sorptive and desorptive fractionation of dissolved organic...
TRANSCRIPT
TECHNICAL REPORTS
526
Interactions of dissolved organic matter (DOM) with soil minerals, such as metal oxides and clays, involve various sorption mechanisms and may lead to sorptive fractionation of certain organic moieties. While sorption of DOM to soil minerals typically involves a degree of irreversibility, it is unclear which structural components of DOM correspond to the irreversibly bound fraction and which factors may be considered determinants. To assist in elucidating that, the current study aimed at investigating fractionation of DOM during sorption and desorption processes in soil. Batch DOM sorption and desorption experiments were conducted with organic matter poor, alkaline soils. Fourier-transform infrared (FTIR) and UV-Vis spectroscopy were used to analyze bulk DOM, sorbed DOM, and desorbed DOM fractions. Sorptive fractionation resulted mainly from the preferential uptake of aromatic, carboxylic, and phenolic moieties of DOM. Soil metal-oxide content positively aff ected DOM sorption and binding of some specifi c carboxylate and phenolate functional groups. Desorptive fractionation of DOM was expressed by the irreversible-binding nature of some carboxylic moieties, whereas other bound carboxylic moieties were readily desorbed. Inner-sphere, as opposed to outer-sphere, ligand-exchange complexation mechanisms may be responsible for these irreversible, as opposed to reversible, interactions, respectively. Th e interaction of aliphatic DOM constituents with soil, presumably through weak van der Waals forces, was minor and increased with increasing proportion of clay minerals in the soil. Revealing the nature of DOM-fractionation processes is of great importance to understanding carbon stabilization mechanisms in soils, as well as the overall fate of contaminants that might be associated with DOM.
Sorptive and Desorptive Fractionation of Dissolved Organic Matter by Mineral Soil Matrices
Adi Oren and Benny Chefetz*
Interactions of dissolved organic matter
(DOM) with mineral constituents infl uence the stabiliza-
tion of organic matter in soil (Torn et al., 1997; Kaiser
and Guggenberger, 2000; Kleber et al., 2005; Rumpel and
Kögel-Knabner, 2011; Sanaullah et al., 2011). Input sources
of DOM can be natural (e.g., litter leachate and its decom-
position products, as well as root and microbial exudates) or
also exogenous, in the case of agricultural systems (e.g., soluble
compounds released from composted materials or biosolids, or
arriving in the dissolved state with treated-wastewater irriga-
tion). Since DOM is characteristically a heterogeneous mix-
ture of chemical groups in various physical sizes and spatial
conformations (Chefetz et al., 1998a), its mineral-surface
reactivity is competitive among its subcomponents (Gu et al.,
1996). Th is leads to molecular-based fractionation of DOM
upon sorption (Navon et al., 2011). Reports on the preferential
sorption of hydrophobic (Kaiser et al., 1996; Kaiser and Zech,
1997; Guo and Chorover, 2003), large (Ochs et al., 1994; Gu
et al., 1995; Zhou et al., 2001; Hur and Schlautman, 2003),
and aromatic (McKnight et al., 1992; Polubesova et al., 2008)
DOM fractions, as well as of carboxylic functional groups (Gu
et al., 1994; Kleber et al., 2007), are common from experi-
mental systems using model soil minerals. Th e involvement of
carboxylic functional groups in ligand-exchange complexation
reactions with hydroxylated Fe- and Al-oxide surfaces is con-
sidered a major interaction pathway by several groups (Parfi tt
et al., 1977; Tipping, 1981; McKnight et al., 1992; Gu et al.,
1994; Kaiser et al., 1997; Chorover and Amistadi, 2001; Kaiser
and Guggenberger, 2007) and is linked with high-energy
bonding of DOM to mineral surfaces (Afi f et al., 1995; Kaiser
and Guggenberger, 2000; Kleber et al., 2005). It is only under
alkaline conditions that phenolic structures (having high pKa
values) can participate in similar complexation reactions (Gu
et al., 1995).
Th e trivalent cations Fe3+ and Al3+ are also involved in the
binding of DOM moieties that are able to donate electrons to
them (Zhu et al., 2004), such as aromatic structures. Together
with ligand-exchange complexation, such interactions may
render metal-oxide surfaces catalysts for humifi cation processes
(Arfaioli et al., 1997). For example, Polubesova et al. (2010)
demonstrated the role of surface Fe3+ in the oxidative trans-
formation of various organic acids, resulting in the formation
Abbreviations: DOM, dissolved organic matter; FTIR, Fourier-transform infrared;
SUVA, specifi c UV absorbance; TOC, total organic carbon.
Dep. of Soil and Water Sciences, Faculty of Agriculture, Food and Environment, The
Hebrew Univ. of Jerusalem. Assigned to Associate Editor Myrna Simpson.
Copyright © 2012 by the American Society of Agronomy, Crop Science Society
of America, and Soil Science Society of America. All rights reserved. No part of
this periodical may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information
storage and retrieval system, without permission in writing from the publisher.
J. Environ. Qual. 41:526–533 (2012)
doi:10.2134/jeq2011.0362
Posted online 3 Feb. 2012.
Received 29 Sept. 2011.
*Corresponding author ([email protected]).
© ASA, CSSA, SSSA
5585 Guilford Rd., Madison, WI 53711 USA
Journal of Environmental QualityORGANIC COMPOUNDS IN THE ENVIRONMENT
TECHNICAL REPORTS
www.agronomy.org • www.crops.org • www.soils.org 527
and polymerization of free radicals. It appears, therefore, that
the involvement of polyvalent metal cations in such energetic
interactions may largely account for the irreversible sorption
of DOM to soil/oxides. In fact, the limited number of studies
examining the desorption of DOM following its sorptive inter-
action with soil or model minerals have reported high levels of
sorption irreversibility (Kaiser and Zech, 2000; Kahle et al.,
2003; Joo et al., 2008).
