Post-AGB-stars: The final stage in the evolution of solar type stars. The AGB-phase is the final evolutionary phase in the life of stars with low or intermediate masses ranging from 1 to 8 M(sol). In this chapter, we will describe the different stages solar type stars go through at the end of their life, when they have ascended the so-called Asymptotic Giant Branch (AGB). 1a. From the Main Sequence to the AGB-branch: low-mass (1-3 M(sol)) stars. When the core of a solar type star has exhausted all its core-hydrogen it starts to evolve off the Main Sequence. In the Hertzsprung-Russell diagram we see it moving towards the upper right, the so-called Red Giant Branch, meaning that the star is cooling down and becoming brighter. Its mass stays the same of course, so it becomes of low average density. In this phase, the star is burning hydrogen in a shell around the helium core. At a certain moment, the energy supplied by fusing hydrogen to helium is no longer sufficient to maintain pressure balance: the star becomes unstable and begins to contract. When the pressure in the core is high enough, it becomes electron-degenerate, and a new fusion process can start: the triple alpha process which fuses helium to carbon. Because of the electron degenerated state, the pressure is hardly sensitive to temperature; however, the triple alpha process is very sensitive to temperature which makes this process an explosive one, called the helium-flash. The star now quickly moves away from the RGB towards the blue in the HR-diagram. It has reached a stable phase which corresponds to the Horizontal Branch. The star is now burning helium in the core. Helium burning proceeds until it is completely converted to carbon and oxygen. Now shell burning of hydrogen and helium provide the energy, increasing luminosity and decreasing temperature. Because of this, the core is compressed and the electron gas becomes degenerate. Our star has now become an Asymptotic Giant Branch or AGB star. The AGB got its name because in the HR diagram it is located asymptotically along the first giant branch (also called RGB). 1b. Intermediate-mass (3-8 M(sol)) stars. For slightly more massive stars, the evolution is mostly similar to that of the lower mass stars as described above. The only difference is that such stars evolve off the RGB earlier, because the required pressure in the core to begin helium fusion is obtained earlier in the evolution, when the core is still not degenerate. Therefore, these stars don't experience a helium core-flash. 2. The AGB-phase. When the star evolves up the AGB-branch, instabilities called thermal pulses appear, caused by interactions between the hydrogen- and helium-shells. The burning of hydrogen leads to a mass increase of the helium layer. The base of the shell is then compressed to a point where it becomes electron-degenerate, causing a helium shell flash. The flash pushes the hydrogen shell outward to areas with low temperatures, where H fusion becomes impossible and the energy production in the shell ceases temporarily. After that, the helium shell drops in luminosity and the hydrogen shell reforms. With the next pulse, this cycle appears again. The luminosity of AGB stars depends on their core mass through the so-called core mass-luminosity relation. The physical reason for this relation is simple: as the degenerate core of the star increases in mass due to the helium shell burning in the helium shell flash, it decreases its radius and the result is an increase in pressure in the H- burning shell surrounding the core. Thus, after each flash the luminosity of the AGB star increases by a small percentage. Eventually solar type stars can reach luminosities of the order of 5000 L(sol)! The luminosity of stars does not change immediately after the AGB ends because the H-burning layer can still maintain its energy production. However, this H-burning will eventually cease and the newly born white dwarf lowers its luminosity by cooling down. Mass-loss also starts to appear near the end of the AGB-phase. When this accelerates the AGB-phase comes to an end, simply because the stellar wind removes the entire H-rich outer envelope in a relatively short period of time, thus terminating the AGB evolution: the star enters a new phase called the post-AGB phase. During this very high mass loss phase, the spectrum of the star changes dramatically: the stellar wind completely obscures the star from view and it can only be observed at long, infrared wavelengths. We will return to this point below. 3. The mass loss mechanism of AGB stars. As said earlier, stars lose mass when they ascend the AGB. The reason why this happens, is probably related to a combination of stellar radial pulsations and the radiation pressure that is exerted on the material in the outer layers of the stellar atmosphere. Note that these stellar pulsations, that have periods of the order of 50 to 1000 days, should not be confused with the helium shell flashes occurring near the core of the star, and which have much longer timescales of 10**4 to 10**5 years (depending on the luminosity of the star). During stellar pulsations, a shock wave is formed which travels outwards and pushes material out. Most of this gas will fall back but will encounter the next shock wave coming in from below. This causes the scale-height of the atmosphere to increase dramatically: the density at a few stellar radii can be orders of magnitude larger than in a star with an atmosphere in hydrostatic equilibrium. Matter is injected into the outer layers of the star and cools down, and new molecules start to form. Eventually the gas condenses and small solid particles, referre to as dust, are formed. The radiation pressure causes the dust to experience an outward force. This causes it to accelerate away from the star, moving out of the molecular layer. If there is enough momentum coupling between the dust and the gas, the collisions between gas and dust drag the gas along with it and a stellar outflow, driven by the radiation pressure, is formed. When