Search

Protostellar jets are an important agent of star formation feedback, tightly connected with the mass-accretion process. The history of jet formation and mass-ejection provides constraints on the mass accretion history and the nature of the driving source. We want to characterize the time-variability of the mass-ejection phenomena at work in the Class 0 protostellar phase, in order to better understand the dynamics of the outflowing gas and bring more constraints on the origin of the jet chemical composition and the mass-accretion history. We have observed the emission of the CO 2-1 and SO N_J=5_4-4_3 rotational transitions with NOEMA, towards the intermediate-mass Class 0 protostellar system Cep E. The CO high-velocity jet emission reveals a central component associated with high-velocity molecular knots, also detected in SO, surrounded by a collimated layer of entrained gas. The gas layer appears to accelerate along the main axis over a length scale delta_0 ~700 au, while its diameter gradually increases up to several 1000au at 2000au from the protostar. The jet is fragmented into 18 knots of mass ~10^-3 Msun, unevenly distributed between the northern and southern lobes, with velocity variations up to 15 km/s close to the protostar, well below the jet terminal velocities. The knot interval distribution is approximately bimodal with a scale of ~50-80yr close to the protostar and ~150-200yr at larger distances >12". The mass-loss rates derived from knot masses are overall steady, with values of 2.7x10^-5 Msun/yr (8.9x10^-6 Msun/yr) in the northern (southern) lobe. The interaction of the ambient protostellar material with high-velocity knots drives the formation of a molecular layer around the jet, which accounts for the higher mass-loss rate in the north. The jet dynamics are well accounted for by a simple precession model with a period of 2000yr and a mass-ejection period of 55yr., Comment: 13 pages, 9 figures, 3 table. Accepted in A&A