While it is clear that the interaction of DOM with soil
minerals involves a degree of irreversibility, it is still unclear
which structural components of DOM correspond to the irre-
versibly bound fraction and which factors may be considered
determinants. Revealing the nature of sorptive and desorptive
DOM fractionation is of special consequence for the fate of
contaminants and other organic and inorganic species in soil,
since these will diff erentially interact with DOM fractions in
the sorbed and mobile phases. Th erefore, the main objective of
the current work was to study molecular-level physicochemical
DOM fractionation processes that might be occurring during
both soil sorption and desorption of DOM.
Materials and MethodsSoils and Dissolved Organic Matter
Soil cores (90–120 cm) were obtained from three sites in
Israel located along the Mediterranean Coast, in areas that
are traditionally used for intensive agriculture. Th e sites,
from north to south, are Akko (33°07′37″N, 35°44′17″E),
Basra (32°25′59″N, 35°32′56″E), and Nir-Oz (31°27′25″N, 35°02′28″E). Th e soil samples were air dried and sieved (<2
mm). Fundamental chemical and physical analyses of the soils
were performed, according to Sparks (1996) and Dane and
Topp (2002), respectively. Soil pH was determined in equili-
brated soil aqueous extract (1:10 v/v) using double-distilled
water. Total organic carbon (TOC) content was determined by
oxidation with 1 N potassium dichromate in acidic medium,
according to Rowell (1994). Metal elements were extracted
from both amorphous and crystalline oxides and hydroxides
by the method of Mehra and Jackson (1958), using dithionite–
citrate–bicarbonate reagent. Quantifi cation of Fe, Al, and, Mn
was performed using ICP/AES (Acros-EOP, Spectro).
Th e DOM was extracted from mature composted biosolids,
obtained from a commercial compost factory (Dlila, Israel), by
mixing a homogeneous compost sample with distilled water
(1:10 v/v) and shaking (200 rpm) for 12 h. Th e DOM solution
was isolated by centrifugation (10,000 g for 20 min) and sub-
sequent pressure fi ltration through a 0.45-μm membrane fi lter.
Th e DOM solution exhibited pH, electrical conductivity, and
TOC content (measured with a VCPH model TOC analyzer,
Shimadzu Scientifi c Instruments) of 9.9, 2.3 dS m−1, and 300
mg L−1, respectively. Th e isolated DOM was subsequently freeze
dried. Th e DOM was analyzed for content of acidic functional
groups, following the procedure of Inbar et al. (1989).
Dissolved Organic Matter–Soil Sorption
and Desorption ExperimentsSorption–desorption processes were investigated in batch
experiments. Aliquots (20 mL) of DOM solution were mixed
with 2-g portions of soil in 40-mL centrifuge tubes. Seven
DOM concentrations were implemented (from ~3 to ~75 mg C
L−1), corresponding to the wide range of environmental soil con-
ditions. Experiments were performed in duplicate. Th e DOM
solutions were prepared by dissolving freeze-dried DOM mate-
rial in a background solution containing 3.7 mM CaCl2 (for
maintaining a constant ionic strength in all experiments) and 1.5
mM NaN3 as a bacteriostatic agent (Chefetz et al., 2006). Th e
solutions were passed through a 0.45-μm membrane fi lter before
their introduction to the soils. An equilibration time of 4 d was
determined in preliminary kinetic experiments. Th e experiments
were performed at 25°C. Th e possibilities of DOM-vessel inter-
action and/or microbial degradation of DOM were excluded by
verifying that TOC of a DOM solution run without added soil
did not change during a similar equilibration period. In addi-
tion, an organic carbon mass balance evaluation was performed
for an equilibrated DOM–soil sorption system. Th e mass bal-
ance (considering solution and solid phase measurements)
accounted for >90% of the organic carbon in the system. At the
end of the sorption-equilibration period, the tubes were centri-
fuged and a 10-mL aliquot of the supernatant was drawn, fi l-
tered (0.45 μm), and analyzed for TOC content. A fresh 10-mL
aliquot of DOM-free background solution was then added and a
4-d desorption phase begun. Four similar sequential desorption
steps were performed.
Dissolved Organic Matter CharacterizationTo characterize the fractionation processes undergone by
DOM in the course of sorptive and desorptive interactions in the
soils, a sorption–desorption procedure identical to that described
above was repeated with a single initial DOM concentration
(~75 mg C L−1). Th e bulk (i.e., initial) DOM, nonbound DOM,
and desorbed DOM fractions were analyzed by UV-Vis and
Fourier-transform infrared (FTIR) spectroscopy. For the UV-Vis
analysis, DOM solutions were diluted to <30 mg C L−1 to keep
them in the linear concentration range of the absorption values.
Absorption was measured over a wavelength range of 200 to 800
nm with a Spectrophotometer Evolution 300 model instrument
(Th ermo Scientifi c). Th e specifi c UV absorbance value at 280 nm
(SUVA280
), which serves as an index of sample aromaticity, was
calculated as the ratio between absorption intensity at 280 nm and
the TOC concentration of the sample. Th e ratio E2/E
3 serves as an
index for the relative molecular size of a material and was calcu-
lated based on the respective absorption values at 250 and 365 nm
(Polubesova et al., 2008). Pellets for FTIR spectroscopic analysis
of the DOM fractions were assembled from 1.5-mg aliquots of
dried DOM samples and 98.5 mg of IR-grade KBr. Th e FTIR
spectra were acquired with a Nicolet 6700 spectrometer (Th ermo
Scientifi c) over a wavenumber range of 400 to 4000 cm−1, with
64 scans per spectrum at a resolution of 4 cm−1. Baseline correc-
tion was performed with Omnic software. Subtraction of the non-
bound DOM spectrum from the bulk DOM spectrum provided
an assessment of the bound DOM fraction.
