w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/watres

Effects of charging on the chromophores of dissolved organic matter from the Rio Negro basin Mingquan Yan a,*, Gregory V. Korshin b, Francis Claret f, Jean-Philippe Croue´ e, Massimiliano Fabbricino c, Herve´ Gallard e, Thorsten Scha¨fer g, Marc F. Benedetti d a

Department of Environmental Engineering, Peking University, Key Laboratory of Water and Sediment Sciences, Ministry of Education, Beijing 100871, China b Department of Civil and Environmental Engineering, University of Washington, Seattle, WA 98195-2700, United States c Dipartimento di Ingegneria Idraulica ed Ambientale “Girolamo Ippolito”, Universita´ degli Studi di Napoli Federico II, Via Claudio 21, 80125 Naples, Italy d Institut de Physique du Globe de Paris, Sorbonne Paris Cite´, Universite´ Paris-Diderot, UMR CNRS 7154, Paris, France e Equipe Chimie de l’Eau et Traitement des Eaux, Institut de Chimie des Milieux et Mate´riaux de Poitiers, UMR 7285, CNRS, Ecole Nationale Supe´rieure d’Inge´nieurs de Poitiers, Universite´ de Poitiers, 86022 Poitiers Cedex, France f Bureau des Recherches Ge´ologiques et Minie`res, Environment and Process Division 3, Avenue Claude Guillemin, F-45060 Orleans Cedex 2, France g Forschungszentrum Karlsruhe, Institut fu¨r Nukleare Entsorgung (INE), P.O. Box 3640 76021, Karlsruhe, Germany

article info

abstract

Article history:

This study demonstrates that the deprotonation of dissolved organic matter (DOM) origi-

Received 14 December 2013

nating from a small creek characteristic for DOM-rich waters located in the Rio Negro basin

Received in revised form

can be quantified based on measurements of pH effects on its absorbance spectra. The

17 February 2014

method was ascertained by the data of Near-Edge X-Ray Absorbance Spectroscopy (NEX-

Accepted 17 March 2014

AFS), potentiometric titration to quantify the structural and compositional differences

Available online 19 April 2014

between the colloidal and hydrophobic fractions that contribute 91% of black-water creek DOM. Changes in the absorbance spectra of the DOM fractions caused by deprotonation

Keywords:

quantified via numeric deconvolution which indicated the presence of six well-resolved

Amazon river

Gaussian bands in the differential spectra. The emergence of these bands was deter-

Absorbance

mined to be associated with the engagement of carboxylic and phenolic functionalities and

Deprotonation

changes of inter-chromophore interactions in DOM molecules. Interpretation of the data

Dissolved organic matter

based on the NICA-Donnan approach showed that behavior of DOM chromophores was

NEXAFS

consistent with results of potentiometric titrations. Similar trends were observed for

NICA-Donnan model

changes of the spectral slope of the DOM absorbance spectra in the range of wavelengths 325e375 nm (DSlope325e375). The behavior of DSlope325e375 values was modeled based on the NICA-Donnan approach and correlated with potentiometrically-estimated charges

* Corresponding author. Department of Environmental Engineering, College of Environmental Sciences and Engineering, Peking University, Beijing 100871, China. Tel.: þ86 10 62755914x81; fax: þ86 10 62756526. E-mail addresses: [email protected], [email protected] (M. Yan). http://dx.doi.org/10.1016/j.watres.2014.03.044 0043-1354/ª 2014 Elsevier Ltd. All rights reserved.

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

155

attributed to the carboxylic and phenolic groups. The correlations between DSlope325e375 and charges of low- and high-affinity protonation-active groups in DOM were monotonic but not linear and had important differences between the colloidal and hydrophobic fractions. ª 2014 Elsevier Ltd. All rights reserved.

1.

Introduction

Soil organic matter and water-borne dissolved organic matter (DOM) are fundamentally important components of all environmental systems. Because the generation of aquatic DOM is affected by local biogeochemical conditions, many of its properties, e.g. its affinity to the proton and metal ions, sizes and charges of DOM molecules, their surface activity, the presence of redox active functionalities all of which frequently play a crucial role in environmental processes, are site-specific (Milne et al., 2001; Lenoir et al., 2010; Aeschbacher et al., 2012) and affected by seasonal cycles (Milne et al., 2001; Leenheer and Croue´, 2003; Ellis et al., 2012). Remarkable progress has been made in the exploration of DOM sitespecificity but effects of local environmental conditions and processes on its structure and reactivity remain to be understood on more detail. In this context, the understanding of properties of DOM from the Amazon River basin is especially important because this area contributes ca. 7% of the global flux of DOM to the oceans while transformations of this DOM have been shown to generate a considerable fraction of the regional flux of CO2 (Richey et al., 2002; Mayorga et al., 2005). Due these factors, elucidation of the intrinsic properties of Amazonian DOM as well as DOM from other fluvial systems is essential for understanding global and local carbon cycles (Hedges et al., 1997; Ellis et al., 2012; Ward et al., 2013). Extensive prior research has addressed many important details the genesis and fate of Amazonian DOM (Hedges et al., 1994; Patel et al., 1999; Krusche et al., 2002; Moreira-Turcq et al., 2003; Amaral et al., 2013; Ward et al., 2013), its role in the speciation of major and trace constituents (Maurice-Bourgoin et al., 2003; Rocha et al., 2003; Allard et al., 2004; de Oliveira et al., 2007; Fritsch et al., 2009; Perez et al., 2011; Kim et al., 2012), its photochemical transformations (Patel-Sorrentino et al., 2004; RodriguezZuniga et al., 2008; Remington et al., 2011; Amaral et al., 2013) and longitudinal, seasonal or anthropogenicallyinduced changes of its properties (McClain et al., 1997; Aufdenkampe et al., 2001; Bernardes et al., 2004; de Oliveira et al., 2007; Salisbury et al., 2011; Amaral et al., 2013). For instance, (Hedges et al., 2000) and ensuing studies (Aufdenkampe et al., 2007) presented a regional “chromatographic” model to account for the evolution of DOM from alluvial soils to the Amazon’s main stem and concluded that selective sorption of DOM onto minerals was the key process that affects the properties of different fractions of organic carbon of the rivers. Prior studies concerned with the evolution of DOM also concluded that further exploration of its composition and reactivity, especially its deprotonationeprotonation and charging processes is necessary to