the dust moves outward, the acceleration decreases, because the radiation pressure decreases with distance squared from the star. Eventually, the dust reaches a so-called terminal velocity, which is usually between 10 and 20 km/s. A typical AGB star loses about 10^-8 to 10^-6 M(sol)/yr. But near the end of the AGB, the mass loss rate rises to 10^-5 to 10^-3 M(sol)/yr, effectively removing the hydrogen-rich envelope in about 10^4 yrs. This phase is often referred to as the superwind phase. The star starts to shed its outer layers, eventually causing hotter layers closer to the core to appear: the star leaves the AGB and the hot white dwarf emerges. The very high mass loss rates during the superwind phase result in a very high density of the outflow. Dust will efficiently absorbe all stellar photons and the star disappears from sight at optical and even at near-infrared wavelengths. Of course there is energy conservation: the dust grains are heated and radiate at infrared wavelengths. The star becomes an Infrared-bright star, only detectable with IR-sensitive detectors. 4. The post-AGB-phase. When a star enters this phase, its evolution is almost at an end. The mass loss rate drops to very low values, because there is no envelope mass left to feed the outflow. The radius becomes smaller and its luminosity remains the same (it is still determined by core mass, the so-called core mass-luminosity relation), this is causing the star to move straight to the blue in the HR diagram. The AGB wind will continue to expand and will now be detached from the star, often observed as a detached shell of gas and dust around the star. As soon the star is hot enough, it may ionize the expelled AGB material which becomes visible as a Planetary Nebula. The central star now has become a white dwarf, and begins cooling down, following the WD-cooling track. Interestingly, the hot white dwarf develops a fast but very low density wind, with expansion velocity in excess of 1000 km/s. This wind will collide with the slowly expanding AGB remnant, which produces the beautiful shapes of planetary nebulae that are observed by e.g. the Hubble Space Telescope. 5. The chemical evolution of AGB and post-AGB stars. When a star enters the AGB phase, it has a carbon-oxygen core that only recently became degenerate, surrounded by two shells where fusion takes place. The first shell has helium fusion, the second fuses hydrogen. These two shells are surrounded by a large, low density hydrogen envelope, in which the C/O ratio is about 0.5 by number. However, every helium shell flash (also called thermal pulse) will suddenly inject a large amount of C into the H-rich envelope. This will gradually increase the C/O ratio to unity and even higher than unity: a so-called carbon star is formed. AGB stars with C/O less than 1 are called M stars, and objects with C/O larger than 1 are C stars. The change in C/O ratio has large implications for the kind of molecules that can form. This is because the CO molecule is the most stable of all, and forms first. Since every C takes up one O atom to form CO, the kind of molecules that can form from the remainder depends on whether there is excess carbon or oxygen. In the case of excess oxygen (as in the sun and in M giants), oxygen-rich molecules can form: H2O, CO2 SO2, OH, SiO etc. In the case of excess carbon, molecules as HCN, C2H2, C2, C3, SiC, TiC, etc can form. This will of course have a large effect on the spectral appearance of the stars, as well as on the kind of dust that can form from the gas-phase molecules. f. 6. How do we recognize post-AGB stars? It is good to know which objects can be classified as post-AGB stars. There is a large difference between a star that just left the AGB and one that is at the brink of forming a PN. A good way to identify post-AGB stars is by means of spectral type and luminosity class. Suggested are spectral types B to K, and luminosity class I to III. Including M-stars herein causes confusion with normal AGB stars and red giants. O-stars are hot enough to ionize the dust cloud remnants, and are therefore central stars of PNe. Luminosity classes below III are not in the range for post-AGBs. Another way to do it is by studying the chemical composition of the photosphere, which is significantly different from that of the Sun. Another good criterion is the galactic latitude. The spectra of post-AGB stars are often classified as supergiants, and these stars are only found in the galactic plane. If a "supergiant" is found at a high galactic latitude (which means in the halo), it is most likely a post-AGB star. Another detection method is looking at the star in IR, where the cool dust of the envelope has thermal emission. There are two classes of variable stars classified as post-AGB stars, RV Tau and R CrB stars. The R CrB stars show extreme low abundancy of H, which suggests a mechanism which removes the hydrogen. the major ideas for these mechanisms are either a binary system or a very late thermal pulse during the AGB phase. Not al PAGB stars have a massive detached AGB remnant, the very low mass stars take very long to reach the O-type (may be up to 10^5 yr), and in these systems the star blows the dust away and leaves practically nothing left at the end when the central star is hot enough to ionize the envelope, the PN that is expected to be formed is then hardly visible. Supergiants at high galactic latitude are not expected: massive stars all form in the disk of the galaxy and live too short to move far away from their birthplace; most of these high galactic latitude supergiants turn out to be PAGB stars. The spectra of these stars show a metal deficiency. Therefore these stars are old and a high mass is thus ruled out: these are likely PAGB stars. Most of the PAGB stars we know are low mass objects, because we can most easily observe these stars through their expanding AGB remnants. More massive, and thus more luminous PAGB stars have such dense shells that these stars turn into PNe before the central star becomes readily observable.