Content uploaded by Jon Chorover
Author content
All content in this area was uploaded by Jon Chorover on May 16, 2016
Content may be subject to copyright.
Adsorption of perfluorooctanoic acid and
perfluorooctanesulfonic acid to iron oxide surfaces as
studied by flow-through ATR-FTIR spectroscopy
Xiaodong Gao
A
,
B
and Jon Chorover
A
,
C
A
Department of Soil, Water and Environmental Science, University of Arizona, Tucson,
AZ 85721, USA.
B
Present address: Department of Earth Science, Rice University, Houston, TX 77251, USA.
Email: xdgao@rice.edu
C
Corresponding author. Email: chorover@cals.arizona.edu
Environmental context. Perfluoroalkyl compounds are organic contaminants that exhibit strong resistance to
chemical- and microbial-degradation. As partitioning between solid and aqueous phases is expected to control
the transport of perfluoroalkyl compounds, we studied the molecular mechanisms of their adsorption–
desorption at a representative Fe oxide surface using in situ molecular spectroscopy. The results provide
valuable information on the types of bonds formed, and enable a better understanding of the transport and fate of
these organic contaminants in natural environments.
Abstract. The kinetics and mechanisms of perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS)
adsorption to nanoparticulate hematite (a-Fe
2
O
3
) from aqueous solutions were examined using in situ, flow-through
attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy. Results indicate that both PFOA and
PFOS molecules are retained at the hydrophilic hematite surface and the adsorption shows strong pH dependence.
However, ATR-FTIR data reveal that PFOA and PFOS are bound to the iron oxide by different mechanisms. Specifically,
in addition to electrostatic interactions, PFOA forms inner-sphere Fe–carboxylate complexes by ligand exchange, whereas
the PFOS sulfonate group forms outer-sphere complexes and possibly hydrogen-bonds at the mineral surface. Both
solution pH and surface loading affect adsorption kinetics. Faster adsorption was observed at low pH and high initial PFC
concentrations. Sorption kinetics for both compounds can be described by a pseudo-second-order rate law at low pH (pH
3.0 and 4.5) and a pseudo-first-order rate law at high pH (pH 6.0). Sorption isotherm data for PFOA derived from
spectroscopic results exhibit features characteristic of ionic surfactant adsorption to hydrophilic charged solid surfaces.
Received 29 September 2011, accepted 15 February 2012, published online 30 April 2012
Introduction
Perfluoroalkyl compounds (PFCs) such as perfluorooctanoic
acid (PFOA) and perfluorooctanesulfonic acid (PFOS) (Fig. 1)
are surfactants that have been used as fire extinguishing agents
in aqueous film forming foam (AFFF) for decades.
[1]
AFFF
products usually comprise 1–5 % w/w PFCs, including PFOS,
PFOA and their short-chain homologues.
[2]
The unique chemi-
cal and thermal stabilities of PFCs are the principal reasons for
their use in AFFFs.
[3]
The physicochemical properties of PFCs
differ significantly from their hydrocarbon analogues. Substi-
tution of H by electronegative F in the aliphatic chain structure
gives rise to very high C–F bond covalency. Relative to their
hydrocarbon analogues, fluorination increases the rigidity of the
carbon chains, compound thermal stability and Brønsted acidity
of the acid group (Table 1).
[4–7]
Historically, effluents from AFFF fire-fighting activities
were not pre-treated before discharge to waste-water treatment
systems or to the environment, and military activities are an
important contributor. For example, because of the use of large
quantities of flammable liquids, the USA military consumes
,75 % of total nationwide AFFF production each year, resulting
in numerous PFC contaminated sites within the Department of
Defence complex that require cleanup.
[3,8]
It is estimated that a
total of 50–100 t of PFOA and 3–30 t of PFOS have been
released to the environment from historical use of AFFF
products in the USA.
[1]
PFCs can cause developmental toxicity,
immunotoxicity, hepatotoxicity and hormonal effects even at
very low concentrations.
[9,10]
Thus, it is essential to establish a
mechanistic understanding of the geochemical processes that
govern the transport and fate of PFCs in soil and water systems,
and that can be used in remediation technologies for effective
contaminant removal.
The global occurrence and distribution of PFCs has only
recently become a focus of study. Despite the fact that
PFCs have been widely identified in surface water,
[11–13]
groundwater,
[8,13]
public drinking water systems
[14,15]
and
sea water,
[11,16]
little is known about their transport and fate in
the environment. Because of their remarkable resistance
to chemical- and microbial-degradation in geomedia,
[17]
sorption–desorption dynamics are expected to be a key control
on environmental fate. Upon release to soils and sediments, the
reactive transport of PFCs in terrestrial systems is governed
largely by surface interactions with soil and sediment particles.
A recent soil column study reported that PFOS in street runoff
CSIRO PUBLISHING
Environ. Chem. 2012,9, 148–157
http://dx.doi.org/10.1071/EN11119
Journal compilation ÓCSIRO 2012 www.publish.csiro.au/journals/env148
Research Paper
was not removed during the infiltration process and in fact the
concentration of PFOS was observed to increase along the flow
path, which was attributed to its formation from precursor
compounds.
[13]
This study and several other recent reports
[12,18]
suggest that PFCs can be relatively mobile in the subsurface and
may cause groundwater contamination. PFCs are strong acids
that dissociate to anionic form in most natural environments.
Thus, anionic PFCs are expected to adsorb to positively charged
surfaces (e.g. Al and Fe oxides) by electrostatic or covalent
bonding interactions. In addition, because of its unique compo-
sition, the fluorocarbon tail exhibits both hydrophobic and
lipophobic properties, and fluorocarbon chain length controls
the extent to which the compound undergoes intermolecular
aggregation and adsorbs to hydrophobic surfaces such as natural
organic matter (NOM).
[19]
The energetics of surfactant adsorption result from the
additive effects of multiple bonding modes, including the Gibbs
energies of molecular and surface hydration, and intermolecular
associations. In general, molecular mechanisms of adsorption
can include covalent (inner-sphere) or electrostatic (outer-
sphere) complexation at surface sites, hydrophobic interaction
(including intermolecular associations such as surface admicelle
and hemimicelle formation), ion or water bridging and hydrogen
bonding.
[20]
Prior macroscopic studies indicate that adsorption
of PFCs is strongly influenced by aqueous geochemistry and
sorbent properties. Adsorption of PFCs to several natural sedi-
ments, clays, synthetic iron oxide (goethite, no organic carbon)
and activated carbon shows positive correlation with organic
matter content
[21,22]
and compound fluorocarbon chain
length,
[21,23]
indicating that PFC hydrophobicity contributes to
sorption energetics. Adsorption of PFCs to sediments also
increases with decreasing pH and charge fraction of bivalent
cation (Ca
2þ
),
[21,22,24,25]
suggesting that electrostatic interaction
of the polar anionic PFC head with charged solids and bridging
cations plays an important role in adsorption. Sorption mechan-
isms, however, cannot be unambiguously determined from these
macroscopic experiments. Molecular-scale adsorption mechan-
isms of PFCs at solid surfaces remain largely unexplored.