understand the partitioning processes in soils and riparian zones (Amon and Benner, 1996; Hedges et al., 2000; Alasonati et al., 2010). While many advanced structure- and compound-specific ex situ methods, e.g. potentiometric titrations have been used to examine the composition, genesis and reactivity of DOM from the Amazon basin and other environmental systems (Hedges et al., 1994; Benner et al., 1995; Hedges et al., 2000; Aufdenkampe et al., 2001; Bernardes et al., 2004; Aufdenkampe et al., 2007; Mopper et al., 2007; Kujawinski et al., 2009; Ellis et al., 2012; Ward et al., 2013), results of these studies can be augmented by data of techniques that allow quantification of DOM properties in situ. Examination of absorbance and fluorescence of DOM can play this role since these methods use unaltered waters to produce optical spectra that are sensitive to DOM molecular weight, aromaticity and fluorophore and chromophore speciation (Hoge et al., 1993; Green and Blough, 1994; Peuravuori and Pihlaja, 1997; McKnight et al., 2001; Chen et al., 2003; Del Vecchio and Blough, 2004; Helms et al., 2008; Boyle et al., 2009). Studies that employed these techniques to examine Amazonian DOM have demonstrated the presence of fluorophore and chromophore groups associated with DOM molecules of varying sizes and chemical natures (Mounier et al., 1999; PatelSorrentino et al., 2002, 2004; Rodriguez-Zuniga et al., 2008). Variations of pH prominent in the Amazon basin (Do Nascimento et al., 2008) affect the fluorescence of Amazonian DOM (Mounier et al., 1999; Patel-Sorrentino et al., 2002) but the nature of such changes that are common to freshwater DOM (Tam and Sposito, 1993; Patel-Sorrentino et al., 2002; Spencer et al., 2007; Do Nascimento et al., 2008) has not been unambiguously determined. Effects of pH variations on the absorbance of DOM (Tam and Sposito, 1993; Andersen et al., 2000; Andersen and Gjessing, 2002; Spencer et al., 2007) have been addressed but because the absorbance spectra of DOM are featureless, these studies has been limited. DOM absorbance spectra can be made more feature-rich via the use of a differential approach that quantifies the evolution of the spectra as a function of any desired reaction parameter, for instance metal complexation, oxidant dose or pH (Korshin et al., 1999; Dryer et al., 2008; Janot et al., 2010; Yan et al., 2013b). In this paper, we present results of the examination of DOM from the basin of the Rio Negro River, one of the most important tributaries of the Amazon, using the method of differential absorbance (DA) and compare its data with those generated using potentiometric titrations and structuresensitive methods. This study’s objective is to establish in situ measurable spectroscopic markers of the important intrinsic properties of molecules Amazonian DOM without preconcentration and altering DOM properties, notably

156

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

relationships between pH, charge and, on the other hand, absorbance of these molecules and, ultimately, their reactivity in more complex systems, for instance in their interactions with mineral phases (Perez et al., 2011; Janot et al., 2012) that define the evolution of DOM in the soil-river continua.

2.

Materials and methods

2.1.

Isolation and fractionation of samples

DOM samples were collected from the Igarape´ Bonito, a small creek that flows into the Jau River. The GPS coordinates of the Jau station were S 01 52.3250 , W 61 35.0270 . The Jau River, a tributary of the Rio Negro, is fed by groundwater seepages and creeks similar to the Igarape´ Bonito (Alasonati et al., 2010). The dissolved organic carbon (DOC) concentration, pH and conductivity in the Igarape´ Bonito at the time of sampling were 56.6 mg L1, 3.6 and 65 ms cm1, respectively. Preparation of the samples included filtration through 1 mm GF/C filters followed by reverse osmosis (RO). RO concentrates were dialyzed against 0.1 M HCl and 0.2 M HF to isolate a colloidal (COLL) fraction with a 3500 D nominal cutoff from the other DOM fractions. The fraction passing through the dialysis membrane was fractionated using XAD-8 and XAD-4 resin columns to obtain the fractions of hydrophobic (HPO) and transphilic (TPH) DOM, respectively (Croue et al., 2000). Prior to this separation, DOM solutions were acidified to pH 2. After pumping each sample through the columns, they were rinsed with a formic acid solution at pH 2. DOM retained on them was eluted using an acetonitrile/water (75%/25% v/v) mixture. The eluent was evaporated under vacuum at 35e45  C. 200 mL of acetonitrile were added two or three times during the evaporation to eliminate traces of formic acid. The dialysis/XAD resins procedure isolated over 90% of the DOM in the RO concentrate. The COLL, HPO and TPH fractions constituted 51%, 40% and 9%, respectively, of the total weight of DOM extracted from the sample. DOC concentrations were determined with a Shimadzu TOC-Vcsh carbon analyzer. Effects of acidic agents, e.g. HCl and HF used in DOM isolation on intrinsic properties of the DOM were assumed to be negligible for the purposes in this study (Hamilton-Taylor et al., 2011; Ahmed et al., 2013) .

2.2.

Potentiometric titrations

Proton titrations were performed using a computer-controlled system in a thermostated vessel (25  C) under 99.99% nitrogen (Janot et al., 2010, 2012). DOM solutions were prepared at a 1 g L1 DOC concentration in the presence of a 0.04 M ionic strength. The pH was read using two pH Metrohm electrodes (6.0133.100) and a Ag/AgCl glass reference Metrohm electrode (6.0733.100) with a salt bridge (same as the solution). The electrodes were calibrated with CO2-free KOH base (0.099 M) and HNO3 (0.100 M) at a 0.1 M ionic strength. The pH values read by the duplicate electrodes were averaged. After addition of acid or base, the rate of drift for both electrodes was measured over 1 min and readings were accepted when the drift was less than 0.1 mV min1. For each data point the maximum time for monitoring pH drift was equal to 20 min.