Results and DiscussionSorption–Desorption Isotherms
Th e DOM sorption isotherms obtained for the three studied
soils (Fig. 1) illustrate that the content of metal-oxide miner-
als is an important factor in determining DOM sorption. Th e
528 Journal of Environmental Quality • Volume 41 • March–April 2012
clay- as well as metal-oxide-rich Akko soil (Table 1) sorbed sig-
nifi cantly larger amounts of DOM than the other two soils at
any given equilibrium concentration. Th is was expressed by the
twofold greater sorption capacity value (Langmuir Model) cal-
culated for the Akko soil compared with the Basra and Nir-Oz
soils (996, 531, and 479 mg C kg−1, respectively). Greater
sorption to the Basra soil than to the Nir-Oz soil, despite the
55% smaller specifi c surface area in the former soil, provided
further support for the important role of metal-oxide miner-
als in the sorption of DOM (Kaiser et al., 1996). Th e DOM
sorption affi nity values are expressed as distribution coeffi cients
(Kd) extracted from the Freundlich equation for an equilibrium
DOM concentration of 40 mg C L−1. Th e Kd values were 6.82,
4.69, and 2.99 L kg−1 with the Akko, Basra, and Nir-Oz soils,
respectively, and followed the trend obtained for sorption
capacity values. Th e Kd values were close to those reported by
Navon et al. (2011) for sorption of DOM to a vertisol and its
isolated clay fraction, and by Ling et al. (2006) for the sorption
of DOM to 10 diff erent soils.
Dissolved organic matter desorption isotherms revealed
strong sorption–desorption hysteresis with all three soils (Fig. 1).
It is important to note that up to 83% of sorbed dissolved organic
carbon was retained in the soil after four sequential desorption
steps. Th e sorption–desorption hysteresis levels recorded in the
current study were comparable to those reported by Kahle et al.
(2003) and were mostly in the range of 60 to 80%.
Bulk Dissolved Organic Matter PropertiesTh e FTIR spectrum of the bulk DOM (Fig. 2, top) is indic-
ative of a high aromatic-to-aliphatic ratio. Th is was mostly
apparent from the very low absorbance in the 2930 to 2970
cm−1 region of the spectrum (sp3 C-H stretches), which is typi-
cally very high in spectra of compost DOM (Chefetz et al.,
1998a, 1998b). Although peaks exclusively attributable to par-
affi nic C-H deformations were observed (1420 and 1435 cm−1;
Chefetz et al., 1998a), they were limited in intensity. In con-
trast, considerable aromaticity of DOM was demonstrated by
major peaks at the 1450 to 1496 cm−1 range (a strongly absorb-
ing region in manure composts that corresponds to aromatic
C = C stretches; Niemeyer et al., 1992), at 1111 to 1172 cm−1 (a strongly absorbing region in hydrophobic fractions of waste-
water DOM corresponding to aromatic C-H vibrations; Ilani
Fig. 1. Dissolved organic matter (DOM) sorption and desorption iso-therms for the Akko (top), Basra (middle), and Nir-Oz (bottom) soils.
Table 1. Major soil characteristics.
Akko Basra Nir-Oz
Clay (%) 67.5 ± 1.4 20.0 ± 0.1 17.5 ± 0.7
Silt (%) 15.0 ± 0.3 5.0 ± 0.1 7.5 ± 0.2
Sand (%) 17.5 ± 0.3 75.0 ± 0.9 75.0 ± 1.1
Texture Clay Sandy loam Sandy loam
TOC† (%) 0.145 ± 0.02 0.078 ± 0.01 0.081 ± 0.01
CEC† (meq 100 g−1) 30.8 ± 3.1 3.2 ± 0.7 1.5 ± 0.5
SSA† (m2 g−1) 327 ± 2 61 ± 1 135 ± 1
Fed‡ (g kg−1) 6.54 ± 0.23 4.46 ± 0.27 1.90 ± 0.07
Ald‡ (g kg−1) 1.44 ± 0.22 0.83 ± 0.09 0.64 ± 0.19
Mnd‡ (g kg−1) 0.90 ± 0.05 0.13 ± 0.01 0.08 ± 0.00
CaCO3 (%) 3.5 ± 0.2 0.0 ± 0.0 8.4 ± 0.3
pH§ 8.5 ± 0.1 7.8 ± 0.0 8.9 ± 0.1
† TOC = total organic carbon; CEC = cation exchange capacity; SSA = specifi c surface area.
‡ Extracted with dithionite–citrate–bicarbonate reagent.
§ Obtained in soil aqueous extract (1:10 with distilled water).
www.agronomy.org • www.crops.org • www.soils.org 529
et al., 2005), and at 817 to 879 cm−1 (typical of humic acids
and corresponding to aromatic out-of-plane C-H bending;
Niemeyer et al., 1992).
An especially high abundance of carboxylic functional
groups was evidenced by a peak at wavenumber 1390 cm−1,
which was exceptionally elevated relative to the rest of the
spectrum. Additional indications were provided for a signifi -
cant contribution of carboxyl functionalities (1635 and 1650
cm−1) and phenol groups (1111, 1343, and 1374 cm−1) to the
DOM spectrum. Th e high pH (~9.9) of DOM used in this
study, which was far above the typical pKa range of carbox-
ylic groups, was undoubtedly responsible for the prevalence of
peaks assigned to carboxylate groups (1390, 1635, and 1650
cm−1) and the lack of undissociated forms of carboxyls (1740
cm−1, Gu et al., 1995; 1725 cm−1, Eusterhues et al., 2011).
Carboxylic acidity of the studied DOM (16.6 meq g C−1) was
in the high range of values reported for compost DOM (Hunt
et al., 2007). Th e studied DOM sample also exhibited an
elevated phenolic acidity value (14.5 meq g C−1), which
was corroborated with a substantial SUVA280
value (Table
2). Th e high carboxylic and phenolic content of the DOM
might be explained by the high level of maturity of the
compost used in the current study (Chefetz et al., 1998b).
Furthermore, it may suggest the presence of hydroxyben-
zoic structures in the DOM (Guan et al., 2006a).
