A comprehensive structural evaluation of humic substances using several fluorescence techniques before and after ozonation. Part I: structural characterization of humic substances.

Science of the Total Environment 476–477 (2014) 718–730

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

A comprehensive structural evaluation of humic substances using several fluorescence techniques before and after ozonation. Part I: Structural characterization of humic substances Francisco J. Rodríguez a,⁎, Patrick Schlenger b, María García-Valverde c a b c

Department of Chemistry, Higher Polytechnic School, University of Burgos, Av. Cantabria s/n, 09006 Burgos, Spain Department of Chemistry & Biology, Faculty of Mathematics and Natural Science, University of Wuppertal, Germany Department of Chemistry, Faculty of Sciences, University of Burgos, Pz. Misael Bañuelos s/n, 09001 Burgos, Spain

H I G H L I G H T S

G R A P H I C A L

• TLS, SFS, ESF, fluorescence index and λ0.5 are useful to characterize humic substances. • EEM spectra of natural humic substances show 2 peaks: A (230/437 nm) and C (335/460 nm). • Synchronous spectra allowed the identification of a protein-like peak (λsyn = 290 nm). • Good correlations were obtained between 13C NMR aromaticity and fluorescence index and λ0.5. • ALHA shows fluorescence spectra completely different to those of natural humic substances.

Total luminescence spectra (EEM contour map and 3-D spectrum) of the humic substance SUFA.

a r t i c l e

a b s t r a c t

i n f o

Article history: Received 3 August 2013 Received in revised form 26 November 2013 Accepted 29 November 2013 Available online 21 December 2013 Keywords: Humic substances Total luminescence spectroscopy Synchronous fluorescence spectroscopy Emission scan fluorescence Fluorescence index Lambda 0.5

A B S T R A C T

The main objective of this work (Part I) is to conduct a comprehensive structural characterization of humic substances, using all the current fluorescence techniques: emission scan fluorescence (ESF), synchronous fluorescence spectroscopy (SFS), total luminescence spectroscopy (TLS or EEM) through the use of both 2-D contour maps and 3-D plots, fluorescence index and the λ0.5 parameter. Four humic substances were studied in this work: three of them were provided by the International Humic Substances Society (Suwannee River Fulvic Acid Standard, Suwannee River Humic Acid Standard and Nordic Reservoir Fulvic Acid Reference) and the other one was a commercial humic acid widely used as a surrogate for aquatic humic substances in various studies (Aldrich Humic Acid: ALHA). The EEM spectra for the three natural aquatic substances were quite similar, showing two main peaks of maximum fluorescence intensity: one located in the ultraviolet region and centered at around Ex/Em values of 230/437 nm (peak A) and another one in the visible region, centered at around 335/ 460 nm (peak C); however, the EEM spectrum of ALHA is completely different to those of natural aquatic humic substances, presenting four poorly resolved main peaks with a high degree of spectral overlap, located at 260/462, 300/479, 365/483 and 450/524 nm. The synchronous spectra at Δλ = 18 and 44 nm (especially at Δλ = 18 nm) allowed the identification of a protein-like peak at λsyn around 290 nm, which was not detected in the EEM spectra; as it happened with EEM spectra, the synchronous spectra of ALHA are quite different from those of the aquatic humic substances, presenting a higher number of bands that suggest greater structural

⁎ Corresponding author. Tel.: +34 947258937; fax: +34 947258910. E-mail addresses: [email protected] (F.J. Rodríguez), [email protected] (P. Schlenger), [email protected] (M. García-Valverde). 0048-9697/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2013.11.150

F.J. Rodríguez et al. / Science of the Total Environment 476–477 (2014) 718–730

719

complexity and a higher degree of polydispersity. Good correlations were achieved between 13C NMR aromaticity and both fluorescence index and λ0.5 parameter. The different spectra presented by ALHA compared to those shown by the natural aquatic humic substances for all the fluorescence techniques studied suggest an important structural difference between them, which cast doubt on the use of commercial humic acids as surrogates for natural humic substances. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Aquatic natural organic matter (NOM) composition clearly depends on the environmental source (Aiken and Costaris, 1995); a range of organic compounds are found in natural waters, from low molecular weight hydrophilic acids, carbohydrates, proteins and amino acids to higher molecular weight compounds such as humic substances (fulvic and humic acids) (Choudry, 1984). Most of the NOM found in natural waters are humic substances (30–50%) (Pernet-Coudrier et al., 2011; Thurman, 1985), which result from elutriation of the surrounding soils and from microbiological, chemical and photochemical reactions (humification process) that occur during the degradation and polymerization of vegetable organic matter in water (Galapate et al., 1997; Langlais et al., 1991; Rodríguez et al., 2012a). Humic substances compete for adsorption sites with target compounds in activated carbon adsorption (Rodríguez et al., 2011), contribute to the fouling of membranes, form soluble complexes with many heavy metal ions and organo-pollutants, promote the formation of bio-film in water distribution pipelines and are known precursors for disinfection by-products in chlorination (Allpike et al., 2005; Liu et al., 2008; Zhou et al., 2000). Humic substances are complex mixtures of high to low molecular weight species, so they are polydisperse systems with a specific distribution of molecular weights (Cabaniss et al., 2000, Myllykangas et al., 2002; Rodríguez and Núñez, 2011); fulvic and humic acids make up the two main fractions of humic substances and they can be distinguished by their different solubility at pH 1: the precipitated fraction is humic acid and the part remaining in solution is fulvic acid. Fulvic acids always represent the larger fraction (the fulvic acid/humic acid mass ratio is generally around 9:1) and are more soluble than humic acids, since they have a lower average molecular weight and a higher acidity (especially carboxylic acidity) than humic acids (humic acids are often in colloidal form due to their large size), whereas humic acids show more aromaticity and UV absorbance and have more color than fulvic acids (Andrews and Huck, 1996; Langlais et al., 1991; Rodríguez and Núñez, 2011); moreover, humic acids generally have a greater trihalomethane formation potential (Rodríguez et al., 2012b) and are more readily coagulated by aluminum and iron (III) salts than fulvic acids (Rodríguez et al., 2012a). The following analytical techniques are prominent among those used for the characterization of humic substances (Rodríguez and Núñez, 2011): high performance size-exclusion chromatography (HPSEC) for the study of the distribution of molecular weights (Chin et al., 1994; McDonald et al., 2007), elemental analysis (EA) (Langlais et al., 1991), organic acidity analysis (Chandrakanth and Amy, 1996), nuclear magnetic resonance (NMR)—both 1H NMR (Kim et al., 2006; Ma et al., 2001) and 13 C NMR (Mao et al., 2007; McDonald et al., 2007; Muller et al., 2004; Thorn et al., 2010; Tsuda et al., 2010), Fourier-transform infrared (FTIR) (Chen et al., 2002; Chiang et al., 2009; Kim et al., 2006; Ma, 2004) and UV/Vis spectroscopy (Rodríguez and Núñez, 2011). In addition to those techniques, fluorescence techniques have also been used in the study of humic substances (Kalbitz et al., 2000). Fluorescence spectroscopy provides important information on the chemical nature of the humic substances: the position, shift and intensity of fluorescence peaks can be correlated to structural information such as functional groups (electron-donating/withdrawing groups), polycondensation, aromaticity, heterogeneity and dynamic properties related to their intramolecular and intermolecular interactions (Chen et al., 2003; Mobed et al., 1996; Zhang et al., 2008); moreover, it is a simple, rapid, sensitive and non-destructive method requiring only a small volume of

aqueous sample at a low concentration (usually b20 mg/L) (Swietlik and Sikorska, 2004). There are several fluorescence techniques, from the simplest and conventional emission (emission scan fluorescence— ESF) to the most recent and complete synchronous fluorescence spectroscopy (SFS) and total luminescence spectroscopy—TLS (also known as excitation–emission matrix—EEM). SFS presents several advantages over conventional ESF: it provides better sensitivity and improved peak resolution (Chen et al., 2002) as well as additional information on structural signatures of humic macromolecules (Chen et al., 2003); it also offers a potentiality to reduce overlapping interferences and a possibility for each fluorescent component to be identified in a specific spectral range (Miano and Senesi, 1992; Peuravuori et al., 2002). TLS (in which repeated emission scans are collected at numerous excitation wavelengths) is at present the most complete technique, as it provides unique “finger prints” (in the form of an excitation–emission matrix) for single compounds or a mixture of fluorescent components (Alberts et al., 2002; Chen et al., 2003; Coble, 1996; Henderson et al. 2009; Her et al., 2003; Hudson et al., 2008; Peiris et al., 2011; Sierra et al., 2005). Even though the various structural units present in the humic macromolecules can have very variable effects on the wavelength and intensity of fluorescence, some general behaviors may be described (Coble, 1996; Peuravuori et al., 2002; Senesi, 1990; Swietlik and Sikorska, 2004): the fluorescence intensity decreases with increasing molecular size of the humic macromolecule; electron-withdrawing groups (COOH) decrease and electron-donating groups (OH, NH2, OCH3) increase the fluorescence intensity in aromatic compounds; carbonyl-containing substituents, hydroxyl, alkoxyl and amino groups tend to shift fluorescence maxima to longer wavelengths (red-shift), whereas a blue-shift (fluorescence maxima shift towards shorter wavelengths) can be caused by a reduction in the degree of the π-electron system (such a decrease in the number of aromatic rings), by a reduction of conjugated bonds in a chain structure or by a conversion of a linear ring system to a non-linear system and finally, the presence of fluorescence bands (peaks) at long wavelengths with low intensity can be attributed to linearly condensed aromatic rings and other unsaturated bond systems capable of a high degree of conjugation within the humic macromolecule (Peuravuori et al., 2002). Direct fluorescence TLS measurement of natural waters is not suitable for studying humic and fulvic acids individually, since their fluorescences in most cases overlap (Baker, 2001; Hudson et al., 2008; Sierra et al., 2005), making an accurate identification of humic acids in the presence of fulvic acids difficult (the comparatively weaker fluorescence signals of the less abundant humic acids are overshadowed by the stronger fluorescence signals of the more abundant fulvic acids) (Peiris et al., 2011); that is why isolated humic and fulvic acids have been used in this study. The main objective of this work is to conduct a comprehensive structural characterization of humic substances, employing all the current fluorescence techniques: emission scan fluorescence (ESF), synchronous fluorescence spectroscopy (SFS) and total luminescence spectroscopy (TLS or EEM) through the use of both 2-D contour maps and 3-D plots. In addition to the three earlier techniques, other fluorescencerelated parameters have been studied, such as the fluorescence index (Chen et al., 2003; Kim et al., 2006; McKnight et al., 2001) and the λ0.5 parameter (Kim et al., 2006), both related with the aromaticity of humic substances. Humic substances provided by the International Humic Substances Society (IHSS), which are considered as reference materials at an

