Composition of organic sulfur in riverine and marine sediments: Insights from sulfur stable isotopes and XANES spectroscopy

Accepted Manuscript Composition of organic sulfur in riverine and marine sediments: Insights from sulfur stable isotopes and XANES spectroscopy Mao-Xu Zhu, Liang-Jin Chen, Gui-Peng Yang, Xiang-Li Huang, Yi-Dong Zhao PII: DOI: Reference:

S0146-6380(16)30090-0 http://dx.doi.org/10.1016/j.orggeochem.2016.07.002 OG 3428

To appear in:

Organic Geochemistry

Received Date: Revised Date: Accepted Date:

9 October 2015 4 July 2016 11 July 2016

Please cite this article as: Zhu, M-X., Chen, L-J., Yang, G-P., Huang, X-L., Zhao, Y-D., Composition of organic sulfur in riverine and marine sediments: Insights from sulfur stable isotopes and XANES spectroscopy, Organic Geochemistry (2016), doi: http://dx.doi.org/10.1016/j.orggeochem.2016.07.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Composition of organic sulfur in riverine and marine sediments: Insights from sulfur stable isotopes and XANES spectroscopy

Mao-Xu Zhu a∗, Liang-Jin Chen a, Gui-Peng Yang a, Xiang-Li Huang a, Yi-Dong Zhao b a

Key Laboratory of Marine Chemistry Theory and Technology, Ministry of Education, College of Chemistry and Chemical Engineering, Ocean University of China, Qingdao 266100, China b

Beijing Synchrotron Radiation Facility, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing 100049, China

ABSTRACT Sulfur isotope and X-ray absorption near edge structure (XANES) spectroscopy were combined to characterize/compare three operational organic sulfur (OS) pools, i.e. fulvic acid sulfur (FA-S), humic acid sulfur (HA-S) and non-chromium reducible organic sulfur (non-CROS) in marine [the East China Sea (ECS) and Jiaozhou Bay (JZB)] vs. riverine [Yangtze River (YR) and JZB tributaries] sediments. XANES results indicate that in marine sediments high valency S was the dominant OS functionality in both HA-S and FA-S, while non-CROS was dominated by low valency OS (80−92%). In riverine sediments FA-S was dominated by high valency OS, while the average fractions of low and high valency OS in HA-S were comparable. The isotopic composition of FA-S (δ34SFA-S) and HA-S (δ34SHA-S) indicated that a substantial fraction of sulfide was incorporated into FA via sulfurization in the marine sediments, whereas terrigenous OS was almost the sole ∗

Corresponding author. Tel.: +86 532 66782513; fax: +86 532 66782540. E mail address: [email protected] (M.-X. Zhu).

1

important source of HA-S. Compared with the ECS sediments, JZB sediments had more depleted 34SFA-S and a higher fraction of highly reduced FA-S due to eutrophication induced sulfurization. Relative to riverine FA-S, substantial sulfurization had not resulted in an increase in the fraction of highly reduced FA-S in marine sediments. This implies that the terrestrial systems may be much more favorable for the formation and/or preservation of highly reduced biogenic FA-S than the marine settings. The fraction of highly reduced HA-S in the JZB and its tributaries was similar, whereas the fraction was much lower in the ECS than in the YR. This indicates that highly reduced HA-S moieties in the YR may have been subject to extensive mineralization loss during transport in the large riverine/estuarine systems, whereas the process in the JZB tributaries may be much weakened due to a relatively small catchment area. Distinct differences in isotopic and structural composition between humic-S (FA-S + HA-S) and non-CROS in the ECS sediments indicate that a combination of S isotopes and S-XANES is needed for characterizing the two operational pools for a better understanding the nature of OS in the ocean. Keywords: Organic sulfur; sulfur isotopes; XANES; sediments; East China Sea; Jiaozhou Bay

1. Introduction Sulfur in the natural environment occurs in both inorganic and organic forms with oxidation state varying from −2 to + 6. In unpolluted soils, peatlands and riverine sediments, organic S (OS) is often the dominant S form, accounting for > 90% of the

2

total S. In marine sediments it is the second most abundant S fraction (only after pyrite, FeS2) and frequently accounts for 35% of total sedimentary S (Anderson and Pratt, 1995; Werne et al., 2004, 2008; Amrani, 2014). However, in some specific localities, OS can represent the major S fraction (up to 85%; Zaback and Pratt, 1992; Werne et al., 2004). Reduced OS is often the dominant S form in soils and sediments, while highly oxidized OS, such as ester sulfates, is also quantitatively important in both oxic and anoxic settings, although mechanisms for their formation remain poorly understood (Vairavamurthy et al., 1994, 1997; Xia et al., 1998; Morgan et al., 2012). Reduced OS is formed through two basic pathways. The first is by assimilatory sulfate reduction and subsequent formation of S-requiring cellular components (Francois, 1987a; Canfield et al., 2005; Werne et al., 2008). The second is through incorporation of various reduced inorganic S species (particularly HS− and

Sx2−)

originally formed by way of microbial sulfate reduction in anoxic conditions and later becoming bound to organic matter (OM) via a range of possible mechanisms during early diagenesis, i.e. sulfurization of OM (Brüchert, 1998; Werne et al., 2003; Canfield et al., 2005; Bottrell et al., 2010). Sulfurization has important implications for OM burial since macromolecules formed through di- and polysulfide cross linking appear to be resistant to microbial degradation (Sinninghe Damsté et al., 1989; Ferdelman et al., 1991), providing a mechanistic interpretation for enhanced preservation of OM in marine sediments (see Sinninghe Damsté and de Leeuw, 1990 for a review). This also implies that reduced OS moieties formed through sulfurization are much less vulnerable to oxidation than metal sulfides. In addition, reduced OS

3

functional groups such as thiol are highly effective in bonding to a number of pollutants such as Hg(II), Zn(II), Cd(II) and As(III) and therefore play an important role in controlling speciation and biotoxicity of these pollutants in aquatic environments (Xia et al., 1999; Langner et al., 2012; Manceau et al., 2015). In turn, the bonding of metals to OS moieties can retard or prevent oxidation of these OS functional groups when exposed to oxidizing conditions (Hutchison et al., 2001). Thus, quantitatively establishing the speciation of OS is crucial for an insight into its source, fate and environmental impact in aquatic systems. Extraction has been widely used for quantitative speciation of S in natural samples. One technique for OS speciation is extraction of humic substances with 0.5 M NaOH and subsequent quantification of humic-bound S (humic-S), given the fact that humic substances are usually the major components (up to 60−70%) of OM in soils and sediments, and humic-S is generally the most important OS species (Rashid, 1985; Buscail et al., 1995; Ferdelman et al., 1991; Henneke et al., 1997; Yücel et al., 2010). Another technique is quantification of residual OS after removal of pyrite by way of acidic Cr(II) reduction, i.e. non-Cr reducible OS (non-CROS) according to Canfield et al. (1998). This technique has been often used for marine sediments because it can circumvent possible contamination of coextracted pyrite in the base extraction (see below; Francois, 1987a, b; Passier et al., 1999; Zhu et al., 2013a). Both the above techniques are imperfect for quantification of bulk sedimentary OS. With respect to base extraction, a fraction of OS is non-extractable, and this pool is often non-trivial in marine settings (Ferdelman et al., 1991). Another potential

4

pitfall of base extraction is that pyrite may be coextracted with humic substances (Francois, 1987b). With respect to the non-CROS, coextraction of OS with acidic Cr(II) reductant has been found insignificant in earlier studies (Canfield et al., 1986; Passier et al., 1999), so non-CROS has long been assumed to provide a reasonable estimate of total OS (Passier et al., 1999). However, a later study demonstrated that acidic Cr(II) solution could coextract a fraction of OS from an OM-enriched marine sapropel (Mangrove Lake, Bermuda; Canfield et al., 1998). Despite the above inherent shortcomings, the two techniques are still widely used for the quantification of OS in marine sediments because of their ease of operation. However, they do not yield information that can resolve the sources (or pathways of formation), functionality and oxidation state of OS. In addition, no effort has been made to compare the compositional and structural differences between humic-S and non-CROS. Specific sources and pathways of S cycling in the natural environment impart their products with characteristic S isotopic composition. S isotopes are therefore widely used to trace S cycling in soils and aquatic systems (Zhu et al., 2013a, b, 2014; Amrani, 2014; Kang et al., 2014; Fike et al., 2015). The technique is especially powerful for tracing the pathways of OS formation (biogenic vs. diagenetic) and the timing of diagenetic sulfurization of OM (Brüchert and Pratt, 1996; Brüchert, 1998; Werne et al., 2003, 2004). Assimilatory sulfate reduction and subsequent biogenic OS formation result in only minor S isotope fractionation (1−3‰) relative to that of coexisting dissolved sulfate (Kaplan and Rittenberg, 1964; Peterson and Howarth,