Dissolved Organic Matter Fractionation
during SorptionTh e FTIR spectra of the bulk DOM, nonbound, and
bound fractions of DOM in the course of its interaction
with the soil are presented in Fig. 2–4. Th e changes in DOM
during the interaction resulted in fairly similar sorbed frac-
tions with the three soils. As noted from Fig. 5, the peak
at 1390 cm−1, which dominated the spectrum of the bulk
DOM, became even more pronounced in the spectra of the
bound fractions. Th e role of carboxylic groups in the bind-
ing of DOM to model soil minerals has been repeatedly
demonstrated using FTIR spectroscopy. Notably, the min-
eral binding of DOM constituents designated band wave-
numbers of ?1390 cm−1 (±10) was reported in most of the
studies and was explicitly interpreted as a carboxylate group
complexed with metal oxide (Biber and Stumm, 1994;
Chorover and Amistadi, 2001; Eusterhues et al., 2011;
Guan et al., 2006a; Kaiser and Guggenberger 2007; Parfi tt
et al., 1977; Sharma et al., 2010; Tejedor-Tejedor et al.,
1992; Yost et al., 1990). Th e carboxylate group designated
by the peak at 1635 cm−1 exhibited only limited sorption
and prevailed mainly in the nonbound fractions. Th e shoulder at
1650 cm−1 in the bulk DOM spectrum changed to a signifi cant
peak in the sorbed fraction from the Akko soil. In contrast, this
peak was rather small in the Basra soil and approached a pla-
teau in the Nir-Oz soil. Th is band was interpreted by Chorover
and Amistadi (2001) as an asymmetric carboxylate stretching
involved in the formation of a stable Fe-carboxylate complex
on a goethite surface. In contrast to the interaction with metal
oxides, those authors argued that the interaction of this carbox-
ylate group with montmorillonite occurs via a weaker associa-
tion between carboxylate and hydrated surface cations, i.e., via
a water- or cation-bridging mechanism on the layer silicate. Th e
diff erential pathway of carboxylate interaction with metal oxides
versus clay minerals may be supported by the trend recorded for
this specifi c moiety in the current study (Akko > Basra > Nir-
Oz), which was positively correlated with the relative contribu-
tions of metal oxides to the mineral compositions of these soils.
According to Kaiser et al. (1997), a DOM band at 1625 cm−1
(1620 cm−1 in our spectra) when accompanied by a shoulder
at 1580 cm−1 (or 1573 and 1589 cm−1 herein) or 1510 cm−1
(1512 cm−1 herein) may result from C = C vibration of aro-
matic structures, rather than from a carboxylic group. Biber and
Stumm (1994) similarly attributed a peak at 1620 cm−1 in the
spectrum of salicylic acid to aromatic ring C = C vibration. Kung
and McBride (1991) attributed major bands at 1582, 1474, and
1441 cm−1 (a highly absorbing region in the current spectra) to
aromatic ring C = C stretching vibrations and demonstrated
their involvement in the sorption of chlorinated phenols on
Fig. 2. Fourier-transform infrared spectra of the dissolved organic matter (DOM) at the various phases of its interaction with the Akko soil.
Table 2. Specifi c UV absorbance value at 280 nm (SUVA280
) and E2/E
3
indices calculated for the dissolved organic matter (DOM) fractions produced during the various phases of DOM interaction with the soils.
SUVA280
(L mgC−1 m−1) E2/E
3
Bulk DOM 2.94 ± 0.02 4.69 ± 0.0047
Akko nonbound 2.29 ± 0.02 5.76 ± 0.0255
Basra nonbound 2.51 ± 0.03 5.59 ± 0.0024
Nir-Oz nonbound 2.76 ± 0.03 5.23 ± 0.0034
Akko desorbed 3.07 ± 0.05 4.89 ± 0.0053
Basra desorbed 2.80 ± 0.05 4.75 ± 0.0090
Nir-Oz desorbed 3.27 ± 0.02 4.41 ± 0.0749
530 Journal of Environmental Quality • Volume 41 • March–April 2012
various types of Fe- and Al-oxides. In addition, the latter
authors demonstrated that peaks at 1378, 1338, and 1198
cm−1 (O-H bending and deformation vibrations) accom-
pany the spectra of the sorbed phenols. Analogous peaks
in the current spectra of the sorbed DOM fractions were
noted at 1374 and 1343 cm−1, whereas a peak at 1111 cm−1
might correspond to a phenolic structure similar to those
demonstrated by Gu et al. (1995) as dominant participants
in DOM complexation to hematite. Interestingly, this latter
peak’s intensity was greatest in the Akko soil spectrum,
followed by Basra soil. It was practically absent from the
Nir-Oz spectrum. Th is follows the order of the metal-oxide
content. It also matches the trend recorded in the SUVA280
measurements (Table 2) for the preferential uptake of aro-
matic moieties. All of the above imply that the contribution
of phenolic groups to DOM sorption in the currently stud-
ied soil systems could certainly be important and might be
associated with interactions with metal oxides.
Th e alkaline conditions in which the current investiga-
tion was conducted are expected to encourage participation
of phenolic groups in the sorptive interaction of DOM.
Due to the signifi cantly elevated pKa range of phenolic
groups (typically >9) relative to carboxylic groups (1.4–4.5),
the main obstacle for phenol surface reactivity (i.e., disso-
ciation) diminishes as pH increases. Under conditions that
allow dissociation, the coordination step is more favored by
the Lewis basicity of the phenolic oxygen than by that of
the carboxylic oxygen (Guan et al., 2006a). Furthermore,
several studies have demonstrated that the sorption of ben-
zoic acid derivatives to metal oxides is enhanced by the pres-
ence of additional phenolic groups (Evanko and Dzombak
1998; Tejedor-Tejedor et al., 1992; Yost et al., 1990). Th e
electron-donating resonance eff ect of phenol groups can
increase the electron density within an adjacent carboxyl
group, thereby favoring the metal-carboxylate complex-
ation. An alternative explanation might be the simultane-
ous binding of a carboxylic oxygen and its respective ortho
phenolic oxygen to a single metal atom (an inner-sphere
bidentate mononuclear complex), or an especially energetic
association of two adjacent phenolic groups ortho posi-
tioned to each other in bidentate complexation (Guan et
al., 2006a). Th ese fi ndings suggest that the high carboxylic
and phenolic contents of the studied DOM, along with the
high pH conditions of both the DOM and soils, are the
main factors underlying the high levels of sorption irrevers-
ibility recorded with this DOM (Fig. 1).