720


international level, were selected with the aim of comparing the results of this work with those found in the literature and with future studies; in particular, Suwannee River Fulvic Acid Standard (SUFA), Suwannee River Humic Acid Standard (SUHA) and Nordic Reservoir Fulvic Acid Reference (NOFA) were used in this study. Additionally, Aldrich Humic Acid (ALHA) has also been studied, which is a commercial humic acid widely used as a surrogate for aquatic humic substances in various studies (coagulation–flocculation, activated carbon adsorption, ozonation, formation of disinfection by-products, etc.). Its inclusion in the present work has the purpose of studying whether its behavior is similar to that of natural humic substances and for comparative purposes with studies by other researchers. 2. Material and methods 2.1. Humic substances The humic substances (HS) used in this study were provided by the International Humic Substances Society (IHSS) and included two fulvic acids and a humic acid: Suwannee River Fulvic Acid Standard (SUFA), Suwannee River Humic Acid Standard (SUHA) and Nordic Reservoir Fulvic Acid Reference (NOFA). The Suwannee River rises in the Okefenokee Swamp in South Georgia (USA) and flows southwest to the Gulf of Mexico. The Okefenokee Swamp contains extensive peat deposits; however, decomposing vegetation is believed to provide most of the dissolved organic carbon (DOC) to its waters; at its headwaters in the Okefenokee Swamp, the Suwannee River is a blackwater river, with DOC concentrations ranging from 25 to 75 mg/L and pH values of less than pH 4.0. The Nordic Reservoir water was obtained from a drinking water reservoir at Vallsjøen, Skarnes, Norway. The reservoir is at 225 m above sea level and has a maximum depth of about 14 m; the sample was obtained from the Sør-Odal County Waterworks intake pipe that draws water from at depth of 10 m (pH = 5.6, EC = 2.1 mS/m and DOC = 10.7 mg/L) (IHSS, 2012). Additionally, a commercially supplied humic acid: ALHA (Aldrich Chemical Co., UK) was used in this work in order to make comparisons with the structural characteristics of the natural aquatic humic substances. ALHA supplier does not provide much information on the origin of this product, a drawback reported by other researchers. According to some investigators (Malcolm and McCarthy, 1986; Monteil-Rivera et al., 2000), the 13C NMR spectrum of ALHA is very close to those of humic acids derived from brown coal, such as Leonardite humic acid (a reference material from the IHSS) and Wyoming dopplerite. The great similarity between Leonardite humic acid and ALHA has also been reported through X-ray photoelectron spectroscopy (XPS) measurements (Monteil-Rivera et al., 2000), so it can be reasonably argued that ALHA is expected to be produced by the natural oxidation of lowgrade coals. All HS samples were dissolved in 0.01 M KCl solutions with a final concentration of 20 mg/L TOC and adjusted to pH 7 using dilute HCl or NaOH (Chen et al., 2003). This TOC concentration was found to be the optimum for this study, although the literature is not consistent with regard to the influence of TOC concentration on fluorescence measurements (Kim et al., 2006; Miano and Senesi, 1992; Mobed et al., 1996; Peiris et al., 2009, 2011; Peuravuori et al., 2002; Rosa et al., 2005; Sierra et al., 2005; Swietlik and Sikorska, 2004). pH adjustment is required since pH variation is known to cause peak shifts in fluorescence spectra (Alberts and Takács, 2004b; Her et al., 2003); pH also affects fluorescence intensity, although the literature is not consistent with regard to this effect: some studies indicated an increase in intensity with increasing pH (Mobed et al., 1996) and others the other way round (Miano and Senesi, 1992). Concerning the influence of the samples ionic strength on fluorescence measurements, the literature also shows contradictory results in this case: some studies indicated no significant effect of ionic strength

in the range of 0–1 M KCl (Mobed et al., 1996) or 0–0.1 M Na2SO4 (Her et al., 2003), whereas other studies reported a decrease in fluorescence intensity with increasing ionic strength (Ghosh and Schnitzer, 1980). 2.2. Fluorescence spectral analysis All fluorescence spectra were recorded on a fluorescence spectrophotometer equipped with both excitation and emission monochromators (Varian Cary Eclipse, Palo Alto-CA, USA); the main instrument parameters were: photomultiplier tube voltage = 800 V, scan rate = 600 nm/min and excitation/emission slit width = 10 nm, conditions found to be optimal in some studies (Peiris et al., 2009, 2011). Corrections for inner filter effects were not applied as inner filtering effects are not expected to be significant at the concentrations levels used in this study; moreover, since all the spectra were recorded on the same instrument using the same experimental conditions, a comparative discussion on the spectra is acceptable although no corrections for fluctuation of instrumental factors and for scattering effects (primary and secondary inner filter effects) were applied to the data (Peuravuori et al., 2002; Sierra et al., 2005; Uyguner and Bekbolet, 2005). Fluorescence intensity variation in triplicate experiments was less than 6%. 2.2.1. Total luminescence spectroscopy (TLS) A series of emission spectra (range from 350 to 550 nm) were collected over a range of excitation wavelengths (from 220 to 450 nm); spectra were then concatenated into an excitation–emission matrix (EEM) to provide a complete representation of the fluorescence of the sample in the form of three-dimensional contour plots of fluorescence intensity as a function of excitation an emission wavelengths (Chen et al., 2003; Kim et al., 2006). A wavelength step of 10 nm was used for the collection of EEM spectra and the equipment was auto-zeroed prior to each analysis. 2.2.2. Synchronous fluorescence spectroscopy (SFS) Synchronous-scan excitation spectroscopy consists in measuring the fluorescence intensity while simultaneously scanning over both the excitation (λex) and emission (λem) wavelengths while keeping a constant optimized wavelength difference (offset): Δλ = λem − λex (Chen et al., 2003). For single compounds having well-defined fluorescence maxima, the optimal condition is obtained when the selected Δλ matches the wavelength interval between the maxima of the emission and the excitation peaks of the fluorophore, which is known as Stoke's shift (Miano and Senesi, 1992); however, in multi-component systems (as is the case with humic substances), the chances of spectral overlap increase, which may lead to problems such as distortion of the synchronous signal (Peuravuori et al., 2002). The possibility of modifying selectively Δλ is the great advantage of SFS; most studies on HS synchronous fluorescence have used Δλ = 18–20 nm, but recently some researchers have explored the usefulness of other Δλ values, such as 44 and 66 nm (Swietlik and Sikorska, 2004; Zhang et al., 2008). In this study SF spectra were collected with constant offsets of 18, 44 and 66 nm; the excitation wavelengths increased from 230 to 650 nm and the spectra were smoothed with a Savitzky–Golay filter afterwards. 2.2.3. Emission scan fluorescence (ESF) Emission spectra of HS samples were collected at emission wavelengths ranging from 360 to 650 nm using a fixed excitation wavelength of 340 nm (the most usual excitation wavelength used for HS studies). 2.2.4. Fluorescence index The fluorescence index was proposed by McKnight et al. (2001) and is defined as the ratio of emission intensity at 450/500 nm for emission spectra measured at a excitation wavelength of 370 nm.


2.2.5. The λ0.5 parameter The λ0.5 parameter was proposed by Fabbricino and Korshin (2004) and is defined as the wavelength corresponding to the normalized emission band at its half-intensity (for wavelengths Nλmax). 2.3. Additional analysis TOC (total organic carbon) was measured with a carbon analyzer (Shimadzu TOC-5050) based on the combustion-infrared method and SUVA (Specific UV-absorbance: UV254/TOC) was determined in 1-cm pathlength quartz cells by a spectrophotometer (Hitachi 100–10) at a wavelength of 254 nm. 3. Results and discussion 3.1. Total luminescence spectroscopy Fig. 1 shows the fluorescence spectra (EEM contour maps) of the humic substances in this study (see the corresponding 3-D spectra in Fig. S1, online Supplementary materials). The spectra were not corrected for Rayleigh scattering peaks, showing them in the form of black bands; first order (FORS) and second order (SORS) Rayleigh scattering peaks occur at the same wavelength and twice the wavelength of the excitation light, respectively (Zhang et al., 2008). These peaks originate from an interaction between the excitation light and water molecules and provide information related to the particulate/colloidal matter

721

present in water (Peiris et al., 2011). The first notable fact observed in that figure is that the spectra corresponding to the natural aquatic humic substances clearly differ from that of the commercial humic acid (ALHA): the natural humic substances (fulvic and humic acids) show two clearly differentiated peaks of maximum intensity, one is located at excitation wavelengths in the ultraviolet region (around 230 nm) and the other one in the visible region (at excitation wavelengths around 335 nm), whereas the commercial humic acid presents four peaks with a higher degree of spectral overlap; this last case (ALHA) will be studied in further detail later on. With regard to the aquatic humic substances, our results coincide with the main body of results reported in the literature (Sierra et al., 2005): the peak located in the ultraviolet region is referred to as peak A according to the terminology proposed by Coble (1996) or peak α according to the terminology proposed by Parlanti et al. (2000), whereas the peak located in the visible region of the spectrum is referred to as peak C or peak β. In this article, the terminology proposed by Coble (peaks A and C) will be followed. However, there is no agreement in the literature on the exact location of the peak present in the ultraviolet region (peak A), with regard to its excitation wavelength: – Some researchers situate that peak at excitation wavelengths in the range of 240–260 nm (slightly above those obtained in the present work), such as Her et al. (2003), Kim et al. (2006), Lanxiu et al. (2006), Parlanti et al. (2000), Peiris et al. (2011), Sierra et al. (2005), Swietlik and Sikorska (2004), and Zhang et al. (2008).

Fig. 1. Total luminescence spectra (EEM contour maps) of the humic substances.

