PG 1610$+$062: a runaway B star challenging classical ejection mechanisms

TL;DR

This paper reports the discovery and detailed analysis of PG 1610+062, a B star ejected at velocities exceeding classical limits, challenging existing models of star ejection mechanisms in the Milky Way.

Contribution

The study presents the first detailed chemical and kinematic analysis of PG 1610+062, a runaway B star with an ejection velocity beyond classical limits, suggesting alternative ejection processes.

Findings

PG 1610+062 is a late B-type main-sequence star with low rotational velocity.

The star was ejected from the Carina-Sagittarius arm at about 550 km/s.

It is one of the most extreme known disk runaway stars.

Abstract

Hypervelocity stars are rare objects, mostly main-sequence (MS) B stars, traveling so fast that they will eventually escape from the Milky Way. Recently, it has been shown that the popular Hills mechanism, in which a binary system is disrupted via a close encounter with the supermassive black hole at the Galactic center, may not be their only ejection mechanism. The analyses of Gaia data ruled out a Galactic center origin for some of them, and instead indicated that they are extreme disk runaway stars ejected at velocities exceeding the predicted limits of classical scenarios (dynamical ejection from star clusters or binary supernova ejection). We present the discovery of a new extreme disk runaway star, PG 1610$+$062, which is a slowly pulsating B star bright enough to be studied in detail. A quantitative analysis of spectra taken with ESI at the Keck Observatory revealed that PG 1610$+$062 is a late B-type MS star of 4-5 $M_\odot$ with low projected rotational velocity. Abundances (C, N, O, Ne, Mg, Al, Si, S, Ar, and Fe) were derived differentially with respect to the normal B star HD 137366 and indicate that PG 1610$+$062 is somewhat metal rich. A kinematic analysis, based on our spectrophotometric distance (17.3 kpc) and on proper motions from Gaia's second data release, shows that PG 1610$+$062 was probably ejected from the Carina-Sagittarius spiral arm at a velocity of $550\pm40$ km s$^{-1}$, which is beyond the classical limits. Accordingly, the star is in the top five of the most extreme MS disk runaway stars and is only the second among the five for which the chemical composition is known.

Tables

Table 1. