LETTER doi:10.1038/nature10855 Preservation of organic matter in sediments promoted by iron Karine Lalonde1, Alfonso Mucci2, Alexandre Ouellet1 & Yves Gélinas1 The biogeochemical cycles of iron and organic carbon are strongly interlinked. In oceanic waters, organic ligands have been shown to control the concentration of dissolved iron1. In soils, solid iron phases shelter and preserve organic carbon2, but the role of iron in the preservation of organic matter in sediments has not been clearly established. Here we use an iron reduction method previously applied to soils3 to determine the amount of organic carbon associated with reactive iron phases in sediments of various mineralogies collected from a wide range of depositional environments. Our findings suggest that 21.5 6 8.6 per cent of the organic carbon in sediments is directly bound to reactive iron phases. We further estimate that a global mass of (19–45) 3 1015 grams of organic carbon is preserved in surface marine sediments as a result of its association with iron4. We propose that these associations between organic carbon and iron, which are formed primarily through co-precipitation and/or direct chelation, promote the preservation of organic carbon in sediments. Because reactive iron phases are metastable over geological timescales, we suggest that they serve as an efficient ‘rusty sink’ for organic carbon, acting as a key factor in the long-term storage of organic carbon and thus contributing to the global cycles of carbon, oxygen and sulphur5. Evidence of interactions between iron and organic carbon in marine sediments was reported nearly 40 yr ago, where concentrations of iron and organic carbon were found to co-vary6. Because both iron and organic carbon are commonly associated with clay mineral surfaces, it was simply stated that ‘‘where there is more deposited fine-grained material with high surface area for adsorption, we find more organic matter and more Fe’’6. It is still not clear whether this correlation stems from the strong affinity of both species for solid surfaces or whether it reflects enhanced preservation of organic carbon by iron. Iron’s preservative effect on organic matter was previously demonstrated in laboratory studies7,8, which reported that the presence of iron-rich solid substrates or the formation of organoferric complexes hampers microbial degradation of simple organic compounds. Iron also imparts a protective effect to organic carbon in soil systems2, but this preservation mechanism has never been explored in sediments. In modern sediments, reactive iron phases (defined here as the solid iron phases that are reductively dissolved by sodium dithionite) are typically found as nanospheres of goethite ,10 nm in diameter9,10. These phases accumulate or are formed within the oxic sediment layer through oxidation and precipitation of dissolved iron(II) produced during weathering and diagenetic recycling within the sediment11. Over time, reactive iron phases become more crystalline, resulting in reduced surface area, reactivity and solubility. Crystallization is, however, hindered by the active diagenetic recycling of iron12 and by coating of iron phases by organic matter13. Accordingly, reactive iron phases have been shown to survive in sediments for hundreds of thousands of years14. We examined sediments collected from a wide range of environments, including fresh waters, estuaries, river deltas, continental margins and the deep sea and encompassing various depositional environments and mineralogies. These samples include organic-carbon-rich, sulphidic Black Sea sediments and organic-carbon-rich sediments from dioxygendeficient zones along the Indian and Mexican (sampling station 306) margins. Also included are sediments from the Arabian Sea, the Saanish Inlet and a boreal lake (Lake Brock), which have a productivity-driven seasonal pattern of dioxygen-deficient waters. Estuarine, deltaic and margin deposits accumulating below well-oxygenated waters of the Arctic margin, the St Lawrence estuary and gulf, the Mexican margin (stations 303–305), the Eel River basin, the Washington coast and the adjacent Columbia River delta are also examined along with pelagic sediments from the Southern Ocean, the Santa Barbara basin (station M) and the equatorial Pacific Ocean. This sample set comprises fresh water, estuarine and marine clastic sediments, carbonate