722


– Other researchers situate peak A at excitation wavelengths of around 225–230 nm, which coincides with our results, such as Alberts et al. (2002), Alberts and Takács (2004a,b), IHSS (2012), McKnight et al. (2001), Mopper and Schultz (1993), and Stedmon et al. (2003). In some cases, these differences may be due to the fact that in many of those works the spectra were not collected up to excitation wavelengths as low as in our case (220 nm): for example, Peiris and Parlanti used excitation wavelengths from 250 nm and Zhang and Swietlik from 240 nm. Except for that difference with regard to the excitation wavelength of peak A, various researchers have reported that the position of both peaks (A and C) stays relatively constant regardless of the origin of the natural surface water (rivers, lakes, reservoirs); for example, Coble reported similar values for peak C in a study that included natural waters from 21 rivers, obtaining the following average values: 340/448 nm. Alberts et al. (2002) also obtained similar values for both peaks (225/ 427 nm and 335/440 nm) in a comparison of two surface waters of different origin (from a lake and a reservoir). Her et al. (2003) also found similar values for peak C in different waters, in a study that included various natural waters from rivers, lakes, ground water and even a secondary effluent from a wastewater treatment plant. The spectra obtained in the present work for the three aquatic humic substances under study (Fig. 1) coincide very precisely with the corresponding spectra published by IHSS on its website (www. humicsubstances.org); the position of the peaks is very similar for the three humic substances (SUFA, SUHA and NOFA), regardless of whether they are fulvic or humic acids, as shown in Table 1. Higher fluorescence intensities for the fulvic acids (both for peaks A and C) than for the humic acids may also be observed in that table, more than double when SUFA and SUHA (humic substances from the same origin) are compared. This result is widely reported in the literature (Chen et al. 2002; Hautala et al., 2000; Korshin et al., 1999; McKnight et al., 2001; Mobed et al., 1996; Peuravuori et al. 2002; Senesi et al., 1989; Sierra et al., 2005) and is attributed to the fact that larger molecular size macromolecules (humic acids) having more highly conjugated systems are also capable of absorbing light and are more likely to reabsorb the emitted fluorescence than smaller macromolecules (fulvic acids) (Alberts and Takács, 2004b). Likewise, Alberts et al. (2002) reported in studies on ultrafiltration and HPSEC of aquatic NOM that the fractions of lower molecular weight presented higher values of relative fluorescence intensity (per mass of carbon) than those of higher molecular weight. It can be observed in Table 1 that the intensities of peaks A and C for the three aquatic humic substances follow this order: SUFA N NOFA N SUHA, which is in agreement with some studies on average molecular weights (Mw) reported in the literature for those humic substances, which follow a reverse order to that of fluorescence intensities: SUFA (800–2280 Da) b NOFA (2138 Da) b SUHA (3820 Da) (Chin et al., 1994; Hongve et al., 1996; Leenheer et al., 1989; McDonald et al., 2007); these values are within the range reported in the most recent studies (Rodríguez and Núñez, 2011), although there is still a wide margin of variation for the Mw values

Table 1 Location and fluorescence intensity of the main peaks found in humic substances.

Peak A Peak C Other peaks

SUFA

SUHA

NOFA

ALHA

230/437 (54.90)a 335/457 (52.85) –

230/437 (20.17) 335/465 (22.33) –

230/436 (40.67) 335/461 (43.59) –

– – 260/462 (30.23) 300/479 (25.67) 365/483 (19.60) 450/524 (18.54)

a Ex/Em (Int): excitation wavelength (nm)/emission wavelength (nm) (fluorescence intensity (F.I.) units).

measured for humic substances (Perminova et al., 2003), depending on the instrumental technique used (high performance size exclusion chromatography: HPSEC, vapor pressure osmometry, colligative property measurements and small-angle X-ray diffraction) (Assemi et al., 2004). Alberts et al. (2002) also proposed some chemical structures that could be responsible for NOM fluorescence. He proposed specifically that the most probable structures for peak C (fluorescence in the visible region) would be based on oxygenated aromatic compounds such as substituted hydroxybenzoic acids, cinnamic acid, coumarins and coumestrol, coming from the decomposition of plants. In that study, the author proposed no chemical structure linked to peak A (fluorescence in the ultraviolet region), indicating that the majority of the most common oxygenated and nitrogenated organic compounds present no fluorescence in that region. Sierra et al. (2005) also proposed the following organic structures as possible chromophore groups responsible for the fluorescence of peak C: hydroxybenzoic acids and other substituted phenolic units originating from lignin, hydroxycoumarin-like structures, Schiff-base Systems and chromone, xanthone and/or quinoline derivatives originated from degraded plant materials. In a subsequent study, Alberts and Takács (2004a) proposed specific organic structures that could represent classes of fluorophores present in the humic macromolecules. He specifically proposed hydroxycoumarins as fluorophore types for the fulvic acids and, in general, for aquatic NOM; for instance, 7hydroxy-coumarin, presenting Ex/Em values of 343/450 nm (similar to those reported in the literature for aquatic NOM and to those found in this study) is the most commonly occurring isomer present in plants (Wolfbies, 1985) and it has a molecular weight of approximately 150 Da, which agrees with the result widely reported in the literature that fulvic acids and aquatic NOM are mixtures of relatively small molecules with low degrees of condensation (Rodríguez and Núñez, 2011). Alberts also proposed a possible fluorophore type for terrestrial humic acids, in this case phytochlorin (a chlorophyll degradation by-product that is a common component of plants); this compound presents a basic structure similar to bilirubin, which when complexed with protein presents Ex/Em at around 455/520 nm (Budavari, 1996; Wolfbies, 1985), similar values to the peak located at the longest excitation wavelength in the visible region of the spectra for terrestrial humic acids and in the present study for ALHA (450/524 nm). Phytochlorin and other similar chlorophyll degradation by-products are expected to have average molecular weights of around 500 Da (and higher if complexed with protein residues), therefore they would coincide with the high condensed structures described in the literature for terrestrial humic acids (Alberts and Takács, 2004a). Table 2 shows a literature review of the location of the peaks for the specific case of the humic substances under study in the present work, where the main fact to highlight is the different position assigned by several researchers for peak A, as mentioned earlier. Some researchers (Belin et al., 1993; Her et al., 2003; Mobed et al., 1996; Sierra et al., 2005) have also observed that, for humic substances from the same source, the humic acids present a “red-shift” (longer excitation and emission wavelengths) in relation to the corresponding fulvic acids, above all for peak C; in the present study, SUHA (230/437 nm and 335/ 465 nm) presents a red-shift compared to SUFA (230/437 nm and 335/ 457 nm) for peak C only in the emission wavelength (see Table 2). This red-shift has been attributed in the literature to the greater apparent molecular weight and to the higher degree of aromatic condensation of humic acids in comparison with fulvic acids (Belin et al., 1993; Mobed et al., 1996). Other authors have also related the position of peak C to certain functional groups content in the humic macromolecule. For example, Silva et al. (1994) suggested that when peak C is located in the region of excitation wavelengths of 300–350 nm, there is a predominant presence of chromophore groups with high carboxylic groups content, whereas chromophore structures containing phenolic groups would predominate if peak C is located in the region of 350–400 nm. In the present work, peak C is located in the region of excitation wavelengths

F.J. Rodríguez et al. / Science of the Total Environment 476–477 (2014) 718–730 Table 2 Comparison of several peak locations reported in the literature. SUFA

SUHA

NOFA

ALHA

This study

230/437 335/457

230/437 335/465

230/436 335/461

Alberts and Takács (2004a)a

229/432 332/442 255/455 320/450 240–255/ 426–443 325/449 –

229/431 339/449 260/485 330/470 240–255/ 426–443 330/456 –

226/431 335/447 –

260/462 300/479 365/483 450/524 –

Sierra et al. (2005) Kim et al. (2006)

Peiris et al. (2011)

–

265/525 360/520 300/443

–

295/490

Excitation wavelength (nm)/emission wavelength (nm). a IHSS has included in its webpage the fluorescence spectra provided by Alberts and Takács (2004a).

of 300–350 nm for the three aquatic humic substances studied, which is in agreement with the higher carboxylic acidity that they all present: according to data from IHSS (2012), the values of carboxylic acidity for SUFA, NOFA and SUHA are 11.44, 11.16 and 9.13 meq/g C respectively, while those of phenolic acidity are 2.91, 3.18 and 3.72 meq/g C, respectively. Other researchers have reported that the position of peaks A and C (in particular, their emission wavelength) might be used to identify the different origin of aquatic fulvic acids. For example, McKnight et al. (2001) studied four fulvic acids of different origins and found that terrestrially derived fulvic acids (terrestrial plant and soil organic matter are the dominant sources of NOM) showed higher emission wavelengths for peaks A and C (230/426 nm and 320/440 nm) than microbially derived fulvic acids (NOM derived from autochthonous microbial processes): 230/ 412 nm and 320/406 nm. In the present work, the two fulvic acids under study (SUFA and NOFA) showed wavelength values for peaks A and C (see Table 1) that allowed to place them in the category of terrestrially derived fulvic acids, a result that has been reported in the literature at least for SUFA (IHSS, 2012; McKnight et al., 2001). Alberts and Takács (2004a) also studied numerous correlations between the fluorescence intensity (F.I.) of peak C for several humic substances and various types of functional groups typical of those macromolecules, finding statistically significant correlations for the following parameters: peak C F.I./carboxyl groups content (positive correlation) and peak C F.I./phenolic groups content (negative correlation), only for aquatic and terrestrial fulvic acids (carboxylic and phenolic groups content determined by 13C NMR); those correlations were confirmed in the present study, even with the incorporation of the humic acid SUHA, as the fluorescence intensity values (in units of fluorescence) determined in this study for peak C of the aquatic humic substances are in the following order (see Table 2): SUFA (52.59) N NOFA (43.97) N SUHA (22.33), the values of carboxylic and phenolic acidity having been presented earlier (IHSS data). Since those correlations apparently contradict the finding commonly reported in the literature (see Section 1) that electron withdrawing groups (e.g., carboxylic groups) decrease and electron donating groups (such as phenolic groups) increase the intensity of fluorescence, it may be argued that in the case of aquatic humic substances, the effect of the molecular weight of the macromolecule (the lower the molecular weight, the higher the fluorescence intensity) dominates over the reverse effect produced by the presence of carboxylic and phenolic groups in the macromolecule (lower fluorescence intensity values with increasing the content in carboxylic groups and decreasing the content in phenolic groups). Alberts and Takács (2004a) also found a positive correlation for the parameters peak C F.I./SUVA but only for aquatic and terrestrial humic acids; however, a negative correlation (r2 = 0.995) was obtained in this study for the parameters peak C F.I./SUVA in the particular case of the three aquatic humic substances under study (including fulvic and

