interactions between natural organic matter and organic ... · for both solution-state and...

Special issue mini‐review

Received: 20 November 2014 Revised: 13 December 2014 Accepted: 16 December 2014 Published online in Wiley Online Library

(wileyonlinelibrary.com) DOI 10.1002/mrc.4209

Interactions between natural organic matterand organic pollutants as revealed by NMRspectroscopyPierluigi Mazzei and Alessandro Piccolo*

Natural organic matter (NOM) plays a critical role in regulating the transport and the fate of organic contaminants in the environ-ment. NMR spectroscopy is a powerful technique for the investigation of the sorption and bindingmechanisms between NOMandpollutants, as well as their mutual chemical transformations. Despite NMR relatively low sensibility but due to its wide versatilityto investigating samples in the liquid, gel, and solid phases, NMR application to environmental NOM–pollutants relations enablesthe achievement of specific and complementary molecular information. This report is a brief outline of the potentialities of thedifferent NMR techniques and pulse sequences to elucidate the interactions between NOM and organic pollutants, with andwithout their labeling with nuclei that enhance NMR sensitivity. Copyright © 2015 John Wiley & Sons, Ltd.

Keywords: NMR; NOM; interactions; organic pollutants; DOSY; relaxation times; saturation transfer difference

* Correspondence to: Alessandro Piccolo, Centro Interdipartimentale per laRisonanza Magnetica Nucleare per l’Ambiente, l’Agro-Alimentare ed i NuoviMateriali (CERMANU), Università di Napoli Federico II, Via Università 100, 80055Portici, Italy. E-mail: [email protected]

Centro Interdipartimentale per la Risonanza Magnetica Nucleare per l’Ambiente,l’Agro-Alimentare ed i Nuovi Materiali (CERMANU), Università di Napoli FedericoII, Via Università 100, 80055 Portici, Italy

Introduction

The prevention of environmental contamination by organic pollut-ants (OPs) and the cleanup of contaminated sites have become aworldwide priority, requiring an advanced knowledge as to boththe pollutants reactivity and fate within environmental matrices.Such an in-depth understanding is essential to design efficientstrategies for decontamination, reduction of pollutants toxicityand bioavailability, and prediction of their sequestration/migration in the environment.[1–3] There are different types oforganic compounds of both anthropogenic and natural origin thatare potentially toxic to plants and animals, even at lowconcentrations.[4] Most of anthropogenic organic contaminantsare industrial products, such as pesticides, pharmaceuticals, endo-crine disrupting compounds, and flame retardants. Depending ontheir chemical structure and reactivity, they are typically classifiedas polycyclic aromatic hydrocarbons, polychlorinated biphenyls(PCBs), nitroaromatic compounds, phenols, and anilines.[4–6]

Polycyclic aromatic hydrocarbons are a large group of organiccompounds with two or more fused aromatic rings and includemolecules such as naphthalene, anthracene, and phenanthrene.They result prevalently from pyrolytic processes and incompletecombustion of organic materials during industrial and other humanactivities, including the processing of coal and crude oil and thecombustion of natural gas.[7] PCBs are persistent OPs differing inthe number of chlorine atoms attached to their biphenyl rings. Mix-tures of PCBs congeners have been widely manufactured from1930 to the 1970s and used as coolants in transformer oil, dielectricfluids, and lubricants.[8] Although the usage of PCBs has beenbanned over 30 years ago in many countries, they still exist in oldelectrical equipment and environmental media to which humanscan be exposed.[6] Nitroaromatic compounds, which are organicmolecules made of at least one nitro group attached to an aromaticring, are largely used in chemical manufacturing, oil refining, pesti-cides, bactericides, explosives, and intermediates in the synthesis of

Magn. Reson. Chem. (2015)

dyes and other high volume.[5] Phenols and anilines are toxicconstituents of dyestuff wastewater, and pentachlorophenol, inparticular, represents one of the most diffused phenols, as it iswidely used as a wood preservative or biocide.[9]

Because of the serious health and ecological risks deriving fromthe continuous and uncontrolled release of pollutants in the envi-ronment and their transformation products of even higher toxicity,more research has been progressively devoted to better under-stand the environmental reactivity of these persistent OPs.[10–13]

In particular, interactions of pollutants with natural organic matter(NOM) have been increasingly studied. A relevant body of literaturehas shown that NOM associated with aquifers, soils, and sedimentsplays a principal role in regulating the transport and fate of contam-inants in the environment.[14–20] NOM is themost abundant form oforganic carbon on the earth’s surface, ubiquitous in terrestrial andaquatic environments,[21,22] and widely heterogeneous, dependingon the climatic, morphological, and microbiological features of itsdevelopment.[23–25] NOM significantly affects the quality of drinkingwater, the yield of crop production, and the environmental mobilityof chemicals and heavy metals.[26–28] NOM is also an important fac-tor in the carbon biogeochemical cycle, and its abiotic and bioticdynamics in soil is directly related to either carbon sequestrationor CO2 emission in the atmosphere.[29–32]

The multifunctionality of NOM is ascribed to its peculiar chemicalnature. In fact, NOM is reckoned to be a supramolecular associationof relatively small (<1000Da) and heterogeneous molecules,

Copyright © 2015 John Wiley & Sons, Ltd.

P. Mazzei and A. Piccolo

derived from the environmental transformation of biomolecules re-leased from dead cells, and held together in metastable conforma-tions by weak bonds, such as hydrophobic and hydrogen bonds[33]

or metal bridges.[34] NOMmetastable conformations can be revers-ibly disrupted by interactions with small amounts of organicacids,[35–38] while the same amount of acids does not alter the con-formation of true macropolymers stabilized by covalent bonds.[39]

The supramolecular nature of NOM and its response to organicacids has been advocated to explain the bioactivity of humicmatteron plants,[40] its molecular dynamics,[41] and the slow release ofsorbed contaminants.[42] Furthermore, the recognition that NOMis composed by supramolecular associations rather thanmacropolymers has allowed the development of a fractionationstrategy, called Humeomics, that enables the analytical detectionof most of the single molecules that constitute the supramolecularassembly.[43–46]

Despite these recent achievements on the NOM chemical under-standing, the mode of interactions between pollutants and NOM isstill unclear and needs further studies.[47] The following principalmechanisms proposed to explain such interactions are non-covalent: (i) adsorption driven by weak forces, such as hydrophobic(van der Waals, π–π, CH–π) and hydrogen bonds, and (ii) electro-static linkages, such as charge transfer, ion exchange, or ligandexchange.[21] However, OPs are also believed to form covalentbondings with NOM, as enabled by specific environmental bioticor abiotic agents, such as microbial or enzymatic activity.[48–51]

Several different analytical techniques have been applied toelucidate the pollutants–NOM interactions, although no singleanalytical technique enables a complete structural or functionalclarification of such interactions. However, NMR spectroscopy iscertainly one of the most powerful techniques for the investiga-tions of the sorption mechanisms occurring between NOM andpollutants.[52] NMR is a nondestructive technique that is capableto unravel themolecular complexities of the environmental hetero-geneous matrixes through the large versatility of its applications inthe solid, semisolid, and liquid state.[53,54] Despite its relatively lowsensibility, NMR offers a number of unique advantages, includingthe possibility to use different, complementary, and multidimen-sional experimental strategies and pulse sequences, which can becombined to show specific intermolecular and intramolecularinteractions.[16,55–58]

NMR spectroscopy in environmental research

The NMR phenomenon occurs when the nuclei of magnetically ac-tive atoms (spin number≠0) are immersed in a static magnetic fieldand exposed to a second oscillating magnetic field. Following thepulse interaction between the applied electromagnetic radiationand the dipolar moments of the nuclei subjected to the static field,the resonance energy is relaxed, producing an FID that contains de-tailed information about the structure, dynamics, reaction state, andchemical environment of the molecular material under study. Thecombination of phase-modulated and length-modulated pulsesspecific of the selected NMR pulse sequences enables a controlledmanipulation of nuclear spins that ultimately provides structural dy-namics information and reveals through-bonding or through-spacecorrelations among nuclei, in either monodimensional or multidi-mensional systems. A number of pulse sequences are commonlyavailable to perform single or multiple selective excitations ofNMR signals and, when required, to suppress specific undesiredsignals without incurring in relevant spectral distortions.[53,54]

wileyonlinelibrary.com/journal/mrc Copyright © 20

Several and diversified NMR pulse sequences are today availablefor both solution-state and solid-state NMR, which allow to facespecific acquisition problems in environmental studies as a functionof samples and reaction conditions. In particular, 1D and multidi-mensional NMR experiments have been employed to structurallycharacterize OPs and functional groups of NOM,[59,60] as well as tostudy the relationships between NOM and organic contaminants.