and siliceous oozes, as well as pelagic red clay sediments. We focused on determining the amount of organic carbon associated with reactive iron phases by applying the citrate–dithionite iron reduction method of ref. 15, which simultaneously dissolves from the sediment matrix all solid reactive iron phases and the organic carbon associated with these phases (OC-Fe). The reduction reaction is conducted at circumneutral pH using sodium bicarbonate as a buffer, thus preventing the hydrolysis of organic matter as well as its protonation and re-adsorption onto sediment particles, which occur under acidic conditions. Whereas the extraction of the same samples with artificial sea water released a negligible fraction of the total organic carbon (less than 3%; results not shown), samples treated under the same experimental conditions after substituting trisodium citrate (complexing agent) and sodium dithionite (reducing agent) for sodium chloride (equivalent ionic strength) released on average 7.2 6 5.4% of the total organic carbon (Supplementary Table 2). Because the organic carbon released in these control experiments is not associated with iron, results of individual control experiments were subtracted from the amount of organic carbon released from the dithionite extractions (see Supplementary method for results and a discussion on contamination and specificity for the OC-Fe fraction). We determined that for all sediments tested, an average of 20.5 6 7.8% of the total organic carbon is directly associated with iron, with the highest OC-Fe concentrations in the uppermost sediment layers, where most of the reactive iron phases accumulate (Fig. 1). Considering organic carbon burial within different depositional settings—deltaic and continental margin sediments respectively account for 44% and 45% of global organic carbon burial, whereas pelagic sediments and high productivity zones, including anoxic basins, respectively account for 5% and 6% (ref. 16)—we estimate that the global pool of organic carbon specifically associated with iron corresponds to 21.5 6 8.6% of the total sedimentary organic carbon, or (19–45) 3 1015 g of organic carbon. Even in mature sediments (1,000–1,500 yr old), 23–27% of the total organic carbon remains bound to reactive iron oxide phases, suggesting that the strong association between iron and organic carbon may inhibit microbial organic carbon degradation and enhance organic carbon preservation. 1 GEOTOP and Department of Chemistry and Biochemistry, Concordia University, 7141 Sherbrooke Street West, Montreal, Quebec H4B 1R6, Canada. 2GEOTOP and Earth and Planetary Sciences, McGill University, 3450 University Street, Montreal, Quebec H3A 2A7, Canada. 1 9 8 | N AT U R E | VO L 4 8 3 | 8 M A R C H 2 0 1 2 ©2012 Macmillan Publishers Limited. All rights reserved 40 35 Deep sea Deltaic and estuarine Continental margin 45 Anoxic 50 Sulphidic LETTER RESEARCH 45 40 35 30 25 25 20 20 15 15 10 5 0 0 Equatorial Pacific 0° N Equatorial Pacific 9° N Station M Eel River basin –10 –12 –14 –16 –18 Equatorial Pacific 9° N (1) Equatorial Pacific 0° N (1) Station M (2) Southern Ocean (1) MacKenzie River delta (1) Eel River (2) Wash. coast 201 (1) St Lawrence 25 (12) Wash. coast 206 (1) organic carbon to iron (OC:Fe) of 1.0 for the co-extracted species, based on the maximal sorption capacity of reactive iron oxides for natural organic matter3. However, co-precipitation and/or chelation of organic compounds with iron generates low-density, organic-rich structures with OC:Fe ratios between 6 and 10 (ref. 3). According to the results of our dithionite extractions, typical continental margin sediments overlain by oxic bottom waters yield an average OC:Fe ratio of 4.0 6 2.8 (Supplementary Table 3), which greatly exceeds the maximum sorption capacity of iron oxides but is consistent with the formation of OC-Fe chelates. These chelates are predominantly organic structures that probably resemble those depicted by the ‘onion model’18, where organic molecules are ‘glued’ together by iron ions or nanophases of iron oxides. The formation of such chelates from solution is possible when the molar porewater OC:Fe ratio is approximately 10 (refs 19, 20). This molar ratio is typically observed in anoxic sediment pore waters such as that in the St Lawrence estuary (K.L., unpublished data) and in the nearby Saguenay fjord21. The diffusion of dissolved