5

1987; Fry et al., 1988), whereas OS formed via sulfurization is usually much more 34S depleted (from −5‰ to −30‰) due to the large S isotope fractionation associated with dissimilatory sulfate reduction (Werne et al., 2003; Amrani, 2014; Fike et al., 2015). Recent developments in compound specific S isotope analysis have put more precise limits on S sources/pathways, the reactivity of some specific organic compounds towards sulfurization and the timing of sulfurization (Werne et al., 2008; Raven et al., 2015). X-ray absorption near edge structure (XANES) spectroscopy is sensitive to electronic structure and oxidation state of the elements of interest and the geometry of neighboring atoms, and is therefore an element-selective technique capable of providing characteristic “fingerprint” information. The technique has emerged as a powerful tool for the direct speciation of elements at a molecular level (Henderson et al., 2014) and has been employed in the identification and quantification of electron oxidation states (EOSs) and the major functional groups of the OS in a variety of natural samples such as coal, petroleum, biosolids, soils, marine sediments and carbonaceous chondrites (Waldo et al., 1991; Morra et al., 1997; Huffman et al., 1991; Vairavamurthy et al., 1997; Xia et al., 1998; Hundal et al., 2000; Orthous-Daunay et al., 2010; Prietzel et al., 2011). S-XANES has been regarded as complementary to S isotope analysis for better deciphering complex S cycling (Schäfer et al., 2005; Einsiedl et al., 2007, 2008). It would be expected that a combination of traditional chemical extraction, S isotopes and S-XANES spectroscopy would greatly improve insights into S speciation,

6

S sources and the geochemical cycling in aquatic systems. Here, S isotopes and S-XANES were employed to characterize and compare humic-acid sulfur (HA-S) and fulvic-acid sulfur (FA-S) in marine [the East China Sea (ECS) and Jiaozhou Bay (JZB)] and riverine [Yangtze River (YR) and JZB tributaries] sediments. It was also used to characterize non-CROS in the ECS sediments. The objectives were to unravel: (i) the differences in source and structural composition of humic-S in/between riverine and marine sediments and (ii) the difference in structural composition between humic-S and non-CROS.

2. Study areas and sampling 2.1. Background of study area The YR is the longest in Eurasia, with a well developed tributary system, which covers a catchment area of 1.8×106 km2 (Gao and Wang, 2008). Before massive construction of dams, annual discharge of sediment from the river to the ECS was ca. 470 million tonnes, but the discharge has decreased to ca. 180 million tonnes due to dam construction and irrigation (Milliman and Farnsworth, 2011). The river dominated ECS is one of the world’s largest shelf seas (Fig. 1). Driven by the Jiangsu Coastal Current (JCC) and Zhejiang-Fujian Coastal Current (ZFCC), southward dispersal of the YR sediments occurs mainly along the inner shelf, developing an elongated mud wedge from the YR mouth into the Taiwan Strait (Xu et al., 2009). As a result, the muddy inner shelf is generally enriched in OM relative to the mid shelf and outer shelf (Kao et al., 2003; Zhu et al., 2012) and is thus an

7

important site for the deposition and mineralization of terrestrial and marine OM (Aller et al., 1985; Yao et al., 2014). In contrast to the large river dominated ECS, JZB is a semi-enclosed water body surrounded by Qingdao City and linked by a narrow channel (maximum width 3.1 km) to the Yellow Sea (Fig. 1). About 10 rivers, notably Dagu, Moshui, Baisha, Licun and Yanghe rivers, flow seasonally into the bay with variable freshwater discharge and sediment load (Liu et al., 2005). Since the 1980s, JZB has received excess riverine input of nutrients, mainly from the Dagu, Moshui, Baisha and Licun rivers (Liu et al., 2005), which has resulted in seasonal eutrophication of the bay and frequent red tide events since 1997 (Wu et al., 2005).

2.2. Sampling Surface sediments at four sites in the ECS inner shelf and JZB (Fig. 1), respectively, were collected using a box corer. Upon retrieval of each core, < 2 cm of surface sediment was collected with a plastic trowel and delivered to a zip-lock plastic bag where it was immediately frozen at −18 ℃. This approach was followed for four YR sediment samples (A, B, C and D; 118.78°E−118.82°E, 32.14°N−32.22°N) collected from the Nanjing section of the lower YR (Fig. 1), and four sediment samples sourced from the JZB tributaries of the Baisha, Dagu, Moshui and Yanghe rivers, respectively (3−4 km upstream from the river mouths).

3. Analytical methods

8

3.1. Humic-S extraction and S isotope analysis The procedure for extraction of humic-S by NaOH has been detailed by Zhu et al. (2013b, 2014). In short, wet sediment samples of known weight (ca. 5 g) in duplicate were washed with 1 M HCl 2× under N2 to remove SO42− and acid volatile S2−, and then washed 2× with Me2CO to remove elemental S. The washed pellets were treated immediately with 15 ml NaOH (0.5 M) under N2 for extraction of humic substances. After 24 h with stirring, the extracted humic substances were separated using centrifugation (4800 rpm) and filtration (0.45 µm). The extraction was repeated 6×. Separated humic substances were treated with dilute HCl to pH 2 to precipitate HAs. Precipitated HAs were separated via centrifugation for HA-S analysis, and the supernatant was saved for FA-S analysis. Subsamples for XANES analysis were stored at −18 °C until analysis. For isotope analysis of HA-S, subsamples were combusted at 800 °C to convert HA-S to SO42− following Eschka’s procedure, and the resultant SO42− was dissolved in deionized water at 60 °C for 30 min and collected via filtration. SO42− in the filtrate was quantitatively precipitated as BaSO4 by adjusting the pH to 2 and adding excess 10% (w/v) BaCl2 (10 ml). Precipitated BaSO4 was saved after drying at 60 °C for determination of HA-S isotopes. For isotope analysis of FA-S, the FA supernatant was treated with 20 ml H2O2 (30% v/v) at 60 °C for 1 h to oxidize the FA-S to SO42− and this procedure was repeated 3× for completion of the oxidation. The SO42− was precipitated as BaSO4 and recovered. The resultant BaSO4 was used for determination of FA-S isotopic composition. The S isotopic composition was analyzed using a Finnigan MAT-252 mass

9

spectrometer at the Institute of Geochemistry, Chinese Academy of Sciences (Xiao and Liu, 2011) and is reported in the δ notation normalized to Vienna Canyon Diablo Troilite (V-CDT). (34 S/ 32 S)sample δ SV-CDT ( 00) = ( 34 32 − 1) × 1000 ( S/ S) V-CDT 34

0

(1)

where (34S/32S)sample and (34S/32S)V-CDT are S isotope ratio values of the sample and standard, respectively. Linear scale correction was applied to yield δ34S of 0.5‰ and 20.3‰ for IAEA SO-5 and NBS-127 (BaSO4 standards). The standard deviation for δ34S analysis of NBS-127 was better than ± 0.2‰ (n = 5).

3.2. Non-CROS extraction and analysis Prior to extraction of non-CROS, pyrite was first removed with cold acidic Cr(II) reductant (Kallmeyer et al., 2004; Burton et al., 2008). The procedure is detailed by Zhu et al. (2012, 2013a). Residual pellets after pyrite reduction were washed 3× with deionized water to remove all dissolvable inorganic S. The residues were then stored at −18 °C until XANES analysis.