NMRworks dealing with organic compounds involved directly orindirectly in environmental processes are based onmagnetically ac-tive nuclei with spin number of one half, such as 1H, 13C, 31P, 15N,and 19F. Basically, the most common drawbacks rise from the fol-lowing: (i) the scarce natural abundance of the magnetically activenucleus being examined, (ii) the low gyromagnetic ratio (γ) of thelatter (responsible for a relatively weak NMR sensitivity), and (iii)the intrinsic low concentration of certain nuclei in natural systems.Fortunately, a number of techniques have been developed in orderto circumvent most of these problems. For example, in case of 13Cor 15N nuclei, whose natural isotopic abundance corresponds onlyto 1.1% and 0.366%, respectively, a solution is to employ 13C or15N-labeled compounds.[49,61–63] This strategy to enhance NMRsensitivity is exploited in structural biology, wherein very complexbiomacromolecules, such as proteins, are biosynthesized andoverexpressed by microorganisms that have been fed with 13C or15N-labeled compounds.[64,65]

In the case of NOM, this principle was first applied to verify bybroadband decoupled 13C-NMR, the occurrence of covalentbinding between a 13C-labeled 2,4-dichlorophenol (DCP) and peathumic molecules, upon the action of a peroxidase enzyme.[66] Thealteration of the DCP chemical structure due to coupling withhumic matter resulted in changes in the 13C chemical shifts of thelabeled carbons, which were indicative of the site and type of bond-ing interaction. Similarly, Piccolo and coworkers[48] applied high-resolution NMR to follow the transformation of a 13C-labeledphenoxyalkanoic acid (2,4-D) herbicide in the presence of bothhumicmatter and horseradish peroxidase. They showed for the firsttime that the abiotic degradation of 2,4-D can be catalyzed bydissolved humic substances at neutral pH. In fact, NMR spectrarevealed that three different humic acids (HAs) catalyzed the abioticsplitting of 2,4-D into DCP and acetic acid at pH7 but not at pH4.7.Conversely, peroxidase did not catalyze the oxidative degradationof 2,4-D at any pH and even inhibited the effect of humic sub-stances (Fig. 1). Catalytic degradation by humic matter was attrib-uted to free-radical reactions enhanced by the stereochemicalcontribution of large conformational structures formed by the asso-ciation of heterogeneous humic molecules at neutral pH. Inhibitionof 2,4-D degradation, when humic substances were combined withperoxidase, was explained by modification of both chemical andconformational humic structure due to peroxidase-promotedoxidative cross-coupling among humic molecules.[67]

An increased importance of NMR in environmental studies isprovided by 19F tagging in halogenated compounds that allowsto understand and predict the behavior of chlorinated pollutants.The advantage consists in the high sensitivity of the fluorinenucleus and its typical large chemical shift range, which helps toidentify site-specific interactions.[68–70] An example is representedby the use of a 19F-labeled pentachorophenol (PCP) to ascertainits oxidative copolymerization with NOM due to a biomimeticoxidative catalyst, such as a water-soluble iron porphyrin.[71] The19F-NMR spectra showed that after the catalyzed oxidative reactionin the presence of NOM, the 19F signals were more abundantand scattered over a larger chemical shift interval than for the19F-labeled PCP alone (Fig. 2). This was explained by the covalent

15 John Wiley & Sons, Ltd. Magn. Reson. Chem. (2015)

Figure 2. 19F-NMR spectra of oxidative reaction solutions containing ahumic acid (HA) from lignin and pentafluorophenol (PFP); iron porphyrin(FeP): (A) HA + PFP +H2O2; (B) PFP + FeP +H2O2 (no HA); (C) HA + PFP+ FeP+H2O2 under exposition to daylight for 24 h. (Reprinted withpermission from Fontaine and Piccolo,[71] Copyright 2011 by Springer).

Figure 1. Liquid-state 13C-NMR spectra of reaction mixtures of 13C-labeled2,4-D in buffered solution kept at pH 7.2. (A) [13C]2,4-D alone; (B) [13C]2,4-Dwith leonardite humic acid (HA1); (C) [13C]2,4-D with HA1 and 64 units ofhorseradish peroxidase (HRP); (D) [13C]2,4-D with HA1 and 128 units ofHRP; (E) [13C]2,4-D with HA1 and 256 units of HRP. (Reprinted withpermission from Piccolo and coworkers,[48] Copyright 2001 by Springer).

Interactions between natural organic matter and organic pollutants

coupling that occurred between PCP and humic molecules andconsequently varied the chemical surroundings of 19F nuclei inrespect to PCP alone, thus generating a large number of NMRresonances.

Similar studies were conducted by Thorn and coworkers byusing 15N-labeled dinitrotoluene and trinitrotoluene and theirderivatives, in order to follow their degradation by 15N-NMRspectroscopy in the presence of NOM and in different experimentalconditions.[72–75] Among these experiments, the horseradishperoxidase was again found to promote the covalent binding of4-methyl-3-nitroaniline-15N to NOM.[75] Liquid-state 15N-NMRspectra of a HA solution with the 15N-labeled nitroaniline showedthat this pollutant became covalently bound to HA both beforeand after addition of the enzyme to solution (Fig. 3).

In case of nuclei with low gyromagnetic ratio, different strategieshave been developed to enhance their NMR sensitivity. One ofthese relies on the transfer of magnetization from a very sensitivenucleus, such as hydrogen, to an insensitive nucleus. A number ofdifferent pulse sequences have been designed to induceheteronuclear magnetization transfer via through-bonding (intra-molecular polarization) or through-space (both intramolecular andintermolecular polarization) correlations. In particular, an importantapplication of through-space magnetization transfer is representedby the cross-polarization (CP) technique that is widely applied forsolid-state experiments, especially for studies on NOM in soils andsediments.[76–79]

NMR sensitivity may also be improved by special probes andequipment, such as cryoprobes and dynamic nuclear polarization(DNP) systems. Because the NMR noise is predominantly generatedby electrical resistance occurring in the receiver coil, the cryoprobeis designed to reduce noise by cooling the NMR coil and

Magn. Reson. Chem. (2015) Copyright © 2015 John Wiley

preamplifier with liquid helium but still maintaining the sample atroom temperature.[80] DNP is a recent technique applicable for bothliquid-state and solid-state NMR that has revolutionized many as-pects of NMR spectroscopy. It consists in a substantial nuclear hy-perpolarization achieved through a magnetization transfer frommicrowaves-excited electrons to nuclei, whose NMR sensitivitymay be enhanced up to 400-fold ormore in particular conditions.[81]

Chemical shift and signal broadening insolution-state NMR

Although modern solution-state NMR spectroscopy provides anumber of sophisticated monodimensional or multidimensionalpulse sequences, in many cases, traditional 1D spectra may be suf-ficient to show reliable evidence of intermolecular interactions. Infact, a ligand–sorbent association perturbs the chemical environ-ment of interacting nuclei, whose NMR signals are varied depend-ing on several factors such as ligand affinity, extent of boundligand, and chemical exchange kinetic. Because the interactionsbetween NOM and other molecules are mostly driven by weakforces, such as hydrophobic (van der Waals, π–π, CH–π) and hydro-gen bonds,[3,59,82–84] chemical shift drifts or signal broadening ofpollutants signals may reveal such associations, assuming that thealteration in solution viscosity may be ignored. The chemical shiftdrifts account for the change in the overall magnetic field

& Sons, Ltd. wileyonlinelibrary.com/journal/mrc

Figure 3. Liquid-state quantitative ACOUSTIC 15N-NMR spectra of IHSSElliot soil humic acid (HA) reacted with 15N-labeled 4-methyl-3-nitroaniline(4M3NA) in DMSO-d6 solvent, without and with horseradish peroxidase(HRP) catalyst. (Reprinted with permission from Thorn and coworkers,[75]

Copyright 2008 by American Chemical Society).


experienced by the observed nuclei, while signal broadening is dueto the reducedmobility of ligand when it is associated with sorbentin larger sized adducts, which implies a less efficient minimizationof molecular dipolar couplings and an increased anisotropy.[85]

The evaluation of specific variations in 1H signals of 2,4-DCP and2,4,6-trichlorophenol (TCP) in the presence of a lignite HA repre-sents an example on how to study the extent of interactionbetween NOM and biotoxic chemicals of highly environmentalconcern.[86] In this case, the 1H aromatic peaks of the pollutantphenols undergo a progressive upfield drift as a function of humicconcentration (Fig. 4). This chemical shift variation was assumed toreflect an increased electronic shielding exerted on H nuclei, as thepollutants become progressively incorporated in humic hydropho-bic domains by non-covalent interactions. The concomitant in-crease of signal broadening was interpreted by the authors as asign of restricted ligand molecular mobility, and an evidence ofnon-covalent association to the humic molecules (Fig. 4). Interest-ingly, the broadening of aromatic multiplets into singlets indicatedthat the chemical shift had become slightly anisotropic followingcomplexation of phenols with humic matter. Similarly, sorption ofthe fluorinated 2,4,6-trifluorophenol (TFP) homologue was studiedby 19F spectroscopy. This not only showed that fluorine resonancesin TFP became broadened and shifted downfield as a function ofhumic addition as for TCP (Fig. 4) but also suggested that TCPwas more extensively bound to HA than TFP, as confirmed by otherNMR parameters such as relaxation times and diffusion measure-ments. Although the increased binding with decreasing solutionpH indicated a role of the phenols protonated fraction, it was foundthat the hydrophobicity conferred to phenols by chlorine atoms onaromatic rings was a stronger drive than protonation toward thephenols repartition in the HA hydrophobic domains.