Polysaccharide moieties exhibited limited sorption
(1065 and 1095 cm−1 bands corresponding to C-O
stretches; Niemeyer et al., 1992). Similarly, Kaiser et al.
(1997) observed only moderate preferential uptake of
carbohydrates during interaction of DOM with various
types of metal oxides. In contrast, Eusterhues et al. (2011),
studying the sorption of forest-fl oor-extracted DOM and
lignin on ferrihydrite, observed sorbed fractions that were
rich in carbohydrate peaks. However, the spectra of the
sorbed fractions in that study were not much diff erent
from that of the original material, preventing a clear evalu-
ation of sorptive fractionation.
Fig. 3. Fourier-transform infrared spectra of the dissolved organic matter (DOM) at the various phases of its interaction with the Basra soil.
Fig. 4. Fourier-transform infrared spectra of the dissolved organic matter (DOM) at the various phases of its interaction with the Nir-Oz soil.
www.agronomy.org • www.crops.org • www.soils.org 531
Dissolved Organic Matter Fractionation
during DesorptionFollowing the sorption step, the equilibrium solution was
replaced with a background solution, thus stimulating DOM
release from the soil. Th e desorbed DOM fractions included
molecular structures that were also present in the bulk and sorbed
DOM fractions, although they diff ered markedly in their relative
proportions (Fig. 6). In each of the soils’ desorption spectra, the
peak at 1635 cm−1 was the most pronounced. In sharp contrast,
the peak at 1390 cm−1, which manifestly dominated the sorption
spectra, was reduced to merely an undersized shoulder. Th e fact
that the carboxylate moiety associated with the 1635 cm−1 peak
constituted much of the desorbed fraction hints to its fairly weak
association with the soil mineral matrix. In contrast, the pro-
nounced sorption irreversibility of the 1390 cm−1 moiety implies
high-energy binding. Eusterhues et al. (2011) showed that a
DOM peak at 1631 cm−1 representing a carboxylate stretch-
ing vibration was a major element in the interaction of DOM
with ferrihydrite but considered this interaction to be occurring
via outer-sphere complexation. Characteristically, outer-sphere
interactions are weaker than inner-sphere ones. Regarding the
peak at 1390 cm−1, Guan et al. (2006a) reported this band as
the most prominent in the spectra of various dihydroxybenzoic
acids. Moreover, those authors showed that gradually increasing
pH from 2 to 12 results in increasing dominance of this peak in
the spectrum. Th is strongly suggests the occurrence in our study
of a binding mechanism involving sorption of a benzoic acid
group that is enhanced by the presence of additional phenolic
groups (Evanko and Dzombak, 1998; Tejedor-Tejedor et
al., 1992; Yost et al., 1990). Th e sorption strength for com-
pounds containing carboxylic groups has been previously
shown to depend strongly on the position of hydroxyls
with respect to the carboxylic group (Gu et al., 1995; Guan
et al., 2006a). In the evaluation of specifi c complexation
mechanisms of dihydroxybenzoic acids with Al-hydroxide,
Guan et al. (2006a) concluded that the carboxylate group
signifi ed by the 1390 cm−1 peak is involved in bidentate
mononuclear complex formation.
Th e peak at 1589 cm−1 made a signifi cant contribu-
tion to the spectra of sorbed DOM for all soils (Fig. 5).
As in the case for 1390 cm−1, it was practically absent
from the desorption spectra (Fig. 6). Th e 1589 cm−1 peak
signifi es a C = C aromatic stretch, rather than a carboxyl-
ate group, as evidenced by the occurrence of dominant
bands (1585–1595 cm−1) in the FTIR spectra of phenol
(Tejedor-Tejedor et al., 1992), various chlorophenols
(Kung and McBride, 1991), and lignin (Eusterhues et
al., 2011). Th e joint contribution of aromatic and car-
boxylic moieties to irreversible sorption may suggest the
surface complexation of aromatic compounds having car-
boxyl functionalities to metal oxides as demonstrated by
McKnight et al. (1992) and Kaiser et al. (1997).
Th e mutualism between aromatic backbones and
acidic functional groups in the sorption of DOM to
metal oxides is also implicated from the more intense
sorption of humic acids as compared with other DOM
sources. Sharma et al. (2010) quantifi ed the sorption of
Pahokee peat humic acid to ferrihydrite as 1- to 10-fold
higher than the Qmax
evaluated for sorption of various ter-
restrial and aquatic sources of DOM on metal-oxide miner-
als (Chorover and Amistadi, 2001; Gu et al., 1994; Kaiser et
al., 1997; Zhou et al., 2001). Similarly, Guan et al. (2006b)
measured 2.6 and 1.3 times greater sorption of Aldrich humic
acid as compared with phthalic acid and 2,3-dihydroxybenzoic
acid, respectively, to Al-hydroxide at pH 8. Remarkably, the
FTIR spectra of both Aldrich and Pahokee peat humic acids
were shown to be exclusively dominated by two peaks, which
precisely matched the 1589/1390 cm−1 pair recorded in the
current study. Th is strongly supports an aromatic/carboxylic
interrelation in the sorption process.
Aldrich humic acid was further shown by Guan et al.
(2006b) to fractionate during the reaction with Al-hydroxide,
exhibiting the preferential sorption of high-molecular-weight
aromatic molecules. Even so, these same components were
more easily desorbed than their lower molecular weight coun-
terparts during a competition with phosphate for the sorption
sites. Th ese results imply that while high-molecular-weight
humifi ed material, characteristically rich in aromatic carbox-
ylic structures, is highly signifi cant in soil surface complex-
ation reactions, it is just the lower molecular weight humic
fractions—selectively enriched in acidic functional groups—
which are more sorption competitive and potentially less
inclined to be released. Th e E2/E
3 molecular-size index cal-
culations from the current study (Table 2) only support this
notion with respect to sorption. Indeed, the sorbed fractions
were considerably larger than the bulk DOM. Yet, the frac-
tions desorbed from the soil were somewhat smaller than
Fig. 5. Fourier-transform infrared spectra of soil-bound dissolved organic matter (DOM) fractions following sorption to the various soils.
