2014A&A...561A.145R

Proto-planetary nebulae (pPNe) are thought to represent the transitory phase of the late stages of low- to intermediate-mass stars. Most pPNe show a bipolar or multipolar geometry. The process(es) that transforms the spherical asymptotic giant branch (AGB) ejecta into a bipolar/multipolar geometry is not known in detail. Interactions between stars in a binary system are suspected to cause the departure from spherical symmetry. We aim to determine whether the existence of a binary source that ejects a wind (from the primary) and a bipolar precessing jet with a time-dependent ejection velocity (from the orbiting companion) could produce the morphology and kinematics of the well-known multipolar pPN CRL 618. We applied an anisotropic wavelet analysis to the [S ii] Hubble Space Telescope (HST) images of CRL 618 to determine the charateristic sizes of various small-scale structures across and along the optical jets of CRL 618. From the archival [S ii] HST images of CRL 618 with a 10.7 yr time base, we carried out proper-motion measurements of the emission structures observed in the lobes of CRL 618. We computed six 3D numerical simulations of a precessing, time-dependent ejection velocity jet launched from a hypothetical companion star of a binary system. We found the proper motions to be well aligned with the jet axis with tangential velocities ranging from 60 to 430 km/s. The tangential velocity is a monotonically increasing function of the distance to the central source. We found that our numerical simulations reproduce the morphology and proper-motion measurements of CRL 618 when we considered a trend of decreasing jet velocity with time to reproduce the tangential velocity vs. distance dependence. From the comparison we made between the structure and proper motions observed in CRL 618 and predictions from 3D jet simulations, we found that the size and morphology of the lobes and the proper motion behavior of CRL 618 can be explained in terms of a well-collimated bipolar ejection, a precession of the outflow axis, an approximately periodic time-variability of the outflow velocity (with a period of ∼15 yr), and a long-term trend of decreasing outflow velocities from dynamical timescales of 50 yr towards more recent times.