723

humic acids), as the measured values of SUVA (units in a.u./mg C/L) were in the following order: SUFA (0.045) b NOFA (0.052) b SUHA (0.065), a result that is in agreement with those reported by other researchers (Stewart and Wetzel, 1980), who indicated that larger molecular weight aquatic humic fractions had a greater absorbance but lower fluorescence than smaller molecular weight fractions. The SUVA values obtained in this study are similar to those reported in the literature for the humic substances studied (Alberts and Takács, 2004a), the fulvic acids having lower SUVA values than the humic acids (Rodríguez and Núñez, 2011); the commercial humic acid (ALHA) had the highest SUVA (0.070) of all the humic substances in the present study. With respect to the commercial humic acid (ALHA), its fluorescence spectrum is completely different to those of the aquatic humic substances, as mentioned earlier (see Fig. 1): the spectrum reveals the presence of four poorly resolved peaks with a high degree of spectral overlap. The peak of maximum intensity was located in the ultraviolet region at Ex/Em of 260/462 nm, two poorly resolved shoulder-like peaks at Ex/Em of 300/479 nm and 365/483 nm respectively, and a peak located at longer excitation wavelengths within the visible region of the spectrum (Ex/Em of 450/524 nm), which was not found in the aquatic humic substances (see Table 1). The peak at Ex/Em of 300/479 nm was not clearly detected in the contour map, but it was precisely identified in the data matrix and in the 3-D spectrum (see Fig. S1, online Supplementary materials). There is no agreement in the literature on the number of peaks assigned to ALHA (see Table 2): for example, Peiris et al. (2011) and Kim et al. (2006) reported only one peak for ALHA, located at Ex/Em of 295/490 nm and Ex/Em of 300/443 nm respectively, a peak that would be comparable to that at Ex/Em of 300/479 nm found in this study. Other authors, such as Sierra et al. (2005), reported two peaks of maximum intensity located at Ex/Em of 265/525 nm and at Ex/Em of 360/520 nm, which would be comparable to those at Ex/Em of 260/ 462 nm and at Ex/Em of 365/483 nm in the present work. The peak identified in the present study, located at high excitation wavelengths in the visible region (Ex/Em of 450/524 nm) was not identified by the earlier authors, although in those studies the spectra were not acquired up to excitation wavelengths as long as in the present work. For example, Peiris only recorded the spectrum up to an excitation wavelength of 380 nm, Sierra up to 410 nm and Kim just up to 450 nm. This apparent lack of agreement on the spectra reported in the literature for the case of ALHA may be due to several causes: differences in the characteristics of the aqueous samples in the tests (pH, concentration of TOC, ionic force), differences in the acquisition conditions of the fluorescence spectrum (range of wavelengths tested, scan speed, slits and bandwidths chosen, subtraction of water blank, etc.) and even differences among ALHA batches supplied by the manufacturer (Malcolm and McCarthy, 1986). The four peaks found in the present work for ALHA have been reported in the literature (with regard to their location through their excitation wavelength) either for the particular case of this humic acid or for other humic acids of terrestrial origin, although the corresponding emission wavelengths show greater variability, as shown in the following paragraphs: – Peak at Ex/Em of 260/462 nm: reported by Sierra et al. (2005) for ALHA (265/525 nm), Alberts and Takács (2004a) for Leonardite Standard Humic Acid from IHSS (265/458 nm), Stedmon et al. (2003) for terrestrial humic substances (270/478 nm) and Murphy et al. (2008) for terrestrial humic substances (260/490 nm). – Peak at Ex/Em de 300/479 nm: reported by Peiris et al. (2011)) for ALHA (295/490 nm), Kim et al. (2006) for ALHA (300/443 nm), Yamashita and Jaffe (2008) for terrestrial humic substances (305/ 428nm) and Alberts and Takács (2004a) for a soil humic acid (278/ 464 nm). – Peak at Ex/Em of 365/483 nm: reported by Sierra et al. (2005) for ALHA (360/520 nm), Alberts and Takács (2004a) for Leonardite

724


Standard Humic Acid from IHSS (356/459 nm) and Stedmon et al. (2003) for terrestrial humic substances (360/478 nm). – Peak at Ex/Em of 450/524 nm: reported by Alberts and Takács (2004a) for Leonardite Standard Humic Acid from IHSS (454/511 nm), Alberts and Takács (2004b) for a soil humic acid (459/513 nm), Stedmon et al. (2003) for a soil fulvic acid (455/521 nm) and Chen et al. (2003) for a soil humic acid (465/530 nm). Some authors (Sierra et al., 2005) have suggested that emission wavelengths as high as those presented by this peak (N500 nm) can only be attributed to the presence of greatly conjugated systems in the macromolecule. Other authors also support that hypothesis (Peuravuori et al., 2002), indicating that fluorescence at long wavelengths with low intensity could be related to the presence of linearly condensed aromatic rings and other unsaturated bond systems capable of a high degree of conjugation, which is in agreement with the high value of SUVA (greater than those of aquatic humic substances) presented by ALHA. Malcolm and McCarthy (1986), as a result of characterizing a variety of commercial humic acids by means of 13C NMR measurements and elemental analysis, reported that these commercial materials are all quite similar regardless of the supplier and are clearly different from aquatic and terrestrial humic acids; this author also indicated that these commercial products (ALHA among them) have similar characteristics to Wyoming dopplerite and to Leonardite humic acid. This result is confirmed in the present study, this time with fluorescence data, since three out of the four peaks reported by Alberts for Leonardite humic acid (265/458, 356/459, 380/465 and 454/511 nm) correspond quite well with peaks found in the present study for ALHA (260/462, 365/ 483 and 450/524 nm); the only ALHA peak that does not have its counterpart in Leonardite humic acid is that at 300/479 nm. 3.2. Synchronous scan spectra 3.2.1. Synchronous scan spectra at Δλ = 18 nm Fig. 2 shows the synchronous fluorescence spectra of the humic substances recorded at Δλ = 18 nm, which is the most common offset reported in the literature. The three aquatic humic substances showed a well-defined maximum peak and two or three poorly resolved additional peaks with a high degree of spectral overlap (Senesi et al., 1989), the fulvic acids (SUFA and NOFA) presenting slightly higher intensity values than the humic acid (SUHA), in the same way as in EEM spectra but in this case the intensity values are very similar for the three compounds.

200

300

400

500

600

700

Fluorescence Intensity (F. units)

14

14 SUFA SUHA NOFA ALHA

12

ALHA

12

10

10

8

8

6

6

NOFA

SUFA

4

It can also be seen that the fulvic acids presented the peak of maximum intensity at shorter excitation wavelengths (SUFA: 392/410 nm and NOFA: 441/459 nm) than the humic acid (SUHA: 468/486 nm); according to the literature (Coble, 1996; Miano and Senesi, 1992; Senesi et al., 1989; Swietlik and Sikorska, 2004), peaks of maximum fluorescence intensity located at long λsyn are generally attributed to the presence of highly substituted aromatic nuclei and/or conjugated unsaturated systems capable of high degree of resonance, which is in agreement with the typical characteristics of the humic acids, whereas maxima peaks located at lower λsyn are indicative of macromolecules containing simpler structural components, for example, a low degree of aromatic polycondensation and/or low levels of conjugated chromophores, characteristics that fit in well with fulvic acids. Some authors (Peuravuori et al., 2002) have proposed that (in the synchronous spectra at Δλ = 18 nm) the peaks located at around Ex/Em of 400/418 nm correspond to polycyclic aromatics with approximately five fused benzene rings, while the peaks at around 460/478 nm are characteristic of polycyclic aromatics consisting of about seven fused benzene rings. Likewise, Chen et al. (2003) proposed two different regions in the synchronous spectra at Δλ = 18 nm: fluorescence maxima in the region of 450–480 nm would be indicative of the predominant presence of high molecular weight humic materials, whereas increased fluorescence in the 380–450 nm region would indicate the presence of large amounts of simple, dissociated phenolic and quinone types of organic compounds. The positions of the maximum intensity peaks found in the present work for the humic substances (Fig. 2 and Table 3) coincide with the values reported in the literature: the maximum peak found for SUFA in this study (392/410 nm) was reported by other researchers for the same SUFA (400/420 nm, Chen et al., 2003) and for a soil fulvic acid (400/418 nm, Kalbitz et al., 2000), the maximum peak found for NOFA (441/459 nm) was reported by Senesi et al. (1989) for the same NOFA (440/458 nm) and the one found for SUHA (472/490 nm) was also reported by Senesi for the same humic acid (472/490 nm) and by Uyguner and Bekbolet (2010) for another aquatic humic acid (470/488 nm). The three aquatic humic substances share a common small peak at 288/306 nm that was not observed in the EEM spectrum (results not shown in Fig. 1; although the EEM spectra shown in Fig. 1 start from a λem = 350 nm, it was proven with EEM spectra acquired from λem = 300 nm that no peak was detected at 288/306 nm); this peak has been assigned in the literature to various structures (Swietlik and Sikorska, 2004): according to De Souza et al. (1994) this band can appear due to the presence of amino acids like tryptophan and tyrosine, according to Peuravuori et al. (2002) it can be assigned to mainly aromatic amino acids and some other volatile acids containing highly conjugated aliphatic structures and finally, according to Duarte et al. (2003) a band at Ex about 280 nm in humic and fulvic acids spectra can be attributed to lignin-derived structural moieties and its intensity increases toward low molecular sizes (Swietlik and Sikorska, 2004). However, the most abundant references in the literature are those that attribute this fluorophore to aromatic amino acid functionalities (Coble, 1996; Elliot et al., 2006; Liu et al., 2007; Mounier et al., 1999; Persson and Wedborg, 2001; Stedmon et al., 2003; Yamashita and Tanoue, 2003; Zhang et al., 2008), the most likely being the tryptophan-like one, since this fluorophore appears at Ex of 270–290 nm whereas the tyrosine-like one is found at Ex of 240–260 nm (Zhang et al., 2008).

4

2

2

Table 3 Comparison of peaks maxima locations for synchronous spectra at different offsets (Δλ).

SUHA

0

0 200

300

400

500

600

700

Excitation wavelength (nm) Fig. 2. Synchronous fluorescence spectra (Δλ = 18 nm) of the humic substances.

Sync. at Δλ = 18 nm Sync. at Δλ = 44 nm Sync. at Δλ = 66 nm a

SUFA

SUHA

NOFA

ALHA

392/410a 364/408 358/424

468/486 380/424 361/427

441/459 381/425 361/427

470/488 455/499 447/513

λsyn: excitation wavelength (nm)/emission wavelength (nm), where em = ex + Δλ.