2.3.

Carbon NEXAFS analysis

Carbon K-edge Near Edge X-Ray Absorption Fine Structure (NEXAFS) spectra (Jacobsen et al., 1991) were measured at the Scanning Transmission X-ray Microscopy (STXM) beamline X1A1 (NSLS) operated by the State University of New York at Stony Brook. STXM sample preparation was performed by drying 1 mL solution of resuspended freeze-dried DOM on a 100 nm thick Si3N4 window. The spectra were measured using the “point spectra” procedure consisting of measurement between 280 and 310 eV in 0.1 eV steps using 120 ms dwell time (Christl and Kretzschmar, 2007). Five consecutive point spectra of a region on the Si3N4 window without sample were averaged to obtain the I0 (E) information. I (E) is the average of 15 spectra taken on three different sample locations. Energy calibration of the spherical grating monochromator was achieved by using the photon energy of the CO2 gas adsorption band at 290.74 eV. To compare NEXAFS spectra, they were baseline corrected and normalized to 1 at 295 eV prior to peak fitting. The spectra were then deconvoluted as described in (Schafer et al., 2005). Precision of determinations of carbon functionalities’ contributions based on carbon NEXAFS data is estimated at 2%.

2.4.

Spectrophotometric titrations

The method of differential spectrophotometric titration of DOM and interpretation of its data have been described in sufficient detail in preceding relevant publications (Dryer et al., 2008; Janot et al., 2010; Yan et al., 2013b). DOM solutions were prepared at 2 or 5 mg L1 DOC concentrations in the presence of NaClO4 with ionic strength 0.04 M. Absorbance spectra were recorded with PerkineElmer Lambda 18 UV/Vis spectrometer. Dilution effects due to addition of acid and base were corrected for in the final data. Fitting of model calculations was performed using Matlab 2010a.

3.

Results and discussion

3.1.

NEXAFS data

The NEXAFS spectra for the COLL, HPO and TPH fractions of Igarape´ Bonito DOM are shown in Fig. S1 in the Supporting Information (SI) section. Results of their deconvolution that allowed determining contributions of different functionalities are compiled in Table S1 in the SI section. In view the NEXAFS spectra for HPO and TPH being similar, only the deconvoluted NEXAFS spectra for the two major fractions, COLL and HPO are shown in Fig. 1. These data indicate that the COLL fraction is richer in aromatic groups than HPO and TPH (34, 26 and 22%, respectively), while the contributions of carboxylic groups exhibit an opposite trend (30, 38 and 42%, respectively). Contributions of other functionalities (Fig. 1) do not exhibit prominent changes.

3.2.

Potentiometric data

Results of potentiometric titration for the COLL, HPO and TPH fractions of Igarape´ Bonito DOM are shown in Fig. 2. Modeling

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

157

Fig. 1 e C(1s) NEXAFS spectra of COLL and HPO fractions of Igarape´ Bonito DOM.

of the data shown in Fig. 2 based on the NICA-Donnan theory (Milne et al., 2001) to determine concentrations of the low and high affinity proton binding sites (denoted henceforth as LAS and HAS, respectively) and other parameters of DOM protonation are compiled in Table S2. They show that in the COLL fraction the HAS (largely associated with phenolic-type groups) are more abundant. The average values of the LAS

and HAS protonation constants of the examined fractions of Igarape´ Bonito DOM were determined to have identical average pK values (specifically, the pK values were fixed at average 4.43 and 8.10 respectively while the other 4 parameters were optimized). The heterogeneity parameters for the examined fractions were also close, having average values of 0.78 and 0.28 for the LAS (mostly carboxylic type groups) and HAS, respectively. The 0.28 value of the heterogeneity parameter for the HAS indicates a larger chemical heterogeneity. This will be explored in more detail based on the spectroscopic data. Trends in the concentrations of the HAS and LAS discerned based on the potentiometric data agreed with those indicated by the carbon NEXAFS data (Fig. 3). This observation reinforces the notion that while the protonation properties of the HPO and TPH fractions were very close, the colloidal fraction was quite distinct.

3.3.

Fig. 2 e Results of potentiometric titration for COLL and HPO fractions of Igarape´ Bonito DOM and its modeling using NICA-Donnan model.

Differential absorbance results

Absorbance experiments were carried out for the COLL and HPO fractions. The TPH fraction was not studied by optical spectroscopy because it was unavailable for the differential absorbance experiments. In addition, its properties determined by NEXAFS and potentiometric titrations were very close to those of the HPO fraction and it had a small contribution (i.e. 9%) to the total concentration of organic carbon in Igarape´ Bonito DOM. The absorbance of both COLL and HPO fractions changed in response to pH variations. As the pH increased, the absorbance increased at all wavelengths, but the spectra remained

158

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

Fig. 3 e Correlations between percentage of carboxylic and phenolic carbon estimated based on carbon NEXAFS data and concentrations of the protonation-active carboxylic and phenolic sites determined based on the results of potentiometry.

featureless, as demonstrated in Fig. S2 in the SI section. To obtain more information on how pH variations affected the behavior of chromophore groups of the DOM, DA spectra were calculated using the equation below: DAl ðpHÞ ¼

 i 1 h Al ðpHÞ  Al pHref DOC$l

(1)