In addition, PFCs can potentially form surface-induced hemi-
micelles at solid–water interfaces well below the critical micelle
concentration (CMC) (0.1–1.0 % of the CMC).
[22]
As a result,
the adsorption mechanisms of PFCs are likely to vary signifi-
cantly across the environmentally relevant concentration range.
Studies are clearly needed to develop a comprehensive under-
standing of molecular-scale sorption processes as affected by
geochemical conditions as they control the transport and fate of
PFCs in natural soil and water systems.
Attenuated total reflection Fourier-transform infrared (ATR-
FTIR) spectroscopy is an established in situ technique for
interrogating molecular sorption mechanisms, redox processes
and reaction kinetics in aqueous systems.
[26–29]
Recently, ATR-
FTIR has been increasingly applied to the determination of
adsorption kinetics and detailed molecular-level information of
the coordination and structure of various adsorbates at the solid–
water interfaces, including inorganic anions (e.g. sulfate, phos-
phate and arsenate),
[30,31]
low-molecular-weight (LMW) car-
boxylic acids,
[32]
extracellular polymeric substance (EPS) and
intact microbial cells
[28,33]
and hydrocarbon ionic surfac-
tants.
[29,34,35]
To our knowledge, however, there are no prior
ATR-FTIR studies of adsorption of PFCs from aqueous solution
onto environmental solid surfaces. The primary objective of this
study, therefore, was to examine the binding mechanisms and
kinetics of PFOA and PFOS, the two most common components
of PFCs in AFFF products, onto the surface of a representative
Fe
III
oxide (hematite, a-Fe
2
O
3
) across a gradient in pH that is
characteristic of natural pore waters present in soils and sedi-
ments. Hematite was chosen as the model surface because it is
one of the most abundant Fe (oxyhydr)oxides in highly weath-
ered soils or sediments, and it often occurs naturally in nano-
particulate, high specific surface area forms, which can serve as
a high affinity sorbent for PFCs.
[36]
ATR-FTIR spectroscopy
was employed in conjunction with a flow cell technique to
elucidate in situ dynamic changes in spectroscopic data of
PFOA and PFOS as affected by solution chemistry and binding
to solid surfaces.
Experimental methods
Chemicals and materials
PFOA (96 %) and PFOS as the potassium salt (.98 %) were
purchased from Sigma–Aldrich Co. (St Louis, MO) and used as
received. Synthesis of nanoparticulate a-Fe
2
O
3
was performed
following the methods of Schwertmann and Cornell.
[36]
The
detailed synthetic procedure and characterisation have been
given previously.
[28]
The synthetic hematite has a particle size of
,10–20 nm in diameter, a Brunauer–Emmett–Teller (BET) N
2
specific surface area of 69.3 0.3 m
2
g
1
, and a measured iso-
electric point (IEP) of ,7.7–7.8 in 1 mM NaNO
3
background
electrolyte.
[28,29]
All solutions were prepared using Barnstead
Nanopure (BNP) water (Thermal Scientific, Waltham, MA;
18.2 MOcm) in a background electrolyte solution containing
10 mM NaCl (except for the ionic strength study) with pH
adjustment by addition of NaOH or HCl.
FTIR spectroscopy measurements
ATR-FTIR spectra were obtained with a Magna-IR 560 Nicolet
spectrometer (Madison, WI) equipped with a purge gas
generator and a deuterated triglycine sulfate (DTGS) detector.
(b)
(a) FFFFFF
FFFFFF
F
F
F
F
FFFFFFF
O
O
OH
S
FF FF FF F
F
F
O
OH
Fig. 1. Chemical structure of (a) perfluorooctanoic acid (PFOA,
C
7
F
15
COOH) and (b) perfluorooctanesulfonic acid (PFOS, C
8
F
17
SO
3
H).
Table 1. Physico-chemical properties of the perfluoroalkyl com-
pounds (PFCs) perfluorooctanoic acid (PFOA) and perfluorooctanesul-
fonic acid (PFOS)
[4]7]
CMC, critical micelle concentration
PFCs Molecular
formula
Water solubility
(mg L
1
)
pK
a
CMC (mM) log K
ow
PFOA CF
3
(CF
2
)
6
COO
3400 2.6 8.7–9.0 4.3
PFOS CF
3
(CF
2
)
7
SO
3
570 ,1 2.0 5.3
PFOA and PFOS adsorption to hematite
149
A458trapezoidal germanium internal reflection element (IRE)
(56 10 3mm
3
) in a flow-through cell (Pike Technologies,
Madison, WI) was employed. The Ge IRE in the flow cell was
coated with ,1.65 mg nanoparticulate hematite as described by
Gao et al.
[28]
and placed on a horizontal ATR sample stage
inside the IR spectrometer. The thickness of the hematite coat-
ing was calculated to be ,0.6 mm. A new coating was prepared
for each experiment, and spectra of dry films were collected to
determine the consistency of the coating. The cell was con-
nected to a reaction vessel containing 1 L of background elec-
trolyte solution, with and without PFCs, continuously stirred
with a magnetic bar. A peristaltic pump was used to deliver the
background electrolyte or PFC solutions from the reaction
vessel through the flow cell at a constant flow rate of 0.5 mL
min
1
. The effluent from the flow cell was collected to waste.
All spectra were acquired at 4.0 cm
1
resolution with 400
scans over the spectroscopic range of 4000–800 cm
1
using the
autogain function and aperture set at 100 at room temperature.
For each experiment, background electrolyte solution (10 mM
NaCl) was first pumped through the cell, allowing the hematite
coating to equilibrate with the background solution. Successive
background spectra of 10 mM NaCl at the hematite-coated Ge
IRE were collected during the equilibrium time under continu-
ous flow conditions. The final background spectrum was col-
lected when no further changes in the spectra were observed and
this background was used for the remainder of the experiment.
PFOA or PFOS in 10 mM NaCl electrolyte solution was
then injected into the cell to initiate the adsorption experi-
ment. Spectra were collected as a function of pH (3.0–6.0)
and compound concentration (50–1000 mM for PFOA and
50–250 mM for PFOS). Experiments were also conducted at
pH 9.0, which is higher than the point of zero net proton charge
(pH
pznpc
) of hematite (pH 7.7–7.8), but spectroscopic intensities
were observed to decrease with increasing pH and became
undetectable by FTIR at pH 9.0. PFOS adsorption was examined
over a lower concentration range due to its lower water solu-
bility and CMC (Table 1). The pH of the solution in the reaction
vessel was monitored throughout the experiments, and adjusted
as necessary by addition of 10 mM NaOH or HCl. Adsorption
kinetics of PFOA or PFOS on hematite were monitored by
collecting spectra at 15 min intervals untiladsorption equilibrium
was attained as indicated by no further changes between succes-
sive spectra. Following the adsorption experiment, PFC-free
background electrolyte solution was introduced again to the
ATR cell to assess desorption kinetics. Spectra were collected
every 15 min until no change was observed in subsequent
spectroscopic data over a time scale of ,120 min and this was
defined as the apparent desorption equilibrium. The effect of
ionic strength on PFOA adsorption was also examined in a
separate experiment in 1, 10 and 100 mM NaCl background
electrolyte using the same method described above.