532 Journal of Environmental Quality • Volume 41 • March–April 2012
those that were sorbed. It is possible that aromatic
structures capable of serving as electron donors to
trivalent metal cations experience oxidative trans-
formation and subsequent polymerization and sta-
bilization on metal-oxide surfaces during sorptive
interactions (Arfaioli et al., 1997; Polubesova et al.,
2010; Zhu et al., 2004). Th is might support the
tendency toward smaller molecules in the desorbed
relative to sorbed fractions.
Moderate peaks at 1496, 1481, 1466, and 1450
cm−1, which typify aromatic C = C stretches (though
the last two signals may also be assigned to aliphatic
functionalities), were observed in the desorption
spectrum of the Akko soil and to a lesser extent, in
those of the Nir-Oz and Basra soils (in that order).
Corresponding peaks denoting aromatic C-H vibra-
tions (1172–1111 cm−1), as well as aromatic C-H
stretches (879–833 cm−1), were perceived, too. Th e
signifi cant release of aromatic DOM components can
be similarly appreciated from the SUVA280
data (Table
2). Nevertheless, the released fractions were less aro-
matic than the sorbed ones.
Th e degree of release of aliphatic moieties from soil
was also considerable in view of the signifi cant absor-
bance at 1435 cm−1 and the shoulder at 1420 cm−1
(bands exclusively attributed to paraffi nic C-H defor-
mations). Th is was especially notable with the Akko
soil and the least so with the Basra soil. Th e affi nity
of aliphatic structures of DOM to soils and minerals
has been shown to be low (Kaiser et al., 1997) and it can be
assumed that such interactions would mainly derive from weak
van der Waals forces. Since the Akko soil has a signifi cantly
greater proportion of clay minerals than the others, we sug-
gest that uncharged, hydrophobic patches of siloxane surfaces
(Sposito et al., 1999) allow for a greater degree of interaction of
aliphatic moieties and hence their greater release.
ConclusionsTh e main soil factors aff ecting DOM sorption have been
broadly recognized in past research (i.e., positive eff ects of
metal oxides and clays, and negative eff ects of soil organic
matter). Chemical and physical modifi cations occurring in
DOM during interaction with soil have also been generalized
in terms of the preferential uptake of high molecular weight,
aromatic, and carboxylic moieties over others. Yet, data from
soils, rather than model mineral systems, are needed.
Overall DOM sorption was shown in the current investi-
gation to positively depend on soil metal-oxide content but
not on the soil surface area available for interaction. Ligand-
exchange complexation reactions of carboxylic and phenolic
DOM constituents with metal-oxide surfaces were supported
by FTIR analysis. Furthermore, correlations were established
between the relative uptake of specifi c DOM ligands and the
metal-oxide content of the sorbents. Individual carboxylic and
phenolic structures demonstrated irreversible binding, advo-
cating the potential contribution of such DOM structures to
carbon sequestration.
Revealing the nature of DOM-fractionation processes is of
special consequence for the fate of contaminants, which may
diff erentially interact with DOM fractions in the sorbed and
mobile phases, particularly in agricultural systems that are
exposed to high input of DOM (Drori et al., 2005; Anagu et
al., 2011).
AcknowledgmentsTh e research was supported by the Canada-Israel BARD program
(CA 9114-09) and Vallazi-Pikovsky Post-Doctoral Fellowship of the
Hebrew University of Jerusalem. Special thanks to Yonatan Keren,
Ziva Hochman, and Yehuda Yebelevkovitch for their assistance in the
soil and DOM analyses.
ReferencesAfi f, E., V. Barron, and J. Torrent. 1995. Organic matter delays but does
not prevent phosphate sorption by Cerrado soils from Brazil. Soil Sci. 159:207–211.
Anagu, I., J. Ingwersen, Y. Drori, B. Chefetz, and T. Streck. 2011. Modeling con-centration-dependent sorption-desorption hysteresis of atrazine in a sandy loam soil. J. Environ. Qual. 40:538–547. doi:10.2134/jeq2010.0294
Arfaioli, P., G.G. Riston, M. Bosetto, and P. Fusi. 1997. Humic-like com-pounds formed from L-tryptophan and D-glucose in the presence of Cu (II). Chemosphere 35:575–584. doi:10.1016/S0045-6535(97)00121-5
Biber, M.V., and W. Stumm. 1994. An in-situ ATR-FTIR study: Th e surface coordination of salicylic acid on aluminum and iron (III) oxides. Envi-ron. Sci. Technol. 28:763–768. doi:10.1021/es00054a004
Chefetz, B., Y. Hadar, and Y. Chen. 1998a. Dissolved organic carbon frac-tions formed during composting of municipal solid waste: Properties and signifi cance. Acta Hydrochim. Hydrobiol. 26:172–179. doi:10.1002/(SICI)1521-401X(199805)26:3<172::AID-AHEH172>3.0.CO;2-5
Chefetz, B., P.G. Hatcher, Y. Hadar, and Y. Chen. 1998b. Characteriza-tion of dissolved organic matter extracted from composted mu-nicipal solid waste. Soil Sci. Soc. Am. J. 62:326–332. doi:10.2136/sssaj1998.03615995006200020005x
Chefetz, B., K. Stimler, M. Shechter, and Y. Drori. 2006. Interactions of so-dium azide with triazine herbicides: Eff ect on sorption to soils. Chemo-sphere 65:352–357. doi:10.1016/j.chemosphere.2006.03.006
Fig. 6. Fourier-transform infrared spectra of the dissolved organic matter (DOM) fractions following desorption from the various soils
www.agronomy.org • www.crops.org • www.soils.org 533
Chorover, J., and M.K. Amistadi. 2001. Reaction of forest fl oor organic matter at goethite, birnessite and smectite surfaces. Geochim. Cosmochim. Acta 65:95–109. doi:10.1016/S0016-7037(00)00511-1
Dane, J.H., and G.C. Topp. 2002. Methods of Soil Analysis. Part 4: Physical Methods. SSSA, Madison, WI.