3.2.2. Synchronous scan spectra at Δλ = 44 nm The synchronous fluorescence spectra of the humic substances at Δλ = 44 nm (Fig. 3) showed a lower number of bands than the spectra at Δλ = 18 nm, especially for the case of ALHA, although this commercial humic acid continues to show the highest degree of polydispersity compared to the aquatic humic substances and SUFA the lowest. All the humic substances continue to present the small band at 288/ 332 nm, which also appeared in the spectrum recorded at Δλ = 18 nm at the same λsyn, probably due to amino acid functionalities (mainly tryptophan-like: 270–290/330–350 nm) (Zhang et al., 2008). The maxima peaks for the four humic substances appear in the following order of intensity: ALHA (455/499) N SUFA (364/408) N NOFA (381/425) N SUHA (380/424); unlike the spectrum recorded at Δλ = 18 nm where

200

Fluorescence Intensity (Fluorescence units)

In addition to the peak in common at 288/306 nm, the three aquatic humic substances also showed various poorly resolved shoulder-like subpeaks: these are found for SUFA at 440/458 nm and 505/523 nm, for NOFA at 468/486 nm and 505/523 nm and for SUHA at 407/425 nm and 505/523 nm. The presence of this series of shoulders is indicative of the polydispersity of humic substances, in which low- and highmolecular weight organic substituents are present (Chen et al., 2003). With regard to the commercial humic acid (ALHA), the large number of peaks shown by its synchronous spectrum is indicative of its greater structural complexity and its greater degree of polydispersity; in addition, the peaks are in general better resolved (especially in the central region of the spectrum) and clearly present a greater relative intensity in comparison with aquatic humic substances. The peak of maximum intensity is located at 470/488 nm and its position coincides with that of the maximum peak found for SUHA; this peak has been reported by various researchers as the maximum peak (in synchronous spectra at Δλ = 18 nm) for various humic acids: Uyguner and Bekbolet (2005) for ALHA (473/491 nm) and for another commercial humic acid: Roth humic acid (470/488 nm), Senesi et al. (1989) for a soil humic acid (475/493 nm), for a peat humic acid (477/495 nm) and for Leonardite humic acid (473/491 nm), Chen et al. (2003) for a soil humic acid (466/486 nm) and Miano and Senesi (1992) for another soil humic acid (473/491 nm); as mentioned earlier for the case of SUHA, this peak is assigned by some authors (Peuravuori et al., 2002) to polycyclic aromatic compounds of about seven fused benzene rings. The great variety of functionalities associated with all these peaks indicates that ALHA is a macromolecule that is clearly different to the aquatic substances, although no conclusive results can be drawn in this case, as the synchronous spectrum of ALHA at Δλ = 18 nm shows a partial overlap with the FORS scattering line (see explanation in Section 3.2.4). Finally, synchronous fluorescence spectra recorded at Δλ = 18 nm have also been used by some authors to estimate the degree of humification of humic substances: as the peaks of the spectrum located at longer λsyn are associated with an increasing number of highly substituted aromatic nuclei, Kalbitz et al. (2000) proposed a parameter called “humification index”, defined as the ratio of fluorescence intensities between a peak of the spectrum located at higher λsyn and another at lower λsyn, always using two clearly distinguishable bands; high values of this index are indicative of a high degree of polycondensation (and therefore of humification) of the humic substances. Kalbitz used in his study terrestrial fulvic acids derived from different types of soil (degraded and relatively intact peatlands), which showed three main peaks in the synchronous spectrum at λsyn = 360, 400 and 470 nm, respectively. The three ratios studied by Kalbitz (400/360 nm, 470/360 nm and 470/400 nm) correlated satisfactorily with the degree of humification of those fulvic acids. In this work, as the humic substances under study showed no peak at λsyn = 360 nm, the ratio of fluorescence intensities 470/390 nm was chosen as humification index, achieving the following values: SUFA (0.59) b NOFA (1.32) b SUHA (1.54) b ALHA (1.61), which is in principle in accordance with the higher degree of aromatic polycondensation attributed to humic acids compared to fulvic acids.

15

300

725

400

500

600

700

SUFA SUHA NOFA ALHA

15

SUFA NOFA

10

10

ALHA

5

5 SUHA

0 200

0 300

400

500

600

700

excitation wavelength (nm) Fig. 3. Synchronous fluorescence spectra (Δλ = 44 nm) of the humic substances.

the intensities of the three aquatic humic substances were similar, in this case the humic acid (SUHA) clearly showed a lower fluorescence intensity than the fulvic acids (SUFA and NOFA), which is in agreement with its higher molecular weight and is a result similar to that obtained in the EEM spectra. ALHA continued to show the maximum peak at higher λsyn (around 455/499 nm) than the aquatic humic substances, which indicates its greater content in highly substituted aromatic systems, as mentioned in Section 3.2.1. In this case, the spectra of NOFA and SUHA had a similar appearance (two broad overlapping bands), even showing the maximum peak at the same λsyn (at around 380 nm), although the fluorescence intensities of both bands were so similar that it may be accepted that both fluorophores contribute equally to the fluorescence observed in the spectra. SUFA presented a different spectrum to those of NOFA and SUHA, showing only a single broad band (apart from the small shoulder at λsyn = 288 nm), the maximum intensity for which was located at lower λsyn (364/408 nm); this would be indicative of the lower degree of polydispersity of SUFA in comparison with the other two aquatic humic substances. In comparison with the synchronous spectra at Δλ = 18 nm, the maxima peaks showed in the spectra at Δλ = 44 nm were located at lower λsyn (see Table 3). There are relatively few works reported in the literature regarding the study of humic substances through the use of synchronous fluorescence spectra at Δλ = 44 nm; one of them is that by Swietlik and Sikorska (2004), who studied molecular fractions from aquatic NOM. In particular, the synchronous fluorescence spectrum of the dissolved hydrophobic acids fraction (fulvic acids) is similar (a broad band centered at around λsyn = 370 nm and a small protein-like peak at around λsyn = 280 nm) to that found in the present study for the fulvic acid SUFA. Swietlik obtained the same type of spectrum for both humic fractions (humic and fulvic acids), unlike the results obtained in this study, where the spectrum of the humic acid SUHA showed two broad overlapping bands rather than only one. 3.2.3. Synchronous scan spectra at Δλ = 66 nm The synchronous fluorescence spectra of the humic substances at Δλ = 66 nm (Fig. 4) are quite similar to those shown in Fig. 3 for Δλ = 44 nm, with ALHA showing the greatest degree of polydispersity (highest number of peaks), NOFA and SUHA exhibiting two broad overlapping bands (the main one and another in the form of a small shoulder) located at the same wavelengths in both humic substances, and SUFA again showing a single band and therefore the lowest degree of polydispersity. Once again, the locations of the maxima peaks of the humic substances were shifted toward lower λsyn in comparison with the

726



200

25

300

400

SUFA SUHA NOFA ALHA

500

600

SUFA

700

25

20

20 ALHA NOFA

15

15

10

10 SUHA

5

0 200

5

300

400

500

600

0 700

excitation wavelength (nm) Fig. 4. Synchronous fluorescence spectra (Δλ = 66 nm) of the humic substances.

synchronous spectra at Δλ = 44 nm and Δλ = 18 nm (Table 3), although in this case the order of intensities of the maxima peaks differed from the earlier cases, SUFA showing in this case the peak with the highest intensity out of all the humic substances: SUFA (358/424 nm) N NOFA (361/427 nm) N ALHA (447/513 nm) N SUHA (361/427 nm). Other authors (Alberts and Takács, 2004b) have studied the synchronous spectra of the overall NOM from the Suwannee River at Δλ = 20, 40 and 60 nm, obtaining quite similar spectra to those for SUFA in this study (a predictable result, as the fulvic acids usually represent the majority fraction of the NOM) and the maximum peak shifted in their tests to lower λsyn as Δλ increased, a result that coincides with the present work. NOFA and SUHA once again presented quite similar spectra (although in this case SUHA showed a little additional shoulder at around λsyn = 463 nm), but unlike the spectra at Δλ = 44 nm the intensities of the two bands were not similar, this time consisting in a main peak (361/427 nm) and a small shoulder (392/458 nm). The presence of peaks at longer λsyn (429/495 nm and 447/513 nm) in the ALHA spectrum was indicative of a significant contribution of high molecular weight fractions to the overall macromolecule. The maximum peak found in the present study for ALHA (447/513 nm) has been reported in the literature (Stedmon et al., 2003) for soil humic substances (455/ 521 nm). In this case, the interesting fact is that the three aquatic humic substances (SUFA, NOFA and SUHA) showed their maxima peaks at the same location (at around 360/426 nm), SUHA having the lowest relative intensity, which is in agreement with its greater molecular weight compared to the fulvic acids. However, unlike the earlier cases, the protein-like peak that appeared in the synchronous spectra at Δλ = 44 nm and Δλ = 18 nm at around λsyn = 290 nm was not observed in this case. With regard to the works published by other researchers on the synchronous fluorescence spectra of aquatic NOM at Δλ = 66 nm, some further references appear in the literature (Duarte et al., 2003; Parlanti et al., 2000; Sierra et al., 2005; Zhang et al., 2008) compared to the earlier case (Δλ = 44 nm), although once again the most widely studied offset in this field is Δλ = 18–20 nm. For example, Parlanti reported a synchronous spectrum at Δλ = 65 nm for a freshwater sample very similar to that obtained in the present study for SUFA, consisting in a single broad band centered at around λsyn = 350 nm; in this case, the small shoulder corresponding to the protein-like fluorescence (λsyn = 270–280 nm) was also detected, which is not surprising since the sample studied was aquatic NOM and not a pure humic substance.

