Magnetic fields of AGB and post-AGB stars

TL;DR

This paper reviews magnetic field observations in AGB and post-AGB stars, discusses their potential origins and roles, and highlights recent findings suggesting intrinsic magnetic activity and rotation in these evolved stars.

Contribution

It provides an updated overview of magnetic field measurements and explores the possible intrinsic origins of magnetic fields in AGB stars, including recent observational evidence.

Findings

Magnetic fields are present in AGB and post-AGB star envelopes.

Recent observations suggest intrinsic magnetic activity and rotation.

Magnetically aligned dust detected around a supergiant star.

Abstract

There is ample evidence for strong magnetic fields in the envelopes of (Post-)Asymptotic Giant Branch (AGB) stars as well as supergiant stars. The origin and role of these fields are still unclear. This paper updates the current status of magnetic field observations around AGB, post-AGB stars and describes their possible role during these stages of evolution. The discovery of magnetically aligned dust around a supergiant star is also highlighted. In our search for the origin of the magnetic fields, recent observations show the signatures of possible magnetic activity and rotation, indicating that the magnetic fields might be intrinsic to the AGB stars.

Tables

Table 1. Table 1: Energy densities in AGB envelopes

		Photosphere	SiO	H2O	OH	CO/CN
$B$	[G]	$\sim 1-10$?	$\sim 3.5$	$\sim 0.3$	$\sim 0.003$	$\sim 0.003-0.008$
$R$	[AU]	-	$\sim 3$	$\sim 25$	$\sim 50$	$\sim 50-100$
$V_{\exp }$	[km s-1]	$\sim 20$	$\sim 5$	$\sim 8$	$\sim 10$	$\sim 10$
$n_{H_2}$	[cm-3]	$\sim {10}^{11}$	$\sim {10}^{10}$	$\sim {10}^8$	$\sim {10}^6$	$\sim {10}^5$
$T$	[K]	$\sim 2500$	$\sim 1300$	$\sim 500$	$\sim 300$	$\sim 150$
$B^2/8\pi$	[dyne cm-2]	${\mathrm{𝟏𝟎}}^{-1.4,+0.6}$	${\mathrm{𝟏𝟎}}^{+0.1}$	${\mathrm{𝟏𝟎}}^{-2.4}$	${10}^{-6.4}$	${\mathrm{𝟏𝟎}}^{-6.0,-6.4}$
$nKT$	[dyne cm-2]	${10}^{-1.5}$	${10}^{-2.7}$	${10}^{-5.2}$	${10}^{-7.4}$	${10}^{-8.7}$
$\rho V_{\exp }^2$	[dyne cm-2]	${10}^{-0.3}$	${10}^{-2.5}$	${10}^{-4.1}$	${\mathrm{𝟏𝟎}}^{-5.9}$	${10}^{-6.9}$
$V_A$	[km s-1]	$\sim 20$	$\sim 100$	$\sim 300$	$\sim 8$	$\sim 8$

Full text

††volume: S343††journal: Title of your IAU Symposium††editors: A.C. Editor, B.D. Editor & C.E. Editor, eds.

1 Introduction

Magnetic fields are ubiquitous throughout the Universe and play an important role across a wide range of scales. Primordial magnetic fields could have played a role in the formation of the first stars just as magnetic fields in molecular clouds are an important ingredient in current star formation. Magnetic fields have also been detected in almost all stellar types and in almost all phases of stellar evolution (e.g. Berdyugina 2009), and have significant effects on stellar evolution through, e.g. their influence on the internal mixing. The magnetic field of stars can have either a dynamo origin, i.e. be generated by a dynamo process in the star itself (e.g. Charbonneau 2014), or can be the result of a remnant ’fossil’ field, which are fields that originate from the star formation process (e.g. Braithwaite & Spruit 2004). The stellar magnetic field is affected by the changes of physical properties during stellar evolution and, because of flux conservation, becomes increasingly difficult to observe at the stellar surface when the star expands in the final phases of its life. However, in stellar end products, such as white dwarfs and neutron stars, magnetic fields are also shown to be significant.

The role of magnetic fields around AGB stars is not clear. In principle, they could help levitate material off the stellar surface through Alfv’en waves (e.g. Falceta-Gonçalves & Jatenco-Pereira 2002), or through the creation of cool spots on the surface above with dust can form easier (Soker 1998). A specific model for the AGB star o Ceti (Mira A) has shown that a hybrid magnetohydrodynamic-dust-driven wind scenario can explain its mass loss (Thirumalai & Heyl 2013). In such a model, Alfvén waves add energy to lift material before dust forms and radiation pressure accelerates a wind. Magnetic fields also play an important role in the internal mixing required for s-process (slow) neutron capture reactions that define the stellar yields (e.g. Trippella et al. 2016).