All spectra (adsorption and desorption) were obtained by
subtracting the final background spectrum of hematite in the
corresponding NaCl background electrolyte. Data collection
and spectroscopic processing, including background subtraction
and baseline correction, were performed using the OMNIC
program (Thermo Nicolet, Co., Madison, WI). The GRAMS/
AI software (Thermo Electron Corp., Madison, WI) was used
for peak deconvolution and to determine the position and area of
the IR bands. A linear baseline was applied to the raw spectra.
The baseline corrected spectra were then fitted with Lorentzian
peaks. No constraints were placed on any of the fitting para-
meters (e.g. peak position, width or intensity).
Results and discussion
FTIR spectra of dissolved and hematite-adsorbed PFOA
The ATR-FTIR spectrum of aqueous PFOA on the Ge IRE in the
absence of nanohematite coatings represents that of the dis-
solved species. Adsorption to the Ge IRE is expected to be
negligible because of electrostatic repulsion between the nega-
tively charged Ge IRE and the anionic surfactant. The spectrum
is dominated by strong asymmetric and symmetric stretching
bands of the CF
2
and CF
3
groups of the hydrophobic tail and the
deprotonated carboxylate head groups (COO
) (Fig. 2).
Detailed peak position and mode assignments are given in
Table 2.
[26,28,37,38]
As PFOA, a monocarboxylic acid with rel-
atively high acid strength (pK
a
¼1.31–2.8 depending on
experimental conditions),
[39,40]
is nearly fully deprotonated
throughout the pH range probed by this study ($pH 3.0), the
carboxylate stretching bands in aqueous form are not expected
to exhibit systematic changes with solution pH.
In the presence of the hematite coating, absorbance intensi-
ties in the PFOA spectrum are much greater (,10-fold higher)
than those for the aqueous form (Fig. 2). The positive effect of
hematite on PFOA adsorption can be attributed to favourable
electrostatic interactions between the positive-charged hematite
surface (pH
pznpc
¼7.7–7.8)
[28]
and the anionic surfactant, in
addition to an increase in total surface area of the solid–water
interface probed by the IR beam when nanoparticulate hematite
is present.
[29]
The spectrum of hematite-adsorbed PFOA is very
similar to its aqueous form, exhibiting major bands correspond-
ing to the stretching bands of COO
,CF
2
and CF
3
groups.
Despite overall similarities between the two spectra (Fig. 2),
close examination reveals that significant changes in the car-
boxylate stretching region results from the presence of hematite,
indicating that the negatively charged COO
groups play a role
in interfacial reaction. For example, compared with the aqueous
PFOA spectrum, the peak position of n
as
(COO
) in the presence
of hematite is shifted to a higher wavenumber, from 1667 to
1677 cm
1
, whereas the n
s
(COO
) is shifted to a slightly lower
wavenumber (from 1414 to 1408 cm
1
), resulting in an increase
1800 1600 1400 1200 1000
Wavenumber (cm1)
1677 1408
1363 &
1319
1242
1208
1149
1102 & 1016
1667
1414
@ Ge IRE
(intensity
10)
@ Hematite
1238
1204
Fig. 2. Attenuated total reflection Fourier-transform infrared spectra of
500-mM perfluorooctanoic acid adsorbed on Ge internal reflection element
(IRE) and hematite surface in 10 mM NaCl solution at pH 6.0. The
absorbance intensity of the spectrum at Ge IRE was increased 10 times for
comparison.
X. Gao and J. Chorover
150
in separation (D~
n) between these two stretching bands. Similar
shifts in the n
as
and n
s
peak positions have been previously
observed for LMW carboxylic acid (L-lactate) adsorbed on
hematite nanoparticles.
[41]
These shifts were attributed to the
formation of inner-sphere complexes between lactate carboxyl-
ate groups and hematite surface Fe atoms. In addition, the
intensity ratio of n
s
to n
as
for COO
stretching increased upon
PFOA adsorption to hematite (Fig. 2). Prior studies indicate that
increased values of the n
s
(COO
)ton
as
(COO
) intensity ratio
upon adsorption are a diagnostic feature of inner-sphere com-
plexation with mineral surface metal centres.
[28,32,42]
Thus, the
results suggest that the carboxylate head groups of PFOA form
inner-sphere (i.e. covalent) bonds at the hematite surface. The
asymmetric stretching bands of CF
2
groups were also shifted
slightly to higher wavenumbers, apparently because of local
changes in the PFOA molecular environment encountered at the
solid–water interface upon adsorption (Fig. 2).
Although ATR-FTIR data clearly indicate that PFOA is
adsorbed to hematite by inner-sphere complexes, the additional
contribution of outer-sphere complexes cannot be excluded
under the experimental conditions, as the hematite surface and
surfactant molecules are oppositely charged. Variation in ionic
strength is often used as an indirect test of the contribution of
inner- and outer-sphere complex modes of metal or anion
adsorption to charged surfaces.
[43]
If inner-sphere complexation
is the sole sorption mechanism, adsorption should not be
significantly affected by a change in the ionic strength of
‘indifferent’ (non-surface complexing, e.g. NaCl) background
electrolyte, whereas if outer-sphere complexation is a predomi-
nant sorption mechanism, the extent of adsorption should
decrease with increasing ionic strength. In the present study,
PFOA adsorption to hematite was measured as a function of
ionic strength (NaCl as background electrolyte) at three differ-
ent pH values (3.0, 4.5 and 6.0). The adsorption, as represented
by the absorbance of the n
as
(CF
2
) band centred at 1208 cm
1
,
decreases with increasing ionic strength at all pH values,
indicating that, in addition to ligand exchange reactions,
outer-sphere complexation also plays a role in PFOA adsorption
(Fig. 3). A negative effect of ionic strength on adsorption was
also observed in a batch study of PFOS adsorption on goe-
thite.
[25]
In the pH range of the present study (3.0–6.0), the
electrostatic attractive force between the positively charged
hematite surface and negatively charged PFOA molecules was
evidently diminished by charge screening or adsorptive compe-
tition with the background ions (i.e. Na
þ
and Cl
). Hence,
although the ATR-FTIR data clearly show a ligand exchange
reaction for PFOA at the hematite surface, decreased adsorption
with increasing ionic strength suggests that a portion of PFOA
may be adsorbed by electrostatic interaction. The fact that a
distinct second n
as
(COO
) peak characteristic of outer-sphere
complexation was not observed at a lower wavenumber suggests
that both complexed and non-complexed carboxylate vibrat-
ional modes may be overlapping to give rise to the broad
1677 cm
1
band.
FTIR spectra of dissolved and hematite-adsorbed PFOS
The structure of PFOS is similar to that of PFOA, but with
sulfonate replacing carboxylate in the polar head group, along
with the addition of a single CF
2
group (Fig. 1). Consequently,
PFOS exhibits stronger acidity (pK
a
,1) and a lower CMC
(2.0 mM) than PFOA.
[5,7]
Spectra of PFOS at the Ge IRE (dis-
solved) and hematite surface (adsorbed) are shown in Fig. 4. The
two spectra are very similar except for the intensity difference.