iron(II) from St Lawrence 23 (15) Wash. coast 204 (1) Wash. coast 202 (1) St Lawrence gulf (2) St Lawrence 20 (15) Wash. coast 203 (1) Wash. coast 213 (1) Wash. coast 215 (1) Mexican margin 303 (3) Arabian Sea (2) Wash. coast 205 (1) Mexican margin 304 (3) Mexican margin 305 (3) Madeira turbidite 7–9 (1) Saanish Inlet (1) Indian margin (1) Lake Brock (2) Mexican margin 306 (3) Black Sea (2) In agreement with the calculations in ref. 3, our measurements reveal that reactive iron phases do not provide sufficient surface area (,5% of the total surface area of sediments; Supplementary Table 3) for adsorption of the entire OC-Fe pool onto iron oxides. As an alternative explanation, we propose the existence of largely organic OC-Fe macromolecular structures that are dissolved and dislodged from the sediment during iron reduction. Transmission electron microscopy studies describe sedimentary organic matter as ‘‘discrete, discontinuous blebs’’ that adhere to the surface of sediment clay particles17. These blebs are consistent with our proposed structure of OC-Fe, as are the findings of ref. 16, where it was reported that sedimentary organic matter is not spread evenly over clay particles but covers only about 15% of the surface. We believe that iron or iron oxides are critical in providing cohesion to these macromolecular structures, possibly fixing them to clay particles through strong covalent bonds. Calculations indicate that simple sorption of organic matter on reactive iron oxide surfaces results in a maximum molar ratio of Mackenzie River delta Southern Ocean Wash. coast 202 Wash. coast 204 St Lawrence 23 Wash. coast 206 Wash. coast 203 St Lawrence 25 Wash. coast 201 St Lawrence gulf Wash. coast 205 Wash. coast 215 Wash. coast 213 St Lawrence 20 Mexican margin 303 Arabian Sea Mexican margin 304 Mexican margin 305 Lake Brock Saanish Inlet Madeira turbidite Indian margin Mexican margin 306 0–0.5 9–11 0–0.5 3–4 16–19 0–3 0–1 9–11 0–20 –– 0–0.5 4–5 22–27 0–0.5 5–6 19–22 0–0.5 10–12 0–0.5 14–16 22–27 11–12 11–12 11–12 0–35 1.5–5 470–500 11–12 11–12 0–35 11–12 11–12 0–35 11–12 0–1 15–20 0–1 8–12 0–1 15–20 0–0.5 10–12 5 Black Sea Depth (cm) 10 δ13C (‰) OC:Fe OC-Fe (%) 30 Figure 1 | Control-corrected percentage of the total sediment organic carbon bound to reactive iron phases. The OC-Fe considered here (black line) is that dislodged from the sediment during the reductive dissolution of reactive iron oxides. Depth intervals are indicated below the x axis, along with the corresponding geographical area and, if relevant, sampling station. Molar OC:Fe ratios of the uppermost surface sediment layer are also shown (black squares). The iron reduction was carried out following the method of ref. 15 without adding agents that promote flocculation of the dissolved organic matter after the reduction step. Error bars, s.d.; n 5 12–15 for the St Lawrence samples and n 5 3 for the others. Figure 2 | Carbon isotopic signatures of non-iron-bound and iron-bound organic carbon for all sediment samples. The samples (non-iron-bound, black; iron-bound, grey) were depth-integrated whenever possible; the number of depth intervals integrated is indicated in parentheses above the sample name. d13C 5 (13C/12C)sample/ (13C/12C)VPDB 2 1; VPDB, Vienna Pee Dee Belemnite. Error bars, s.d.; n 5 12–15 for the St Lawrence samples and n 5 3 for the others. –20 –22 –24 –26 –28 –30 –35 –37 8 M A R C H 2 0 1 2 | VO L 4 8 3 | N AT U R E | 1 9 9 ©2012 Macmillan Publishers Limited. All rights reserved RESEARCH LETTER anoxic to surficial oxic sediments would trigger the oxidation of iron(II) to iron(III) and the formation of very stable organic complexes22,23 (the formation constant ranges from 1014 M21 for natural dissolved organic carbon to 1052 M21 for siderophores). Sediments bathed by oxygen-depleted bottom waters, such as in the Black Sea, the Mexican margin (station 306) and the Indian margin, host structures with high OC:Fe ratios (7 to 32). These organometallic structures seem to be particularly stable under anaerobic conditions and survive degradation. By contrast, in oxic environments, the organic lining of these structures is progressively degraded, reducing the OC:Fe ratio to values observed in typical continental margin sediments (Fig. 1). Long exposure to oxic conditions increases the fraction of the total sedimentary organic carbon pool that is tightly adsorbed to