Zhang et al. (2008) also reported a synchronous spectrum (Δλ = 66 nm) for the hydrophobic acid fraction (humic substances as a whole) extracted from a river water, characterized by a maximum peak located at λsyn = 380–420 nm and three small shoulders at λsyn = 290–305 nm, 340–360 nm and 460–480 nm. Except for the small protein-like shoulder that was not detected in the present work, the synchronous spectrum obtained by Zhang is quite similar to that of SUHA in this study, as the three remaining peaks coincide (SUHA: 361/427 nm, 392/458 nm and 463/529 nm), although the peak of maximum intensity in Zhang's study was at λsyn = 380–420 nm whereas in this study it was at λsyn = 361 nm, which suggests that the humic substances studied by Zhang (extracted from filtered river water) presented a greater proportion of more complex structures (with higher molecular weight and a higher degree of aromatic conjugation) than the humic substances used in the present study. In the specific case of synchronous spectra at Δλ = 66 nm, Zhang indicated that the peaks at λsyn = 340–360 nm are typical of humic-like fluorophores and other authors (Smith and Kramer, 1999) have pointed out that the peak at λsyn = 392 would be characteristic of polycyclic aromatic structures like flavone and coumarine (with Ex/Em of 400–420/460– 480 nm) and that the peak at λsyn = 463 nm could be attributed to other polycyclic aromatic structures with more aromatic rings and a higher degree of conjugation (Chen et al., 2003). 3.2.4. Critical review of the synchronous scan spectra Fig. S2 (see online Supplementary materials) shows the positions of the synchronous spectra (lines a, b and c) on the EEM of the samples, where it can be observed that synchronous spectra actually represent slices of the total EEM spectra which emphasize the shoulders of peak C rather than real peak maxima (Sierra et al., 2005). In that figure, the synchronous spectra at Δλ = 18 nm of the three aquatic humic substances are also seen to be very close to the scattering line (FORS) and taking into account the low fluorescence intensities recorded in that spectral zone, there is a risk of misinterpreting the spectrum if the FORS end overlap with the synchronous spectra. Actually, this does happen with ALHA (see Fig. S2), where its synchronous spectrum at Δλ = 18 nm can be seen to partially overlap with FORS, thus casting doubt on the reliability of this spectrum in the case of ALHA, which in addition was the one that showed a higher number of peaks. For that reason, although the offset at 18 nm has been the most widely reported to date in the literature for the study of humic substances through synchronous scan fluorescence, the results should be interpreted carefully and it should be confirmed in each case that the peaks found present no overlap with the FORS line of the EEM spectrum. With regard to the synchronous spectra at Δλ = 44 and 66 nm, it can be seen in Fig. S2 that in these cases there is no overlap of those spectra with the FORS line for any of the humic substances, therefore their interpretation in the earlier sections is perfectly valid and is in accordance with the results obtained in the EEM spectra. For example, in the particular case of ALHA, its synchronous spectrum at Δλ = 66 nm (see Fig. 4) showed the peak of maximum fluorescence intensity located at 447/513 nm, which corresponds perfectly well with the zone of maximum intensity covered by line c in the EEM spectrum of ALHA (see Fig. S2). Finally, although the EEM technique presents a global and complete view of the fluorescence spectra of the humic substances, some authors have indicated that the synchronous scan fluorescence technique can have an additional usefulness, revealing the existence of peaks corresponding to protein-like constituents of NOM that are often difficult to detect in EEM spectra (Swietlik and Sikorska, 2004; Zhang et al., 2008), as demonstrated in the present study. 3.3. Emission scan spectra Fig. 5 shows the scan emission fluorescence spectra (at λex = 340 nm, the most widely reported in the literature) for the four humic


substances studied in this work. In the case of the aquatic humic substances the spectra consisted of a single broad band (sharper for the fulvic acids), while ALHA showed some additional poorly defined shoulders. In general, the emission spectrum of humic substances is intrinsically broad, because of severe overlapping occurring for different signals (Miano and Senesi, 1992). It can also be seen that the fulvic acids showed greater fluorescence intensities than the humic acids (in the following order: SUFA N NOFA N SUHA N ALHA, although in this case SUHA and ALHA showed similar intensities); likewise, the fulvic acids showed their maxima peaks at lower λem than the humic acids, in the following order: SUFA (460 nm) b NOFA (463 nm) b SUHA (464 nm) b ALHA (470 nm). The explanation is similar to that provided for the case of the EEM spectra; in fact, it can be observed in Fig. S2 (line d) that the emission spectra for the aquatic humic substances actually correspond to cross sections passing through peak C in EEM spectra. In fact, given that the excitation wavelength selected for the acquisition of the spectra (λex = 340 nm) almost exactly coincided with the excitation wavelength corresponding to peak C of aquatic humic substances (λex = 335 nm), the maxima peaks of the emission spectra may be considered equivalent to peak C of the EEM spectra. The situation is different for ALHA, as it presents an EEM spectrum that is clearly different from those of the aquatic humic substances (there is no peak C in this case). The position of the maxima peaks found in the present study for the aquatic humic substances coincides with the literature: Senesi reported λem values of 461 nm for SUFA, 465 nm for NOFA and 471 nm for SUHA (Senesi et al., 1989; Senesi, 1990) and McKnight reported a value of 460–461 nm for SUFA (McKnight et al., 2001). McKnight also proposed that the position of the emission maxima could be used to predict the origin (either terrestrially or microbially derived) of aquatic fulvic acids (he also proposed that the position of peaks A and C in EEM spectra could also be used for that purpose: see Section 3.1), indicating that fulvic acids from microbially derived environments had emission maxima at shorter wavelengths (between 442 and 448 nm) than those from terrestrially derived environments (between 457 and 461 nm), which would suggest a terrestrial origin for SUFA and NOFA according to the results obtained in the present study, a fact that has previously been reported in the literature for the case of SUFA (IHSS, 2012; McKnight et al., 2001). As mentioned earlier, the emission spectra of the fulvic acids were sharper than those of the humic acids (see Fig. 5), which is in agreement with the degrees of polydispersity reported in the literature for the four humic substances (polydispersity calculated as the Mw/Mn ratio, where Mw is the weight-averaged molecular weight and Mn is the number-


350

50

400

450

500

550

600

650 SUFA SUHA NOFA ALHA

SUFA

40

50

40

NOFA

30

30

20

20

10 SUHA

0 350

0 400

450

500

averaged molecular weight): SUFA (1.50–1.77) b NOFA (1.58) b SUHA (1.86–2.11) b ALHA (2.50) (Chin et al., 1994, 1997; Perminova et al., 2003). With regard to ALHA, it again showed a different behavior and its emission spectrum did not coincide with those reported in the literature for terrestrial humic acids: the latter showed the maximum peak at λem around 500–510 nm (Chen et al., 2002; Senesi et al., 1989) whereas ALHA showed it at λem = 470 nm. Other researchers (Uyguner and Bekbolet, 2010) have also found the same result for another commercial humic acid (Roth humic acid), which showed its maximum peak at λem = 450 nm. 3.4. Fluorescence index McKnight et al. (2001) also proposed the use of a new parameter for the distinction between terrestrially and microbially derived fulvic acids, which is referred to as “fluorescence index” and defined as the ratio of emission intensity at 450/500 nm for emission spectra measured at an excitation wavelength of 370 nm. McKnight obtained values for that parameter around 1.9 for microbially derived fulvic acids and around 1.4 for terrestrially derived fulvic acids. In the present study, the fluorescence index values obtained for the four humic substances were in the following order, as shown in Table 4: SUFA (1.44) N NOFA (1.30) N SUHA (1.19) N ALHA (0.99), which allows to include the fulvic acids SUFA and NOFA in the category of terrestrially derived fulvic acids, a result that coincides with those described in Sections 3.1 and 3.3. The fluorescence indices found in this work coincide with those reported in the literature for the humic substances under study, since values of 1.3–1.4 (McKnight et al., 2001), 1.34 (Kim et al., 2006) and 1.41 (Chen et al., 2003) have previously been reported for SUFA, values of 1.06 for SUHA and values of 0.93 for ALHA (Kim et al., 2006). Once again, ALHA presented significantly different values to those shown by terrestrial humic acids, as fluorescence indices in the range of 0.60–0.80 have been reported for the latter in the literature (Chen et al., 2003). The existence of a negative correlation between the fluorescence index and the aromaticity of humic substances has also been reported in the literature (McKnight et al., 2001; Kim et al., 2006). That possible correlation was also evaluated in this study by taking the ratio of the area of aromatic carbon region (110–165 ppm) to the total area of the 13 C NMR spectra (data from IHSS for the aquatic humic substances and from Ashley, 1996 for ALHA) and SUVA (UV254/TOC) obtained in this study as parameters that are indicative of aromaticity. The results of these correlations are shown in Fig. S3 (see online Supplementary materials), observing a good linear relationship of the fluorescence index with both 13C NMR aromaticity (R2 = 0.94) and SUVA (R2 = 0.92), supporting the suggestion that the fluorescence index might serve as a surrogate for the aromaticity of humic substances (both fulvic and humic acids), although the limited number of humic substances studied in the present work (two natural fulvic acids, a natural humic acid and a commercial humic acid) does not allow to draw final conclusions. McKnight et al. (2001) used 18 natural aquatic fulvic acids in his study, in which he also obtained a good correlation (R2 = 0.85), although no humic acid was included in his work. Finally, Kim et al. (2006) reported that a difference in the fluorescence index of at least 0.1 might be

Table 4 Fluorescence index, λ0.5 parameter, SUVA and aromaticity values for the humic substances.

10

ALHA

550

600

emission wavelength (nm) Fig. 5. Emission scan spectra of the humic substances.

650

727

Fluorescence index λ0.5 parameter (nm) SUVA (L/mg-m) % aromaticity (13C NMR)a

SUFA

SUHA

NOFA

ALHA

1.44 516 4.5 24

1.19 534 6.5 31

1.30 523 5.2 31

0.99 563 7.0 41

a Values are from IHSS (2012) (for SUFA, SUHA and NOFA) and from Ashley (1996) (for ALHA).

728


450

indicative of a difference in aromaticity of humic substances. In the present work, the differences between the fluorescence indices of the humic substances were always greater than 0.1, pointing to significant differences in their aromatic character.