Similar to PFOA, PFOS exhibits a much higher affinity for the
hematite surface as a result of electrostatic attraction and
increased interfacial area produced by the hematite film. The
peak positions of major sulfonate and CF
2
stretching bands
occur with consistent relative intensities and peak locations in
110100
0
5
10
15
20
25
30
pH 3.0
pH 4.5
pH 6.0
Absorbance
ν
as
(CF
2
)
Ionic stren
g
th (mM)
Fig. 3. Effect of ionic strength on perfluorooctanoic acid adsorption on
hematite at pH 3.0, 4.5 and 6.0 shown by the absorbance of the n
as
(CF
2
) band
centred at 1208 cm
1
.
Table 2. Positions (cm
21
) and mode assignments of major IR bands in attenuated total reflection Fourier-
transform infrared spectra of aqueous perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS)
(pH 6.0) on the Ge internal reflection element
n
as,
asymmetric stretch; n
s,
symmetric stretch; n
ax,
axial stretch
PFOA PFOS
Wavenumber (cm
1
) Mode assignment Wavenumber (cm
1
) Mode assignment
1667 n
as
(COO
) 1372 n
ax
(CF
2
)
1414 n
s
(COO
) 1329 n
ax
(CF
2
)
1363 n
ax
(CF
2
) 1283 n(CF
2
)
1319 n
ax
(CF
2
) 1267 n(CF
2
)
1238 n
as
(CF
2
) 1243 n
as
(CF
2
)þn
as
(R–SO
3
)
1204 n
as
(CF
2
)þn
as
(CF
3
) 1215 n
as
(R–SO
3
)
1149 n
s
(CF
2
) 1152 n
s
(CF
2
)
1102 n(C–C) 1066 n
s
(R–SO
3
)
1014 n(C–C) 1037 n(C–C)
PFOA and PFOS adsorption to hematite
151
the spectroscopic region 1300–1100 cm
1
. Precise assignment
of individual IR bands was obtained by peak fitting and
deconvolution. As shown in Fig. 5a, the set of bands determined
by peak deconvolution is consistent with those previously
reported in the literature.
[26,29,37,38]
The absorption bands at 1283,
1267, 1243 and 1202 cm
1
correspond to asymmetric and sym-
metric CF
2
stretching bands that are also present in the spectrum
of PFOA. The additional band at ,1215 cm
1
can be assigned to
n
as
(R–SO
3
).
[29,34]
Sulfonate groups are very sensitive to changes
in local coordination geometry. For example, a symmetry change
of sulphate and sulfonate groups may result in splits or shifts of
the asymmetric vibrations (n
3
).
[30,44]
The sulfonate group in the
PFOS structure has C
3v
symmetry, which is expected to exhibit
doublet peaks corresponding to a doubly degenerate band at a
higher wavenumber (,1240 cm
1
) and a non-degenerate band at
a lower wavenumber (,1210 cm
1
).
[29,34,35]
Thus, we attribute
the strong band centred at 1243 cm
1
to a combination of
n
as
(CF
2
)andn
as
(R–SO
3
). This assignment is also supported by
the fact that the band became the strongest band in the spectrum
whereas the other n
as
(CF
2
) band (i.e. 1204 cm
1
) is the strongest
band in the spectrum of PFOA in the absence of interference
from sulfonate stretching. Band assignments are summarised
in Table 2.
Peak fitting results of the hematite-adsorbed PFOS spectrum
are very similar to those obtained for the aqueous phase species
(Fig. 5b). The n
as
(R–SO
3
) manifests as a doublet corresponding
to the C
3v
symmetry, indicating that the symmetry of the
sulfonate group was not changed upon adsorption, and therefore,
no direct chemical bond formed between the sulfonate groups
and hematite surface. Nevertheless, we observed slight spectro-
scopic changes in the region of asymmetric sulfonate stretching.
Compared with the spectrum of aqueous PFOS, one of the
n
as
(R–SO
3
) bands shifted from 1215 to 1220 cm
1
, and the
intensity of this band was diminished relative to other bands in
the spectrum (Fig. 5). Similar changes were observed in a
previous study of sodium dodecyl sulfonate (SDS, a hydrocar-
bon anionic surfactant) adsorption to hematite.
[29]
The two
n
as
(OSO
3
) bands correspond to a doubly degenerate E vibration
and a second non-degenerate A vibration with the directions of
the transition dipole moment perpendicular to each other.
[29,35]
The A vibration at 1220 cm
1
is more sensitive to direct contact
with charged solid surfaces, whereas the E vibration mode at
1243 cm
1
is sensitive to lateral surfactant–surfactant interac-
tions.
[35]
Thus, the change in the A vibration mode of
n
as
(R–SO
3
) upon adsorption to hematite indicates a change in
the local environment of the sulfonate head, possibly attribut-
able to hydrogen-bond formation between the sulfonate group
and hematite surface hydroxy groups.
[29,34]
Effect of solution pH on PFOA and PFOS adsorption
Several recent ATR-FTIR spectroscopic studies of adsorption of
LMW carboxylic acids on various hydroxylated mineral sur-
faces (e.g. a-Fe
2
O
3
,Al
2
O
3
and TiO
2
) indicate that carboxylate
groups can form either inner-sphere or outer-sphere complexes
at the oxide–water interface, and that solution pH plays an
important role in the process.
[32,42,45]
The effect of pH on PFOA
adsorption was, therefore, examined in this study. Adsorption of
1400 1300 1200 1100 1000 900
Wavenumber (cm1)
Absorbance
@ Hematite
@ Ge IRE (intensity 5)
Fig. 4. Attenuated total reflection Fourier-transform infrared spectra of
250-mM perfluorooctanesulfonic acid adsorbed on Ge internal reflection
element (IRE) and hematite surface in 10 mM NaCl solution at pH 5.0. The
absorbance intensity of the spectrum at Ge IRE was increased five times.
1400 1300 1200 1100 1000
Absorbance
Wavenumber (cm1)
Wavenumber (cm1)
νas(CF2) νas(R–SO3
)
νas(CF2) νas(R–SO3
)
νas(R–SO3
)
νs(R–SO3
)
νs(R–SO3
)
νas(CF2)
νs(CF2)
νas(R–SO3
)
νas(CF2)
νs(CF2)
ν(CC)
ν(CC)
ν(CF2)
ν(CF2)
(a) @ Ge IRE
1283
1267
1243
1215
1202
1152
1066 1037
1372
νax(CF2)
1329
1400 1300 1200 1100 1000
Absorbance
(b) @ Hematite
1283
1266
1243
1220
1206
1153
1066 1037
1370 1329
νax(CF2)
Fig. 5. Peak deconvolution of the perfluorooctanesulfonic acid spectra
(250 mM) on (a) Ge internal reflection element (IRE) and (b) hematite in
10 mM NaCl solutions at pH 5.0. The solid line is the experimental data. The
red dash line represents the non-linear least square fit. The blue lines are the
deconvoluted peaks.