3.3. XANES spectroscopy K-edge XANES spectra of HA-S, FA-S and non-CROS were collected in fluorescence mode at beamline 4B7A of the Beijing Synchrotron Radiation Facility. A double-crystal Si (111) monochromator was used to diffract the X-ray beam with an energy resolution (∆E/E) of ca. 1.4×10−4. Elemental S was used as reference for energy calibration at 2472 eV. Procedure for the collection of XANES spectra has 10

been described elsewhere (Zhu et al., 2014). Briefly, freeze-dried and ground sample powder was scattered carefully onto S-free Kapton tape mounted on a stainless steel holder to minimize self-absorption, and run under ultra-high vacuum. XANES spectra were recorded in a range of 2460−2540 eV with a step size of 0.2 eV. The ATHENA software package (Ravel and Newville, 2005) was used for normalization, calibration and spectrum deconvolution by way of Gaussian curve fits. XANES data between 2465 and 2490 eV were used for the deconvolution. A series of Gaussian (white lines) functions (G1, G2, G3, G4, G5, and G6) and two arctangent functions (step height) were applied for the fitting following the optimized strategies of Manceau and Nagy (2012). EOS values of individual S functionalities were determined according to the linear correlation between EOS of various standard S compounds and their white-line peak energy compiled by Zeng et al. (2013): y = 0.6179x −1529

(2)

where y is the EOS and x the white-line peak energy. The EOS reflects the relative electronic charge density of S atoms in the valence shell in their actual bonding environments. The difference between the EOS and the formal oxidation state can be substantial, especially for reduced S species in complex organic compounds, depending on whether the S is bonded to S, hydrogen, carbon or metals (Vairavamurthy et al., 1997, 1998; Xia et al., 1998; Martínez et al., 2002). For high valency (EOS > 4) S species or S atoms bound to multiple O atoms, the difference is not significant due to the high electronegativity of O. In this study, non-integer EOS values are reported as calculated for low valency S (EOS ≤ 4) species; for high

11

valency S species calculated EOS values were, however, always rounded to integer values (Xia et al., 1998; Solomon et al., 2005, 2011; Zhao et al., 2006). The relative abundance of individual S functional groups was determined from the area under the respective Gaussian peak relative to the total area under all the Gaussian peaks. The areas were corrected for EOS-dependent absorption cross section using the generic calibration curve of Eq. 3 (Manceau and Nagy, 2012): y = 0.3841x − 909.97

(3)

where x is the energy of absorption maximum (i.e. corresponding to the oxidation state of the S functionality) and y is scaling factor normalized to y = 1 at the energy of elemental S (E 2472.70 eV). It has been reported that the error for OS fraction with the Gaussian curve fit is 4−8% for reduced and < 5% for oxidized S species (Huffman et al., 1991; Manceau and Nagy, 2012).

4. Results 4.1. Isotopic compositions of HA-S and FA-S The isotopic composition of HA-S (δ34SHA-S) for the ECS and JZB sediments was within a narrow range (Table 1) and had a similar average of 4.38‰ and 4.10‰, respectively. The δ34SFA-S values for the ECS and JZB sediments had a wide range and the average (−11.8 ± 7.6‰ and −18.5 ± 3.0‰, respectively) was much lower than the δ34SHA-S values. For the YR and JZB tributary sediments, both δ34SHA-S and δ34SFA-S values were within a narrow range (Table 1), while FA-S was slightly more 34

S-depleted, on average, than the corresponding HA-S. The δ34SHA-S values for the

12

riverine sediments were roughly comparable with those of HA-S in terrestrial systems from a wide variety of geographic distributions (2.3−4.6‰; Nissenbaum and Kaplan, 1972).

4.3. XANES spectroscopy Representative S-XANES spectra and Gaussian curve fits for HA-S, FA-S and non-CROS are shown in Fig. 2. Generally, the spectra could be fitted using five or six different Gaussian functions, indicating the presence of multiple EOSs. Shown inTables 2 and 3 are the EOS value and fraction of S functionalities in various OS pools of the marine and riverine sediments, respectively. Given the error (up to 8%) in the Gaussian curve fitting, some S functionalities with a small fraction, say < 5%, may be only nominally present. With this caveat in mind, OS functionality was evaluated only in groups, i.e. highly reduced S (EOS ≤ 1), weakly reduced S (1 < EOS ≤ 4), weakly oxidized S (EOS 5) and the most oxidized S (EOS 6). The highly reduced S with EOS ≤ 1 includes thiols, monosulfides, disulfides, polysulfides and thiophene (Vairavamurthy et al., 1994, 1997; Zhao et al., 2006; Solomon et al., 2011). Oxidized S with EOS 5 and 6 can be unambiguously assigned to sulfonates and ester sulfates, respectively (Vairavamurthy et al., 1994, 1997; Zhao et al., 2006; Solomon et al., 2011). High valency S (EOS 5 and 6) was the dominant OS functionality in both the HA-S and FA-S in the marine sediments (Table 2, Fig. 3a). For the ECS sediments, the fraction of high valency OS was 61−74% and 60−78% in the HA-S and FA-S,

13

respectively; for JZB sediments, it was 52−59% and 61−65%, respectively. For the riverine sediments, FA-S was dominated by high valency OS (Table 3, Fig. 3b), with the fraction ranging from 70−75% in the YR and 54−60% in the JZB tributarie. However, the fractions of the low and high valency OS from the riverine HA-S were not significantly different (Table 3, Fig. 3b) and the average fraction of the high valency OS was 50% and 52% in the YR and JZB tributaries, respectively. For both the marine and riverine sediments, the fractions of highly reduced S in HA-S and FA-S was much higher than that in the weakly reduced OS, with the latter always being marginal (Fig. 3). Non-CROS in the ECS sediments was dominated by low valency OS (80−92%) (Table 2, Fig. 3a), with highly reduced (G1 + G2) and weakly reduced S (G3 + G4) accounting for 33−58% and 25−48%, respectively. This feature was distinctly different from S functionalities in humic-S of the sediments.

5. Discussion 5.1. S isotope constraints on sources of humic-S 5.1.1. S isotope constraints on sources of riverine humic-S Terrestrial OS is almost all of biological origin, so has an isotopic composition similar to δ34SO4 of dissolved SO42− in the catchment basin since OS formed through assimilatory SO42− reduction has only minor S isotope fractionation (1−3‰; Kaplan and Rittenberg, 1964; Peterson and Howarth, 1987; Trust and Fry, 1992). Dissolved SO42- has multi-sources, including oxidative weathering of S2− bearing minerals,

14

dissolution of evaporates and atmospheric deposition in the catchment basins (Pawellek et al., 2002; Rock et al., 2009). Roughly similar values of δ34SHA-S for the YR sediments and δ34SO4 (5.3−5.6‰) of the river water in the Nanjing section (Li et al., 2011) seemingly indicate that the δ34SHA-S is essentially controlled by the river δ34SO4. It would be expected that δ34SHA-S in JZB tributaries would also be controlled by δ34SO4 in the catchment basin, though δ34SO4 was not determined. Similar δ34SHA-S in the YR and JZB tributary sediments (Table 1) indicates that the isotopic composition of HA-S does not exhibit regional differences in the two contrasting terrestrial systems. With respect to the YR, the similarity between δ34 SHA-S and the river δ34SO4 may indicate that HA-S has not been subjected to appreciable isotope variation after its formation, whereas lower 34SFA-S values may indicate incorporation of isotopically lighter S2− into FAs or preferential decomposition loss of isotopically heavier FA-S. Incorporation of S2− into FAs has been identified in reducing microenvironments of terrestrial systems (Schäfer et al., 2005; Einsiedl et al., 2007, 2008). Consistently lower 34SFA-S than δ34SHA-S values in both the YR and JZB tributary sediments may imply that the isotopic difference between HA-S and FA-S is a common feature of all terrestrial systems, regardless of geographic locality.

5.1.2. δ34SFA-S in riverine and marine sediments 34

SFA-S values for the ECS and JZB sediments were much lower than those for the

riverine sediments and also lower than δ34SO4 (ca. 21‰) of modern seawater. This

15

unequivocally indicates that a substantial fraction of 34S depleted S2− has been incorporated into FAs via sulfurization since, if terrestrial and/or marine biogenic FA-S were the sole source, δ34SFA-S for the marine sediments would be similar to either the riverine δ34SFA-S or seawater δ34SO4, or be somewhere between the two. Dissimilatory SO42− reduction in upper marine sediments usually results in extensive 34

S depletion in the product S2− and a substantial incorporation of the S2− imparts the

34

S depleted signal to sulfurized OM. Highly 34S depleted, reduced inorganic S [e. g.