500

550

600

1

650 1

SUFA SUHA NOFA ALHA

SUHA

0,8

There is still another parameter related to fluorescence techniques that has been used (although less widely than EEM, synchronous and emission spectra) in order to study the humic substances and was introduced by Fabbricino and Korshin (2004): that parameter is the wavelength corresponding to the normalized emission band at its half-intensity (for wavelengths N λmax) and is denoted as λ0.5 (Kim et al., 2006). In this work, that parameter was determined in the fluorescence emission spectra (excitation at 340 nm, emission in the range 450–650 nm) and the results for the four humic substances under study are shown in Fig. 6 and in Table 4, where it can be observed that λ0.5 values appear in the following order: SUFA (516 nm) b NOFA (523 nm) b SUHA (534 nm) b ALHA (563 nm). According to the literature (Kim et al., 2006), a decrease in λ0.5 is likely to be associated with a blue-shift (the emission shifts towards a shorter wavelength), which can be attributed to a reduction in the πelectron system, such as a decrease in the number of aromatic rings (Coble, 1996), therefore a positive correlation should be expected between λ0.5 and aromaticity, a result that was confirmed in this study (see Fig. S4, online Supplementary materials) with R2 = 0.92 for the correlation λ0.5 vs. 13C NMR aromaticity (see Section 3.4 on data sources) and R2 = 0.82 for the correlation λ0.5 vs. SUVA (data from this study), although as mentioned in Section 3.4, the limited number of humic substances studied in the present work is not enough to extrapolate the results to the whole of humic substances (both aquatic and terrestrial). 4. Conclusions – A comprehensive structural characterization of humic substances has been conducted in this study using all the current fluorescence techniques: total luminescence spectroscopy (TLS, also referred to as excitation emission matrix—EEM), synchronous scan fluorescence (SFS), emission scan fluorescence (ESF), fluorescence index and λ0.5 parameter, which allowed us to compare the results and to establish interrelations between the different techniques. In particular, fluorescence has been applied to the characterization of the following humic substances: three natural aquatic humic substances supplied by IHSS (Suwannee River Fulvic Acid Standard—SUFA, Suwannee River Humic Acid Standard—SUHA, Nordic Reservoir Fulvic Acid Reference—NOFA) and a commercial humic acid (Aldrich Humic Acid: ALHA). – Concerning the natural humic substances, their EEM spectra (2-D contour maps and 3-D plots) were quite similar, showing two main peaks of maximum fluorescence intensity: one located in the ultraviolet region at around Ex/Em values of 230/437 nm (peak A) and another one in the visible region at around 335/460 nm (peak C). The positions of the peaks have been related in this study to some structural characteristics of the humic substances, such as the predominant presence of either carboxylic groups or phenolic groups in the macromolecule and also to identify the origin of aquatic fulvic acids (terrestrially or microbially derived fulvic acids). Correlations have also been obtained between peak C intensity and the following parameters: carboxylic acidity, phenolic acidity and SUVA. Synchronous scan fluorescence at different offsets (Δλ = 18, 44 and 66 nm) allowed the identification of a proteinlike peak at λsyn around 290 nm, which was not detected in the EEM spectra. The emission scan spectra of the humic substances consisted of a single broad band and the fulvic acids showed sharper

F.I/F.Imax

3.5. The λ0.5 parameter

0,8 SUFA

0,6

0,6 534 516

λ 0.5

563

523

0,4

0,4 ALHA

0,2

0,2

NOFA

0 450

0 500

550

600

650

Emission wavelength (nm) Fig. 6. λ0.5 parameter for the humic substances.

spectra, maxima peaks located at lower λem and higher fluorescence intensities than the humic acids. Both fluorescence index and λ0.5 parameter showed good correlations with typical parameters indicative of humic substances aromaticity, such as 13C NMR aromaticity and SUVA. – The commercial humic acid ALHA, very often used as a surrogate for natural humic substances in studies on coagulation–flocculation, filtration, activated carbon adsorption and disinfection by-products formation, showed very different spectra to those shown by the natural aquatic humic substances for all the fluorescence techniques studied in the present work. This suggests an important structural difference between ALHA and the natural humic substances, which casts doubt on its use as a surrogate for those substances. – Although some interesting results were achieved in this study, the limited number of humic substances studied in this work (two natural aquatic fulvic acids, a natural aquatic humic acid and a commercial humic acid) does not allow us to extrapolate the results to the whole of humic substances. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2013.11.150. Conflict of interest No conflict of interest. References Aiken G, Costaris E. Soil and hydrology: their effect on NOM. J AWWA 1995;87:36–44. Alberts JA, Takács M. Total luminescence spectra of IHSS standard and reference fulvic acids, humic acids and natural organic matter: comparison of aquatic and terrestrial source terms. Org Geochem 2004a;35:243–56. Alberts JA, Takács M. Comparison of the natural fluorescence distribution among size fractions of terrestrial fulvic and humic acids and aquatic natural organic matter. Org Geochem 2004b;35:1141–9. Alberts JA, Takács M, Egeberg PK. Total luminescence spectral characteristics of natural organic matter (NOM) size fractions as defined by ultrafiltration and high performance size exclusion chromatography (HPSEC). Org Geochem 2002;33:817–28. Allpike BP, Heitz A, Joll CA, Kagi RI. Size exclusion chromatography to characterize DOC removal in drinking water treatment. Environ Sci Technol 2005;39:2334–42. Andrews SA, Huck PM. Using fractionated natural organic matter to study ozonation by-product formation. In: Minear RA, Amy GL, editors. Disinfection by-products in water treatment—the chemistry of their formation and control. Florida: Lewis Publishers; 1996. p. 126–40. Ashley JTF. Adsorption of Cu(II) and Zn(II) by estuarine, riverine and terrestrial humic acids. Chemosphere 1996;33:2175–87. Assemi S, Newcombe G, Hepplewhite C, Beckett R. Characterization of natural organic matter fractions separated by ultrafiltration using flow field-flow fractionation. Water Res 2004;38:1467–76. Baker A. Fluorescence excitation–emission matrix characterization of some sewageimpacted rivers. Environ Sci Technol 2001;35:948–53.

F.J. Rodríguez et al. / Science of the Total Environment 476–477 (2014) 718–730 Belin C, Quellec C, Lamotte M, Ewald M, Simon Ph. Characterization by fluorescence of the dissolved organic matter in natural water. Application to fractions obtained by tangential ultrafiltration and XAD resin isolation. Environ Technol 1993;14:1131–44. Budavari S, editor. The Merck Index. NJ (USA): Merck & Co; 1996. Cabaniss SE, Zhou Q, Maurice PA, Chin YP, Aiken GR. A log-normal distribution for the molecular weight of aquatic fulvic acids. Environ Sci Technol 2000;34: 1103–9. Chandrakanth MS, Amy GL. Effects of ozone on the colloidal stability and aggregation of particles coated with natural organic matter. Environ Sci Technol 1996;30: 431–43. Chen J, Gu B, LeBooeuf E, Pan H, Dai S. Spectroscopic characterization of the structural and functional properties of natural organic matter fractions. Chemosphere 2002;48: 59–68. Chen J, LeBooeuf E, Pan H, Dai S, Gu B. Fluorescence spectroscopic studies of natural organic matter fractions. Chemosphere 2003;50:639–47. Chiang PC, Chang EE, Chang PC, Huang CP. Effects of pre-ozonation on the removal of THM precursors by coagulation. Sci Total Environ 2009;407:5735–42. Chin YP, Aiken G, O'Loughlin G. Molecular weight, polydispersity and spectroscopic properties of aquatic humic substances. Environ Sci Technol 1994;28:1853–8. Chin YP, Aiken G, Danielsen KM. Binding of pyrene to aquatic and commercial humic substances: the role of molecular weight and aromaticity. Environ Sci Technol 1997;31: 1630–5. Choudry GG. Humic substances—structural, photophysical, photochemical and free radical aspects of interactions with environmental chemicals. New York: Gordon and Breach; 1984. Coble PG. Characterization of marine and terrestrial DOM in seawater using excitation– emission matrix spectroscopy. Mar Chem 1996;51:325–46. De Souza MM, Donard OFX, Lamotte M, Belin C, Ewald M. Fluorescence spectroscopy of coastal and marine waters. Mar Chem 1994;47:127–44. Duarte R, Santos E, Duarte A. Spectroscopic characteristics of ultrafiltration fractions of fulvic and humic acids isolated from a eucalyptus bleached Kraft pulp mill effluent. Water Res 2003;37:4073–80. Elliot S, Lead JR, Baker A. Characterization of the fluorescence from freshwater, planktonic bacteria. Water Res 2006;40:2075–83. Fabbricino M, Korshin GV. Probing the mechanisms of NOM chlorination using fluorescence: formation of disinfection by-products in Alento River water. Water Sci Technol Water Supply 2004;4:227–33. Galapate RP, Kitanaba A, Kazuaki I. Origin of thihalomethane precursors in Kurose River, Hiroshima, Japan. Water Sci Technol 1997;35:15–20. Ghosh K, Schnitzer M. Fluorescence excitation spectra of humic substances. Can J Soil Sci 1980;60:373–9. Hautala K, Peuravuori J, Pihlaja K. Measurement of aquatic humus content by spectroscopic analyses. Water Res 2000;34:246–58. Henderson RK, Baker A, Murphy KR, Hambly A, Stuetz RM, Kahn SJ. Fluorescence as a potential monitoring tool for recycled water systems: a review. Water Res 2009;43: 863–81. Her N, Amy G, McKnight D, Sohn J, Yoon Y. Characterization of DOM as a function of MW by fluorescence EEM and HPLC-SEC using UVA, DOC and fluorescence detection. Water Res 2003;37:4295–303. Hongve D, Baann J, Becher G. Characterization of humic substances by means of high performance size exclusion chromatography. Environ Int 1996;22:489–94. Hudson N, Baker A, Ward D, Reynolds D, Brunsdon C, Carliell-Marquet C, Browning S. Can fluorescence spectrometry be used as a surrogate for the Biochemical Oxygen Demand (BOD) test in water quality assessment? An example from South West England. Sci Total Environ 2008;391:149–58. International Humic Substances Society: IHSS. http://www.humicsubstances.org, 2012. Kalbitz K, Geyer S, Geyer W. A comparative characterization of dissolved organic matter by means of original aqueous samples and isolated humic substances. Chemosphere 2000;40:1305–12. Kim HC, Yu MJ, Han I. Multi-method study of the characteristic chemical nature of aquatic humic substances isolated from the Han River, Korea. Appl Geochem 2006;21: 1226–39. Korshin GV, Kumbe MU, Li CW, Frimmel FH. Influence of chlorination on chromophores and fluorophores in humic substances. Environ Sci Technol 1999;33: 1207–12. Langlais B, Reckhow A, Brink D. Ozone in water treatment: application and engineering. Michigan (USA): AWWA Research Foundation and Compagnie Générale des Eaux; 1991. Lanxiu Y, Fengchang W, Congquiang L, Wen L, Zhicheng L. Relationship between fluorescence characteristics and molecular weight distribution of natural dissolved organic matter in Lake Hongfeng and Lake Baihua, China. Chin Sci Bull 2006;51:89–96. Leenheer JA, Brown PA, Noyes TI. Implications of mixture characteristics on humic-substance chemistry. In: Suffet IH, Mac&thy PA, editors. Aquatic humic substances, influence on fate and treatment of pollutantsAdv Chem Ser; 1989. p. 25–39. Liu R, Lead JR, Baker A. Fluorescence characterization of cross flow ultrafiltration derived freshwater colloidal and dissolved organic matter. Chemosphere 2007;68: 1304–11. Liu S, Lim M, Fabris R, Chou C, Chiang K, Amal R. Removal of humic acid using TiO2 photocatalytic process—fractionation and molecular weight characterization studies. Chemosphere 2008;72:263–71. Ma YS. Reaction mechanisms for DBPs reduction in humic acid ozonation. Ozone Sci Eng 2004;26:153–64. Ma H, Allen H, Yin Y. Characterization of isolated fractions of dissolved organic matter from natural waters and a wastewater effluent. Water Res 2001;35:985–96. Malcolm RL, McCarthy P. Limitations in the use of commercial humic acids in water and soil research. Environ Sci Technol 1986;20:904–11.