The drift of 19F chemical shifts was also followed to verify the oc-currence of associations between the 4′-fluoro-1′-acetonaphthoneorganic contaminant and the Suwannee River fulvic acid.[82] Aprogressive downfield drift of fluorine signals was detected whenthe fluorinated contaminant was dissolved in a 55% CH3OD/45%D2O solution at an increasing fulvic acid concentration. Moreover,the occurred interactions between a forest soil HA and several13C-labeled PAHs, such as naphthalene, 1-naphtol, and quinoline,were shown by Simpson and coworkers[62] by means of 13C-NMRspectroscopy. In particular, both proton and carbon signals of thestudied pollutants showed an enhanced line broadening in thepresence of HA, whereas no change in pollutants chemical shiftscould be detected.

Relaxation times

NMR relaxation is the process bywhich the total energy of a nuclearsystem, acquired after excitation by a radio-frequency pulse, is pro-gressively lost and the equilibrium state is recovered.[85,87] The im-portance of measuring the relaxation velocity of a nucleus is dueto the fact that, given the absence of interfering factors, it is relatedto either structural variations or interactions undergone by the nu-cleus. Generally, NMR relaxation may be divided into longitudinaland transverse components, which generate spin–lattice (T1) andspin–spin (T2) relaxation times, respectively. These NMR parametersare very sensitive and reliable in revealing indirectly the changes inthe overall molecular mobility, including translational, rotational,and vibrational motion. Therefore, NMR relaxation times are excel-lent means to assess the non-covalent association between a smallsorbate molecule and a larger sorbent through the consequent re-duction of the sorbate translational and rotational motion.[82] T1and T2 values can also be employed to extrapolate the molecularcorrelation time τc that is defined as the average time taken by a nu-clear spin to rotate through one radian. The τc correlation timemayprovide a further indication on the change in ligand molecular mo-bility, because the shorter the tumbling rate of a ligand involved inthe interaction, the longer becomes its correlation time.[82,88,89]

However, it is also to be noted that in a mixture that still containsthe free ligand, the extrapolated T1 or T2 relaxation timescorrespond to an intermediate value ranging between that of thefree ligand and that of the sorbed form.[83] Furthermore, relaxationtimes remain also dependent on solution viscosity, magnetic fieldstrength and homogeneity, and temperature.[59]

In the late 1990s, a number of works have related the interactionsbetween organic contaminants and NOM with the help of NMR re-laxation data. Bortiatynski and Hatcher[90] employed a 13C-labeledphenol at the C-1 position in order to follow its presumed associa-tion with a HA. These authors showed that the T1 relaxation timeof the 13C-labeled phenol varied linearly with the concentration ofhumic matter, and on this basis, they characterized and quantifiednon-covalent interactions in the humic–phenol association.Analogously, Nanny et al.[83] proved in different solvents, such aschloroform, methanol and methanol/water, the occurrence ofnon-covalent interactions between 13C-labeled acenaphthenone,a microbial metabolic by-product of acenaphthene, and theSuwannee River fulvic acid. Interestingly, these authors observedan enhanced solubilization of acenaphthenone that was accountedto the encapsulation of the pollutant in the three-dimensional fulvicacid structure, predominantly solvated with methanol.

Deuterated compounds were also employed to investigate pol-lutants interactions with NOM.[91] In fact, application of 2H-NMR


Figure 4. 1H chemical shifts of 2,4-dichlorophenol (DCP); 2,4,6-trichlorophenol (TCP); phenol (P); and 2,4,6-trifluorophenol (TFP) with and without 1, 3, and7mg of humic acids form lignite. (Reprinted with permission from Smejkalova and coworkers,[86] Copyright 2009 by American Chemical Society).


spectroscopy is advantageous because deuterium is a relativelysensitive nucleus for NMR experiments, while its relaxationmechanism is predominantly quadrupolar, thereby greatly simplify-ing interpretation of T1 data. The interaction between benzene-d6,pyridine-d5, and phenol-d5 and three different HAs was shown byliquid-state 2H-NMR to be non-covalent and significantly influencedby solution pH, degree of aromaticity, and functionality on the aro-matic ring.[91] Interestingly, this work also reported the calculationof the spin correlation time (τc) for a quadrupolar nucleus inmonoaromatic compounds, by taking into account the changesin solution viscosity. Furthermore, the NMR measurements of 19Frelaxation and correlation times were also applied to assess thenature of interactions between 4′-fluoro-1′-acetonaphthone andthe Suwannee River fulvic acid.[82]

Different NMR experiments related to relaxation times mea-surements were applied by Smejkalova and Piccolo[60] to studythe nature and extent of interactions between 2,4-DCP andhumic and fulvic acids. Both 1H relaxation and correlation values,


as well as the derived binding constants, showed that DCP wasmore strongly complexed by humic than fulvic acids. Theauthors attributed this phenomenon to the predominant hydro-phobic character of HA that favored a more effective trapping ofhydrophobic DCP molecules, thereby establishing host–guestassociations.[60]

A further approach based on nuclear relaxation parameters anduseful to identify the mechanisms behind pollutants–NOM interac-tions is related to the use of paramagnetic molecular probes, whichinduce an enhanced nuclear relaxation of excited spins through in-teractions between electrons and nuclear dipoles.[85] The efficiencyof such paramagnetic relaxations derives from the fact that theelectron magnetic moment is 658 times larger than the nuclearmagnetic moment and results in long-range interactions capableto relevantly influence both nuclear spin–lattice and spin–spin re-laxation rates.[87] Paramagnetic probes have been largely employedas contrasting agents inmagnetic resonance imaging studies, espe-cially in medical research, with the aim to enhance detection of



dead tumor cells in vivo or to assess early tumor responses to a drugtreatment.[92,93]

An application of this technique to environmental studieswas proposed by Chien et al.[68] who investigated the interac-tion of trifluoromethylated atrazine with a soil HA by measur-ing the relaxation of 19F-NMR signals in the presence of twowell-known paramagnetic probes, namely, gadolinium-ethylene-diaminetetraacetate (Gd-EDTA) and 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO). The F-enriched atrazine was dissolvedin aqueous humic solutions, and the changes in 19F atrazine relax-ation induced by paramagnetic probes were then followed. Whilethe hydrophilic anionic Gd-EDTA did not exert any significantparamagnetic relaxation in atrazine, the hydrophobic neutralTEMPO induced a rapid change in the nuclear relaxation rate ofthe herbicide. The authors thus inferred that humic moleculesformed ‘micelle-like’ aggregates containing hydrophobic coresinto which nonpolar organic compounds may partition togetherwith the hydrophobic TEMPO, thereby leading to an efficientparamagnetic relaxation of the fluorinated atrazine.Recently, two paramagnetic spin probes, such as TEMPO and

4-hydroxy-2,2,6,6-tetramethylpiperidine-N-oxyl, were subjectedto interactions with several different NOM samples in order todetermine whether the sorption of these probes depended onspecific domains or functional groups.[94] In particular, solid-state13C-NMR spectra of NOM were used to identify a specific molec-ular region in which T1ρH, T1H, and T1C relaxation times were se-lectively reduced as a function of the proximity of fluctuatingmagnetic fields produced by the unpaired electrons of probes.They found that the paramagnetic effect was distributed alongall functional intervals of NOM spectra and no selective interac-tions were observed, thereby failing to reveal any preferentialsorption on specific molecular structures of NOM.[94]

In binding studies between small ligands and large sorbents, sig-nals of the large molecule frequently overlap those of the ligand,thus preventing a clear detection of the latter. This problem is suc-cessfully overcome by applying pulse sequences, which introducedifferent NMR filters and enable simplification of mixture spectraby selectively attenuating the undesired signals.[95] The most com-mon strategies to achieve this purpose are based on T1 or T2 filtersor on their combination and may be used in both one-dimensionaland two-dimensional NMR spectroscopy.[96,97] Indisputably, thesefilters may represent an efficient method to reach a clear detectionof small contaminants signals, when they are masked by theintense and broad resonances produced by NOM.

