Post-common envelope PN, fundamental or irrelevant?
Orsola De Marco, T. Reichardt, R. Iaconi, T. Hillwig, G. H. Jacoby, D. Keller, R. G. Izzard, J. Nordhaus, E. G. Blackman
TTitle of your IAU SymposiumProceedings IAU Symposium No. xxx, 2016A.C. Editor, B.D. Editor & C.E. Editor, eds. c (cid:13) Post-common envelope PN,fundamental or irrelevant?
Orsola De Marco , , T. Reichardt , , R. Iaconi , , T. Hillwig , G. H.Jacoby , D. Keller , R. G. Izzard , J. Nordhaus and E. G. Blackman Department of Physics & Astronomy, Macquarie University, Sydney, NSW 2109, Australiaemail: [email protected] Astronomy, Astrophysics and Astrophotonics Research Centre, Macquarie University Department of Physics & Astronomy, Valparaiso University, Indiana, USA Lowell Observatory, Flagstaff, AZ, USA KU Leuven, Belgium Institute of Astronomy, Cambridge University, UK National Technical Institute for the Deaf, Rochester Institute of Technology, NY, USA University of Rochester, NY, USA
Abstract.
One in 5 PN are ejected from common envelope binary interactions but
Kepler resultsare already showing this proportion to be larger. Their properties, such as abundances can bestarkly different from those of the general population, so they should be considered separatelywhen using PN as chemical or population probes. Unfortunately post-common envelope PNcannot be discerned using only their morphologies, but this will change once we couple our newcommon envelope simulations with PN formation models.
Keywords. planetary nebulae: general, hydrodynamics, stars: AGB and post-AGB, binaries(including multiple): close, stars: evolution, stars: statistics
1. The fraction and chemistry of post-common envelope PN
At least one in 5 planetary nebulae (PN) derives from a common envelope (CE) binaryinteraction, where the PN is the ejected CE and the inner close binary is the result of thecompanion’s in-spiral into the progenitor AGB stars (Ivanova et al. 2013). This fraction isa lower limit because we cannot detect photometric variabilities ∼ < KeplerSpace Telescope resulted in the detection of 4 variables out of 5 objects with data (DeMarco et al. 2015). Of the 4 variables, one is a double-degenerate binary central star, andone is likely to be a pole-on binary. With variability amplitudes of 0.0007 and 0.0005 mag,neither would have been detected from the ground.A binary search using data from
Kepler II (K2), campaigns 0, 2, 7 and 11 is underway.Campaigns 0, 2 and 7 include a grand total of 15 PN, two of which have periodic variabil-ity consistent with binarity. The variability amplitudes, 0.05 and 0.02 mag, respectively,are at the low end of what could be detected from the ground. However, the sensitivityof K2 is up to 5 times worse than for
Kepler and neither of the 2 binaries detected with
Kepler would have been detected by K2. In addition, 11 of the 15 K2 targets analysed arecompact PN, while all the
Kepler targets are extended: the variability detection thresholdis therefore much worse for the bulk of the K2 targets. K2, Campaign 11 has 139 PN onsilicon of which 22 are extended. Hopefully those data will return better statistics. The
Kepler data show that low amplitude post-CE PN central binaries exist, as predictedand that the fraction of post-CE PN could be substantially larger than the lower limit of1 a r X i v : . [ a s t r o - ph . S R ] D ec O. De Marco et al.one in 5. This lower limit is already larger than the prediction from population synthesisof 9% for Solar metallicity of Izzard & Keller (2015, which are consistent with previouswork, e.g., Yungelson et al. (1993); Nie et al. (2012)). An over-abundance of post-CE PNcould be explained either if some single stars make under-luminous PN (Soker & Subag2005) or if some post-CE PN are mimics, such as for example Stroemgren spheres aroundhot stars (Frew & Parker 2010).Expectations of the properties of post-CE PN are different from those of PN from otherevolutionary channels. Post-CE PN have a curbed AGB evolution, resulting in reducedC/O ratios and lower s-process abundances. A secondary effect is that more massivestars are in binaries more often, something that could inflate the relative number ofmore massive stars in the post-CE PN population resulting in relatively larger N/Oratios. In Fig. 2 we can compare the top-left and bottom-right panels where we see therelatively broader distributions in the post-CE channel and also the larger, though stillsmall, proportion of PN at high N/O and very low C/O not present in the single starchannel. The observations are still insufficient for a proper comparison, but it is alreadyclear that the observed bipolar PN at low C/O and mostly but not exclusively high N/O(more massive central stars) can only be (approximately) reproduced by the CE channel.