δ34S of pyrite (δ34Spyrite) ca. −30‰] was observed for both the ECS and JZB sediments (Zhu et al., 2013a, 2014) and would be expected to be the source of the sulfurization. Sulfurization of FAs has also been inferred on the basis of δ34SFA-S in estuarine sediments of Callaway Bayou, Florida (Brüchert, 1998). Note that, in the marine sediments, δ34SFA-S was much higher than δ34Spyrite, implying that S2− could not be the sole source, and other isotopically heavier OS also contributed to FA-S. Biogenic OS of both terrigenous and marine sources is 34S-enriched, and thus a potential contributor. In most studies, only sulfurization (i.e. diagenetic S2−) and biogenic OS of marinesource have been considered as the important OS contributors to marine sediments, while terrigenous OS has long been ignored by assuming that most of the pool was extensively decomposed before reaching the seabed due to its high lability. In that circumstance, a two end member S isotope mixing model has been applied to estimate the relative contributions of sulfurization and biogenic OS of marine source (Brüchert and Pratt, 1996; Brüchert, 1998; Passier et al., 1999; Canfield et al., 1998;

16

Zhu et al., 2013a, 2013b, 2014). Actually, there is no justifiable reason to date to exclude the contribution from terrigenous OS. This notion is supported by XANES analysis of the non-CROS (see below). Future studies to establish a three end member mixing model, including terrestrial OS, marine biogenic OS and sulfurization for better quantification of FA-S sources in marine sediments are warranted.

5.1.2. δ34SHA-S in riverine and marine sediments Our previous studies indicated that in the ECS and JZB sediments HAs were basically inert and their sulfurization was largely negligible (Zhu et al., 2013b, 2014), though minor HA sulfurization in some other marine sediments has been inferred (Nissenbaum and Kaplan, 1972; Brüchert, 1998). It follows that biogenic OS of both terrigenous and marine sources is a potential source of HA-S in the sediments. The very similar average δ34SHA-S values between the riverine and marine sediments (Table 1) clearly indicate that terrigenous HA-S is nearly the sole important contributor to bulk HA-S in the marine sediments. This implies that the HA-S distribution and δ34SHA-S in the ECS and JZB may be used to trace the input and dispersal of terrigenous HAs in the two coastal seas.

5.2. XANES characterization of humic-S functionalities in marine and riverine sediments Both oxidized (EOS 5 and 6) and highly reduced OS moieties (EOS ≤ 1) are the main S functionalities in FA-S and HA-S in the riverine and marine sediments, with

17

other S functionalities being only marginal (Fig. 3). This feature is consistent with previous studies (Vairavamurthy et al., 1997; Xia et al., 1998; Schäfer et al., 2005; Zhu et al., 2014). Although two oxidized S moieties, namely sulfonates and ester sulfates, are often among the main fractions of humic-S in soils and sediments, the pathways for their formation remain poorly understood. Ester sulfates are speculated to be formed via biotic rather than abiotic pathways, whereas both biotic and abiotic pathways have been considered possible for sulfonate formation (Vairavamurthy et al., 1994, 1995, 1997; Bottrell et al., 2010; Morgan et al., 2012; Zeng et al., 2013). Given that much remains unclear about oxidized OS components, our efforts focus below only on reduced OS moieties.

5.2.1. FA-S in riverine and marine sediments The ECS sediments have wider variability in both δ34SFA-S and the fraction of the highly reduced FA-S than the JZB sediments (Table 1, Fig. 4). The wider variability in the ECS sediments may reflect more complex sources of FA-S and an unsteady depositional environment, which may have reprocessed FA-S differentially as a result of highly hydrodynamic conditions (Aller et al., 1985; DeMaster et al., 1985; Yao et al., 2014). Average 34SFA-S for the ECS sediments is less depleted than for JZB sediments, and the average fraction of highly reduced FA-S in the former is lower than in the latter (Fig. 4). These differences can be ascribed to eutrophication-induced sulfurization in JZB (Chen et al., 2014; Zhu et al., 2014). FA-S in the ECS sediments

18

seemingly also follows this pattern, namely that the FA-S, which had a largr fraction of highly reduced OS was generally more 34S depleted (Table 1, Fig. 4). This feature demonstrates that differential sulfurization of FAs is in the δ34SFA-S and S-XANES. As speculated above, a substantial fraction of S2− was incorporated into FAs in the marine sediments. Despite this, FA-S was dominated not by the highly reduced OS but by an oxidized one. This implies that the highly reduced biogenic OS may have been initially very low or subjected to substantial mineralization loss. As addressed above, the highly reduced FA-S in the marine sediments is mainly the product of sulfurization, with a minor fraction of terrigenous OS. In the YR sediments, the average fraction of the highly reduced FA-S (24.5%) is similar to that in the ECS sediments (26.8%), and the case is the same for JZB (33.3%) vs. its tributaries (38.5%). The similarity indicates that the relative importance of the highly reduced FA-S in the marine sediments has not been enhanced despite the sulfurization. Thus one may conclude that terrestrial systems are more favorable for the formation and/or preservation of highly reduced biogenic FA-S than marine environments.

5.2.2. HA-S in riverine and marine sediments Both the similar δ34SHA-S and fraction of highly reduced HA-S among the riverine sediments (YR vs. four JZB tributaries) indicate that there is no significant regional difference in S isotopic composition of HA-S and the relative importance of highly reduced HA-S in the two catchment basins with distinct sizes. Note that the fraction of highly reduced HA-S in the ECS sediments (avg. 28.1%) is much lower than in the

19

YR sediments (avg. 43.8%), whereas the fraction is roughly similar between the JZB (avg. 41.4%) and its tributaries (avg. 40.8%). This may indicate that the highly reduced HA-S moieties in the YR sediments have been subject to greater mineralization loss during transport through the large riverine/estuarine systems, whereas mineralization may be much less in the JZB tributaries, probably due to relatively short transport distance in the limited catchment area. Further work is needed to examine whether enhanced mineralization of highly reduced HA-S is a common feature of all large riverine/estuarine systems.

5.3. Contrasting humic-S and non-CROS based on S isotopes and XANES As shown in Table 2 and Fig. 3a, there are distinct differences in S functionality between the humic-S and the non-CROS: (i) the humic-S (HA-S and FA-S) is dominated by oxidized S functionalities (59−78%), whereas reduced S is the most important moiety in the non-CROS (80−92%); (ii) weakly reduced OS iss always a minor fraction of the humic-S (< 5%), but is nontrivial in the non-CROS (25−48%). The high fraction of weakly reduced OS (up to 48%) in the non-CROS suggests that this OS pool is actually of importance in the bulk marine sediments. An extremely low fraction of this OS pool in the humic-S may imply that it is essentially resistant to base extraction. This may be the reason why previous XANES studies focusing on humic-S failed to recognize its importance. In the same vein, a high fraction of the oxidized OS in the humic-S but a low one in the non-CROS might suggest that this OS pool is susceptible to both base and acidic Cr(II) extraction. Loss of OS during

20

pyrite removal by acidic Cr(II) reduction has been reported for OM-rich marine sediments (Canfield et al., 1998). The speculation above means that characterization of humic-S only may be biased towards oxidized OS moieties but characterization of non-CROS only towards reduced ones. It follows that neither of the two operational OS pools could provide a whole picture of OS functionality for bulk sediments. A combination of the two promises to offer complementing and comprehensive insights. Beside the above differences in structural composition, humic-S and non-CROS also differ in isotopic composition. δ34S for the non-CROS (δ34Snon-CROS) was not determined, but our previous studies found δ34Snon-CROS in surface sediments of the ECS and JZB within a narrow range of 6.6−9.6‰ (Zhu et al., 2013a, Huang et al., 2014). These values are higher than δ34SHA-S and δ34SFA-S for the sediments. In comparison with HA-S and FA-S, the higher δ34Snon-CROS and larger fraction of reduced non-CROS suggest that the reduced biogenic non-CROS is indeed quantitatively important and has survived early diagenesis. This conclusion is at odds with the long held notion that most biogenic reduced OS is decomposed during early diagenesis in the upper sediment due to its highly labile nature and, as a result, only a minor fraction of the OS pool has a chance of being preserved in marine sediments. Note that this new finding could not be gleaned from examination of humic-S and has thus been missing from previous studies concentrating on humic-S. Characterization of non-CROS with S isotope and XANES spectroscopy can therefore provide more comprehensive information on sources/pathways and fate of reduced biogenic OS in marine sediments. The structural and isotopic differences between humic-S and

21

non-CROS here suggest that S isotope and S-XANES should be combined to characterize the two operational OS pools for better deciphering of S cycling in the ocean.