Molecular diffusion

Diffusion-ordered NMR spectroscopy (DOSY-NMR) is an importanttool for studying the changes in molecular dimension of biomole-cules and small compounds when in complexes with NOM.[60,98,99]

DOSY-NMR is based on a pulsed field gradient sequence that en-ables measurement of translational diffusion (conventionally de-fined as self-diffusion) of dissolved molecules and provides directinformation on molecular dynamics, including intermolecularinteractions[100,101] and conformational changes.[102,103] Followingthe Einstein–Stokes theory, the diffusivity of a molecule decreaseswhen its hydrodynamic radius increases, and information on inter-molecular associations may be obtained bymeasuring the diffusioncoefficient in amolecular system.[104,105] The diffusion rate of a smallligand thus changes considerably when it becomes associated witha largemolecule. As a result, the diffusion coefficients of bound and


unbound ligands differ substantially, thus enabling the diffusionNMR technique to selectively detect the bound ligand signals.[16]

The processing of DOSY spectra is unique in enhancing the reso-lution of the diffusions of molecular components in complex sam-ples, because it provides a direct correlation of translationaldiffusion to the chemical shift in a second NMR dimension.[106] Asdiscussed by Simpson,[47] there are different methods to processDOSY spectra. The CONTIN technique was introduced by Morriset al.[107] in NOM studies as a means to solve the inverse of theLaplace transform and provide a distribution of diffusivities for anumber of soil fulvic acid and HA.[107,108] Thus, for mixtures contain-ing NOM and a contaminant, DOSY-NMR has the potential to mapthe changes in diffusion of both the contaminant and the surround-ing molecules, thus providing a direct evidence of the occurredinteractions.[98] However, analogously to relaxation parameters,also in case of self-diffusion, one has to take into account that theself-diffusion measured by DOSY-NMR corresponds to an interme-diate value between those of both the free and bound ligands.

DOSY-NMR has been up to date largely employed in different re-search fields to show the occurrence of ligand–sorbent interactionsand conduct molecular screenings based on ligand affinity. Thetechnique was also efficiently applied in environmental studiesand focused on interactions between NOM and organicmolecules.[28,60,86,88,98] An early example was the application ofDOSY-NMR to compare the extent of interactions between a peatHA and both the methyl tert-butyl ether, an important petroleumadditive, and chlorsulfuron, a widely used and toxic herbicide.[47]

In this work, the two contaminants were mixed with HA, and aseries of 1D bipolar pulse pair longitudinal eddy-current delaydiffusion experiments at increasing gradient strength was thenacquired. The strongest interaction of the contaminant with humicmatter was then directly related to the rapidity by which the pollut-ant signals were suppressed by the gradient strength. Moreover,the same work showed that DOSY may also efficiently reveal theassociations between NOM and environmentally hazardousmetals,such as cadmium.[47]

Remarkably, useful thermodynamic parameters may also be ob-tained by DOSY-NMR spectroscopy.[104] The self-diffusion constantsof pollutants in interaction with NOMwere employed to determinethe fraction of contaminant bound to NOM and calculate both theassociation constants and Gibbs free energy for a number of pollut-ants such as DCP,[60] simple phenol and a number of halogenatedphenols,[86] and the ionic and hydrophilic glyphosate herbicide.[99]

Diffusion-edited techniques are among the NMR filtersemployed to simplify the spectra of mixtures composed bymolecules with different characteristics and dimensions. Differentlyfrom relaxation-edited filters, diffusion-based filters exploit therapid diffusion of small molecules, thus enabling the isolation andcharacterization of signals due only to large molecules oraggregate.[109,110] Interestingly, some diffusion-edited sequenceswere also developed to isolate signals of small molecules. In partic-ular, the pulse sequence called gradient modified spin-echo-bipolargradient pulse pairs stimulated echo, introduced by Otto andLarive,[111] aimed to isolate diffusion signals of pollutants in amixture with NOM components. This 1H-NMR pulse sequence wascapable, by first using a spin echo to null the resonances of thecoupled spins, to elegantly neglect the spectral resonances ofcoupled spins and detect only those of singlets.

A recent pulse sequence based on a diffusion filter is representedby reverse diffusion.[95] It consists in the subtraction of a controlspectrum from another one achieved with suppression of signalsof small molecule by means of a diffusion filter. It is advocated that



this sequence may be even superior to a traditional relaxation-based filter, because the extent of losses for small molecule signalsbecomes less influenced by relaxation times.[95]

Figure 5. 1H saturation transfer difference (STD) spectra of glyphosate withhumic matter at pH 5.2 and 7: (A) reference STD spectra acquired withoutirradiation, (B) STD difference spectra of a sample treated with soil fulvicacids, and (C) STD difference spectra of a sample treated with lignitehumic acid. A vertical expansion (64×) was applied to the differencespectra. (Reprinted with permission from Mazzei and Piccolo,[99] Copyright2012 by American Chemical Society).

Saturation transfer difference (STD) methods

Dipole–dipole coupling is one of the NMR processes through whichthe relaxation of a nuclear spin can occur. The excess of energyacquired by a nucleus subjected to a radio frequency may be trans-ferred from one to another spin system, provided that internucleardistance is lower than 5Å.[53] Such a through-space magnetizationtransfer is identified as the nuclear Overhaüser effect (NOE), and ithas been largely employed as follows: (i) to identify intermolecularinteractions, by detecting changes in the relaxation rate of a mole-cule bound to a larger macromolecule; (ii) to build up a detectablemagnetization in a partner spin system; and (iii) to enhance signalintensity for relatively insensible nuclei.

Because NOE is due to dipole–dipole couplings, it is highly sensi-tive to internuclear distances and can be used not only to ascertainthe existence of intermolecular interactions but also to measureboth intramolecular and intermolecular distances. An NMR experi-ment based on NOE through-space transfer was applied to studythe interactions between a fluorinated acetonaphthone and aSuwannee River fulvic acid.[82] In this work, a NOE-differenceexperiment has been conducted to selectively irradiate fulvic acidresonances in the presence of the contaminant. Such irradiationinduced a heteronuclear 1H–19F NOE effect, and the 19F signal ofacetonaphthone resulted intensively enhanced.[82]

The saturation transfer difference (STD) is one of the most robustand reliable methods among all through-space NMR experimentsto investigate NOM–pollutants interactions. This technique, origi-nally introduced by Mayer andMeyer (1999)[112] to evaluate the ex-tent of non-covalent protein–ligand interactions, permits the fastdetection of ligand bindings to large molecules, such as proteins,macropolymers, and aggregates, and allows to easily discriminateligands from nonbinding molecules. In STD experiments, the fre-quency of large molecules, which are generally characterized byshort tumbling rate, is saturated by a selective shaped-pulse irradi-ation. The ligand interacting with the irradiated molecule receives apartial saturation transfer through the NOE effect, and it is NMRdetected. A second similar pulse sequence is then applied by anoff-resonance irradiation to avoid any saturation of the large mole-cule. Finally, the subtraction of the second spectrum from the firstone generates a difference spectrum that shows only signals forthe protons involved in non-covalent interactions. Only the ligandprotons in interaction with the large molecule can experience apolarization transfer and produce an enhanced signal intensity.The efficiency of the STD technique largely depends on the chem-ical exchange between the ligand bound and free states, whichshould be faster than the relaxation time. While only a relativelysmall amount of sorbent is sufficient for this experiment, a largeexcess of ligand is recommended in order to magnify the bindingprocess and ensure sufficiently intense NMR signals.[57,113]

The STD technique has been already employed to elucidate theinteractions occurring between NOM and xenobiotics.[114,115]

Recently, the interactions between the glyphosate (N-phosphonomethylglycine) herbicide and soluble fulvic acid andHA have been studied at both pH 5.2 and 7, whereby the herbi-cide is present in different ionic forms.[99] The NOM–glyphosateassociation was also studied by the 1H-NMR STD technique thatrevealed that the non-covalent interactions were significantly


enhanced only at pH 5.2 for both humic materials (Fig. 5). This re-sult was explained with the stabilization of the NOM–glyphosatecomplex by the H-bonds established between the herbicide,protonated at the phosphonate group at pH 5.2, and the comple-mentary oxygen-containing functions in humic matter.