In this equation, DAl(pH) is the differential absorbance at any specified wavelength, l is the optical cell’s length, Al(pH) and Al(pHref) are absorbances at any desired pH value and a reference pH, respectively. The spectrum recorded at pH 3.1 was used as the reference, respectively. The intensity of the DA spectra increased monotonically with the pH, and discernible features that had different prominence for the COLL and HPO fractions were present in them (Fig. 4). Specifically, peaks with maxima located approximately at 240, 280 and 315 nm, as well a broad structure located at >350 nm were prominent. Normalized (by the maximum of their intensity at l > 350 nm) DA spectra (Fig. S3) were calculated using the reference spectra recorded at pH 3.3 and 8.3 corresponding to the deprotonation of LAS and HAS chromophores, respectively (Dryer et al., 2008). The data show that the deprotonation of LAS chromophores of Igarape´ Bonito DOM is associated with the emergence of bands with maxima located at 240, 278, 315 and 375 nm (Fig. S3a). These features are similar for the HPO and COLL fractions, except that located for HPO at 240 nm. The most intense feature in the normalized DA spectra of the LAS chromophores is located at l > 350 nm, that is in the range where manifestations of inter-chromophore interactions are located (Del Vecchio and Blough, 2004; Dryer et al., 2008). The engagement of the HAS chromophores is accompanied by the development of two bands with maxima close 247 nm and 315 nm, and a much stronger feature with a maximum located at 385e390 nm; that feature was especially prominent for the COLL fraction (Fig. S3b). This is likely to indicate higher importance of inter-chromophore interactions

Fig. 4 e Development of DOC-normalized pH-differential absorbance spectra of the COLL (a) and HPO (b) fractions of Igarape´ Bonito DOM. Reference pH values 3.1.

in the molecules of the COLL fractions of Igarape´ Bonito DOM due to their larger molecular weights (Green and Blough, 1994; Del Vecchio and Blough, 2004). To examine the structure of the pH-differential spectra of both COLL and HPO fractions in more detail, they were deconvoluted to determine contributions of discrete bands constituting them. In agreement with the approach presented in prior research (Korshin et al., 1997; Gege, 2000; Yan et al., 2013b), such bands were assumed to have a Gaussian shape when represented against photon energy (measured in eV), calculated as EðeVÞ ¼

1240 lðnmÞ

(2)

The fitting procedure allowed determining such characteristics of each Gaussian bands as the location of its maximum (E0i), width (Wi) and intensity at E ¼ E0i (A0i). The overall differential spectra (DA(E)) were modeled as DAðEÞ ¼

X i

DA0i exp 

2 !  E  E0i pffiffiffi Wi 2

(3)

Selected results of the application of this concept to the modeling of pH-differential spectra of the COLL and HPO

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

fractions of Igarape´ Bonito DOM are shown in Fig. 5. They demonstrate a very close fit between the observed and modeled spectra (R2 > 0.995). The pH-differential spectra of Igarape´ Bonito DOM were determined to comprise six Gaussian components whose intensities changed with pH while positions of their maxima were practically constant (Table S3). The maxima of the bands referred to, as in prior relevant publications (Yan et al., 2013b) as A0, A1, A2, A3, A4 and A5, were located at ca. 6.00 eV (207 nm), 5.07 eV (245 nm) 4.45 eV (280 nm), 3.97 eV (313 nm), 3.25 eV (380 nm) and 2.33 eV (530 nm). The locations of the maxima of these Gaussian

159

bands A1, A2, A3 and A4 were close to those of the bands observed previously for SRFA (Yan et al., 2013b) while slight differences were observed for bands A0 and A5. Because the properties of band A0 were difficult to estimate due to the presence of spectroscopic interferences from hydroxyl ions and the intensity of band A5 was lower than that of all other bands, only the data for A1, A2, A3 and A4 will be discussed henceforth. Changes of the intensity of band A1, A2, A3 and A4 with pH for COLL and HPO are presented in Fig. S4. Although the relative contributions of each Gaussian band that comprise

Fig. 5 e Gaussian band fitting of the DOC-normalized differential spectra of COLL and HPO fractions of Igarape´ Bonito DOM at selected pH. (a), (b) and (c) for COLL at pH7.1, pH9.1 and pH11.0; (d), (e) and (f) for HPO at pH7.0, pH9.0 and pH11.0.

160

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

the pH-differential spectra of Igarape´ Bonito DOM were different (Fig. 5 and Fig. S4), trends in the evolution of the intensity of each band caused variations of pH were similar to those typically seen in potentiometric titrations. This was interpreted to indicate that these bands comprise contributions of chromophores associated with both LAS and HAS groups. Prior research has demonstrated that bands A4 especially sensitive to the changes of the intrinsic chemistry DOM molecules caused by their deprotonation or complexation with metal cations (Yan et al., 2013b). However, the relatively low intensity of these bands, especially that of band A5 necessitate that the absorbance spectra of DOM be log-transformed and differential spectra be calculated using the logtransformed data using the expression below: DLnAi ðlÞ ¼ LnAi ðlÞ  LnAref ðlÞ

(4)

The differential log-transformed absorbance spectra of the COLL and HPO fractions at selected pH values are presented in Fig. 6. Similarly to the linear differential spectra show in Fig. 4, the intensity of the log-transformed spectra increases with pH. This increase is especially prominent for wavelengths

>320 nm. The prominence of the features observed in the logtransformed differential spectra and located at wavelengths >320 nm is associated with the development of Gaussian bands A4 that are relatively inconspicuous in terms of their absolute intensities, compared with the intensities of bands A1, A2 and A3. In accord with the approach presented in prior research (Yan et al., 2013a), we used the differential spectral slope in the range of wavelengths 325e375 nm (DSlope325e375) measured at varying pHs examine effects of pH on the protonation of Igarape´ Bonito DOM. DSlope325e375were calculated using the formula given below: DSlope325375 ¼ Slope325375;i  Slope325375;ref

(5)