particle surfaces24, owing to the preferential degradation of organic structures that are more loosely attached to the clay mineral matrix, such as the OC-Fe chelates. Very long exposure to oxic conditions results in the very low OC:Fe ratio observed at the deep-sea equatorial Pacific site (0.36; Fig. 1). We also analysed the isotopic composition (d13C and d15N) and elemental composition (molar ratio of carbon to nitrogen) of the bulk organic matter and the iron-associated organic carbon fractions of all sediment samples. In most cases, we find that OC-Fe is enriched in 13C (d13C increases by 1.7 6 2.8%; Fig. 2) and nitrogen (C:N decreases by 1.7 6 2.8) relative to the rest of the sedimentary organic carbon pool, whereas d15N shows little or no fractionation (Supplementary Figs 1 and 2). Natural organic compounds rich in 13C include proteins and carbohydrates25, which are rich in nitrogen and/or oxygen functionalities that favour the formation of inner-sphere complexes with iron. The preferential binding of such highly labile organic compounds to iron may explain why reactive organic compounds can be preserved in sediments whereas other, more refractory, molecules are degraded4. Our findings have far-reaching implications for our understanding of organic matter cycling in sediments. First, the protection mechanism described above, which preferentially shields 13C- and nitrogen-rich organic compounds from microbial degradation, could help explain a phenomenon that has puzzled organic geochemists for decades: the replacement, seaward of river mouths, of terrigenous organic matter from sediments by compounds bearing a more marine isotopic and elemental signature26. Our data also show that the traditional sorptive stabilization mechanism, which proposes that clay particles have a preservative effect on organic matter through direct adsorption on their surfaces4,27,28, does not describe accurately the mode of stabilization for all organic compounds in sediments. Although more work is needed to elucidate the exact nature of interactions between organic carbon and iron, our data suggest that direct chelation or co-precipitation of macromolecular OC-Fe structures has a significant role. Finally, our results reveal that 21.5 6 8.6% of the organic carbon buried in surface marine sediments (150 3 1015 g of organic carbon4), or a global mass of (19–45) 3 1015 g of organic carbon, is preserved as a result of its intimate association with reactive iron phases. Assuming that our estimate also applies to organic carbon locked in the sedimentary rock reservoir (150,000 3 1018 g of organic carbon4), iron-associated organic carbon would account for (1,900–4,500) 3 1018 g of organic carbon, or roughly 2,900–6,800 times the amount in the atmospheric carbon pool. Hence, reactive iron phases serve as an extremely efficient ‘rusty sink’ for organic carbon and are a key factor in the long-term storage of organic carbon and the global cycles of carbon, oxygen and sulphur. Received 22 February 2011; accepted 10 January 2012. 1. 2. Johnson, K. S., Gordon, R. M. & Coale, K. H. What controls dissolved iron in the world ocean? Mar. Chem. 57, 137–161 (1997). Kaiser, K. & Guggenberger, G. The role of DOM sorption to mineral surfaces in the preservation of organic matter in soils. Org. Geochem. 31, 711–725 (2000). 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. Wagai, R. & Mayer, L. M. Sorptive stabilization of organic matter in soils by hydrous iron oxides. Geochim. Cosmochim. Acta 71, 25–35 (2007). Hedges, J. I. & Keil, R. G. Sedimentary organic matter preservation: an assessment and speculative synthesis. Mar. Chem. 49, 81–115 (1995). Berner, R. A. The long-term carbon cycle, fossil fuels and atmospheric composition. Nature 426, 323–326 (2003). Berner, R. A. Sedimentary pyrite formation. Am. J. Sci. 268, 1–23 (1970). Boudot, J.-P., Bel Hadj, B. A., Steiman, R. & Seigle-Murandi, F. Biodegradation of synthetic organo-metallic complexes of iron and aluminum with selected metal to carbon ratios. Soil Biol. Biochem. 21, 961–966 (1989). Jones, D. L. & Edwards, A. C. Influence of sorption on the biological utilization of two simple carbon substrates. Soil Biol. Biochem. 