A saturation transfer double difference (STDD) technique hasbeen introduced as a variation of normal STD experiment forNOM–pollutants studies.[114] It consists in obtaining a clear spectralsubtraction of the background signals due to NOM, therebyreaching an accurate quantitative STD estimation of epitope mapsfor the pollutant. This STDD technique was successfully employedto quantitatively describe the interactions between a peat HA andseveral water-soluble pesticides, such as diflufenzopyr, imazapyr,acifluorfen, and chlorsulfuron.[114]

Later, a 1H{19F} reverse heteronuclear STD technique was appliedto investigate the sorption of a perfluorinated organic compoundson a peat HA and other biomolecules, such as lignin andalbumin.[115] Such further modification of the original STDtechnique comprises a selective irradiation of 19F nuclei in theperfluorinated compound, whose magnetization is then trans-ferred to NOM nuclei only in the case of direct interaction withthe 19F-labeled molecule. The final output of the application of thisNMR experiment is a 1H spectrum that shows only those NOMprotons that received the through-space magnetization transferdue to the direct association with the irradiated fluorinatedcompound.[115] A combination of the reverse-heternonuclear andforward-heternonuclear STD experiments was attempted byLongstaffe and Simpson.[116] They employed the former techniqueto identify the binding pattern of structurally similar xenobioticsthat interact with NOM, while the latter sequence was used to



reveal the binding orientation of fluorinated xenobiotics during theinteraction.

Figure 6. Quantitative 13C-NMR spectra of a standard prairie soil humic acid,measured at 14 kHz MAS within 5 h each, after 11.5 ms (10 × 1.1ms + 0.5ms)multiCP with tz = 0.4 s (thick black line) and after direct polarization (DP) with30 s recycle delays (dashed red line), scaled to match peak intensities. Thincontinuous (green) and dashed (blue) lines: corresponding multiCP and DPspectra, respectively, of nonprotonated C and mobile segments selected byrecoupled dipolar dephasing. (Reprinted with permission from Johnsonand Schmidt-Rohr,[126] Copyright 2014 by Elsevier).

Solid-state NMR

The MAS technique enables NMR analysis of crystalline or amor-phous solid samples. MAS NMR has been conceived to minimizethe processes that prevent the acquisition of meaningful NMRspectra in the solid state, such as strong homonuclear dipolarinteractions, chemical shift anisotropy, quadrupolar interactions,and enhanced magnetic susceptibility.[117] However, relevantadvantages in NMR studies may originate from dipolar interactions.In fact, strong heteronuclear dipolar couplings can lead to efficientthrough-space magnetization transfer that is called cross-polarization and is capable to induce significant signals enhance-ment. The CP-MAS technique is widely applied in environmentalstudies, because it increases the sensitivity for less abundant nucleiwith low gyromagnetic ratio γ, such as 13C, 15N, and 29Si, via polar-ization transfer from the large γ and most abundant 1H nucleus.These advantages make the CP-MAS technique mostly appropriateto study the composition of NOM and its interactions with OPs inthe solid state.[20,76,78,118–121]

A number of NOM samples with different chemical compositionwere studied by solid-state MAS NMR to evaluate their interactionswith phenanthrene.[122] This work showed that NOM nonpolar ali-phatic components had a relevant role in phenanthrene retention.In particular, the authors acquired solid-state 13C-NMR spectra byboth a Bloch decay experiment (direct polarization) and CP, in orderto investigate the correlations between the Freundlich sorptioncoefficient of phenanthrene and the characteristics of NOM sam-ples such as aromatic, nonpolar aliphatic [including amorphouspoly(methylene)] and polar components. Interestingly, also the 1Hinversion recovery detected by 13C was used to differentiate thepoly(methylene) domains from the branched nonpolar aliphaticsegments and correlate their distribution with the phenanthrenesorption capacity by NOM.[122]

Moreover, some advanced solid-state experiments, such as 1Hwide-line spectra (coupled to T1ρH filters to remove the contribu-tion of mobile components) and inversion recovery to measure1H T1 relaxation times, were employed to relate the domains ofdifferent mobility in several humic substances with their capacityto adsorb OPs.[123] The different mobility of humic domains wasalso assessed from the broadening of signals obtained with the2D 1H–13C wide-line separation (WISE) technique. This sequenceenabled a relationship between the molecular mobility of humicdomains and the polarization transfer that was considered thatthe more efficient and faster, the broader was the signal in theWISE spectrum. As a confirmation of such result, 2D 1H–13Cheteronuclear correlation (HETCOR) spectra were also acquired byvarying the mixing times and, thus, exploring the rigidity of humicdomains as a function of the 1H–13C spin-diffusion rate. It was foundthat both mobile and rigid domains of different compositionscoexist in humic matter, although mobile domains represent theprincipal site whereby organic contaminants may partition.[123]

As for solution-state NMR, the quality of solid-state NMR wasimproved by the application of spectral filters.[117] For example,cross-polarization/total side band suppression (CP/TOSS), chemicalshift anisotropy (CSA), and dipolar dephasing filters were appliedto characterize in details the chemical composition of NOMprior to its interactions with DCP, phenanthrene, and 1,2,4,5-tetrachlorobenzene.[124] The CP/TOSS sequence selectively


removes spinning sidebands from the spectrum, thus enablinga concomitant lower rotor spinning rate and an enhancementof the CP efficiency. The CSA filter is designed to only showsp3-hybridized carbons in the spectrum, by selectively removinganisotropic signals, such as those for aromatic or carbonyl carbons,while increasing the detection of anomeric carbons.[122] By varyingthe dephasing delay, the dipolar dephasing filter allows to selec-tively suppress carbon signals as a function of their protonation,whereby the length of filter delay is generally dictated by the fol-lowing order: CH2<CH<CH3<Cquaternary.

[79,124,125] Nevertheless,it should be recalled that these techniques may also lead to the lossof observable signals, depending on other factors, such as filterduration and specific nuclear relaxation times.

Recently, an advance of CP-MAS, referred to as the multiCP tech-nique, has introduced a simple and robust method that can providequantitative 1D 13C-NMR spectra of NOM with an improved largersignal-to-noise ratio than previously achievable by quantitativedirect-polarization NMR.[126] It consists in long (>10ms) CP contactfrom 1H, without any significant magnetization loss due to relaxa-tion and with a moderate duty cycle of the radio-frequency irradia-tion. This is followed by multiple 1ms CP times, alternated with 1Hspin–lattice relaxation periods to repolarize protons.[126] The quan-titative reliability of this pulse sequence can be appreciated by thecomparison (Fig. 6) of 13C spectra for a prairie soil HA acquired byeither a direct polarization or amultiCP technique, with andwithoutthe application of a dipolar dephasing filter.[126]

The advancedmultiCP pulse technique was applied to character-ize different NOM types, which were then investigated by two-dimensional 1H–13C HETCOR correlation NMR spectroscopyafter their interactions with a fully deuterium-exchanged,carbonyl-13C-labeled benzophenone.[127] The full deuteration forbenzophenone had the objective to eliminate intramolecular1H–13C correlations, thus allowing the benzophenone 13C-carbonylto interact exclusively with the nearby protons in NOM. The spin dif-fusion from benzophenone carbons to NOM protons was followedby HETCOR experiments, which employed three different mixingtimes (0.01, 0.3, and 10.25ms). As for the interactions with a Pakoeepeat, the three HETCOR experiments indicated the moleculardomains with which benzophenone preferentially interacted(Fig. 7a–c). The signals related to benzophenone carbonyl, at198ppm, and aromatic, at 137ppm, were further shown by 1HHETCOR vertical 1D slices, at different mixing times, in Fig. 7e


Figure 7. Two-dimensional HETCORNMR spectra of the Pahokee peat sorbedwith benzophenone-(13C¼O)-d10 with (a) 1ms LG-CP (tm,e = 0.01ms), (b) 1msHH-CP (tm,e = 0.3ms), and (c) 1ms HH-CP with 10ms spin diffusion (tm,e = 10.25ms). (d) 13C-NMR-detected 1H inversion recovery of Pahokee peat sorbedwith benzophenone-(13C¼O)-d10, after the indicated recovery delays (ms). The 1H spectra extracted at the 13C chemical shifts of 13C¼O (197 ppm, panel e) andnonprotonated aromatic C (137 ppm, panel f) at different mixing times. (g) Proton projection along 0–220 ppm at tm,e = 10.25ms. (Reprintedwith permissionfrom Cao and coworkers,[127] Copyright 2014 by American Chemical Society).


and 7f, respectively, whereas the proton projection for theHETCOR experiment at the longest mixing time is reported inFig. 7g. Interestingly, 13C-detected 1H inversion-recovery spectrawere also acquired for the Pahokee peat sorbed with benzophe-none (Fig. 7d), in order to detect the spatial proximity of NOMdomains and predict their availability for interactions.[127]

The 19F spectroscopy was also employed in the solid state toinvestigate the role of lipid composition in geosorbents, such ascoastal sediments, humin, and HAs, on the adsorption of


phenanthrene.[70] The authors used the 2,2,2-trifluoroethyl laurateas a probe to show the relevant contribution of the lipid compo-nents of the selected geosorbents. In fact, they observed that a sig-nificant change occurred in 19F spectra after adding phenanthreneto a humic matter that was previously sorbed with the fluoroprobe.