6. Summary and conclusions δ34SFA-S values for the marine sediments were much lower than δ34SFA-S values for the riverine sediments and also much lower than seawater δ34SO4. This result is indicative of substantial sulfurization of marine FAs by incorporating 34S depleted inorganic reduced S during early diagenesis. As manifested from S-XANES and isotopic composition, an enhancement in the fraction of the highly reduced OS in FA-S and in depletion of 34SFA-S was indicative of increased sulfurization of marine FAs. In comparison with in the ECS, sulfurization of FAs in the JZB was enhanced due to increased eutrophication in the bay. The similarity in δ34SHA-S between the riverine and marine HA-S suggests that terrigenous OS was almost the sole important source of marine HA-S. HA-S distributions and their associated S isotopic composition in the marine sediments may be used to trace the input and dispersal of terrigenous HAs in the two coastal seas. Despite substantial sulfurization of FAs in the marine sediments, FA-S was dominated by the oxidized OS functionalities. This indicates that the highly reduced biogenic OS wa initially very low or was subject to extensive mineralization loss. In comparison with riverine FA-S, substantial sulfurization of marine FAs did not result in an increase in the fraction of highly reduced OS in the FA-S. This implies that

22

terrestrial systems may be more favorable for the formation and/or preservation of highly reduced biogenic FA-S than the marine settings. The fraction of highly reduced HA-S was roughly similar in JZB and its tributary sediments, but the fraction in the ECS marine sediments was much lower than in the YR sediments. The different patterns indicate that the highly reduced HA-S moieties in the YR sediments may have been subject to extensive mineralization loss during transport in the large riverine/estuarine systems, whereas the process may be much weakened in the JZB tributaries, probably due to relatively short transport in the small catchment area. Humic-S and non-CROS in the ECS sediments displayed distinct differences in isotopic and structural composition. Neither of the two operational pools could provide the whole picture of isotopes and functionalities of OS in the bulk sediments. Both S isotope and XANES should be combined to characterize the two pools for a better understanding of the nature of OS in the ocean.

Acknowledgments The authors are indebted to C.-G. Tang of the Institute of Geochemistry, Chinese Academy of Sciences for assistance in isotope analysis. Our gratitude goes also to two anonymous reviewers for critical and constructive comments. The research was jointly supported by the National Science Foundation of China (grant 41576078 to M.X.Z.), the Shandong Province Natural Science Foundation (grant ZR2015DM006 to M.X.Z.) and the Taishan Scholars Programme of Shandong Province (to G.P.Y.).

23

Associate Editor _ L.R. Snowdon

References Aller, R.C., Mackin, J.E., Ullman, W.J., Wang, C.-H., Tsai, S.-M., Jin, J.-C., Sui, Y.-N., Hong, J.-Z., 1985. Early chemical diagenesis, sediment-water solute exchange, and storage of reactive organic matter near the mouth of the Changjiang, East China Sea. Continental Shelf Research 4, 227–251. Amrani, A., 2014. Organic sulfur compounds molecular and isotopic signatures of past environments and post-depositional processes. Annual Review of Earth and Planetary Sciences 42, 733–768. Anderson, T.F., Pratt, L.M., 1995. Isotope evidence for the origin of organic sulfurand elemental sulfur in marine sediments. In: Vairavamurthy, M.A., Schoonen, M.A.A. (Eds.), Geochemical Transformations of Sedimentary Sulfur. ACS Symposium Series 612, Washington, D.C., pp. 378–396. Bottrell, S.H., Hatfield, D., Bartlett, R., Spence, M.J., Bartle, K.D., Mortimer, R.J.G., 2010. Concentrations, sulfur isotopic compositions and origin of organosulfur compounds in pore waters of a highly polluted raised peatland. Organic Geochemistry 41, 55–62. Brüchert, V., Pratt, L.M., 1996. Contemporaneous early diagenetic formation of organic and inorganic sulfur in estuarine sediments from St. Andrew Bay, Florida, USA. Geochimica et Cosmochimica Acta 60, 2325−2332. Brüchert, V., 1998. Early diagenesis of sulfur in estuarine sediments: The roles of sedimentary humic and fulvic acids. Geochimica et Cosmochimica Acta 62, 1567–1586. Burton, E.D., Sullivan, L.A., Bush, R.T., Johnston, S.G., Keene, A.F., 2008. A simple and inexpensive chromium-reducible sulfur method for acid-sulfate soils. Applied Geochemistry 23, 2759–2766. Canfield, D.E., Raiswell, R., Westrich, J.T., Reavse, C.M., Berner, R.A., 1986. The use of chromium reduction in the analysis of reduced inorganic sulfur in

24

sediments and shales. Chemical Geology 54, 149–155. Canfield, D.E., Boudreau, B.P., Mucci, A., Gundersen, J.K., 1998. The earlydiagenetic formation of organic sulfur in the sediments of Mangrove Lake, Bermuda. Geochimica et Cosmochimica Acta 62, 767–781. Canfield, D.E., Kristensensen, E., Thamdrup, B., 2005. Aquatic Geomicrobiology. Elsevier, Amsterdam. Chen, K.-K., Zhu, M.-X., Yang, G.-P., Fan, D.-J, Huang, X.-L., 2014. Spatial distribution of organic and pyritic sulfur in surface sediments of eutrophic Jiaozhou Bay, China: Clues to anthropogenic impacts. Marine Pollution Bulletin 88, 284−291. DeMaster, D.J., McKee, B.A., Nittrouer, C.A., Jiangchu, Q., Guodong, C., 1985. Rates of sediment accumulation and particle reworking based on radiochemical measurements from continental shelf deposits in the East China Sea. Continental Shelf Research 4, 143–158. Einsiedl, F., Schäfer, T., Northrup, P., 2007. Combined sulfur K-edge XANES spectroscopy and stable isotope analysis of fulvic acids and groundwater sulfate identify sulfur cycling in a karstic catchment area. Chemical Geology 238, 268–276. Einsiedl, F., Mayer, B., Schäfer, T., 2008. Evidence for incorporation of H2S in groundwater fulvic acids from stable isotope ratios and sulfur K-edge X-ray absorption near edge structure spectroscopy. Environmental Science and Technology 42, 2439–2444. Ferdelman, T.G., Church, T.M., Luther, G.W., 1991. Sulfur enrichment of humic substances in a Delaware salt marsh sediment core. Geochimica et Cosmochimica Acta 55, 979–988. Fike, D.A., Bradley, A.S., Rose, C.V., 2015. Rethinking the ancient sulfur cycle. Annual Review of Earth and Planetary Sciences 43, 593–622. Francois, R., 1987a. A study of sulfur enrichment in the humic fraction of marine sediments during early diagenesis. Geochimica et Cosmochimica Acta 51, 17–27. 25