729

Mao JD, Tremblay L, Gagné JP, Hohl S, Rice J, Schmidt-Rohr K. Humic acids from particulate organic matter in the Saguenay Fjord and the St. Lawrence Estuary investigated by advanced solid-state NMR. Geochim Cosmochim Acta 2007;71: 5483–99. McDonald S, Pringle JM, Bishop AG, Prenzler PD, Robards K. Isolation and seasonal effects on characteristics of fulvic acid isolated from an Australian floodplain river and billabong. J Chromatogr A 2007;1153:203–13. McKnight DM, Byer EW, Westerhoff PK, Doran PT, Kulbe T, Andersen DT. Spectrofluorometric characterization of dissolved organic matter for indication of precursor organic material and aromaticity. Limnol Oceanogr 2001;46:38–48. Miano TM, Senesi N. Synchronous excitation fluorescence spectroscopy applied to soil humic substances chemistry. Sci Total Environ 1992;117:41–51. Mobed JJ, Hemmingsen SL, Autry JL, McGown LB. Fluorescence characterization of IHSS humic substances: total luminescence spectra with absorbance correction. Environ Sci Technol 1996;30:3061–5. Monteil-Rivera F, Brouwer EB, Masset S, Deslandes Y, Dumonceau J. Combination of X-ray photoelectron and solid-state 13C nuclear magnetic resonance spectroscopy in the structural characterisation of humic acids. Anal Chim Acta 2000;424: 243–55. Mopper K, Schultz CA. Fluorescence as a possible tool for studying the nature and water column distribution of DOC components. Mar Chem 1993;41:229–38. Mounier S, Patel N, Quilici L, Benaim JY, Benamou C. Three-dimensional fluorescence of the dissolved organic carbon in the Amazon River. Water Res 1999;33:1523–33. Muller MB, Fritz W, Lankes U, Frimmel F. Ultrafiltration of non-ionic surfactants and dissolved organic matter. Environ Sci Technol 2004;38:1124–32. Murphy KR, Stedmon CA, Waite TD, Ruiz GM. Distinguishing between terrestrial and autochthonous organic matter sources in marine environments using fluorescence spectroscopy. Mar Chem 2008;108:40–58. Myllykangas T, Nissinen TK, Rantakkoko P, Martikainen PJ, Vertiainen T. Molecular size fractions of treated aquatic humus. Water Res 2002;36:3045–53. Parlanti E, Wörz K, Geoffroy L, Lamotte M. Dissolved organic matter fluorescence spectroscopy as a tool to estimate biological activity in a coastal zone submitted to anthropogenic inputs. Org Geochem 2000;31:1765–81. Peiris BRH, Budman H, Moresoli C, Legge RL. Acquiring reproducible fluorescence spectra of dissolved organic matter at very low concentrations. Water Sci Technol 2009;60: 1385–92. Peiris BRH, Budman H, Moresoli C, Legge RL. Identification of humic acid-like and fulvic acid-like natural organic matter in river water using fluorescence spectroscopy. Water Sci Technol 2011;63:2427–33. Perminova IV, Frimmel FH, Kudryavtsev AV, Kulikova NA, Abbt-Braun G, Hesse S, Petrosyan VS. Molecular weight characteristics of humic substances from different environments as determined by size exclusion chromatography and their statistical evaluation. Environ Sci Technol 2003;37:2477–85. Pernet-Coudrier B, Varrault G, Saad M, Croué JP, Dignac MF, Mouchel JM. Characteristics of dissolved organic matter in Parisian urban aquatic systems: predominance of hydrophilic and proteinaceous structures. Biogeochemistry 2011;106: 89–106. Persson T, Wedborg M. Multivariate evaluation of the fluorescence of aquatic organic matter. Anal Chim Acta 2001;434:179–92. Peuravuori J, Koivikko R, Pihlaja K. Characterization, differentiation and classification of aquatic humic matter separated with different sorbents: synchronous scanning fluorescence spectroscopy. Water Res 2002;36:4552–62. Rodríguez FJ, Núñez LA. Characterization of aquatic humic substances. Water Environ J 2011;25:163–70. Rodríguez FJ, Marcos LA, Núñez LA. Effects of ozonation on natural organic matter reactivity in adsorption and biodegradation processes—a case study: the Úzquiza Reservoir water. Ozone Sci Eng 2011;33:185–93. Rodríguez FJ, Marcos LA, Núñez LA, García M. Effects of ozonation on molecular weight distribution of humic substances and coagulation processes—a case study: the Úzquiza Reservoir water. Ozone Sci Eng 2012a;34:342–53. Rodríguez FJ, Marcos LA, Núñez LA. Influence of preozonation on the formation of chlorination disinfection byproducts—a case study: the Úzquiza Reservoir water. Ozone Sci Eng 2012b;34:213–24. Rosa AH, Simões ML, de Oliveira LC, Rocha JC, Neto LM, Milori DM. Multimethod study of the degree of humification of humic substances extracted from different tropical soil profiles in Brazil's Amazonian region. Geoderma 2005;127: 1–10. Senesi N. Molecular and quantitative aspects of the chemistry of fulvic acid and its interactions with metal ions and organic chemicals. Part II: the fluorescence spectroscopy approach. Anal Chim Acta 1990;232:77–106. Senesi N, Miano TM, Provenzano MR, Brunetti G. Spectroscopic and compositional comparative characterization of IHSS reference and standard fulvic and humic acids of various origin. Sci Total Environ 1989;81:143–56. Sierra MMD, Giovanela M, Parlanti E, Soriano-Sierra EJ. Fluorescence fingerprint of fulvic and humic acids from varied origins as viewed by single-scan and excitation/ emission matrix techniques. Chemosphere 2005;58:715–33. Silva CSPCO, Esteves da Silva JCG, Machado AASC. Evolving factor analysis of synchronous fluorescence spectra of fulvic acids in the presence of aluminium. Appl Spectrosc 1994;48:363–71. Smith DS, Kramer JR. Fluorescence analysis for multi-site aluminium binding to natural organic matter. Environ Int 1999;25:295–306. Stedmon CA, Markager S, Bro R. Tracing dissolved organic matter in aquatic environments using a new approach to fluorescence spectroscopy. Mar Chem 2003;82: 239–54. Stewart AJ, Wetzel RG. Fluorescence: absorbance ratios—a molecular-weight tracer of dissolved organic matter. Limnol Oceanogr 1980;25:559–64.

730


Swietlik J, Sikorska E. Application of fluorescence spectroscopy in the studies of natural organic matter fractions reactivity with chlorine dioxide and ozone. Water Res 2004;38:3791–9. Thorn KA, Younger SJ, Cox LG. Order of functionality loss during photodegradation of aquatic humic substances. J Environ Qual 2010;39:1416–28. Thurman EM. Developments in biogeochemistry: organic geochemistry of natural waters. Dordrecht, The Netherlands: M. Nijhoff and Dr. W Junk Publishers; 1985. Tsuda K, Mori H, Asakawa D, Yanagi Y, Yonebayashi K, Fujitake N. Characterization and grouping of aquatic fulvic acids isolated from clear-water rivers and lakes in Japan. Water Res 2010;44:3837–46. Uyguner CS, Bekbolet M. Evaluation of humic acid photocatalytic degradation by UV–vis and fluorescence spectroscopy. Catal Today 2005;101:267–74. Uyguner CS, Bekbolet M. TiO2-assisted photocatalytic degradation of humic acids: effect of copper ions. Water Sci Technol 2010;61:2581–90.

Wolfbies OS. The fluorescence of organic natural products. In: Schulman SG, editor. Molecular luminescence spectroscopy, part I: methods and applications. New York: Wiley; 1985. p. 167–370. Yamashita Y, Jaffe R. Characterizing the interactions between metals and dissolved organic matter using excitation–emission matrix and parallel factor analysis. Environ Sci Technol 2008;42:7374–9. Yamashita Y, Tanoue E. Chemical characterization of protein-like fluorophores in DOM in relation to aromatic amino acids. Mar Chem 2003;82:255–71. Zhang T, Lu J, Ma J, Qiang Z. Fluorescence spectroscopic characterization of DOM fractions isolated from a filtered river water after ozonation and catalytic ozonation. Chemosphere 2008;71:911–21. Zhou Q, Cabaniss SE, Maurice PA. Considerations in the use of high performance size exclusion chromatography (HPSEC) for determining molecular weights of aquatic humic substances. Water Res 2000;34:3505–14.

Generation of mutagenicity by ozonation of humic substances' components.

Characterization and quantification of humic substances 2D-Fluorescence by usage of extended size exclusion chromatography.

Natural carbon-based dots from humic substances.

Humic substances shorten human plasma prothrombin time.

A fluorescence method for studies of spent sulfite liquor and humic substances in sea water.

Production of humic substances through coal-solubilizing bacteria.

Binding mechanisms of pesticides to soil humic substances.

Determination of the carboxyl content in humic substances by methylation.

Electrocoagulation treatment of peat bog drainage water containing humic substances.

River-derived humic substances as iron chelators in seawater.

Prion protein interaction with soil humic substances: environmental implications.

Effect of humic substances on phosphorus removal by struvite precipitation.

Linking fluorescence spectroscopy to diffuse soil source for dissolved humic substances in the Daning River, China.

Investigating chloroperoxidase-catalyzed formation of chloroform from humic substances using stable chlorine isotope analysis.

Metal distribution and spectroscopic analysis after soil washing with chelating agents and humic substances.

Comparative sorption and desorption behaviors of PFHxS and PFOS on sequentially extracted humic substances.

Photosensitised humic-like substances (HULIS) formation processes of atmospheric significance: a review.

Physicochemical and ion-binding properties of highly aliphatic humic substances extracted from deep sedimentary groundwater.

Enhanced photoproduction of hydrogen peroxide by humic substances in the presence of phenol electron donors.

Monitoring transitory profiles of leachate humic substances in landfill aeration reactors in mesophilic and thermophilic conditions.

BDE-209: kinetic studies and effect of humic substances on photodegradation in water.

Methanogenesis affected by the co-occurrence of iron(III) oxides and humic substances.

Photooxidation-induced changes in optical, electrochemical, and photochemical properties of humic substances.

Mitigation of Fe(0) nanoparticles toxicity to Trichosporon cutaneum by humic substances.