sorptive and desorptive fractionation of dissolved organic...

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).


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


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.


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.


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


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

sorptive and desorptive fractionation of dissolved organic...

Documents