In this expression, Slope325e375,i and Slope325e375,ref are the slopes of the linear fit of the log-transformed DOM absorbance in the wavelengths region 325e375 nm for any selected solution conditions and reference, respectively. The choice of DSlope325e375 parameter reflects the observation that the intensity of the log-transformed spectra undergoes rapid changes in the 325e375 nm region corresponding to the location of Gaussian band A4 and these changes are nearly linear vs. the observation wavelength. To determine whether the evolution of the DA spectra of Igarape´ Bonito DOM could be described by the NICA-Donnan theory developed to model the potentiometric behavior of DOM, we applied the equation developed in literature (Kinniburgh et al., 1999; Ritchie and Perdue, 2003; Dryer et al., 2008; Janot et al., 2010) to model the evolution of differential slopes of the absorbance of Igarape´ Bonito samples: "

DSlopeLAS ðlÞ DSlopeHAS ðlÞ DSlopepH ðlÞ ¼   mLAS þ  þ ~ ~HAS ½Hþ  mHAS 1þ KLAS ½H  1þ K 2

# 3

DSlopeLAS ðlÞ DSlopeHAS ðlÞ 7 6 4 mLAS þ mHAS 5   þ ~ ~ HAS ½Hþ  1þ KLAS ½H ref 1þ K ref

Fig. 6 e Differential log-transformed absorbance spectra of COLL (a) and HPO (b) fractions of Igarape´ Bonito DOM. Reference pH values 3.1.

(6)

where DSlopeLAS(l) and DSlopeHAS(l) correspond to the maximum change of absorbance associated with the deprotonation of the LAS (mostly carboxylic) and HAS (mostly phenolic) groups, respectively, DSlopeLAS(l) and DSlopeHAS(l) referred to DSlope value in the range of wavelengths ~ HAS are the median values ~ LAS and K 325e375 nm in this study. K of the protons affinity distributions for these groups, mLAS and mHAS define the width of these distributions and are measures of the heterogeneity of DOM (Milne et al., 2001; Dryer et al., 2008). The behavior of DSlope325e375 values for COLL and HPO vs. pH and their fitting is shown in Fig. 7 and Table S2. The data show that excellent agreement could be reached between the DSlope325e375 data for both COLL and HPO samples and the predictions made based on the revised NICA-Donnan model (R2 > 0.99). The above observations of the highly interpretable response of the chromophores of Igarape´ Bonito DOM to their deprotonation are enhanced by the observation that changes of the spectral slope of the examined DOM fractions are correlated with the charges of HAS and LAS groups in DOM molecules calculated using the NICA-Donnan parameters

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

161

Fig. 7 e Comparison of effects of pH on the experimental differential log-transformed spectral slope in the range of wavelengths 325e375 nm (DSlope325e375) and their NICAbased fitting for HPO and COLL fractions of Igarape´ Bonito DOM.

obtained using conventional potentiometric titration in Table S2. The correlations between LAS and HAS charges and changes of the spectral slope are compared in Fig. 8b and c, respectively. It demonstrates that while both QLAS and QHAS are correlated with DSlope values, these correlations have distinct differences for the LAS and HAS groups. In the former case, the correlations between QLAS and DSlope values are nearly linear but the slope of the correlations are different for the HPO and COLL fractions reflecting the difference in the responses of carboxylic-type chromophores in these fractions to the accumulation of charge. On the other hand, correlations between QHAS and changes of the DSlope for COLL and HPO fractions are similar but the response of the HAS phenolictype chromophores is characterized by two distinct ranges. In the range of QHAS charges from 0 to ca. 0.7 meq g1, changes of the slope are relatively small while for QHAS values above ca. 0.7 meq g1, spectral slopes of DOM changes prominently but the accumulation of charge is less rapid. This is likely to be indicative of the engagement of distinct sub-groups of the HAS chromophores. Their properties need to be examined in more detail in future studies. Given the complexity of the linear or log-transformed DA spectra, the NICA-Donnan approach to interpret them appears to be an excellent approximation although it does not address the nature of the distinct spectroscopic features, for instance Gaussian bands A1 to A5. The presence of these bands, their association with the charging of DOM molecules in the HPO and COLL fractions highlights the need to expand the examination of responses of chromophores and fluorophores of a wider range of DOM. On the other hand, this can indicate that mechanisms other the deprotonation of the discrete operationally defined LAS and HAS chromophores may define the evolution of the pH-differential spectra. As mentioned above, these alternative mechanisms are likely to reflect changes of inter-chromophore interactions (Hoge et al., 1993; Green and Blough, 1994;

Fig. 8 e Comparison of potentiometric and spectrophotometric results (a) and LAS (b) and HAS (c) in HPO and COLL fractions of Igarape´ Bonito DOM predicted by NICA-Donnan model using the parameters shown in Table S2.

Korshin et al., 1999; Del Vecchio and Blough, 2004) that depend on the molecular weight and conformational status of DOM molecules. pH-differential spectra can also reflect responses of specific functionalities, for instance lignins, terpenoids, bound proteins and others whose deprotonation may yield a distinct signal in the differential spectroscopy.

162

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

More work is needed to explore this issue further as well as to quantify effects of other system parameters (e.g., those of ionic strength) on chromophores in DOM of varying provenance.

4.

Conclusions

The data presented above and their interpretation support the following conclusions: (1) The deprotonation of DOM originating from the Rio Negro basin can be quantified based on measurements of pH effects on its absorbance spectra. (2) Changes in the absorbance spectra of the DOM fractions caused by deprotonation quantified via numeric deconvolution which indicated the presence of six wellresolved Gaussian bands in the differential spectra. The emergence of these bands was determined to be associated with the engagement of carboxylic and phenolic functionalities and changes of inter-chromophore interactions in DOM molecules. (3) Interpretation of the data of spectrophotometric titrations based on the NICA-Donnan approach showed that behavior of DOM chromophores was consistent with results of conventional potentiometric titrations. (4) The behavior of DSlope325e375 values was correlated with charges attributed to the carboxylic and phenolic groups. The correlations between DSlope325e375 and charges of low- and high-affinity protonation-active groups in DOM were monotonic but not linear and had important differences between the HPO and COLL fractions.