After the AGB phase, the stellar envelopes undergo a major modification as they evolve to Planetary Nebulae (PNe). The standard assumption is that the initial slow AGB mass loss quickly changes into a fast superwind, generating shocks and accelerating the surrounding envelope (Kwok et al. 1978). It is during this phase that the typically spherical CSE evolves into a Planetary Nebula. As the majority of of pre-PNe are aspherical, an additional mechanism is needed to explain the departure from sphericity. This mechanism is still a matter of fierce debate. One possibility is that the interaction of the post-AGB star and a binary companion or massive planet supports a strong magnetic field that is capable of shaping the outflow (e.g. Nordhaus et al. 2007).

This paper expands on (and partly reproduces) the reviews presented in [Vlemmings(2018), Vlemmings 2018] and [Vlemmings14, Vlemmings 2014] and I refer interested readers to those review (and references therein) for further background.

2 Overview of magnetic field observations

2.1 AGB stars

Generally, AGB magnetic field measurements come from maser polarization observations (SiO, H2O and OH). These have revealed a strong magnetic field throughout the circumstellar envelope. Figure 1, the magnetic field strength in the regions of the envelope traced by the maser measurements throughout AGB envelopes. The field appears to vary between $B\propto R^{-2}$ (solar-type) and $B\propto R^{-1}$ (toroidal). Although the maser observations trace only oxygen-rich AGB stars, recent CN Zeeman splitting observations ([Duthu et al.(2017)Duthu, Herpin, Wiesemeyer, Baudry, Lèbre, & Paubert, Duthu et al. 2017]) indicate that similar strength fields are found around carbon-rich stars. The envelope magnetic fields are also consistent with thus far the only direct measurement of the Zeeman effect on the surface of an AGB star, the Mira variable star $\chi$ Cyg ([Lèbre et al.(2014)Lèbre, Aurière, Fabas, Gillet, Herpin, Konstantinova-Antova, & Petit, Lèbre et al. 2014]). In Table. 1 an overview is given of the energy densities throughout the AGB envelopes.

The large-scale structure of the magnetic field is more difficult to infer, predominantly because the maser observations often probe only limited line-of-sights. Even though specifically OH observations seem to indicate a systematic field structure, it has often been suggested that there might not be a large-scale component to the field that would be necessary to shape the outflow (Soker 2002). Until recently the only tight shape constraints throughout the envelope had been determined for the field around the supergiant star VX Sgr, where maser observations spanning 3 orders of magnitude in distance are all consistent with a large scale, possibly dipole shaped, magnetic field (Vlemmings et al. 2005, Vlemmings et al. 2011).

Very recent ALMA observations have shown that it will soon be possible to finally overcome the problems with determining the circumstellar magnetic field structure. This involves observations aimed at measuring the Goldreich-Kylafis effect, which allows us to use the polarisation of non-maser molecular lines (in this case CO) to determine the magnetic field morphology in the more diffuse circumstellar gas. The first of these observations, for the post-AGB star OH 17.7-2.0, indicate that the magnetic field structure probed by the CO is consistent with that derived from OH maser observations (Fig. 2, Tafoya & Vlemmings in prep.). This puts to rest the decades old question if maser magnetic field measurements can really be used to probe the large-scale fields. The second set of observations has given us the first view velocity resolved view of the large-scale magnetic field in the AGB stars IRC+10216 (Fig. 3, Vlemmings et al. in prep.).

2.2 post-AGB stars

Similar to the AGB stars, masers are the main source of magnetic field information of post-AGB and P-PNe and even for some PNe. OH maser observations indicate magnetic field strengths similar to those of AGB stars (few mG) and a clear large scale magnetic field structure ([Bains et al.(2003)Bains, Gledhill, Yates, & Richards, Bains et al. 2003], [Gómez et al.(2016)Gómez, Uscanga, Green, Miranda, Suárez, & Bendjoya, Gómez et al. 2016]). Also dust polarization observations indicate a large scale magnetic field ([Sabin et al.(2015a)Sabin, Hull, Plambeck, Zijlstra, Vázquez, Navarro, & Guillén, e.g. Sabin et al. 2015]).