X. Gao and J. Chorover
152
PFOA to hematite, as measured by FTIR absorbance intensities,
exhibits strong pH dependence (Fig. 6). Adsorption increases
substantially with decreasing pH from 6.0 to 3.0 (no detectable
IR signals at pH $7.0 under the flow conditions), consistent
with prior macroscopic adsorption studies.
[21,25]
The stronger
adsorption at low pH can be partially attributed to favourable
electrostatic forces. At pH ,pH
pznpc
of hematite (7.7–7.8),
adsorption is certainly affected by electrostatic attractions
between the negatively charged carboxylate and positively
charged hematite surface functional groups. Adsorption is
negligible at pH .pH
pznpc
of hematite where, in the presence of
an indifferent background electrolyte, PFOA and the hematite
surface are both net negatively charged. Despite similarities
among the spectra across the pH range, small pH-dependent
changes in carboxylate stretching vibrations were observed. The
intensity ratio of n
s
(COO
)ton
as
(COO
) increases with
decreasing pH (Fig. 6, inset), suggesting that inner-sphere
metal–carboxylate complexes are favoured at low pH, similar to
what has been previously reported for hematite adsorption of
LMW carboxylic acids
[32]
and carboxylate groups of microbial
surface biomacromolecules.
[28]
Therefore, in addition to elec-
trostatic attraction, ligand exchange of carboxylate groups at
mineral surface hydroxy groups also contributes to enhanced
adsorption at low pH.
Similar to PFOA, the adsorption of PFOS to hematite also
shows strong pH dependence (Fig. 7). Adsorbate surface excess
increases with decreasing pH, consistent with prior batch sorp-
tion studies.
[21,22,25]
Other than increased intensity, no detect-
able changes were observed in the spectroscopic shape across
the pH range, indicating that electrostatic attraction is domi-
nantly responsible for the strong pH-dependence of adsorption.
Adsorption kinetics of PFOA and PFOS
Integrated absorbances of IR bands have been used previously to
assess surface excess of surfactants at solid surfaces.
[26,46]
The
penetration depth (d
p
) of the IR evanescent wave (0.16–1.6 mm
for the spectroscopic range of 4000–400 cm
1
with the 458Ge
IRE used in this study) may be greater than the layer of inter-
facial sorption and extend into the bulk solution. Thus, in
addition to adsorption at the solid–water interface, the measured
IR absorbance may also contain contributions from isotropic
bulk solution.
[26,46]
However, as discussed above, as the
absorbance intensities of the aqueous PFOA spectrum are ,1/10
of those for the hematite-adsorbed form, we assume that the
contributions from the isotropic bulk media are negligible in this
experiment. Therefore, the kinetics of PFOA adsorption to
hematite were assessed by directly monitoring the intensity
change of its fluorocarbon IR band – specifically the n
as
(CF
2
)
band centred at 1208 cm
1
that does not shift significantly with
solution chemistry and surface loading during the flow through
experiment. Adsorption was relatively rapid at all pH values,
reaching equilibrium within a few hours, but the rate and extent
of the adsorption increased significantly with decreasing pH
(Fig. 8). At pH 6.0, adsorption increased progressively with
reaction time and achieved equilibrium after 6 h. Adsorption at
1800 1600 1400 1200 1000
Wavenumber (cm1)
Absorbance
pH 3.0
pH 4.5
pH 6.0
1800 1700 1500 1400
Wavenumber (cm1)
νs(COO)
νas(COO)
pH 3.0
pH 4.5
pH 6.0
Fig. 6. Attenuated total reflection Fourier-transform infrared spectra of
500-mM perfluorooctanoic acid adsorbed to hematite surface as a function
of pH.
1400 1300 1200 1100 1000 900
Wavenumber (cm1)
Absorbance
250 µM PFOS @ hematite
pH 3.0
pH 4.5
pH 5.0
pH 6.0
Fig. 7. Attenuated total reflection Fourier-transform infrared spectra of
250-mM perfluorooctanesulfonic acid (PFOS) adsorbed to hematite surface
as a function of pH.
0 100 200 300 400
0.1
1
10
100
pH 3.0
pH 4.5
pH 6.0
Absorbance of νas(CF2)
Time (min)
456789
20
10
0
10 Fe
Fe
O
OH2
Fe
OH
20
30
40
Zeta potential (mv)
pH
Fig. 8. Adsorption kinetics of 500-mM perfluorooctanoic acid on hematite
as represented by the IR absorbance of n
as
(CF
2
) band at 1208 cm
1
. Fitting
results of the pseudo-first-order (pH 6.0) and pseudo-second-order models
(pH 4.5 and 3.0) are plotted as dash lines. The inset presents the zeta potential
of hematite (0.1 g L
1
) as a function of pH in 1 mM NaNO
3
solution.
PFOA and PFOS adsorption to hematite
153
pH 3.0 was much more rapid, reaching an adsorption maximum
within 30 min. The difference can be attributed to the pH-
dependent surface charge of hematite (Fig. 8, inset). At pH 3.0,
hematite surface hydroxy groups that can undergo protonation
are in a fully protonated state, giving rise to maximum positive
surface charge and, therefore, exhibit high affinity for anionic
surfactant molecules. The kinetics of ligand exchange are also
expected to be increased under these conditions, where inner-
sphere complexation of the PFOA carboxy group is promoted by
the dissociation of the water molecule from the Fe
III
surface
metal centre
[47]
:
Fe OHþ
2þCF3ðCF2Þ6COO!
Fe OOCðCF2Þ6CF3þH2Oð1Þ
Relative to its neutral (Fe–OH) or deprotonated forms
(Fe–O
), protonation of the hematite surficial hydroxy groups
reduces the activation energy for ligand exchange and, thereby
increases their substitutional lability.
[47]
Thus, a large concen-
tration of protonated sites at low pH evidently increases the
reaction rate of Eqn 1. With increasing pH, the hematite surface
becomes progressively deprotonated, resulting in a decrease in
adsorption kinetics and equilibrium extent. The slower second
step adsorption kinetics at high pH is consistent with a sorption
barrier generated by structural re-arrangement and surfactant
self-assembly of adsorbed PFOA at the solid–water inter-
face.
[26]
In addition, intra- and interparticle diffusion into the
porous hematite film may also contribute to the slow kinetics.
[48]
This observation contrasts with prior studies pertaining to
adsorption of hydrocarbon analogues of PFOA and PFOS.
Adsorption of hydrocarbon ionic surfactants to charged particles
is generally rapid, reaching sorption equilibrium within minutes
irrespective of solution pH.
[29,45]
One possible explanation for
this discrepancy is the size and hydrophobicity differences
between the two types of compounds. Substitution of H by F in
the aliphatic chain structure increases the molecular size and
hydrophobicity of PFCs relative to hydrocarbon analogues. This
may limit their diffusion in the hydrophilic sorbent film and
thereby contribute to the slow sorption kinetics.
Adsorption kinetic data were fit to pseudo-first-order (Eqn 2)
and second-order (Eqn 3) models to facilitate interpretation of
the rate controlling sorption processes.
