Muhammad Akashi

Accretion onto the Companion of η Carinae during the Spectroscopic Event. II. X-Ray Emission Cycle

We calculate the X-ray luminosity and light curve for the stellar binary system η Car for the entire orbital period of 5.54 yr. By using a new approach we find, as suggested in previous works, that the collision of the winds blown by the two stars can explain the X-ray emission and temporal behavior. Most X-ray emission in the 2-10 keV band results from the shocked secondary stellar wind. The observed rise in X-ray luminosity just before minimum is due to the increase in density and subsequent decrease in radiative cooling time of the shocked fast secondary wind. Absorption, particularly of the soft X-rays from the primary wind, increases as the system approaches periastron and the shocks are produced deep inside the primary wind. However, absorption cannot account for the drastic X-ray minimum. The 70 day minimum is assumed to result from the collapse of the collision region of the two winds onto the secondary star. This process is assumed to shut down the secondary wind, and hence the main X-ray source. We show that this assumption provides a phenomenological description of the X-ray behavior around the minimum.

Shaping planetary nebulae by light jets

We conduct numerical simulations of axisymmetrical jets expanding into a spherical asymptotic giant branch (AGB) slow wind. The three-dimensional flow is simulated with an axially symmetric numerical code. We concentrate on jets that are active for a relatively short time. Our results strengthen other studies that show that jets can account for many morphological features observed in planetary nebulae (PNs). Our main results are as follows. (1) With a single jets launching episode, we can reproduce a lobe structure having a ‘front lobe’, that is a small bulge on the front of the main lobe, such as that in the PN Mz 3. (2) In some runs, dense clumps are formed along the symmetry axis, such as those observed in the pre-PN M1-92. (3) The mass-loss history of the slow wind has a profound influence on the PN structure. (4) A dense expanding torus (ring; disc) is formed in most of our runs. The torus is formed from the inflated lobes and not from a separate equatorial mass-loss episode. (5) The torus and lobes are formed at the same time and from the same mass-loss rate episode. However, when the slow wind density is steep enough, the ratio of the distance divided by the radial velocity is larger for regions closer to the equatorial plane than for regions closer to the symmetry axis. (6) With the short jet-active phase, a linear relation between distance and expansion velocity is obtained in many cases. (7) Regions at the front of the lobe are moving sufficiently fast to excite some visible emission lines.

The jet feedback mechanism (JFM): From supernovae to clusters of galaxies

We study the similarities of jet-medium interactions in several quite different astrophysical systems using 2D and 3D hydrodynamical numerical simulations, and find many similarities. The systems include cooling flow (CF) clusters of galaxies, core-collapse supernovae (CCSNe), planetary nebulae (PNe), and common envelope (CE) evolution. The similarities include hot bubbles inflated by jets in a bipolar structure, vortices on the sides of the jets, vortices inside the inflated bubbles, fragmentation of bubbles to two and more bubbles, and buoyancy of bubbles. The activity in many cases is regulated by a negative feedback mechanism. Namely, higher accretion rate leads to stronger jet activity that in turn suppresses the accretion process. After the jets power decreases the accretion resumes, and the cycle restarts. In the case of CF in galaxies and clusters of galaxies we also study the accretion process, which is most likely by cold clumps, i.e., the cold feedback mechanism. In CF clusters we find that heating of the intra-cluster medium (ICM) is done by mixing hot shocked jet gas with the ICM, and not by shocks. Our results strengthen the jet feedback mechanism (JFM) as a common process in many astrophysical objects.

X-ray emission from jet–wind interaction in planetary nebulae

We conduct 2D numerical simulations of jets expanding into the slow wind of asymptotic giant branch stars. We show that the post-shock jets’ material can explain the observed extended X-ray emission from some planetary nebulae (PNs). Such jets are thought to shape many PNs, and therefore it is expected that this process will contribute to the X-ray emission from some PNs. In other PNs (not simulated in this work) the source of the extended X-ray emission is the shocked spherical wind blown by the central star. In a small fraction of PNs both sources might contribute, and a two-temperatures gas will fit better the X-ray properties than a one-temperature gas. A spatial separation between these two components is expected.

X-ray emission from planetary nebulae calculated by 1D spherical numerical simulations

We calculate the X-ray emission from both constant and time-evolving shocked fast winds blown by the central stars of planetary nebulae (PNe) and compare our calculations with observations. Using spherically symmetric numerical simulations with radiative cooling, we calculate the flow structure and the X-ray temperature and luminosity of the hot bubble formed by the shocked fast wind. We find that a constant fast wind gives results that are very close to those obtained from the self-similar solution. We show that in order for a fast shocked wind to explain the observed X-ray properties of PNe, rapid evolution of the wind is essential. More specifically, the mass-loss rate of the fast wind should be high early on when the speed is ∼300-700 km s -1 , and then it needs to drop drastically by the time the PN age reaches ∼ 1000 yr. This implies that the central star has a very short pre-PN (post-asymptotic giant branch) phase.