Francois, R., 1987b. A study of the extraction conditions of sedimentary humic acids to estimate their true in situ sulfur content. Limnology and Oceanology 32, 964–972. Fry, B., Gest, H., Hayes, J.M., 1988. 34S/32S fractionation in sulfur cycles catalyzed by anaerobic bacteria. Applied and Environmental Microbiology 54, 250–256. Gao, S., Wang, Y.P., 2008. Changes in material fluxes from the Changjiang River and their implications on the adjoining continental shelf ecosystem. Continental Shelf Research 28, 1490–1500. Henneke, E., Luther, G.W., de Lange, G.J., Hoefs, J., 1997. Sulphur speciation in anoxic hypersaline sediments from the eastern Mediterranean Sea. Geochimica et Cosmochimica Acta 61, 307–321. Huffman, G.P., Mitra, S., Huggins, F.E., Shah, H., Vaidya, S., Lu, F., 1991. Quantitative analysis of all major forms of sulfur in coal by X-ray absorption finestructure spectroscopy. Energy and Fuels 5, 574–581. Henderson, G.S., de Groot, F.M.F., Moulton, B.J.A., 2014. X-ray absorption near-edge structure (XANES) spectroscopy. Reviews in Mineralogy & Geochemistry 78, 75–138. Huang, X.-L., Zhu, M.-X., Chen, L.J., Li, T. 2014. Sources and formationmechanisms of organic sulfur in Jiaozhou Bay sediments. Acta Oceanologica Sinica 36, 50–57 (in Chinese with English abstract). Hundal, L.S., Carmo, A.M., Bleam, W., Thompson, M.L., 2000. Sulfur in biosolids-derived fulvic acids: characterization by XANES spectroscopy and selective dissolution approaches. Environmental Science and Technology 34, 5184–5188. Hutchison, K. J., Hesterberg, D., Chou, J.W., 2001. Stability of reduced organic sulfur in humic acids as affected by aeration and pH. Soil Science Society of America Journal 65, 704–709 Kallmeyer, J., Ferdelman, T.G., Weber, A., Fossing, H., Jørgensen, B., 2004. A cold chromium distillation procedure for radiolabeled sulfide applied to sulfate reduction measurements. Limnology Oceanography: Methods 2, 171–180. 26

Kang, P.-G., Mitchell, M.J., Mayer, B., Campbell, J.L., 2014. Isotopic evidence for determining the sources of dissolved organic sulfur in a forested catchment. Environmental Science and Technology 48, 11259-11267. Kao, S.J., Lin, F.J., Liu, K.K., 2003. Organic carbon and nitrogen contents and their isotopic compositions in surficial sediments from the East China Sea shelf and the southern Okinawa Trough. Deep-Sea Research II 50, 1203–1217. Kaplan, I.R., Rittenberg, S.C., 1964. Microbiological fractionation of sulfur isotopes.Journal of General Microbiology 34, 195–212. Langner, P., Mikutta, C., Kretzschmar, R., 2012. Arsenic sequestration by organicsulphur in peat.

Nature Geoscience 5, 66–73.

Li, S.-L., Liu, C.-Q., Patra, S., Wang, F., Wang, B., Yue, F., 2011. Using a dual isotopic approach to trace sources and mixing of sulphate in Changjiang Estuary, China. Applied Geochemistry 26, S210–S213. Liu, S.M., Zhang, J., Chen, H.T., Zhang, G.S., 2005. Factors influencing nutrient dynamics in the eutrophic Jiaozhou Bay, North China. Progress in Oceanography 66, 66–85. Manceau, A., Nagy, K.L., 2012. Quantitative analysis of sulfur functional groups in natural organic matter by XANES spectroscopy. Geochimica et Cosmochimica Acta 99, 206–223. Manceau, A., Lemouchi, C., Enescu, M., Gaillot, A.-C., Lanson, M., Magnin, V., Glatzel, P., Poulin, B. A., Ryan, J. N., Aiken, G. R., Gautier-Luneau, I., Nagy, K. L., 2015. Formation of mercury sulfide from Hg(II)−thiolate complexes in natural organic matter. Environmental Science and Technology 49, 9787−9796. Martínez, C.E., McBride, M.B., Kandianis, M.T., Dubury, J.M., Yoon, S.-J., Bleam, W.F., 2002. Zinc-sulfur and cadmium-sulfur association in metaliferous peats: evidence from spectroscopy, distribution coefficients and phytoavailability. Environmental Science and Technology 36, 3683–3689. Milliman, J.D, Farnsworth, K.L., 2011. River Discharge to the Coastal Oceana Global Synthesis. Cambridge University Press. Morra, M.J., Fendorf, M.S., Brown, P.D., 1997. Speciation of sulfur in humic and 27

fulvic acids using X-ray absorption near-edge structure (XANES) spectroscopy. Geochimica et Cosmochimica Acta 61, 683–688. Morgan, B., Burton, E.D., Rate, A.W., 2012. Iron monosulfide enrichment and the presence of organosulfur in eutrophic estuarine sediments. Chemical Geology 296–297, 119–130. Nissenbaum, A., Kaplan, I.R., 1972. Chemical and isotopic evidence for the in situ origin of marine humic substances. Limnology and Oceanography 17, 570–582 Orthous-Daunay, F.R., Quirtico, E., Lemelle, L., Beck, P., deAndrade, V., Simionovici, A., Derenne, S., 2010. Speciation of sulfur in the insoluble organic matter from carbonaceous chondrites by XANES spectroscopy. Earth and Planetary Science Letters 300, 321–328 Passier, H.F., Bottcher, M.E., De Lange, G.J., 1999. Sulphur enrichment in organic matter of eastern Mediterranean sapropels: a study of sulphur isotope partitioning. Aquatic Geochemistry 5, 99–118. Pawellek, F., Frauenstein, F., Veizer, J., 2002. Hydrochemistry and isotopegeochemistry of the upper Danube River. Geochimica et Cosmochimica Acta 66, 3839–3854. Peterson, B.J., Howarth, R.W., 1987. Sulfur, carbon, and nitrogen isotopes used to trace organic matter flow in the salt-marsh estuaries of Sapelo Island, Georgia. Limnology and Oceanography 32, 1195–1213. Prietzel, J., Kögel-Knabner, I., Thieme, J., Paterson, D., McNulty, I., 2011.Microheterogeneity of element distribution and sulfur speciation in an organic surface horizon of a forested Histosol as revealed by synchrotron-based X-ray spectromicroscopy. Organic Geochemistry 42, 1308−1314. Rashid, M.A., 1985. Geochemistry of Marine Humic Compounds. Springer-Verlag, New York, 300 pp. Ravel, B., Newville, M., 2005. ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. Journal of Synchrotron Radiation 12, 537–541. Raven, M.R., Adkins, J.F., Werne, J.P., Lyons, T.W., Sessions, A.L., 2015. Sulfur 28

isotopic composition of individual organic compounds from Cariaco Basin sediments. Organic Geochemistry 80, 53–59. Rock, L., Mayer, B., 2009. Identifying the influence of geology, land use, and anthropogenic activities on riverine sulfate on a watershed scale by combining hydrometric, chemical and isotopic approaches. Chemical Geology 262, 121–130. Schäfer, 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. Organic Geochemistry 36, 567–582. Sinninghe Damsté, J.S., Rijpstra, W.I.C., de Leeuw, J.W., Schenck, P A., 1989.Quenching of labile functionalized lipids by inorganic sulphur compounds at theearly stages of diagenesis. Geochimica et Cosmochimica Acta 53, 1343–1355. Sinninghe Damsté, J.S., de Leeuw, J.W., 1990. Analysis, structure and geochemical significance of organically-bound sulphur in the geosphere: state of the art and future research. Organic Geochemistry 16, 1077–1101. Solomon, D., Lehmann, J., Martinez, C.E., Tveitnes, S., DuPreez, C.C., Ameijing, W.,2005. Sulphur speciation and biogeochemical cycling in long-term arable cropping of subtropical soils: evidence from wet-chemical reduction and S K-edge XANES spectroscopy. European Journal of Soil Science 56, 621–634. Solomon, D., Lehmann, J., Zarruk, K.K., Dathe, J., Kinyangi, J., Liang, B., Machado,S., 2011. Speciation and long- and short-term molecular-level dynamics of soilorganic sulfur studied by X-ray absorption near-edge structure spectroscopy. Journal of Environmental Quality 40, 704–718. Trust, B., Fry, B., 1992. Sulfur isotopes in plants: a review. Plant Cell and Environment 15, 1105−1110. Vairavamurthy, M.A., Maletic, D., Wang, S., Manowitz, B., Eglinton, T., Lyons, T.,1997. Characterization of sulfur-containing functional group in sedimentaryhumic substances by X-ray absorption near-edge structure spectroscopy. Energy and Fuels 11, 546−553. 29