The 19F→ 1H solid-state CP-MAS spectroscopy was also appliedto study the association between organofluorine xenobiotic com-pounds and an unmodified peat.[128] In particular, the 19F nucleiof organofluorine compounds were used to induce observable



transverse magnetization to the 1H nuclei in the peat molecularcomponents, which were in direct interaction with the sorbedxenobiotics. This experiment was conducted with the recentlyconceived comprehensive multiphase MAS probe that involves aspectrometer lock. The latter device provides an enhanced lineshape for 1H detection, as compared to a traditional solid-statehardware, and also includes gradients that allow to conduct NMRdiffusion experiments in the solid state.[129,130]

HR-MAS NMR

The versatility of NMR spectroscopy was further enlarged by theintroduction of the high-resolution MAS (HR-MAS) technique thatenables NMR investigation on semisolid or gel-phase samples.[131]

The sample is placed in a locked-up rotor and added with a smallamount of a deuterated solvent that helps to reduce dipolarinteractions, which are further minimized by spinning the rotor atthe magic angle. The HR-MAS probes are generally provided witha deuterium lock and a gradient system. The latter device permitsexperiments (originally designed for solution-state NMR), such asdiffusion-based and diffusion-edited experiments, or thoseexploiting the selection of spoil gradients blocks for both 1D andmultidimensional spectroscopy. A further singular advantage ofHR-MAS is that it may distinguish between the free and boundligand forms, because the technique detects the latter only by fullyremoving anisotropy through MAS rotation.Simpson and coworkers[28] early applied HR-MAS NMR to study

samples of environmental interest. They showed that the structureof a whole soil at the solid–aqueous interface may be characterizedby HR-MAS spectra, by combining one-dimensional and two-dimensional solution-like experiments, including T2 filters whennecessary. In addition, they observed variations in spectra inducedby specific solvents, such as deuterated water, that producedbroader signals, and dimethylsulphoxide-d6 that conversely led toan improved resolution, prevalently because of the solvent-induceddisplacement of H-bonds. When the herbicide trifluralin was put incontact with the soil, new signals appeared in the spectra and wereattributed to the herbicide bound to the external soil matrix.[28]

The role of the physical conformation of organic matter in theprocesses of pollutants sorption was also explored by HR-MAS.[132] A number of organo-clay complexes were formed bycoating montmorillonite and kaolinite with humic matter andphenanthrene in Na+-dominated or Ca2+-dominated solutions atdifferent pH and ionic strength values. The resulting HR-MASspectra suggested that the sorption of the pollutant onto theorgano-clay complexes occurred as a function of the distributionof hydroxyl sites on the mineral surfaces. In fact, these sites deter-mined the interfacial configuration of the humic coating throughthe bonds formed with the complementary carboxyl groups inhumic molecule, thus varying the accessibility of sorptive sites tophenathrene.[132] The HR-MAS spectroscopy was also applied tostudy the interactions between toluene and HA.[133] This workshowed that a correlation existed between the molecular motionof humic aliphatic components and the attenuation of self-diffusionfor toluene through the humic matrix.

Perspectives

The NMR spectroscopy has so far largely contributed to amolecularunderstanding of the interactions between OPs andNOM in the en-vironment. However, much is left to be done in this sector,


especially because of the complex properties of NOM, which mayvary in different environmental compartments, as well as in differ-ent geographical and climatic conditions. Future efforts should bedevoted to elucidate further not only the molecular content ofNOM but also how themolecular components are stereochemicallyarranged in water and soil. To such an aim, a great contribution tothe unraveling of the molecular composition of NOM can be againprovided by NMR spectroscopy with its fast advances in novel andhighly performing pulse sequences or additional hardware devices(powerful magnets, cryosystems, new probes, hyphenation withchromatographic separations, etc.). Most of the techniquesdescribed here will be made more efficient in the near future withthe advance of NMR technologies.

Nevertheless, coupling of NMR spectroscopy with high-resolution and ultra-high-resolution liquid-chromatography/mass-spectrometry techniques should also enhance considerably thestructural precision in identifying the humic molecules responsiblefor the binding and environmental transport of pollutants in differ-ent conditions. It is also believed that the large amount of refineddata that can be obtained in the future by advanced NMR andmass-spectrometry instrumentation will have to undergo extensiveelaboration by intelligent software and multivariate analyses in or-der to reduce statistical uncertainty in data interpretation. Finally,the stereochemical understanding that is required to predict theadsorption on and diffusion through NOM of pollutants of differentpolarity will be achieved by handling the vast amount of moleculardata with computational chemical software. This computationalsupport will become fundamental to increase the significance ofcollected NMR data, such as those acquired with DOSY orthrough-space correlations techniques, which can be used to feedthe predictive models of pollutant behavior in the environment.

References[1] J. D. Kubicki, S. E. Apitz. Org. Geochem. 1999, 30, 911–927.[2] B. Pan, P. Ning, B. Xing. Environ. Sci. Pollut. Res. 2008, 15, 554–564.[3] F. Sannino, A. Piccolo, in Sustainable Development in Chemical

Engineering: Innovative Technologies, (Eds: A. Basile, V. Piemonte, M.de Falco), John Wiley & Sons, Chichester, United Kingdom, 2013,pp. 267–296.

[4] A. S. Knox, A. P. Gamerdinger, D. C. Adriano, R. K. Kolka, D. I. Kaplan, inSources and practices contributing to soil contamination,Bioremediation of Contaminated Soils, (Ed: D. M. Kral), AmericanSociety of Agronomy, South Segoe Road, Madison, Wisconsin, 1999,pp. 53–80.

[5] K. S. Ju, R. E. Parales. Microbiol. Mol. Biol. Rev. 2010, 74(2), 250–268.[6] R. Gioia, A. J. Akindele, S. A. Adebusoye, K. A. Asante, S. Tanabe,

A. Buekens, A. J. Sasco. Environ. Sci. Pollut. Res. 2014, 21, 6278–6289.[7] A. Mastral, M. A. Callena. Environ. Sci. Technol. 2000, 34(15),

3051–3057.[8] L. Passatore, S. Rossetti, A. A. Juwarkar, A. Massacci. J. Hazard. Mater.

2014, 278, 189–202.[9] W. Zheng, H. Yu, X. Wang, W. Qu. Environ. Int. 2012, 42, 105–116.[10] B. K. Robertson, M. Alexander. Environ. Toxicol. Chem. 1998, 17,

1034–1038.[11] B. J. Reid, K. C. Jones, K. T. Semple. Environ. Pollut. 2000, 108, 103–112.[12] M. Delort, B. Combourieu, N. Haroune. Environ. Chem. Lett. 2004, 1,

209–213.[13] R. Weber, C. Gaus, M. Tysklind, P. Johnston, M. Forter, H. Hollert,

E. Heinisch, I. Holoubek, M. Lloyd-Smith, S. Masunaga, P. Moccarelli,D. Santillo, N. Seike, R. Symons, J. P. Machado Torres, M. Verta,G. Varbelow, J. Vijgen, A. Watson, P. Costner, J. Woelz, P. Wycisk,M. Zennegg. Environ. Sci. Pollut. Res. 2008, 15, 363–393.

[14] R. P. Schwarzenbach, P. M. Gschwend, D. M. Imboden, EnvironmentalOrganic Chemistry, John Wiley and Sons, Inc., New York, 1993.

[15] P. H. Santschi, J. J. Lenhart, B. D. Honeyman. Mar. Chem. 1997, 58,99–125.



[16] A. Chen, M. J. Shapiro. Anal. Chem. 1999, 71(19), 669–675.[17] B. Pan, S. Ghosh, B. Xing. Environ. Sci. Technol. 2007, 41, 6472–6478.[18] M. Borisover, M. Sela, B. Chefetz. Chemosphere 2011, 82, 1454–1460.[19] J. J. Pignatello. J. Soil Sediment 2012, 12, 1241–1256.[20] K. Sun, Y. Ran, Y. Yang, B. Xing, J. Mao. Geoderma 2013, 204, 68–74.[21] A. Piccolo, in H. Substances, in Terrestrial Ecosystems (Ed: A. Piccolo),

Elsevier, Amsterdam, 1996.[22] E. L. Sharp, S. A. Parson, B. Jefferson. Water Sci. Technol. 2006, 53(7),

67–76.[23] A. Piccolo, P. Conte, A. Cozzolino, R. Spaccini, in Handbook of Processes

and Modelling in Soil-plant System, (Eds: D. K. Benbi, R. Nieder),Haworth Press, New York, 2003, pp. 83–120.