2. Observed and simulated post-common envelope PN
Unfortunately we cannot unambiguously recognise post-CE PN based on their mor-phology alone. Approximately 50 post-CE central stars are known to date and abouttwo thirds of them are reasonably well imaged. As shown by Miszalski et al. (2009),many of these PN are approximately bipolar (e.g., A 41, A 79, K 1-2, Fig.2), but there issome diversity. Some PN display only a narrow waist (e.g., HaTr 4), sometimes with jets(A 63; jets not visible in Fig. 2). Ring nebulae (e.g., NGC 6337, Sp 1) are likely to bepole-on views of waists. More complex morphologies are also present: elliptical PN withjets (NGC 6778, Fleming 1), more or less elliptical distributions of faint gas on which we
Figure 1.
Observed vs. predicted C/O and N/O ratios for PN produced via different channels:(1) single stars, (2) wind mass-loss from binary star origin, (3) CE merger + wind mass loss,(4) CE ejection. Observation symbols: round: round PN; star: elliptical PN; square: bipolar PN;triangular: elliptical PN with bipolar core, rhomboid: quadrupolar PN. Observations: Leisy &Dennefeld (2006); Stanghellini et al. (2005). Figure from Keller, Izzard and Stanghellini, in prep. ost-common envelope PN accounting for the common envelope ejec-tion . Simulations of the CE interaction have shown that the ejection of the envelopetakes place primarily on the orbital plane (Sandquist et al. 1998; Passy et al. 2012). Wehave recently carried out a 2-million particle SPH simulation with the code phantom (Iaconi et al. 2017) and the parameters of Passy et al. (2012), but starting the simulationat the time of Roche lobe overflow, something that allows the distribution of ejected gasto be more realistic. In addition, with SPH the entire gas distribution at the end of thesimulation ( ∼
15 years after Roche lobe overflow) is known, albeit with varying resolution(from 0.3 R (cid:12) near the binary to 80 R (cid:12) near the outer parts of the distribution).This simulation has not considered the fast wind, nor the ionising radiation. Wecould plausibly assume that the central star at the end of our simulation is hotter than ∼ − g cm − in the vicinity of the stars (Fig. 3, left panel). Even the evacu-ated, funnel-like regions seen above and below the equatorial plane have high densities( ∼ − g cm − ). By comparison typical PN simulations blow a fast wind into a muchmore tenuous medium with typical densities of 10 − g cm − , e.g., Garc´ıa-Segura (2006).This shows that the CE structure presents more or less a solid wall to the fast wind. On Figure 2.
PN around post-CE central binaries; (a) A 63 (Mitchell et al. 2007); (b) HaTr 4(from the ESO NTT archive); (c) NGC 6778 (Guerrero & Miranda 2012); (d) A 41 (Jones et al.2010); (e) A 65 (Huckvale et al. 2013); (f) ETHOS 1 (Miszalski et al. 2011); (g) NGC 6337(Hillwig et al. 2010); (h) Sp 1 (Hillwig et al. 2016); (i) A 79, (j) He 2-428, (k) M 1-91 (Rodr´ıguezet al. 2001); (l) K 1-2 (Corradi et al. 1999); (m) HFG 1 (De Marco 2009); (n) PN G135.9+55.9(Napiwotzki et al. 2005); (o) M 2-29 (Hajduk et al. 2008); (p) Fleming 1 (Boffin et al. 2012).