Vairavamurthy, A., Zhou, W., Eglinton, T., Manowitz, B., 1994. Sulfonates: A novel class of organic sulfur compounds in marine sediments. Geochimica et Cosmochimica Acta 58, 4681–4687. Vairavamurthy, M.A., Wang, S., Khandelwal, B., Manowitz, B., Ferdelman, T., Fossing, H., 1995. Sulfur transformations in early diagenetic sediments from the Bay of Concepcion, off Chile. In: Vairavamurthy, M.A., Schoonen, M.A.A. (Eds.), Geochemical Transformations of Sedimentary Sulfur. ACS Symposium Series 612. American Chemical Society, Washington D.C., pp. 38–58. Vairavamurthy, A., 1998. Using X-ray absorption to probe sulfur oxidation states in complex molecules. Spectrochimica Acta Part A 54, 2009−2017. Waldo, G.S., Carlson, R.M.K., Moldowan, J.M., Peters, K.E., Penner-Hahn, J.E., 1991. Sulfur speciation in heavy petroleums: information from X-ray absorption near-edge structure. Geochimica et Cosmochimica Acta 55, 801−814. Werne, J.P., Lyons, T.W., Hollander, D.J., Formolo, M.J., Sinninghe Damsté, J.S., 2003. Reduced sulfur in euxinic sediments of the Cariaco Basin: sulfur isotope constraints on organic sulfur formation. Chemical Geology 195, 159−179. Werne, J.P., Hollander, D.J., Lyons, T.W., Sinninghe Damsté, J.S., 2004. Organic sulfur biogeochemistry: recent advances and future research directions. In: Amend, J.P., Edwards, K.J., Lyons, T.W., (Eds.), Sulfur biogeochemistry—Past and present: Boulder, Colorado, Geological Society of America Special Paper 379, pp. 117–134. Werne, J.P., Lyons, T.W., Hollander, D.J., Schouten, S., Hopmans, E.C., Sinninghe Damsté, J.S., 2008. Investigating pathways of diagenetic organic mattersulfurization using compound-specific sulfur isotope analysis. Geochimica et Cosmochimica Acta 72, 3489–3502. Wu, Y.L., Sun, S., Zhang, Y.S., 2005. Long-term change of environment and its influence on phytoplankton community structure in Jiaozhou Bay. Oceanologia et Limnologia Sinica 487–498 (in Chinese with English abstract). Xia, K., Weesner, F., Bleam, W.F., Bloom, P.R., Skyllberg, U.L., Helmke, P.A., 1998. XANES studies of oxidation states of sulfur in aquatic and soil humic substances. 30

Soil Science Society of America Journal 62, 1240–1246. Xia, K., Skyllberg, U.L., Bleam, W.F., Bloom, P.R., Nater, E.A., Helmke, P.A., 1999. X-ray absorption spectroscopic evidence for the complexation of Hg(II) by reduced sulfur in soil humic substances. Environmental Science and Technology 33, 257–261. Xiao, H.-Y., Liu, C.-Q., 2011. The elemental and isotopic composition of sulfur andnitrogen in Chinese coals. Organic Geochemistry 42, 84−93. Xu, K., Milliman, J.D., Li, A., Liu, J.P., Kao, S.-J., Wan, S., 2009. Yangtze- and Taiwan- derived sediments on the inner shelf of East China Sea. Continental Shelf Research 29, 2240−2256. Yao, P., Zhao, B., Bianchi, T.S., Guo, Z., Zhao, M., Li, D., Pan, H., Wang, J., Zhang, T., Yu, Z., 2014. Remineralization of sedimentary organic carbon in mud deposits of the Changjiang Estuary and adjacent shelf: Implications for carbon preservation and authigenic mineral formation. Continental Shelf Research 91, 1–11. Yücel, M., Konovalov, S.K., Moore, T.S., Janzen, C.P., Luther, G.W., 2010. Sulfur speciation in the upper Black Sea sediments. Chemical Geology 269, 364–375. Zaback, D.A., Pratt, L.M., 1992. Isotopic composition and speciation of sulfur in the Miocene Monterey Formation: Reevaluation of sulfur reactions during early diagenesis in marine environments. Geochimica et Cosmochimica Acta 56, 763–774. Zeng, T., Arnold, W.A., Toner, B.M., 2013. Microscale characterization of sulfur speciation in lake sediments. Environmental Science and Technology 47, 1287–1296. Zhao, F.J., Lehmann, J., Solomon, D., Fox, M.A., McGrath, S.P., 2006. Sulfur speciation and turnover in soils: evidence from sulfur K-edge XANES spectroscopy and isotope dilution studies. Soil Biology and Biochemistry 38, 1000–1007. Zhu, M.-X., Chen, L.-J., Yang, G.-P., Huang, X.-L., Ma, C.-Y., 2014. Humic sulfur in eutrophic bay sediments: characterization by sulfur stable isotopes and K-edge 31

XANES spectroscopy. Estuarine and Coastal Shelf Science 138, 121−129. Zhu, M.-X., Hao, X.-C., Shi, X.-N., Yang, G.-P., Li, T., 2012. Speciation and spatial distribution of solid-phase iron in surface sediments of the East China Sea continental shelf. Applied Geochemistry 27, 892−905. Zhu, M.-X., Huang, X.-L., Yang, G.-P., Hao, X.-C., 2013b. Speciation and stable isotopic compositions of humic sulfur in mud sediment of the East China Sea: Constraints on origins and pathways of organic sulfur formation. OrganicGeochemistry 63, 64−73. Zhu, M.-X., Shi, X.-N., Yang, G.-P., Hao, X.-C., 2013a. Formation and burial of pyrite and organic sulfur in mud sediments of the East China Sea inner shelf: constraints from solid-phase sulfur speciation and stable sulfur isotope. Continental ShelfResearch 54, 24−36.

32

Figure captions Fig. 1. Map of sampling sites in the East China Sea and Jiaozhou Bay. Four sediment samples (A, B, C and D) from the Yangtze River were collected from the Nanjing section of the lower Yangtze River (red solid circles). Four samples from Jiaozhou Bay tributaries were collected in the Baisha River, Dagu River, Moshui River and Yanghe River, respectively, 3−4 km upstream from the river mouths (red solid circles). JCC, Jiangsu Coastal Current; KC, Kuroshio Current; TWC, Taiwan Warm Current; ZFCC, Zhejiang-Fujian Coastal Current; YDW, Yangtze diluted water (modified after Xu et al., 2009).

Fig. 2. Representative XANES spectra and Gaussian curve fits of (a) HA-S, (b) FA-S and (c) non-CROS. Individual Gaussian curves are labeled with G1 to G6.

Fig. 3. Fraction of highly reduced (EOS < 1), weakly reduced (1 < EOS > 4), and oxidized OS (EOS 5 and 6) in HA-S, FA-S and (a) non-CROS for the marine and (b) riverine sediments.

Fig. 4. Plot of fraction of highly reduced FA-S (%) vs. isotopic composition of FA-S (δ34SFA-S) in the ECS and JZB sediments; also shown are averages.

33

Table 1 Content and isotopic composition of HA-S and FA-S in sediments from ECS, JZB, YR and JZB tributaries (SD, standard deviation). ECS DH2

DH4

DH5

JZB DH8

E4

Avg. ±

E6

E7

G2

SD 34

δ SFA-S

−17.7

−4.0

−18.9

−6.49

34

δ SHA-S

SD

−11.8 ±

−16.1

−21.0

4.08

4.71

4.00

4.74

0.94

B

1.58

C

1.41

4.38

4.38 ±

3.97

δ SHA-S (‰)

4.09

3.97

0.19

−0.10

Avg. ±

Baisha

Dagu

Moshui

Yanghe

Avg. ±

SD

River

River

River

River

SD

0.96 ±

−2.62

−2.19

−1.25

−2.25

−2.08 ±

0.76 4.88

4.90

4.72

4.10 ±

JZB tributary D

(‰) 34

−18.5 ± 3.0

YR A

δ SFA-S

−15.6

0.40

(‰)

34

−21.2

7.6

(‰)

Avg. ±

4.91

0.58 4.32

4.10 ± 0.19

3.75

4.95

3.85

4.22 ± 0.55

34

Table 2 EOS and fraction (%) of OS functionalities in HA-S, FA-S and non-CROS of ECS and JZB sediments (SD, standard deviation). ECS HA-S G1