[24] A. Matilainen, E. T. Gjessing, T. Lahtinen, L. Hed, A. Bhatnagar,M. Sillanpää. Chemosphere 2011, 83, 1431–1442.

[25] G. Chilom, A. Baglieri, C. A. Johnson-Edler, J. A. Rice. Org. Geochem.2013, 57, 119–126.

[26] A. Piccolo. Sci. Tot. Environ. 1989, 81, 607–614.[27] J. I. Hedges, R. G. Keil. Mar. Chem. 1995, 49, 81–115.[28] A. J. Simpson, W. Kingery, D. R. Shaw, M. Spraul, E. Humpfer,

P. Dvortsak. Environ. Sci. Technol. 2001, 35, 3321–3325.[29] W. H. Schlesinger, J. A. Andrews. Biogeochem. 2000, 48, 7–20.[30] R. Spaccini, P. Conte, A. Piccolo, G. Haberhauer, M. H. Gerzabek. Soil

Biol. Biochem. 2002, 34, 1839–1851.[31] A. Piccolo, R. Spaccini, A. Nebbioso, P. Mazzei. Environ. Sci. Technol.

2011, 45, 6697–6702.[32] A. Piccolo, in Carbon Sequestration in Agricultural Soils (Ed: A. Piccolo),

Springer, Verlag, Heidelberg, 2012.[33] A. Piccolo. Adv. Agronom. 2002, 75, 57–134.[34] A. Nebbioso, A. Piccolo. Environ. Sci. Technol. 2009, 43, 2417–2424.[35] A. Piccolo, S. Nardi, G. Concheri. Eur. J. Soil Sci. 1996, 47, 319–328.[36] A. Piccolo, P. Conte, A. Cozzolino. Eur. J. Soil Sci. 1999, 50, 687–694.[37] A. Piccolo, P. Conte, E. Trivellone, B. Van Lagen, P. Buurman. Environ.

Sci. Technol. 2002, 36, 76–84.[38] A. Piccolo, P. Conte, R. Spaccini, M. Chiarella, Biol. Fert. Soils 2003, 37,

255–259.[39] A. Piccolo, P. Conte, A. Cozzolino. Soil Sci. 2001, 166, 174–185.[40] L. P. Canellas, F. L. Olivares. Chem. Biol. Technol. Agric. 2014, 1, 1–11.[41] M. Orsi. Chem. Biol. Technol. Agric. 2014, 1, 1–14.[42] A. Beiraghi, K. Pourghazi, M. Amoli-Diva. Chem. Eng. Sci. 2014, 108,

103–110.[43] A. Nebbioso, A. Piccolo. Biomacromolecules 2011, 12, 1187–1199.[44] A. Nebbioso, A. Piccolo. Anal. Chim. Acta 2012, 720, 77–90.[45] A. Nebbioso, A. Piccolo, M. Lamshoft, M. Spiteller. RSC Adv. 2014, 4,

23658–23665.[46] A. Nebbioso, P. Mazzei, D. Savy. Chem. Biol. Technol. Agric. 2014,

1(24), 1–9.[47] A. J. Simpson. Magn. Reson. Chem. 2002, 40, S72–S82.[48] A. Piccolo, P. Conte, A. Cozzolino, M. Paci. J. Indust. Microbiol.

Biotechnol. 2001, 26, 70–76.[49] T. Kacker, E. T. K. Haupt, C. Garms, W. Francke, H. Steinhart.

Chemosphere 2002, 48, 117–131.[50] H. M. Bialk, C. Hedman, A. Castillo, J. A. Pedersen. Environ. Sci. Technol.

2007, 41, 3593–3600.[51] A. Gulkowska, M. Sander, J. Hollender, M. Krauss. Environ. Sci. Technol.

2013, 47, 6916–6924.[52] R. L. Cook. Anal. Bioanal. Chem. 2004, 378, 1484–1503.[53] N. E. Jacobsen, in NMR Spectroscopy Explained – Simplified Theory,

Applications and Examples for Organic Chemistry and StructuralBiology, (Ed: JohnWiley and sons), Wiley, New Jersey, Hoboken, 2007.

[54] M. Levitt, in Spin Dynamics: Basics of Nuclear Magnetic Resonance,(2nd edn), (Ed: John Wiley & Sons inc.), Wiley, Chichester, 2008.

[55] B. Meyers, T. Peters. NMR. Angew. Chem. Int. Ed. 2003, 42, 864–890.[56] L. A. Cardoza, A. K. Korir, W. H. Otto, C. J. Wurrey, C. K. Larive. Prog. Nucl.

Magn. Reson. Spectrosc. 2004, 45, 209� 238.[57] L. P. Calle, F. J. Canada, J. Jimenez-Barbero. Nat. Prod. Rep. 2011, 28,

1118–1125.[58] D. Huster. Biochim. Biophys. Acta 2014, 1841, 1146–1160.[59] A. J. Simpson, B. Lefebvre, A. Moser, A. Williams, N. Larin, M. Kvasha,

W. L. Kingery, B. Kelleher. Magn. Reson. Chem. 2004, 42, 14–22.[60] D. Smejkalova, A. Piccolo. Environ. Sci. Technol. 2008, 42, 699–706.[61] E. A. Guthrie, J. M. Bortiatynski, J. D. H. van Heemst, J. E. Richman,

K. S. Hardy, E. M. Kovach, P. G. Hatcher. Environ. Sci. Technol. 1999,33, 119–125.

[62] M. J. Simpson, A. J. Simpson, P. G. Hatcher. Environ. Toxic. Chem. 2004,23(2), 355–362.


[63] U. Wanner, F. Fuehr, A. A. de Graaf, P. Burauel. Environ. Poll. 2005, 133,35–41.

[64] N. K. Goto, L. E. Kay. Cur. Opin. Struc. Biol. 2000, 10(5), 585–592.[65] S. Y. Ohki, M. Kainosho. Prog. Nucl. Mag. Res. Sp. 2008, 53, 208–226.[66] P. G. Hatcher, J. M. Bortlatynski, R. D. Minard, J. Dec, J. M. Bollag.

Environ. Sci. Technol. 1993, 27, 2098–2103.[67] A. Piccolo, P. Conte, A. Cozzolino, R. Spaccini. Naturwissenschaften

2000, 87, 391–394.[68] Y. Y. Chien, E. G. Kim, W. F. Bleam. Environ. Sci. Technol. 1997, 31,

3204–3208.[69] B. E. Herbert, P. M. Bertsch, in NMR Spectroscopy in Environmental

Chemistry, (Eds: M. A. Nanny, R. A. Minear, J. A. Leenheer), OxfordUniversity Press: Oxford, 1997.

[70] L. Tremblay, S. D. Kohl, J. A. Rice, J. P. Gagné. Chemosphere 2005, 58,1609–1620.

[71] B. Fontaine, A. Piccolo. Environ. Sci. Pollut. Res. 2011, 19, 1485–1493.[72] K. A. Thorn, K. R. Kennedy. Environ. Sci. Technol. 2002, 36, 3787–3796.[73] K. A. Thorn, J. C. Pennington, C. A. Hayes. Environ. Sci. Technol. 2002,

36, 3797–3805.[74] K. A. Thorn, P. G. Thorne, L. G. Cox. Environ. Sci. Technol. 2004, 38,

2224–2231.[75] K. A. Thorn, J. C. Pennington, K. R. Kennedy, L. G. Cox, C. A. Hayes,

B. E. Porter. Environ. Sci. Technol. 2008, 42, 2542–2550.[76] C. Preston. Can. J. Soil Sci. 2001, 81, 255–270.[77] J. Chen, B. Gu, E. J. LeBouf, H. Pan, S. Dai. Chemosphere 2002, 48,

59–68.[78] J. Peuravuori, P. Ingman, K. Pihlaja. Talanta 2003, 59, 177–189.[79] P. Conte, R. Spaccini, A. Piccolo. Prog. Nucl. Mag. 2004, 44, 215–223.[80] P. Styles, N. F. Soffe, C. A. Scott. J. Magn. Reson. 1989, 84(2), 376–378.[81] A. B. Barnes, G. De Paepe, P. C. A. van der Wel, K. N. Hu, C. G. Joo,

V. S. Bajaj, M. L. Mark-jurkauskas, J. S. Sirigiri, J. Herzfeld, R. J. Temkin,R. G. Griffin. Appl. Magn. Reson. 2008, 34, 237–263.