O. De Marco et al.the other hand the entire CE is expanding at a range of speeds with the bulk of thevolume expanding at 20 km s − . Hence, in 100-1000 years the densities will decline by 5to 8 orders of magnitude and the wind will start sweeping and penetrating the CE. References
Boffin, H. M. J., Miszalski, B., Rauch, T., Jones, D., et al. 2012, Science, 338, 773Corradi, R. L. M., Perinotto, M., Villaver, E., Mampaso, A., et al. 1999, ApJ, 523, 721De Marco, O. 2009, PASP, 121, 316De Marco, O., Long, J., Jacoby, G. H., Hillwig, T., et al. 2015, MNRAS, 448, 3587Frew, D. J. & Parker, Q. A. 2010, PASA, 27, 129Garc´ıa-Segura, G. 2006, in IAU Symposium, Vol. 234, ed. M. J. Barlow & R. H. M´endez, 297–304Garc´ıa-Segura, G., Langer, N., R´o˙zyczka, M., & Franco, J. 1999, ApJ, 517, 767Guerrero, M. A. & Miranda, L. F. 2012, A&A, 539, A47Hajduk, M., Zijlstra, A. A., & Gesicki, K. 2008, A&A, 490, L7Hillwig, T. C., Bond, H. E., Af¸sar, M., & De Marco, O. 2010, AJ, 140, 319Hillwig, T. C., Jones, D., De Marco, O., Bond, H. E., et al. 2016, ApJ, 832, 125Huarte-Espinosa, M., Frank, A., Balick, B., Blackman, E. G., et al. 2012, MNRAS, 424, 2055Huckvale, L., Prouse, B., Jones, D., Lloyd, M., Pollacco, D., et al. 2013, MNRAS, 434, 1505Iaconi, R., Reichardt, T., Staff, J., De Marco, O., Passy, J.-C., et al. 2017, MNRAS, 464, 4028Ivanova, N., Justham, S., Chen, X., De Marco, O., et al. 2013, A&A Rev., 21, 59Izzard, R. G. & Keller, D. 2015, in ASPCS, Vol. 497, ed. Kerschbaum, Wing, Hron, 161Jones, D., Lloyd, M., Santander-Garc´ıa, M., L´opez, J. A., et al. 2010, MNRAS, 408, 2312Leisy, P. & Dennefeld, M. 2006, A&A, 456, 451Miszalski, B., Acker, A., Parker, Q. A., Boffin, H. M. J., Frew, D. J., et al. 2011, in APN5, 109Miszalski, B., Acker, A., Parker, Q. A., & Moffat, A. F. J. 2009, A&A, 505, 249Mitchell, D. L., Pollacco, D., O’Brien, T. J., Bryce, M., et al. 2007, MNRAS, 374, 1404Napiwotzki, R., et al. 2005, in AIPCS, Vol. 804, ed. Szczerba, Stasinska, & Gorny, p. 173Nie, J. D., Wood, P. R., & Nicholls, C. P. 2012, MNRAS, 423, 2764Passy, J.-C., De Marco, O., Fryer, C. L., Herwig, F., Diehl, S., et al. 2012, ApJ, 744, 52Rodr´ıguez, M., Corradi, R. L. M., & Mampaso, A. 2001, A&A, 377, 1042Sandquist, E. L., Taam, R. E., Chen, X., Bodenheimer, P., & Burkert, A. 1998, ApJ, 500, 909Soker, N. & Subag, E. 2005, AJ, 130, 2717Stanghellini, L., Shaw, R. A., & Gilmore, D. 2005, ApJ, 622, 294Yungelson, L. R., Tutukov, A. V., & Livio, M. 1993, ApJ, 418, 794
Figure 3.
Density (left) and velocity (right, in cm s − ; arrows indicate flow direction) slicesperpendicular to the orbital plane at 14.8 years of an SPH CE simulation between a 0.88 M (cid:12) ,RGB star and a 0.6 M (cid:12) companion, carried out with the phantomphantom