Forming equatorial rings around dying stars

We suggest that clumpy-dense outflowing equatorial rings around evolved giant stars, such as in supernova 1987A and the Necklace planetary nebula, are formed by bipolar jets that compress gas toward the equatorial plane. The jets are launched from an accretion disk around a stellar companion. Using the FLASH hydrodynamics numerical code we perform 3D numerical simulations, and show that bipolar jets expanding into a dense spherical shell can compress gas toward the equatorial plane and lead to the formation of an expanding equatorial ring. Rayleigh-Taylor instabilities in the interaction region break the ring to clumps. Under the assumption that the same ring-formation mechanism operates in massive stars and in planetary nebulae, we find this mechanism to be more promising for ring formation than mass loss through the second Lagrangian point. The jets account also for the presence of a bipolar nebula accompanying many of the rings.

X-ray emission by a shocked fast wind from the central stars of planetary nebulae

We calculate the X-ray emission from the shocked fast wind blown by the central stars of planetary nebulae (PNe) and compare with observations. Using spherically symmetric self-similar solutions, we calculate the flow structure and X-ray temperature for a fast wind slamming into a previously ejected slow wind. We find that the observed X-ray emission of six PNe can be accounted for by shocked wind segments that were expelled during the early-PN phase, if the fast wind speed is moderate, v 2 ∼ 400-600 km s -1 , and the mass-loss rate is a few times 10 -7 . M ⊙ yr -1 . We find, as proposed previously, that the morphology of the X-ray emission is in the form of a narrow ring inner to the optical bright part of the nebula. The bipolar X-ray morphology of several observed PNe, which indicates an important role of jets, rather than a spherical fast wind, cannot be explained by the flow studied here.

Bipolar rings from jet-inflated bubbles around evolved binary stars

We show that a fast wind that expands into a bipolar nebula composed of two opposite jet-inflated bubbles can form a pair of bipolar rings around giant stars. Our model assumes three mass loss episodes: a spherical slow and dense shell, two opposite jets, and a spherical fast wind. We use the FLASH hydrodynamical code in three-dimensions to simulate the flow, and obtain the structure of the nebula. We assume that the jets are launched from an accretion disk around a stellar companion to the giant star. The accretion disk is assumed to be formed when the primary giant star and the secondary star suffer a strong interaction accompanied by a rapid mass transfer process from the primary to the secondary star, mainly a main sequence star. Later in the evolution the primary star is assumed to shrink and blow a fast tenuous wind that interacts with the dense gas on the surface of the bipolar structure. We assume that the dense mass loss episode before the jets are launched is spherically symmetric. Our results might be applicable to some planetary nebulae, and further emphasize the large variety of morphological features that can be formed by jets. But we could not reproduce some of the properties of the outer rings of SN1987A. It seems that some objects, like SN1987A, require a pre-jets mass loss episode with a mass concentration at mid-latitudes.

Forming H-shaped and barrel-shaped nebulae with interacting jets

We conduct three-dimensional hydrodynamical simulations of two opposite jets launched from a binary stellar system into a previously ejected shell and show that the interaction can form barrel-like and H-like shapes in the descendant nebula. Such features are observed in planetary nebulae and supernova remnants. Under our assumption the dense shell is formed by a short instability phase of the giant star as it interacts with a stellar companion, and the jets are then launched by the companion as it accretes mass through an accretion disk from the giant star. We find that the H-shaped and barrel-shaped morphological features that the jets form evolve with time, and that there are complicated flow patterns, such as vortices, instabilities, and caps moving ahead along the symmetry axis. We compare our numerical results with images of 12 planetary nebulae, and show that jet-shell interaction that we simulate can account for the barrel-like or H-like morphologies that are observed in these PNe.

Numerical simulations of wind–equatorial gas interaction in η Carinae

We perform three-dimensional gas-dynamical simulations and show that the asymmetric morphology of the blue and red-shifted components of the outflow at hundreds of astronomical units (AU) from the massive binary system eta Carinae can be accounted for from the collision of the free primary stellar wind with the slowly expanding dense equatorial gas. Owing to the very complicated structure of the century-old equatorial ejecta, that is not fully spatially resolved by observations, we limit ourselves to modelling the equatorial dense gas by one or two dense spherical clouds. Because of that we reproduce the general qualitative properties of the velocity maps, but not the fine details. The fine details of the velocity maps can be matched by simply structuring the dense ejecta in an appropriate way. The blue and red-shifted components are formed in the post-shock flow of the primary wind, on the two sides of the equatorial plane, respectively. The fast wind from the secondary star plays no role in our model, as for most of the orbital period in our model the primary star is closer to us. The dense clouds are observed to be closer to us than the binary system is, and so in our model the primary star faces the dense equatorial ejecta for the majority of the orbital period.