Drori, Y., Z. Izenshtat, and B. Chefetz. 2005. Sorption-desorption behavior of atrazine in soils irrigated with reclaimed wastewater. Soil Sci. Soc. Am. J. 69:1703–1710. doi:10.2136/sssaj2004.0345
Eusterhues, K., T. Rennert, H. Knicker, I. Kögel-Knabner, K.U. Totsche, and U. Schwertmann. 2011. Fractionation of organic matter due to reaction with ferrihydrite: Coprecipitation versus adsorption. Environ. Sci. Tech-nol. 45:527–533. doi:10.1021/es1023898
Evanko, C.R., and D.A. Dzombak. 1998. Infl uence of structural features on sorption of NOM-analogue organic acids to goethite. Environ. Sci. Technol. 32:2846–2855. doi:10.1021/es980256t
Gu, B., T.L. Mehlhorn, L. Liang, and J.F. McCarthy. 1996. Competitive ad-sorption, displacement, and transport of organic matter on iron oxide: I. Competitive adsorption. Geochim. Cosmochim. Acta 60:1943–1950. doi:10.1016/0016-7037(96)00059-2
Gu, B., J. Schmitt, Z. Chen, L. Liang, and J.F. McCarthy. 1994. Adsorption and desorption of natural organic matter on iron oxide: Mechanisms and models. Environ. Sci. Technol. 28:38–46. doi:10.1021/es00050a007
Gu, B., J. Schmitt, Z. Chen, L. Liang, and J.F. McCarthy. 1995. Adsorption and desorption of diff erent organic matter fractions on iron oxide. Geochim. Cosmochim. Acta 59:219–229. doi:10.1016/0016-7037(94)00282-Q
Guan, X.-H., C. Shang, and G.-H. Chen. 2006a. ATR-FTIR investigation of the role of phenolic groups in the interaction of some NOM model compounds with aluminum hydroxide. Chemosphere 65:2074–2081. doi:10.1016/j.chemosphere.2006.06.048
Guan, X.-H., C. Shang, and G.-H. Chen. 2006b. Competitive adsorption of organic matter with phosphate on aluminum hydroxide. J. Colloid In-terface Science 296:51–58. doi:10.1016/j.jcis.2005.08.050
Guo, M., and J. Chorover. 2003. Transport and fractionation of dis-solved organic matter in soil columns. Soil Sci. 168:108–118. doi:10.1097/00010694-200302000-00005
Hunt, J.F., T. Ohno, Z. He, C.W. Honeycutt, and D.B. Dail. 2007. Infl uence of decomposition on chemical properties of plant- and manure-derived dissolved organic matter and sorption to goethite. J. Environ. Qual. 36:135–143. doi:10.2134/jeq2006.0133
Hur, J., and M. Schlautman. 2003. Molecular weight fractionation of hu-mic substances by adsorption onto minerals. J. Colloid Interface Sci. 264:313–321. doi:10.1016/S0021-9797(03)00444-2
Ilani, T., E. Schulz, and B. Chefetz. 2005. Interactions of organic compounds with wastewater dissolved organic matter: Role of the hydrophobic frac-tions. J. Environ. Qual. 34:552–562. doi:10.2134/jeq2005.0552
Inbar, Y., Y. Chen, and Y. Hadar. 1989. Solid-state carbon-13 nuclear magnetic resonance and infrared spectroscopy of composted or-ganic matter. Soil Sci. Soc. Am. J. 53:1695–1701. doi:10.2136/sssaj1989.03615995005300060014x
Joo, J.C., C.D. Shackelford, and K.F. Reardon. 2008. Association of humic acid with metal (hydr)oxide-coated sands at solid-water interfaces. J. Colloid Interface Sci. 317:424–433. doi:10.1016/j.jcis.2007.09.061
Kahle, M., M. Kleber, and R. Jahn. 2003. Retention of dissolved organic mat-ter by illitic soils and clay fractions: Infl uence of mineral phase proper-ties. J. Plant Nutr. Soil Sci. 166:737–741. doi:10.1002/jpln.200321125
Kaiser, K., and G. Guggenberger. 2000. Th e role of DOM sorption to mineral surfaces in the preservation of organic matter in soils. Org. Geochem. 31:711–725. doi:10.1016/S0146-6380(00)00046-2
Kaiser, K., and G. Guggenberger. 2007. Sorptive stabilization of organic matter by microporous goethite: Sorption into small pores vs. surface complex-ation. Eur. J. Soil Sci. 58:45–59. doi:10.1111/j.1365-2389.2006.00799.x
Kaiser, K., G. Guggenberger, L. Haumaier, and W. Zech. 1997. Dis-solved organic matter sorption on subsoils and minerals studied by 13C-NMR and DRIFT spectroscopy. Eur. J. Soil Sci. 48:301–310. doi:10.1111/j.1365-2389.1997.tb00550.x
Kaiser, K., G. Guggenberger, and W. Zech. 1996. Sorption of DOM and DOM fractions to forest soils. Geoderma 74:281–303. doi:10.1016/S0016-7061(96)00071-7
Kaiser, K., and W. Zech. 1997. Competitive sorption of dissolved organic matter fractions to soil and related mineral phases. Soil Sci. Soc. Am. J. 61:64–69. doi:10.2136/sssaj1997.03615995006100010011x
Kaiser, K., and W. Zech. 2000. Sorption of dissolved organic nitrogen by acid subsoil horizons and individual mineral phases. Eur. J. Soil Sci. 51:403–411. doi:10.1046/j.1365-2389.2000.00320.x
Kleber, M., R. Mikutta, M.S. Torn, and R. Jahn. 2005. Poorly crystalline min-eral phases protect organic matter in acid subsoil horizons. Eur. J. Soil
Sci. 56:717–725.