30, 1895–1902 (1998). Poulton, S. W. & Raiswell, R. Chemical and physical characteristics of iron oxides in riverine and glacial meltwater sediments. Chem. Geol. 218, 203–221 (2005). van der Zee, C., Roberts, D. R., Rancourt, D. G. & Slomp, C. P. Nano-goethite is the dominant reactive oxyhydroxide phase in lake and marine sediments. Geology 31, 993–996 (2003). Canfield, D. E. The geochemistry of river particulates from the continental USA: Major elements. Geochim. Cosmochim. Acta 61, 3349–3365 (1997). Burdige, D. J. Geochemistry of Marine Sediments 107 108 (Princeton Univ. Press, 2006). Schwertmann, U. Inhibitory effect of soil organic matter on the crystallization of amorphous hydroxide. Nature 212, 645–646 (1966). Haese, R. R. et al. Iron species determination to investigate early diagenetic reactivity in marine sediments. Geochim. Cosmochim. Acta 61, 63–72 (1997). Mehra, O. P. & Jackson, M. L. Iron oxide removal from soils and clays by a dithionitecitrate system buffered with sodium bicarbonate. Clays Clay Miner. 7, 317–327 (1958). Mayer, L. M. Extent of coverage of mineral surfaces by organic matter in marine sediments. Geochim. Cosmochim. Acta 63, 207–215 (1999). Ransom, B., Bennet, R. H., Baerwald, R. & Shea, K. TEM study of in situ organic matter on continental margins: occurrence and the ‘‘monolayer’’ hypothesis. Mar. Geol. 138, 1–9 (1997). Mackey, D. J. & Zirino, A. Comments on trace metal speciation in seawater or do ‘‘onions’’ grow in the sea? Anal. Chim. Acta 284, 635–647 (1994). Guggenberger, G. & Kaiser, K. Dissolved organic matter in soil: challenging the paradigm of sorptive preservation. Geoderma 113, 293–310 (2003). Baas Becking, L. G. M. & Moore, D. The relation between iron and organic matter in sediments. J. Sedim. Petrol. 29, 454–458 (1959). Deflandre, B., Mucci, A., Gagné, J.-P., Guignard, C. & Sundby, B. Early diagenetic processes in coastal marine sediments disturbed by a catastrophic sedimentation event. Geochim. Cosmochim. Acta 66, 2547–2558 (2002). Rue, E. L. & Bruland, K. W. Complexation of iron(III) by natural organic ligands in the Central North Pacific as determined by a new competitive ligand equilibration/ adsorptive cathodic stripping voltammetric method. Mar. Chem. 50, 117–138 (1995). Kraemer, S. M. Iron oxide dissolution and solubility in the presence of siderophores. Aquat. Sci. 66, 3–18 (2004). Arnarson, T. S. & Keil, R. G. Changes in organic matter-mineral interactions for marine sediments with varying oxygen exposure time. Geochim. Cosmochim. Acta 71, 3545–3556 (2007). Wang, Z.-C., Druffel, E. R. M., Griffin, S., Lee, C. & Kashgarian, M. Radiocarbon studies of organic compound classes in plankton and sediment of the Northeastern Pacific Ocean. Geochim. Cosmochim. Acta 62, 1365–1378 (1998). Hedges, J. I., Keil, R. G. & Benner, R. What happens to terrestrial organic matter in the ocean? Org. Geochem. 27, 195–212 (1997). Mayer, L. M. Surface area control on organic carbon accumulation in continental margin sediments. Geochim. Cosmochim. Acta 58, 1271–1284 (1994). Mayer, L. M. Speculative organic matter preservation: an assessment and speculative synthesis – a comment. Mar. Chem. 49, 123–126 (1995). Supplementary Information is linked to the online version of the paper at www.nature.com/nature. Acknowledgements This work is dedicated to the memory of J. I. Hedges; in addition to being an inspiration to Y.G., he provided several sediment samples used in this work. We thank H. T. Yan for surface area measurements and the captains and crews of RV Coriolis II for their help during sampling missions on the St Lawrence estuary. L. N. Barazzuol is also acknowledged for her work during the first phase of the project. This work was supported by grants (Y.G. and A.M.) and scholarships (K.L.) from NSERC, CFI and FQRNT. This is GEOTOP contribution no. 2012-0001. Author Contributions The original hypothesis was formulated by Y.G., and K.L., Y.G. and A.M. designed the project, interpreted the data and wrote the manuscript. K.L. gathered all the data. Groundwork for this study was carried out by A.O., who also contributed to the writing of the manuscript. Author Information Reprints and permissions information is available at www.nature.com/reprints. The authors declare no competing financial interests. Readers are welcome to comment on the online version of this article at www.nature.com/nature. Correspondence and requests for materials should be addressed to Y.G. (ygelinas@alcor.concordia.ca) 2 0 0 | N AT U R E | VO L 4 8 3 | 8 M A R C H 2 0 1 2 ©2012 Macmillan Publishers Limited. All rights reserved