Acknowledgments This study has been partially supported by National Science Foundation (grants 0504447 and 0931676). G. Korshin expresses gratitude to l’Institut de Physique du Globe de Paris/ Universite´ de Paris VII and French INSU-CNRS program ECCO for support of his work in Paris, and to the Foreign Experts Program of China (GDW20131100008) for support of his work at Peking University. Use of the National Synchrotron Light Source, Brookhaven National Laboratory, was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-98CH10886. Data were collected using the X1A STXM developed by the group of Janoz Kirz and Chris Jacobsen at SUNY Stony Brook, with support from the Office of Biological and Environmental Research, US. DoE under contract DE-FG02-89ER60858, and from the NSF under grant DBI-9605045.

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.watres.2014.03.044.

references

Aeschbacher, M., Brunner, S.H., Schwarzenbach, R.P., Sander, M., 2012. Assessing the effect of humic acid redox state on organic pollutant sorption by combined electrochemical reduction and sorption experiments. Environ. Sci. Technol. 46, 3882e3890. Ahmed, I.A.M., Hamilton-Taylor, J., Lofts, S., Meeussen, J.C.L., Lin, C., Zhang, H., Davison, W., 2013. Testing copperspeciation predictions in freshwaters over a wide range of metal-organic matter ratios. Environ. Sci. Technol. 47, 1487e1495. Alasonati, E., Slaveykova, V.I., Gallard, H., Croue, J.P., Benedetti, M.F., 2010. Characterization of the colloidal organic matter from the Amazonian basin by asymmetrical flow fieldflow fractionation and size exclusion chromatography. Water Res. 44, 223e231. Allard, T., Menguy, N., Salomon, J., Calligaro, T., Weber, T., Calas, G., Benedetti, M.F., 2004. Revealing forms of iron in river-borne material from major tropical rivers of the Amazon Basin (Brazil). Geochim. Cosmochim. Ac 68, 3079e3094. Amaral, J.H.F., Suhett, A.L., Melo, S., Farjalla, V.F., 2013. Seasonal variation and interaction of photodegradation and microbial metabolism of DOC in black water Amazonian ecosystems. Aquat. Microb. Ecol. 70, 157e168. Amon, R.M.W., Benner, R., 1996. Bacterial utilization of different size classes of dissolved organic matter. Limnol. Oceanogr. 41, 41e51. Andersen, D.O., Alberts, J.J., Takacs, M., 2000. Nature of natural organic matter (NOM) in acidified and limed surface waters. Water Res. 34, 266e278. Andersen, D.O., Gjessing, E.T., 2002. Natural organic matter (NOM) in a limed lake and its tributaries. Water Res. 36, 2372e2382. Aufdenkampe, A.K., Hedges, J.I., Richey, J.E., Krusche, A.V., Llerena, C.A., 2001. Sorptive fractionation of dissolved organic nitrogen and amino acids onto fine sediments within the Amazon Basin. Limnol. Oceanogr. 46, 1921e1935. Aufdenkampe, A.K., Mayorga, E., Hedges, J.I., Llerena, C., Quay, P.D., Gudeman, J., Krusche, A.V., Richey, J.E., 2007. Organic matter in the Peruvian headwaters of the Amazon: compositional evolution from the Andes to the lowland Amazon mainstem. Org. Geochem. 38, 337e364. Benner, R., Opsahl, S., ChinLeo, G., Richey, J.E., Forsberg, B.R., 1995. Bacterial carbon metabolism in the Amazon River system. Limnol. Oceanogr. 40, 1262e1270. Bernardes, M.C., Martinelli, L.A., Krusche, A.V., Gudeman, J., Moreira, M., Victoria, R.L., Ometto, J., Ballester, M.V.R., Aufdenkampe, A.K., Richey, J.E., Hedges, J.I., 2004. Riverine organic matter composition as a function of land use changes, Southwest Amazon. Ecol. Appl. 14, S263eS279. Boyle, E.S., Guerriero, N., Thiallet, A., Del Vecchio, R., Blough, N.V., 2009. Optical properties of humic substances and CDOM: relation to structure. Environ. Sci. Technol. 43, 2262e2268. Chen, J., LeBoef, E.J., Dai, S., Gu, B.H., 2003. Fluorescence spectroscopic studies of natural organic matter fractions. Chemosphere 50, 639e647. Christl, I., Kretzschmar, R., 2007. C-1s NEXAFS spectroscopy reveals chemical fractionation of humic acid by cationinduced coagulation. Environ. Sci. Technol. 41, 1915e1920. Croue, J.P., Korshin, G.V., Benjamin, M.M., Foundation, A.R., 2000. Characterization of Natural Organic Matter in Drinking Water. AWWA Research Foundation and American Water Works Association. de Oliveira, L.C., Sargentini, T., Rosa, A.H., Rocha, J.C., Simoes, M.L., Martin-Neto, L., da Silva, W.T.L., Serudo, R.L., 2007. The influence of seasonalness on the structural

