formation constant
(5.18) was almost two orders of magnitude greater than the
formation constants for the group two elements. Based on the
data produced in this study, transition elements in higher
period and group are preferentially complexed to CMCD
compared to group 2 elements. High concentrations of Ca and
Mg or the presence of soluble minerals containing group 2
elements would, however, lessen the solubility enhancement
of heavier target metals such as Pb and Cd.
4. Conclusions
Carboxymethyl--cyclodextrin enhances the solubility of
common metal contaminants making CMCD a potential agent
for in-situ remediation of metals. The solubility enhancement
is greater for heavy metals (greater period and group) than
for group 2 elements, however, the presence of soluble
minerals composed of non-target elements would affect to
what extent the solubility of the target metal would be
enhanced. Conditional formation constants for the complexes
between CMCD and 7 metal ions (Ba, Ca, Cd, Ni, Pb, Sr, and Zn)
range form 10 3.2 to 10 5.1 with condence intervals
within 0.08 log units. Temperature variation between
10 C and 25 C did not affect the metal complexation by
CMCD, but signicantly affected the solubility of the investigated metal salts. These conditional formation constants
incorporated into geochemical software may be used to
screen CMCD as a potential remediation agent. It is recommended that the technique be tested under eld conditions to
further assess the applicability to in-situ remediation.
Acknowledgements
This work was in part sponsored by the Environmental
Science and Engineering Division and the Department of
Geology and Geological Engineering at the Colorado School
of Mines and by the Laboratory for Environmental and
Geological Studies at the University of Colorado at Boulder. The
authors wish to acknowledge the editor and three anonymous
reviewers who greatly improved the quality of this manuscript.

113

References
Alter, S.R., Brusseau, M.L., Piatt, J.J., Ray-Maitra, A., Wang, J.-M., Cain, R.B.,
2003. Use of tracer tests to evaluate the impact of enhanced-solubility
ushing on in-situ biodegradation. Journal of Contaminant Hydrology
34, 191202.
Blanford, W.J., Barackman, M.L., Boving, T.B., Klingel, G.R., Johnson, G.R.,
Brusseau, M.L., 2001. Cyclodextrin-enhanced vertical ushing of a
trichloroethane contaminated aquifer. Ground Water Monitoring and
Remediation 21 (4), 5866.
Boving, T.B., McCray, J.E., 2000. Cyclodextrin-enhanced remediation of
organic and metal contaminants in porous media and groundwater.
Remediation Journal 10, 5983.
Brusseau, M.L., Wang, X., Wang, W.-Z., 1997. Simultaneous elution of heavy
metals and organic compounds from soil by cyclodextrin. Environmental
Science and Technology 31, 10871092.
Chatain, V., Hanna, K., de Brauer, C., Bayard, R., Germain, P., 2004. Enhanced
solubilization of arsenic and 2,3,4,6 tetrachlorophenol from soils by a
cyclodextrin derivative. Chemosphere 57, 197206.
CRC Press, 2004. CRC Handbook of chemistry and physics, 85th Edition. CRC
Press, Cleveland, OH.
Dean, J.R., 1999. Lange's handbook of chemistry, 15th Edition. McGraw-Hill,
New York.
Donia, A.M., 1997. Synthesis, identication and thermal analysis of coprecipitates
of silver-(cobalt, nickel, copper and zinc) oxalate. Polyhedron 16, 30133031.
Martell, A.E., Smith, R.M., 1977. Critical stability constants, Volume 3: Other
organic ligands. Plenum Press, New York.
McCray, J.E., Brusseau, M.K., 1999. Cyclodextrin-enhanced in situ ushing of
multiple-component immiscible organic liquid contamination at the
eld scale: analysis of dissolution behavior. Environmental Science and
Technology 33, 8995.
McCray, E.M., Bryan, K.D., Cain, R.B., Johnson, G.R., Blanford, W.J., Brusseau, M.L.,
1999. Field test of cyclodextrin for enhanced in-situ ushing of multiplecomponent immiscible organic liquid contamination: comparison to
water ushing. In: Brusseau, M.L., Sabatini, D.A., Gierke, J.S., Annable,
M.D. (Eds.), Innovative Subsurface Remediation, ASC Symposium
Series, vol. 725, pp. 136152.
Mulligan, C.N., Yong, R.N., Gibbs, B.F., 2001. Remediation technologies for
metal-contaminated soils and groundwater: an evaluation. Engineering
Geology 60, 193207.
Parkhurst, D.L., Appelo, C.A.J., 1999. User's guide to phreeqc (version 2) A
computer program for speciation, batch-reaction, one-dimensional
transport, and inverse geochemical calculations; U.S. Geological Survey
Water-Resources Investigations Report 99-4259; U.S. Geological Survey:
Denver, CO.
Poeter, E.P., Hill, M.C., 1998. Documentation of UCODE, a computer code for
universal inverse modeling: U.S. Geological Survey, Water-Resources
Investigations Report 98-4080.
Poeter, E.P., Hill, M.C., Banta, E.R., Mehl, S., Christensen, S., 2005. UCODE_2005
and six other computer codes for universal sensitivity analysis, calibration, and uncertainty evaluation: U.S. Geological Survey Techniques and
Methods 6-A11, 283p.
Semer, R., Reddy, K.R., 1996. Evaluation of soil washing process to remove
mixed contaminants from a sandy loam. Journal of Hazardous Materials
45, 4557.
Skold, M.E., Thyne, G.D., Drexler, J.W., McCray, J.E., 2007a. Determining
conditional stability constants for Pb complexation by carboxymethyl-cyclodextrin (CMCD). Journal of Contaminant Hydrology 93, 203215.
Skold, M.E., Thyne, G.D., McCray, J.E., 2007b. Using UCODE_2005 and
PHREEQC to determine thermodynamic constants from experimental
data. Ground Water 45, 368373.
Smedley, P.L., Kinniburg, D.G., 2002. A review of the source, behavior and
distribution of arsenic in natural waters. Applied Geochemistry 17, 517568.
Stumm, W., Morgan, J.J., 1995. Aquatic chemistry, 3rd Edition. John Wiley and
sons, New York.
Tick, G.R., Lourenso, F., Wood, A.L., Brusseau, M.L., 2003. Pilot-scale demonstration of cyclodextrin as a solubility-enhancement agent for remediation of
a tetrachloroethene-contaminated aquifer. Environmental Science and
Technology 37, 58295834.
Wang, X., Brusseau, M.L., 1993. Solubilization of some low-polarity organic
compounds by hydroxypropyl--cyclodextrin. Environmental Science
and Technology 27, 28212825.
Wang, X., Brusseau, M.L., 1995. Simultaneous complexation of organic
compounds and heavy metals by a modied cyclodextrin. Environmental
Science and Technology 29, 26322635.
Wang, X., Yolcubal, I., Wang, W., Artiola, J., Maier, R., Brusseau, M.L., 2004.
Enhanced removal of mercury from soil by cyclodextrin, ramnolipid, and
calcium chloride. Environmental Toxicology and Chemistry 23, 18881892.