[4850]
logðqeqtÞ¼log qek1
2:303 tð2Þ
t
qt
¼1
k2qe2þt
qe
ð3Þ
where q
t
is the adsorbed mass of PFCs at time t(represented by
the absorbance of the n
as
(CF
2
) band); q
e
is the equilibrium
adsorbed mass, and k
1
and k
2
are the first and second-order
adsorption rate coefficients. As shown in Fig. 8 and Table 3,
adsorption at pH 3.0 is well described by a pseudo-second-order
reaction equation with the correlation coefficient R
2
¼0.9998
and the calculated sorbate concentration at equilibrium (q
e,cal
)
close to the measured value (q
e,exp
). Adsorption at pH 4.5 is also
fairly well fit to pseudo-second-order kinetics (R
2
¼0.984).
However, the q
e,cal
is much larger than q
e,exp
in this case, sug-
gesting that adsorption equilibrium may not have been reached
during the experimental time frame. As shown in Table 3,
adsorption kinetics at pH 6.0 are better described by a pseudo-
first-order model (Table 3). However, this model still clearly
over-predicts adsorption at early times (,250 min) (Fig. 8). The
kinetics of PFOS adsorption to hematite exhibits similar trends
(Fig. 9). Increasing pH diminished both the PFOS uptake rate
and extent.
The concentration of surfactant in the bulk solution signifi-
cantly affects PFOA adsorption kinetics. As shown in Fig. 10,
increasing the initial PFOA concentration from 50 to 1000 mM
not only increased the extent of equilibrium uptake, but the
adsorption rate coefficient as well. Time to adsorption equilib-
rium decreased with increasing concentration. In the case
of adsorption to hematite, this observation can be explained
by the specific adsorption mechanisms. The adsorption of
aqueous organic acid anions to hydrous metal oxides often
involves a two-step process with rapid formation of an outer
sphere complex followed by slower ligand exchange reaction
with the elimination of H
2
O molecules.
[42,48]
Increasing the
Table 3. Fitting parameters of the pseudo-first-order and pseudo-second-order models for perfluorooctanoic acid (PFOA) adsorption on hematite
from 500 lM PFOA aqueous solution at different pH
q
e,exp
, experimental equilibrium sorption uptake as represented by IR absorbance (peak area) of n
as
(CF
2
); q
e,cal
, calculated equilibrium sorption uptake
Solution pH q
e,exp
Pseudo-first-order Pseudo-second-order
q
e,cal
k
1
(min
1
)R
2
q
e,cal,
k
2
(min
1
)R
2
3.0 29.337 – – – 29.851 0.0105 0.9998
4.5 7.150 – – – 8.489 0.0022 0.984
6.0 5.190 6.994 0.0092 0.932 7.530 0.0007 0.774
0 60 120 180 240 300
0
2
4
6
250 μM PFOS (pH 3.0)
250 μM PFOS (pH 5.0)
Absorbance of νas(CF2)
Time (min)
Fig. 9. Adsorption–desorption kinetics of perfluorooctanesulfonic acid
(PFOS) on hematite as represented by the IR absorbance of the n
as
(CF
2
)at
pH 3.0 and 5.0
X. Gao and J. Chorover
154
concentration of protonated reactant surface hydroxy groups
(Fe–OH
2
+
) evidently increases the reaction rate of ligand
exchange, resulting in an accelerated second step for high PFOA
aqueous concentration. It is noteworthy, however, that Qu
et al.
[48]
reported similar variation in adsorption kinetics of
PFOA onto powdered activated carbon in response to changing
initial PFOA aqueous concentration, which presumably did not
involve an inner-sphere metal complexation mechanism.
A small decrease in adsorption was observed at early reaction
times for a concentration of 1000 mM PFOA (Fig. 10). This
decrease is likely attributable to a surfactant-induced hematite
dispersion effect whereby PFOA at high concentration formed
complete bilayers on the hematite surface, reversing the charge
on the Fe oxide particles, which resulted in partial displacement
of the hematite film as a result of shear forces associated with
fluid flow. Desorption is relatively rapid at all concentrations.
When PFOA-free electrolyte solution (i.e. 10 mM NaCl) was
introduced to the flow through cell, the adsorbate mass
decreased quickly, and only a small sorbate mass remained after
60 min (Fig. 10).
Adsorption isotherm
The results from the PFOA adsorption experiment at pH 3.0 are
depicted in the form of an adsorption isotherm in Fig. 11 using
the equilibrium absorbance of the n
as
(CF
2
) band at 1208 cm
1
as
a measure of surface excess. The spectra are very similar across
the concentration range from 50–1000 mM, indicating that the
dominant adsorption mechanisms are largely independent of
surface loading. The log–log scale isotherm exhibits charac-
teristic features of ionic surfactant adsorption to charged mineral
surfaces, as reported previously in the literature.
[29,51]
The iso-
therm contains several distinct regions : a low surface excess
region (,100 mM), where the adsorption is dominated by
covalent chemical bonding and electrostatic forces; an inter-
mediate region with very steep slope (from 100 to ,250 mM)
where lateral surfactant–surfactant hydrophobic interactions are
postulated to predominate resulting in the formation of hemi-
micelle-like monolayer clusters on the hematite surface; and a
plateau region at concentrations $500 mM where the surface is
occupied by surfactant bilayers.
Conclusions
The adsorption mechanisms, kinetics and isotherms of PFOA
and PFOS from aqueous solution to the hematite surface were
examined using in situ flow through ATR-FTIR spectroscopy.
To our knowledge, this is the first spectroscopic study of PFC
adsorption to environmental surfaces. Our results suggest that
adsorption of both PFCs to the hydrophilic hematite surface
exhibits strong pH-dependence. The data also reveal different
sorption mechanisms controlling the adsorption of PFOA and
PFOS to hematite. In addition to electrostatic interactions,
PFOA forms inner-sphere complexes at Fe metal centres, and
such direct covalent bond formation may also contribute sorp-
tion energetics at low pH, whereas PFOS forms only outer-
sphere complexes at the mineral surface. The kinetics of the
adsorption of PFCs to hematite was generally rapid, reaching
equilibrium within a few hours. Both solution pH and surface
loading had profound effects on the adsorption kinetics. Faster
adsorption was observed at low pH and high surface loading.
Acknowledgements
Research support was provided by the Binational Agricultural Research and
Development (BARD) fund, Grant # IS-3822-06, the Water Research
Foundation (Award #4269) and the University of Arizona Water Sustain-
ability Program. The comments and views detailed herein may not be nec-
essarily reflecting the views of the Water Research Foundation, its officers,
directors, affiliates or agents. Analyses in the ALEC were supported by NSF
grant CBET-0722579.
References
[1] K. Prevedouros, L. T. Cousins, R. C. Buck, S. H. Korzeniowski,
Sources, fate and transport of perfluorocarboxylates. Environ. Sci.
Technol. 2006,40, 32. doi:10.1021/ES0512475
[2] C. D. Vecitis, Y. J. Wang, J. Cheng, H. Park, B. T. Mader, M. R.
Hoffmann, Sonochemical degradation of perfluorooctanesulfonate in
aqueous film-forming foams. Environ. Sci. Technol. 2010,44, 432.
doi:10.1021/ES902444R
[3] C. A. Moody, J. A. Field, Determination of perfluorocarboxylates in
groundwater impacted by fire-fighting activity. Environ. Sci. Technol.