G2

G3

G5

EOS 5

EOS 6

EOS

EOS

0.19−0.22

0.93−1.06

a

a

DH2

22.2

7.77

2.19

30.4

37.5

DH4

15.9

6.07

4.45

37.0

36.6

DH5

25.4

10.1

3.42

31.5

29.4

DH8

16.6

8.45

3.76

37.8

33.5

20.0 ± 4.6

8.10 ± 1.7

3.46 ± 0.9

34.2 ± 3.8

34.3 ± 3.7

Sample

Avg. ± SD

EOS

G4

2.04−2.17

a

FA-S

Sample

G1

G2

G3

G4

G5

G6

EOS

EOS 1.12

EOS 1.67

EOS

EOS 5

EOS 6

0−0.38

a

3.28−3.59 a

DH2

34.3

1.9

0

4.33

15.1

44.4

DH4

16.3

5.53

0

0.06

28.5

48.1

DH5

28.9

2.92

0

3.02

21.1

44.1

DH8

13.6

3.95

0.35

1.66

30.2

48.2

23.3 ± 9.9

3.58 ± 1.5

0.09 ± 0.18

2.27 ± 1.8

23.7 ± 7.0

G1

G2

G3

G4

G5

G6

EOS

EOS 0.38

EOS

EOS 3.10

EOS 5

EOS 6

Avg. ± SD

46.2 ± 2.3

Non-CROS

Sample

1.74−1.80

−0.07

a

DH2

4.75

53.1

19.4

5.74

4.59

12.4

DH4

0

33.2

33.9

13.6

3.79

15.5

DH5

0

44.1

35.6

12.7

3.05

4.5

DH8

0

44.5

25.8

9.48

4.76

15.4

1.19 ± 2.4

43.7 ± 8.2

28.7 ± 7.5

10.4 ± 3.6

4.05 ± 0.8

G4

G5

G6

EOS

EOS 5

EOS 6

27.9

30.7

Avg. ± SD

12.0 ± 5.2

JZB HA-S

Sample

E4

G1

G2

G3

EOS

EOS

EOS

0.22−0.31

0.81−0.93

a

a

29.1

5.38

1.92−2.04

a

3.22−3.28 a

6.39

35

0.45

E6

30.2

12.5

2.92

0.11

32.8

21.5

E7

33.9

8.03

2.99

0.23

29.4

26.4

G2

32.2

12.0

3.99

0.10

30.6

21.1

31.4 ± 2 .1

9.48 ± 3.4

4.07 ± 1.6

G1

G2

G3

G4

G5

G6

EOS

EOS 1.06

EOS 2.17

EOS 3.34

EOS 5

EOS 6

Avg. ± SD

0.22 ± 0.16

30.2 ± 2.1

24.9 ± 4.5

FA-S

Sample

−0.03 E4

29.5

1.74

1.57

2.70

16.9

47.6

E6

31.2

2.68

0.78

2.48

18.4

44.5

E7

33.3

1.68

0.75

3.67

15.4

45.2

G2

30.3

2.98

0.31

3.00

16.2

47.2

Avg. ± SD a

31.1 ± 1.6

2.27 ± 0.66

0.85 ± 0.52

Range of EOS.

36

2.96 ± 0.52

16.7 ± 1.3

46.1 ± 1.5

Table 3 EOS and fraction (%) of sulfur functionalities in HA-S and FA-S of YR and JZB tributary sediments (SD, standard deviation). YR HA-S

Sample

G1

G2

G3

G4

G5

G6

EOS

EOS 0.75

EOS

EOS

EOS 5

EOS 6

0.01−0.07

2.11−2.17

a

a

3.34−3.37 a

A

30.2

15.6

5.54

0

28.5

20.1

B

33.8

8.83

6.76

0.6

25.0

25.0

C

24.9

17.1

5.67

0

28.7

23.6

D

30.5

14.3

6.95

0.22

26.2

21.7

29.9 ± 3.7

14.0 ± 3.6

6.23 ± 0.73

Avg. ± SD

0.205 ± 0.28

27.1 ± 1.8

22.6 ± 2.1

FA-S

Sample

G1

G2

G3

G4

G5

G6

EOS

EOS 0.75

EOS

EOS

EOS 5

EOS 6

0.01−0.07

2.11−2.17

a

a

3.34−3.37 a

A

18.5

5.75

1.64

1.39

24.8

47.9

B

18.1

4.20

1.34

1.64

21.5

53.2

C

19.2

5.61

0

2.59

22.5

50.2

D

20.4

6.26

2.15

1.50

23.8

45.9

Avg. ± SD

19.0 ± 1.0

5.46 ± 0.88

1.28 ± 0.92

1.78 ± 0.55

23.2 ± 1.4

49.3 ± 3.1

JZB tributary HA-S

Sample

Baisha River Dagu River Moshui River Yanghe River Avg. ± SD

G1

G2

G3

EOS

EOS 0.93

EOS 2.04

G4

G5

G6

EOS

EOS 5

EOS 6

0.19−0.22

3.22−3.28

a

a

32.5

7.83

7.25

0

26.8

25.6

25.7

13.7

3.09

0.25

31.3

26.0

38.3

7.10

9.11

0.23

30.3

14.9

30.3

10.2

8.40

0

27.4

23.7

31.7 ± 5.2

9.71 ± 3.0

6.96 ± 2.7 FA-S

37

0.12 ± 0.14

29.0 ± 2.2

22.6 ± 5.2

G1 Sample

G2

EOS −0.01−0

a

G3

EOS

EOS

0.75−1.06

2.11−2.17

a

a

Baisha River Dagu River Moshui River Yanghe River Avg. ± SD a

G4

G5

G6

EOS

EOS 5

EOS 6

3.34−3.37 a

35.1

5.51

2.27

3.49

16.9

36.8

30.8

5.54

0

4.99

15.3

43.4

38.2

3.32

0.35

3.79

16.0

38.3

30.4

5.41

2.25

1.76

27.8

32.4

33.6 ± 3.7

4.94 ± 1.1

1.22 ± 1.2

3.51 ± 1.3

19.0 ± 5.9

Range of EOS.

38

37.7 ± 4.5

39°

N 38° 37° 36° 35°

Yellow Sea

34° C JC

33°

Cheju Island

Yangtze River

32°

East

Nanjing YDW

31°

China

DH2

30°

Sea DH4

CHINA DH5 ZFC C

28°

TW C

27° DH8

KC

29°

26° 25° 118°

120°

122°

124°

126° E

36.30 Moshui R.

N

Dagu R.

36.25

Hongdao

G2

36.20

E6 E4

36.15 Yanghe R.

Jiaozhou Bay

36.10

36.05

Baisha R.

E7

Licun R.

Qingdao Huangdao

36.00

Yellow Sea

35.95 120.00 120.05 120.10 120.15 120.20 120.25 120.30 120.35 120.40 120.45 E

Figure 1 39

Fig 2a

Fig 2b

Fig 2c 40

AS

AS

-H AS D -H AS AFA -S BFA -S C -F ABa S ish D FA a R ive - S D ag r-H u ARi M S ve os r-H hu A Y a i Ri ve -S ng r -H he ARi Ba S v er ish -H a AR S ive D ag r-F u ARi M S ve os r-F hu Ya i R iv A-S ng er -F he ARi S ve r-F AS

C

BH

AH

Fraction of OS (%)

H2 -H A D H4 -S -H A D H5 -S -H D H8 A-S -H A D H2 -S -F A D H4 -S -F A D H5 -S -F A D D H -S H2 8F -n o AD H4 n-C S R D non O S H5 -C -n RO o D H8 n-C S R -n on O S -C R E4 O S -H A E6 -S -H A E7 -S -H G A-S 2HA E4 -S -F A E6 -S -F A E7 -S -F A G -S 2FA -S

D

Fraction of OS (%)

100 90

80

70

60

50

40

30

20

10

0

Highly reduced OS

Highly reduced OS Weakly reduced OS

Weakly reduced OS

Fig. 3b

Figure 3

41 Oxidized OS

Fig. 3a

100

90

80

70

60

50

40

30

20

10

0

Oxidized OS

Fraction of highly reduced FA-S (% ) 0

5

10

15

20

0

34

δ 3 4 SFA-S (‰)

-5

-10

-15

ECS JZB

-20

ECS average JZB average

-25

Figure 4

42

25

30

35

40

Highlights •Organic sulfur (OS) in sediments characterized from S isotopes and XANES. •Highly reduced fulvic acid S (FA-S) not enhanced by sulfurization. •Humic acid S (HA-S) in marine sediments dominated by terrigenous OS. •Humic-S is structurally and isotopically different from non-Cr reducible OS.

43