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

characteristics of aquatic humic substances extracted from Negro river (Amazon state) waters: Interactions with Hg(II). J. Braz. Chem. Soc. 18, 860e868. Del Vecchio, R., Blough, N.V., 2004. On the origin of the optical properties of humic substances. Environ. Sci. Technol. 38, 3885e3891. Do Nascimento, N.R., Fritsch, E., Bueno, G.T., Bardy, M., Grimaldi, C., Melfi, A.J., 2008. Podzolization as a deferralitization process: dynamics and chemistry of ground and surface waters in an Acrisol-Podzol sequence of the upper Amazon Basin. Eur. J. Soil. Sci. 59, 911e924. Dryer, D.J., Korshin, G.V., Fabbricino, M., 2008. In situ examination of the protonation behavior of fulvic acids using differential absorbance spectroscopy. Environ. Sci. Technol. 42, 6644e6649. Ellis, E.E., Keil, R.G., Ingalls, A.E., Richey, J.E., Alin, S.R., 2012. Seasonal variability in the sources of particulate organic matter of the Mekong River as discerned by elemental and lignin analyses. J. Geophys. Res-Biogeo 117. Fritsch, E., Allard, T., Benedetti, M.F., Bardy, M., do Nascimento, N.R., Li, Y., Calas, G., 2009. Organic complexation and translocation of ferric iron in podzols of the Negro River watershed. Separation of secondary Fe species from Al species. Geochim. Cosmochim. Ac 73, 1813e1825. Gege, P., 2000. Gaussian model for yellow substance absorption spectra. In: Proc. Ocean Optics XV, Monaco 16e20 October 2000. Green, S.A., Blough, N.V., 1994. Optical-absorption and fluorescence properties of chromophoric dissolved organicmatter in natural-waters. Limnol. Oceanogr. 39, 1903e1916. Hamilton-Taylor, J., Ahmed, I.A.M., Davison, W., Zhang, H., 2011. How well can we predict and measure metal speciation in freshwaters? Environ. Chem. 8, 461e465. Hedges, J.I., Cowie, G.L., Richey, J.E., Quay, P.D., Benner, R., Strom, M., Forsberg, B.R., 1994. Origins and processing of organic-matter in the Amazon River as indicated by carbohydrates and amino-acids. Limnol. Oceanogr. 39, 743e761. Hedges, J.I., Keil, R.G., Benner, R., 1997. What happens to terrestrial organic matter in the ocean? Org. Geochem 27, 195e212. Hedges, J.I., Mayorga, E., Tsamakis, E., McClain, M.E., Aufdenkampe, A., Quay, P., Richey, J.E., Benner, R., Opsahl, S., Black, B., Pimentel, T., Quintanilla, J., Maurice, L., 2000. Organic matter in Bolivian tributaries of the Amazon River: a comparison to the lower mainstream. Limnol. Oceanogr. 45, 1449e1466. Helms, J.R., Stubbins, A., Ritchie, J.D., Minor, E.C., Kieber, D.J., Mopper, K., 2008. Absorption spectral slopes and slope ratios as indicators of molecular weight, source, and photobleaching of chromophoric dissolved organic matter. Limnol. Oceanogr. 53, 955e969. Hoge, F.E., Vodacek, A., Blough, N.V., 1993. Inherent optical properties of the ocean: retrieval of the absorption coefficient of chromophoric dissolved organic matter from airborne laser spectral fluorescence measurements. Limnol. Oceanogr. 38, 1394e1402. Jacobsen, C., Williams, S., Anderson, E., Browne, M.T., Buckley, C.J., Kern, D., Kirz, J., Rivers, M., Zhang, X., 1991. Diffraction-limited imaging in a scanning-transmission X-ray microscope. Opt. Commun. 86, 351e364. Janot, N., Reiller, P.E., Korshin, G.V., Benedetti, M.F., 2010. Using spectrophotometric titrations to characterize humic acid reactivity at environmental concentrations. Environ. Sci. Technol. 44, 6782e6788. Janot, N., Reiller, P.E., Zheng, X., Croue, J.P., Benedetti, M.F., 2012. Characterization of humic acid reactivity modifications due to adsorption onto alpha-Al2O3. Water Res. 46, 731e740.

163

Kim, J.H., Zell, C., Moreira-Turcq, P., Perez, M.A.P., Abril, G., Mortillaro, J.M., Weijers, J.W.H., Meziane, T., Damste, J.S.S., 2012. Tracing soil organic carbon in the lower Amazon River and its tributaries using GDGT distributions and bulk organic matter properties. Geochim. Cosmochim. Ac 90, 163e180. Kinniburgh, D.G., van Riemsdijk, W.H., Koopal, L.K., Borkovec, M., Benedetti, M.F., Avena, M.J., 1999. Ion binding to natural organic matter: competition, heterogeneity, stoichiometry and thermodynamic consistency. Colloid Surf. 151, 147e166. Korshin, G.V., Benjamin, M.M., Li, C.W., 1999. Use of differential spectroscopy to evaluate the structure and reactivity of humics. Water Sci. Technol. 40, 9e16. Korshin, G.V., Li, C.W., Benjamin, M.M., 1997. Monitoring the properties of natural organic matter through UV spectroscopy: a consistent theory. Water Res. 31, 1787e1795. Krusche, A.V., Martinelli, L.A., Victoria, R.L., Bernardes, M., de Camargo, P.B., Ballester, M.V., Trumbore, S.E., 2002. Composition of particulate and dissolved organic matter in a disturbed watershed of southeast Brazil (Piracicaba River basin). Water Res. 36, 2743e2752. Kujawinski, E.B., Longnecker, K., Blough, N.V., Del Vecchio, R., Finlay, L., Kitner, J.B., Giovannoni, S.J., 2009. Identification of possible source markers in marine dissolved organic matter using ultrahigh resolution mass spectrometry. Geochim. Cosmochim. Ac 73, 4384e4399. Leenheer, J.A., Croue´, J.-P., 2003. Peer reviewed: characterizing aquatic dissolved organic matter. Environ. Sci. Technol. 37, 18Ae26A. Lenoir, T., Matynia, A., Manceau, A., 2010. Convergenceoptimized procedure for applying the NICA-Donnan model to potentiometric titrations of humic substances. Environ. Sci. Technol. 44, 6221e6227. Maurice-Bourgoin, L., Quemerais, B., Moreira-Turcq, P., Seyler, P., 2003. Transport, distribution and speciation of mercury in the Amazon River at the confluence of black and white waters of the Negro and Solimoes Rivers. Hydrol. Process 17, 1405e1417. Mayorga, E., Aufdenkampe, A.K., Masiello, C.A., Krusche, A.V., Hedges, J.I., Quay, P.D., Richey, J.E., Brown, T.A., 2005. Young organic matter as a source of carbon dioxide outgassing from Amazonian rivers. Nature 436, 538e541. McClain, M.E., Richey, J.E., Brandes, J.A., Pimentel, T.P., 1997. Dissolved organic matter and terrestrial-lotic linkages in the central Amazon basin of Brazil. Glob. Biogeochem. Cy 11, 295e311. McKnight, D.M., Boyer, E.W., Westerhoff, P.K., Doran, P.T., Kulbe, T., Andersen, D.T., 2001. Spectrofluorometric characterization of dissolved organic matter for indication of precursor organic material and aromaticity. Limnol. Oceanogr. 46, 38e48. Milne, C.J., Kinniburgh, D.G., Tipping, E., 2001. Generic NICADonnan model parameters for proton binding by humic substances. Environ. Sci. Technol. 35, 2049e2059. Mopper, K., Stubbins, A., Ritchie, J.D., Bialk, H.M., Hatcher, P.G., 2007. Advanced instrumental approaches for characterization of marine dissolved organic matter: extraction techniques, mass spectrometry, and nuclear magnetic resonance spectroscopy. Chem. Rev. 107, 419e442. Moreira-Turcq, P.F., Seyler, P., Guyot, J.L., Etcheber, H., 2003. Characteristics of organic matter in the mixing zone of the Rio Negro and Rio Solimoes of the Amazon River. Hydrol. Process 17, 1393e1404. Mounier, S., Patel, N., Quilici, L., Benaim, J.Y., Benamou, C., 1999. Three-dimensional fluorescence of the dissolved organ carbon in the Amazon River. Water Res. 33, 1523e1533. Patel-Sorrentino, N., Mounier, S., Benaim, J.Y., 2002. Excitationemission fluorescence matrix to study pH influence on organic matter fluorescence in the Amazon basin rivers. Water Res. 36, 2571e2581.