[82] A. M. Dixon, M. A. Mai, C. K. Larive. Environ. Sci. Technol. 1999, 33,958–964.

[83] M. A. Nanny, J. M. Bortiatynski, P. G. Hatcher. Environ. Sci. Technol.1997, 31, 530–534.

[84] D. Zhu, S. Hyun, J. P. Pignatello, L. S. Lee. Environ. Sci. Technol. 2004, 38,4361–4368.

[85] V. I. Bakhmutov, Practical NMR Relaxation for Chemists (Ed: John Wiley& sons), Wiley, Chichester, West Sussex, England, 2004.

[86] D. Smejkalova, R. Spaccini, B. Fontaine, A. Piccolo. Environ. Sci. Technol.2009, 43(14), 5377–5382.

[87] I. Bertini, C. Luchinat, S. Aime. Coord. Chem. Rev. 1996, 150, 77� 110.[88] P. Mazzei, H. Oschkinat, A. Piccolo. Chemosphere 2013, 93, 1972–1979.[89] P. Mazzei, A. Piccolo. Eur. J. Soil Sci. 2013, 64(4), 508–515.[90] J. M. Bortiatynski, P. G. Hatcher, in Nuclear Magnetic Resonance

Spectroscopy in Environmental Chemistry, (Eds: M. A. Nanny, R. A.Minear, J. A. Leenheer), Oxford University Press, New York, 1996.

[91] M. A. Nanny, J. P. Maza. Environ. Sci. Technol. 2001, 35, 379–384.[92] T. E. Yankeelov, J. C. Gore. Curr. Med. Imaging rev. 2007, 3(2),

91–107.[93] A. S. Krishnan, A. A. N. Meng, M. M. de Backer, D. E. Hu, B. Davletov,

M. I. Kettunen, K. M. Brindle. Radiology 2008, 246, 854–862.[94] C. Lattao, X. Cao, Y. Li, J. Mao, K. Scmidt-Rohr, M. A. Chappell,

L. F. Miller, A. L. Dela Cruz, J. J. Pignatello. Environ. Sci. Technol. 2012,46, 12814–12822.

[95] M. Vega-V’azquez, J. C. Cobas, F. F. Oliveira de Sousa, M. Martin-Pastor.Magn. Reson. Chem. 2011, 49, 464–468.

[96] H. Tang, Y. Wang, J. K. Nicholson, J. C. Lindon. Anal. Biochem. 2004,325, 260–272.

[97] H. Farroq, D. Courtier-Murias, R. Soonf, H. Masoom, W. Maas, M. Fey,R. Kumar, M. Monette, H. Stronks, M. J. Simpson, A. J. Simpson.Magn. Reson. Chem. 2013, 51, 129–135.

[98] A. J. Simpson, W. Kingery, M. Spraul, E. Humpfer, P. Dvortsak,R. Kerssebaum. Environ. Sci. Technol. 2001, 35(22), 4421–4435.

[99] P. Mazzei, A. Piccolo. Environ. Sci. Technol. 2012, 46, 5939–5946.[100] T. Brand, E. J. Cabrita, S. Berger. Prog. Nucl. Mag. Reson. Spectrosc. 2005,

46, 159–196.[101] Y. Cohen, L. Avram, L. Frish. Angew. Chem. 2005, 44, 520–554.[102] S. Viel, L. Mannina, A. Segre. Tetrahedron Lett. 2002, 43, 2515–2519.[103] B. Fontaine, M. Drosos, P. Mazzei. Environ. Sci. Pollut. Res. 2014, 21,

8016–8024.[104] R. Wimmer, F. L. Aachmann, K. L. Larsen, S. B. Petersen. Carbohydr. Res.

2002, 337, 841� 849.[105] G. Zheng, W. S. Price. Environ. Sci. Technol. 2012, 46(3), 1675–1680.



[106] K. E. Price, L. H. Lucas, C. K. Larive. Anal. Bioanal. Chem. 2004, 378,1405–1407.

[107] K. F. Morris, B. J. Cutak, A. M. Dixon, C. K. Larive. Anal. Chem. 1999, 71,5315–5321.

[108] A. J. Simpson, D. J. McNally, M. J. Simpson. Prog. Nucl. Magn. Reson.Spectrosc. 2011, 58, 97–175.

[109] L. Ma, E. Philipp, J. J. Led. J. Biomol. NMR 2001, 19, 199.[110] V. Borlototti, P. Fantazzini, M. Gombia, D. Greco, G. Rinaldin, S. Sykora.

J. Magn. Reson. 2010, 206, 219.[111] W. H. Otto, C. K. Larive. J. Magn. Reson. 2001, 153, 273–276.[112] M. Mayer, B. Meyer. Angew. Chem. Int. Ed. 1999, 38(12), 1784–1788.[113] B. Claasen, M. Axmann, R. Meinecke, B. Meyer. J. Am. Chem. Soc. 2005,

127, 916–919.[114] A. Shirzadi, M. J. Simpson, R. Kumar, A. J. Baer, Y. Xu, A. J. Simpson.

Environ. Sci. Technol. 2008, 42, 5514–5520.[115] J. G. Longsaffe, M. J. Simpson, W. Maas, A. J. Simpson. Environ. Sci.

Technol. 2010, 44, 5476–5482.[116] J. G. Longstaffe, A. J. Simpson. Env. Tox. Chem. 2011, 30(8), 1745–1753.[117] M. J. Duer, in Solid-state NMR Spectroscopy: Principles and Applications

(1st edn) (Ed: M. J. Duer), Blackwell Science, Oxford, 2002, pp. 73–110.[118] J. Peuravuori. Anal. Chim. Acta 2001, 429, 75–89.[119] R. Ahmad, R. S. Kookana, A. M. Alston, J. O. Skjemstad. Environ. Sci.

Technol. 2001, 35, 878–884.[120] F. Fava, A. Piccolo. Biotechnol. Bioeng. 2002, 77, 204–211.[121] P. J. Mitchell, M. J. Simpson. Environ. Sci. Technol. 2013, 47, 412–419.[122] J. D. Mao, L. S. Hundal, M. L. Thompson, K. Schmidt-Rohr. Environ. Sci.

Technol. 2002, 36, 929–936.


[123] J. D. Mao, K. D. Schmidt-Rohr. Environ. Sci. Technol. 2006, 9,1751–1756.

[124] H. Sun, D. Zhu, J. Mao. Environ. Toxicol.Chem. 2008, 27(12), 2449–2456.[125] J. D. Mao, K. Schmidt-Rohr. Solid State Nucl. Magn. 2004, 26, 36–45.[126] R. L. Johnson, K. Schmidt-Rohr. J. Magn. Reson. 2014, 239, 44–49.[127] X. Cao, C. Lattao, J. P. Pignatello, J. Mao, K. Schmidt-Rohr. Environ. Sci.

Technol. 2014, 48, 8645� 8652.[128] J. G. Longstaffe, D. Courtier-Murias, R. Soong, M. J. Simpson,

W. E. Maas, M. Fey, H. Hutchins, S. Krishnamurthy, J. Struppe,M. Alaee, R. Kumar,M. Monette, H. J. Stronks, A. J. Simpson. Environ.Sci. Technol. 2012, 46, 10508� 10513.

[129] D. Courtier-Murias, H. Farooq, H. Masoom, A. Botana, R. Soong,J. G. Longstaffe, M. J. Simpson, W. E. Maas, M. Fey, B. Andrews,J. Struppe, H. Hutchins, S. Krishnamurthy, R. Kumar, M. J. Monette,H. J. Stronks, A. Hume, A. J. Simpson. J. Magn. Reson. 2012, 217,61� 76.

[130] A. J. Simpson, D. Courtier-Murias, J. G. Longstaffe, H. Masoom,R. Soong, L. Lam, A. Sutrisno, H. Farooq, M. J. Simpson, W. E. Maas,M. Fey, B. Andrew, J. Struppe, H. Hutchins, S. Krishnamurthy,R. Kumar, M. Monette, H. J. Stronks. eMagRes 2013, 2(3), 399–414.

[131] F. D. Doty, G. Entzminger, A. Y. Yang. Concept. Magnetic Res. 1998, 10,239–260.

[132] X. Feng, A. J. Simpson, M. J. Simpson. Environ. Sci. Technol. 2006, 40,3260–3266.

[133] K. W. Fomba, P. Galvosas, U. Roland, J. Kaerger, F. D. Kopinke. Environ.Sci. Technol. 2011, 45, 5164–5169.


interactions between natural organic matter and organic ... · for both solution-state and...

Documents