Magnetic fields have also been detected around the so-called ’water-fountain’ sources. These sources exhibit fast and highly collimated H2O maser jets that often extend beyond even the regular OH maser shell. With the dynamical age of the jet of order 100 years, they potentially are the progenitors of the bipolar (P-)PNe. Observations of the arch-type of the water-fountains, W43A, have revealed a strong toroidal magnetic field that is collimating the jet ([Vlemmings et al.(2006), Vlemmings et al. 2006]). For another water-fountain source, IRAS 15445-5449, a synchrotron jet related to strong magnetic fields has been detected ([Pérez-Sánchez et al.(2013)Pérez-Sánchez, Vlemmings, Tafoya, & Chapman, Pérez-Sánchez et al. 2013]). Similar, synchrotron emission has been found from what could be one of the youngest PNe ([Suárez et al.(2015)Suárez, Gómez, Bendjoya, Miranda, Guerrero, Uscanga, Green, Rizzo, & Ramos-Larios, Suárez et al. 2015]).

Finally, recently also surface fields have been measured for 2 post-AGB stars ([Sabin et al.(2015b)Sabin, Wade, & Lèbre, Sabin et al. 2015]). These fields are consistent with the fields inferred from the envelope measurements

2.3 Supergiant stars

Many maser observations show that strong magnetic fields are also present in the envelopes of Red Supergiant stars ([Vlemmings et al.(2002)Vlemmings, Diamond, & van Langevelde, Herpin et al.(2006)Herpin, Baudry, Thum, Morris, & Wiesemeyer, e.g. Vlemmings et al. 2002, Herpin et al. 2006]). The questions about local or large scale fields, are the same as around AGB stars. As noted above, the supergiant VX Sgr is one of the first stars where a large scale magnetic field, with a structure consistent throughout the envelope, was found. At (sub-)millimeter wavelengths it is now possible to simultaneously study the polarization of masers, regular molecular lines, and circumstellar dust using ALMA. Recent observations of VY CMa indicate magnetically aligned dust and consistent structures between the maser and non-maser molecular lines ([Vlemmings et al.(2017b)Vlemmings, Khouri, Martí-Vidal, Tafoya, Baudry, Etoka, Humphreys, Jones, Kemball, O’Gorman, Pérez-Sánchez, & Richards, Fig. 4, Vlemmings et al. 2017]). The observations indicate that magnetic fields could be involved in the mass loss of these massive stars.

3 Indirect tracers and origin of the magnetic field

The origin of AGB magnetic fields is unclear and might require an extra source of angular momentum to maintain a stellar dynamo. This however depends strongly on the magnetic coupling throughout the star itself. If a sufficiently strong magnetic field persist at the AGB stellar surface, it might be possible to detect signs of magnetic activity. Recently, it has been shown that the majority of the AGB stars are UV-emitters ([Montez17, Montez et al. 2017]) which could be a sign of (magnetic) activity. Similarly, recent observation of the surface of the AGB star W Hya show high brightness temperature hotspots ([Vlemmings et al.(2017a)Vlemmings, Khouri, O’Gorman, De Beck, Humphreys, Lankhaar, Maercker, Olofsson, Ramstedt, Tafoya, & Takigawa, Fig 5, Vlemmings et al. 2017]). These spots can arise from strong shocks but could also point to magnetic activity.

As previously noted, the angular momentum imparted by a stellar (or sub-stellar) companion might be needed to maintain a stellar dynamo that can generate the observed magnetic fields. However, rotation is very difficult to measure for the extended AGB stars that are undergoing pulsations and show large convective cells. Only very recently has ALMA been able to measure the fast ($\sim 1$ km s*-1*) rotation of the AGB star R Dor ([Vlemmings2018, Fig 6, Vlemmings et al. 2018]). As the rotation is almost two orders of magnitude larger than otherwise expected, it is a likely sign of interaction with an hitherto unknown companion. Unfortunately, no magnetic field observations exist yet for R Dor and it it is thus not yet possible to establish a link between the generation of a magnetic field and the fast rotation.

4 Conclusions

Magnetic fields are ubiquitous around AGB and post-AGB stars, and several observations indicate a link between the magnetic field and the collimated outflows found in pre-PNe. Additionally, indirect observations of hotspots and UV-emission might point to magnetic activity on the surface of AGB stars. However, it is only now possible to start probing the morphology of the magnetic field in AGB envelopes and to finally determine the role of magnetism around evolved stars.