164

w a t e r r e s e a r c h 5 9 ( 2 0 1 4 ) 1 5 4 e1 6 4

Patel-Sorrentino, N., Mounier, S., Lucas, Y., Benaim, J.Y., 2004. Effects of UV-visible irradiation on natural organic matter from the Amazon basin. Sci. Total Environ. 321, 231e239. Patel, N., Mounier, S., Guyot, J.L., Benamou, C., Benaim, J.Y., 1999. Fluxes of dissolved and colloidal organic carbon, along the Purus and Amazonas rivers ( Brazil). Sci. Total Environ. 229, 53e64. Perez, M.A.P., Moreira-Turcq, P., Gallard, H., Allard, T., Benedetti, M.F., 2011. Dissolved organic matter dynamic in the Amazon basin: sorption by mineral surfaces. Chem. Geol. 286, 158e168. Peuravuori, J., Pihlaja, K., 1997. Molecular size distribution and spectroscopic properties of aquatic humic substances. Anal. Chim. Acta 337, 133e149. Remington, S., Krusche, A., Richey, J., 2011. Effects of DOM photochemistry on bacterial metabolism and CO2 evasion during falling water in a humic and a whitewater river in the Brazilian Amazon. Biogeochemistry 105, 185e200. Richey, J.E., Melack, J.M., Aufdenkampe, A.K., Ballester, V.M., Hess, L.L., 2002. Outgassing from Amazonian rivers and wetlands as a large tropical source of atmospheric CO2. Nature 416, 617e620. Ritchie, J.D., Perdue, E.M., 2003. Proton-binding study of standard and reference fulvic acids, humic acids, and natural organic matter. Geochim. Cosmochim. Ac 67, 85e96. Rocha, J.C., Sargentini, E., Zara, L.F., Rosa, A.H., dos Santos, A., Burba, P., 2003. Reduction of mercury(II) by tropical river humic substances (Rio Negro) e part II. Influence of structural features (molecular size, aromaticity, phenolic groups, organically bound sulfur). Talanta 61, 699e707. Rodriguez-Zuniga, U.F., Milori, D., Da Silva, W.T.L., MartinNeto, L., Oliveira, L.C., Rocha, J.C., 2008. Changes in optical properties caused by UV-irradiation of aquatic humic

substances from the amazon river basin: seasonal variability evaluation. Environ. Sci. Technol. 42, 1948e1953. Salisbury, J., Vandemark, D., Campbell, J., Hunt, C., Wisser, D., Reul, N., Chapron, B., 2011. Spatial and temporal coherence between Amazon River discharge, salinity, and light absorption by colored organic carbon in western tropical Atlantic surface waters. J. Geophys Res-Oceans 116. Schafer, T., Buckau, G., Artinger, R., Kim, J.I., Geyer, S., Wolf, M., Bleam, W.F., Wirick, S., Jacobsen, C., 2005. Origin and mobility of fulvic acids in the Gorleben aquifer system: implications from isotopic data and carbon/sulfur XANES. Org. Geochem. 36, 567e582. Spencer, R.G., Bolton, L., Baker, A., 2007. Freeze/thaw and pH effects on freshwater dissolved organic matter fluorescence and absorbance properties from a number of UK locations. Water Res. 41, 2941e2950. Tam, S.C., Sposito, G., 1993. Fluorescence spectroscopy of aqueous pine litter extracts: effects of humification and aluminum complexation. J. Soil. Sci. 44, 513e524. Ward, N.D., Keil, R.G., Medeiros, P.M., Brito, D.C., Cunha, A.C., Dittmar, T., Yager, P.L., Krusche, A.V., Richey, J.E., 2013. Degradation of terrestrially derived macromolecules in the Amazon River. Nat. Geosci. 6, 530e533. Yan, M., Benedetti, M.F., Korshin, G.V., 2013a. Study of iron and aluminum binding to Suwannee River fulvic acid using absorbance and fluorescence spectroscopy: comparison of data interpretation based on NICA-Donnan and Stockholm humic models. Water Res. 47, 5439e5446. Yan, M., Dryer, D., Korshin, G.V., Benedetti, M.F., 2013b. In situ study of binding of copper by fulvic acid: comparison of differential absorbance data and model predictions. Water Res. 47, 588e596.