The particular behaviour of molybdenum towards the oxygenation potential in aqueous environments makes this element very useful to infer the oxygenation level under which a sediment is deposited. The first part of this PhD thesis is dedicated to the development of a new protocol for the extraction of molybdenum from geological samples. Until now, most extractions are performed using two ion exchange resin columns: an anionic one, then a cationic one. The protocol proposed here, initially set to allow measurements using a classical Sample-Standard-Bracketing (SSB) method, allows good purification and recovery of molybdenum through a single pass in an anion exchange resin. This purification is good enough to perform analyses using the double-spike method. A comparison between these two methods to correct the instrumental mass bias allowed us to better constrain SSB limits versus the double-spike method. The second part of this work aimed at geochemically characterizing an extraordinary fossil assemblage deposit located in North Greenland: the Sirius Passet Lagerstätte. In association with paleontological criteria, this study shows that these sediments were deposited in an environment at least partially oxygenated, in the form of muds enabling little exchange with the water column. The paleogeographical context, chemical analyses and thin-section observations suggest that these muds are mainly aeolian in origin.