1999,33, 2800. doi:10.1021/ES981355þ
0 60 120 180 240 300 360
0.1
1
10
100
50 μM
100 μM
250 μM
500 μM
1000 μM
Absorbance of νas(CF2)
Time (min)
Fig. 10. Adsorption–desorption kinetics of perfluorooctanoic acid (PFOA)
on hematite as represented by the IR absorbance of the n
as
(CF
2
) band at pH
3.0 as a function of PFOA concentration.
100 1000
5
10
15
20
25
30
35
40
Absorbance of ν
as
(CF
2
)
Equilibrium conc. of PFOA (
μ
M)
Fig. 11. Adsorption isotherm of perfluorooctanoic acid (PFOA) on hema-
tite at pH 3.0 in 10 mM NaCl solution as represented by the IR absorbance of
the n
as
(CF
2
) band (critical micelle concentration of PFOA ¼8.7–9.0 mM).
PFOA and PFOS adsorption to hematite
155
[4] H. P. H. Arp, C. Niederer, K. U. Goss, Predicting the partitioning
behavior of various highly fluorinated compounds. Environ. Sci.
Technol. 2006,40, 7298. doi:10.1021/ES060744Y
[5] E. Kissa, Fluorinated surfactants: Synthesis, properties, and applica-
tions 1994 (Marcel Dekker: New York).
[6] USEPA 2002. Revised draft hazard assessment of perfluorooctanoic
acid and its salts.Available at www.ewg.org/files/EPA_PFOA_110402.
pdf [Verified 4 April 2012].
[7] Q. Yuan, R. Ravikrishna, K. T. Valsaraj, Reusable adsorbents for
dilute solutionseparation.5: photodegradation of organic compounds
on surfactant-modified titania. Separ. Purif. Technol. 2001,24, 309.
doi:10.1016/S1383-5866(01)00136-8
[8] C. A. Moody, G. N. Hebert, S. H. Strauss, J. A. Field, Occurrence and
persistence of perfluorooctanesulfonate and other perfluorinated sur-
factants in groundwater at a fire-training area at Wurtsmith Air Force
Base, Michigan, USA. J. Environ. Monit. 2003,5, 341. doi:10.1039/
B212497A
[9] G. L. Kennedy, J. L. Butenhoff, G. W. Olsen, J. C. O’Connor, A. M.
Seacat, R. G. Perkins, L. B. Biegel, S. R. Murphy, D. G. Farrar, The
toxicology of perfluorooctanoate. Crit. Rev. Toxicol. 2004,34, 351.
doi:10.1080/10408440490464705
[10] C. Lau, K. Anitole, C. Hodes, D. Lai, A. Pfahles-Hutchens, J. Seed,
Perfluoroalkyl acids: a review of monitoring and toxicological find-
ings. Toxicol. Sci. 2007,99, 366. doi:10.1093/TOXSCI/KFM128
[11] M. Murakami, C. Morita, T. Morimoto, H. Takada, Source analysis of
perfluorocarboxylates in Tokyo Bay during dry weather and wet
weather using sewage marker. Environ. Chem. 2011,8, 355.
[12] D. Skutlarek, M. Exner, H. Farber, Perfluorinated surfactants in
surface and drinking water. Environ. Sci. Pollut. Res. 2006,13, 299.
doi:10.1065/ESPR2006.07.326
[13] M. Murakami, K. Kuroda, N. Sato, T. Fukushi, S. Takizawa,
H. Takada, Groundwater Pollution by Perfluorinated Surfactants in
Tokyo. Environ. Sci. Technol. 2009,43, 3480. doi:10.1021/
ES803556W
[14] G. B. Post, J. B. Louis, K. R. Cooper, B. J. Boros-Russo, R. L.
Lippincott, Occurrence and potential significance of perfluoroocta-
noic acid (PFOA) detected in New Jersey public drinking water
systems. Environ. Sci. Technol. 2009,43, 4547. doi:10.1021/
ES900301S
[15] O. Quin
˜ones, S. A. Snyder, Occurrence of perfluoroalkyl carboxylates
and sulfonates in drinking water utilities and related waters from the
United States. Environ. Sci. Technol. 2009,43, 9089. doi:10.1021/
ES9024707
[16] N. Yamashita, K. Kannan, S. Taniyasu, Y. Horii, T. Okazawa,
G. Petrick, T. Gamo, Analysis of perfluorinated acids at parts-
per-quadrillion levels in seawater using liquid chromatography-
tandem mass spectrometry. Environ. Sci. Technol. 2004,38, 5522.
doi:10.1021/ES0492541
[17] M. Sa´ez, P. de Voogt, J. R. Parsons, Persistence of perfluoroalkylated
substances in closed bottle tests with municipal sewage sludge.
Environ. Sci. Pollut. Res. 2008,15, 472. doi:10.1007/S11356-008-
0020-5
[18] M. H. Plumlee, J. Larabee, M. Reinhard, Perfluorochemicals in water
reuse. Chemosphere 2008,72, 1541. doi:10.1016/J.CHEMOSPHERE.
2008.04.057
[19] S. Rayne, K. Forest, Perfluoroalkyl sulfonic and carboxylic acids: a
critical review of physicochemical properties, levels and patterns in
waters and wastewaters, and treatment methods. J. Environ. Sci.
Health Part A 2009,44, 1145. doi:10.1080/10934520903139811
[20] R. Zhang, P. Somasundaran, Advances in adsorption of surfactants
and their mixtures at solid/solution interfaces. Adv. Colloid Interface
Sci. 2006,123–126, 213. doi:10.1016/J.CIS.2006.07.004
[21] C. P. Higgins, R. G. Luthy, Sorption of perfluorinated surfactants on
sediments. Environ. Sci. Technol. 2006,40, 7251. doi:10.1021/
ES061000N
[22] R. L. Johnson, A. J. Anschutz, J. M. Smolen, M. F. Simcik, R. L. Penn,
The adsorption of perfluorooctane sulfonate onto sand, clay, and iron
oxide surfaces. J. Chem. Eng. Data 2007,52, 1165. doi:10.1021/
JE060285G
[23] V. Ochoa-Herrera, R. Sierra-Alvarez, Removal of perfluorinated sur-
factantsby sorption onto granular activated carbon, zeolite and sludge.
Chemosphere 2008,72, 1588. doi:10.1016/J.CHEMOSPHERE.2008.
04.029
[24] C. You, C. Jia, G. Pan, Effect of salinity and sediment characteristics
on the sorption and desorption of perfluorooctane sulfonate at
sediment-water interface. Environ. Pollut. 2010,158, 1343.
doi:10.1016/J.ENVPOL.2010.01.009
[25] C. Y. Tang, Q. S. Fu, D. Gao, C. S. Criddle, J. O. Leckie, Effect of
solution chemistry on the adsorption of perfluorooctane sulfonate onto
mineral surfaces. Water Res. 2010,44, 2654. doi:10.1016/
J.WATRES.2010.01.038
[26] R. Xing, S. E. Rankin, Three-stage multilayer formation kinetics
during adsorption of an anionic fluorinated surfactant onto ger-
manium. 1. Concentration effect. J. Phys. Chem. B 2006,110, 295.
doi:10.1021/JP0549069
[27] S. J. Parikh, B. J. Lafferty, D. L. Sparks, An ATR-FTIR spectroscopic
approach for measuring rapid kinetics at the mineral/water interface.
