The biogeochemical significance of gaseous hydrocarbons sorbed to marine sediments

Sorption of hydrocarbon gases (HCs) to marine sediments is a recognized phenomenon that has been investigated mostly in the context of petroleum exploration. However, little is known about the mechanisms of sorption and the importance of biologically produced HCs. In this study, we sought to constrain quantities and sources of sorbed HCs, its major sorbents and sorption mechanisms and its potential importance for sedimentary biogeochemistry. We applied geochemical and mineralogical analysis to 431 sediment samples from different oceanographic settings and geochemical regimes, integrated efficiency tests of extraction pro-tocols for sorbed HCs, and used high pressure equipment to experimentally infer sorption capacities of clay minerals. Significant amounts of biogenic methane were liberated from all samples, regardless of the pore water geochemistry. Alkaline conditions mostly yielded more HCs than the established acidic extraction, whereas both methods indicated the importance of biological methane production. Ethane to hexane were not restricted to cold seep settings or sulfate-free sediments. C2 HCs were selectively retained according to their carbon number, and therefore showed opposite abundance patterns compared to the dissolved gas phase. Methanogenic sediments with a high quartz/phyllosilicate ratio released particularly high amounts of sorbed methane (up to 2.1 mmol kg-1 dry sediment). Both, clay-rich or sulfate-reducing environments displayed strong partitioning of HCs in favor of the sorbed reservoir. The mineralogical data set did not show significant correlations with inorganic minerals or organic carbon, whereas experimental sorption experiments point to the importance of hydro-phobic siloxane patches of tetrahedral silicate sheets as sorbent sites. Microbial production and consumption of methane, even in sulfate-replete environments, affected the sorbed meth-ane pool with resulting stable isotopic compositions ranging from -44 to -84 per mill vs. VPDB (N=225) for carbon, and from -194 to -203 per mill vs. SMOW (N=7) for hydrogen, respectively.