Kleber, M., P. Sollins, and R. Sutton. 2007. A conceptual model of organo-mineral interactions in soils: Self-assembly of organic molecular frag-ments into zonal structures on mineral surfaces. Biogeochemistry 85:9–24. doi:10.1007/s10533-007-9103-5
Kung, K.-H.S., and M.B. McBride. 1991. Bonding of chlorophenols on iron and aluminum oxides. Environ. Sci. Technol. 25:702–709. doi:10.1021/es00016a015
Ling, W., J. Xu, and Y. Gao. 2006. Dissolved organic matter enhances the sorption of atrazine by soil. Biol. Fertil. Soils 42:418–425. doi:10.1007/s00374-006-0085-6
McKnight, D.M., K.E. Bencala, G.W. Zellweger, G.R. Aiken, G.L. Feder, and K.A. Th orn. 1992. Sorption of dissolved organic carbon by hydrous aluminum and iron oxides occurring at the confl uence of Deer Creek with the Snake River, Summit County, Colorado. Environ. Sci. Technol. 26:1388–1396. doi:10.1021/es00031a017
Mehra, O.P., and M.L. Jackson. 1958. Iron oxide removal from soils and clays by dithionite-citrate systems buff ered with sodium bicarbonate. Clays Clay Miner. 7:317–327. doi:10.1346/CCMN.1958.0070122
Navon, R., S. Hernandez-Ruiz, J. Chorover, and B. Chefetz. 2011. Interac-tions of carbamazepine in soil: Eff ects of dissolved organic matter. J. En-viron. Qual. 40:942–948. doi:10.2134/jeq2010.0446
Niemeyer, J., Y. Chen, and J.-M. Bollag. 1992. Characterization of humic acids, composts and peat by diff use refl ectance Fourier-transform in-frared spectroscopy. Soil Sci. Soc. Am. J. 56:135–140. doi:10.2136/sssaj1992.03615995005600010021x
Ochs, M., B. Cosovic, and W. Stumm. 1994. Coordinative and hydrophobic interaction of humic substances with hydrophilic Al
2O
3 and hydropho-
bic mercury surfaces. Geochim. Cosmochim. Acta 58:639–650.
Parfi tt, R.L., A.R. Fraser, and V.C. Farmer. 1977. Adsorption on hydrous ox-ides. III. Fulvic acid and humic acid on goethite, gibbsite and imogolite. J. Soil Sci. 28:289–296. doi:10.1111/j.1365-2389.1977.tb02237.x
Polubesova, T., Y. Chen, R. Navon, and B. Chefetz. 2008. Interactions of hydro-phobic fractions of dissolved organic matter with Fe3+- and Cu2+-montmo-rillonite. Environ. Sci. Technol. 42:4797–4803. doi:10.1021/es8003602
Polubesova, T., S. Eldad, and B. Chefetz. 2010. Adsorption and oxidative transformation of phenolic acids by Fe(III)-montmorillonite. Environ. Sci. Technol. 44:4203–4209. doi:10.1021/es1007593
Rowell, D.L. 1994. Soil science: Methods and applications. Longman Group UK Ltd., London.
Rumpel, C., and I. Kögel-Knabner. 2011. Deep soil organic matter—a key but poorly understood component of terrestrial C cycle. Plant Soil 338:143–158. doi:10.1007/s11104-010-0391-5
Sanaullah, M., A. Chabbi, J. Leifeld, G. Bardoux, D. Billou, and C. Rumpel. 2011. Decomposition and stabilization of root litter in top- and subsoil horizons: What is the diff erence? Plant Soil 338:127–141. doi:10.1007/s11104-010-0554-4
Sharma, P., J. Ofner, and A. Kappler. 2010. Formation of binary and ternary colloids and dissolved complexes of organic matter, Fe and As. Environ. Sci. Technol. 44:4479–4485. doi:10.1021/es100066s
Sparks, D.L. 1996. Methods of soil analysis. Part 3: Chemical methods. SSSA, Madison, WI.
Sposito, G., N.T. Skipper, R. Sutton, S.-H. Park, and A.K. Soper. 1999. Surface geochemistry of the clay minerals. Proc. Natl. Acad. Sci. USA 96:3358–3364. doi:10.1073/pnas.96.7.3358
Tejedor-Tejedor, M.I., E.C. Yost, and M.A. Anderson. 1992. Characteriza-tion of benzoic and phenolic complexes at the goethite/aqueous solu-tion interface using cylindrical internal refl ection Fourier transform in-frared spectroscopy. Part 2. Bonding structures. Langmuir 8:525–533. doi:10.1021/la00038a036
Tipping, E. 1981. Th e adsorption of aquatic humic substanc-es by iron oxides. Geochim. Cosmochim. Acta 45:191–199. doi:10.1016/0016-7037(81)90162-9
Torn, M.S., S.E. Trumbore, O.A. Chadwick, P.A. Vitousek, and D.M. Hen-dricks. 1997. Mineral control of soil organic carbon storage and turn-over. Nature 389:170–173. doi:10.1038/38260
Yost, E.C., M.I. Tejedor-Tejedor, and M.A. Anderson. 1990. In situ CIR-FTIR characterization of salicylate complexes at the goethite/aqueous solution interface. Environ. Sci. Technol. 24:822–828. doi:10.1021/es00076a005
Zhou, Q., P. Maurice, and S.E. Cabaniss. 2001. Size fractionation upon adsorp-tion of fulvic acid on goethite: Equilibrium and kinetic studies. Geochim. Cosmochim. Acta 65:803–812. doi:10.1016/S0016-7037(00)00536-6
Zhu, D., B.E. Herbert, M.A. Schlautman, E.I. Carraway, and J. Hur. 2004. Cation-π bonding: A new perspective on the sorption of polycyclic aro-matic hydrocarbons to mineral surfaces. J. Environ. Qual. 33:1322–1330. doi:10.2134/jeq2004.1322