J. Colloid Interface Sci. 2008,320, 177. doi:10.1016/J.JCIS.2007.
12.017
[28] X. D. Gao, D. W. Metge, C. Ray, R. W. Harvey, J. Chorover, Surface
complexation of carboxylate adheres Cryptosporidium parvum
o¨ ocysts to the hematite-water interface. Environ. Sci. Technol. 2009,
43, 7423. doi:10.1021/ES901346Z
[29] X. D. Gao, J. Chorover, Infrared spectroscopic study of sodium
dodecylsulfate (SDS) adsorption to hematite nano-particles in aque-
ous systems. J. Colloid Interface Sci. 2010,348, 167. doi:10.1016/
J.JCIS.2010.04.011
[30] S. J. Hug, In situ Fourier transform infrared measurements of sulfate
adsorption on hematite in aqueous solutions. J. Colloid Interface Sci.
1997,188, 415. doi:10.1006/JCIS.1996.4755
[31] S. Goldberg, C. T. Johnston, Mechanisms of arsenic adsorption on
amorphous oxides evaluated using macroscopic measurements, vibra-
tional spectroscopy, and surface complexation modeling. J. Colloid
Interface Sci. 2001,234, 204. doi:10.1006/JCIS.2000.7295
[32] Y. S. Hwang, J. Liu, J. J. Lenhart, C. M. Hadad, Surface complexes of
phthalic acid at the hematite/water interface. J. Colloid Interface Sci.
2007,307, 124. doi:10.1016/J.JCIS.2006.11.020
[33] A. Omoike, J. Chorover, Adsorption to goethite of extracellular
polymeric substances from Bacillus subtilis.Geochim. Cosmochim.
Acta 2006,70, 827. doi:10.1016/J.GCA.2005.10.012
[34] K. D. Dobson, A. D. Roddick-Lanzilotta, A. J. McQuillan, An in situ
infrared spectroscopic investigation of adsorption of sodium dode-
cylsulfate and of cetyltrimethylammonium bromide surfactants to
TiO
2
, ZrO
2
,Al
2
O
3
, and Ta
2
O
5
particle films from aqueous solutions.
Vib. Spectrosc. 2000,24, 287. doi:10.1016/S0924-2031(00)00096-5
[35] H. Y. Li, C. P. Tripp, Use of infrared bands of the surfactant headgroup
to identify mixed surfactant structures adsorbed on Titania. J. Phys.
Chem. B 2004,108, 18318. doi:10.1021/JP046741U
[36] U. Schwertmann, R. M. Cornell, Iron Oxides in the Laboratory:
Preparation and Characterization 1991 (Wiley-VCH: Weinheim,
Germany).
[37] C. A. Alves, M. D. Porter, Atomic force microscopic characterization
of fluorinated alkanethiolate monolayer at gold and correlations
electrochemical and infrared reflection spectroscopic structural
descriptions. Langmuir 1993,9, 3507. doi:10.1021/LA00036A027
[38] T. J. Lenk, V. M. Hallmark, C. L. Hoffmann, J. F. Rabolt, D. G.
Castner, C. Erdelen, H. Ringsdorf, Structural investigation of molec-
ular organization in self-assembled monolayers of a semifluorinated
amidethiol. Langmuir 1994,10, 4610. doi:10.1021/LA00024A037
[39] J. L. Lo´ pez-Fonta´n, F. Sarmiento, P. C. Schulz, The aggregation of
sodium perfluorooctanoate in water. Colloid Polym. Sci. 2005,283,
862. doi:10.1007/S00396-004-1228-7
[40] N. O. Brace, Long chain alkanoic and alkenoic acids with perfluor-
oalkyl terminal segments. J. Org. Chem. 1962,27, 4491. doi:10.1021/
JO01059A090
[41] J. Y. Ha, T. H. Yoon, Y. G. Wang, C. B. Musgrave, G. E. Brown,
Adsorption of organic matter at mineral/water interfaces: 7. ATR-
FTIR and quantum chemical study of lactate interactions with
X. Gao and J. Chorover
156
hematite nanoparticles. Langmuir 2008,24, 6683. doi:10.1021/
LA800122V
[42] T. H. Yoon, S. B. Johnson, C. B. Musgrave, G. E. Brown, Adsorption
of organic matter at mineral/water interfaces: I. ATR-FTIR spectro-
scopic and quantum chemical study of oxalate adsorbed at boehmite/
water and corundum/water interfaces. Geochim. Cosmochim. Acta
2004,68, 4505. doi:10.1016/J.GCA.2004.04.025
[43] D. L. Sparks, Sorption, in Encyclopedia of Soils in the Environment
(Ed. D. Hillel) 2004, pp. 532 (Academic Press).
[44] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordi-
nation Compounds 1986 (Wiley: New York).
[45] R. Atkin, V. S. J. Craig, S. Biggs, Adsorption kinetics and structural
arrangements of cationic surfactants on silica surfaces. Langmuir
2000,16, 9374. doi:10.1021/LA0001272
[46] D. J. Neivandt, M. L. Gee, C. P. Tripp, M. L. Hair, Coadsorption
of poly(styrenesulfonate) and cetyltrimethylammonium bromide on
silica investigated by attenuated total reflection techniques. Langmuir
1997,13, 2519. doi:10.1021/LA962047X
[47] Z. Wang, C. C. Ainsworth, D. M. Friedrich, P. L. Gassman, A. G. Joly,
Kinetics and mechanism of surface reaction of salicylate on alumina in
colloidal aqueous suspension. Geochim. Cosmochim. Acta 2000,64,
1159. doi:10.1016/S0016-7037(99)00360-9
[48] Y. Qu, C. J. Zhang, F. Li, X. W. Bo, G. F. Liu, Q. Zhou, Equilibrium
and kinetics study on the adsorption of perfluorooctanoic acid from
aqueous solution onto powdered activated carbon. J. Hazard. Mater.
2009,169, 146. doi:10.1016/J.JHAZMAT.2009.03.063
[49] Q. Yu, R. Q. Zhang, S. B. Deng, J. Huang, G. Yu, Sorption of
perfluorooctane sulfonate and perfluorooctanoate on activated car-
bons and resin: kinetic and isotherm study. Water Res. 2009,43, 1150.
doi:10.1016/J.WATRES.2008.12.001
[50] J. Chorover, M. L. Brusseau, Kinetics of sorption-desorption, in
Kinetics of Water-Rock Interaction (Eds S. L. Brantley, J. D. Kubicki,
A. F. White) 2008, pp. 109–149 (Springer: New York).
[51] M. R. Bohmer, L. K. Koopal, Adsorption of ionic surfactants on
variable-charge surfaces. 1. Charge effects and structure of the
adsorbed layer. Langmuir 1992,8, 2649. doi:10.1021/LA00047A014
PFOA and PFOS adsorption to hematite
157
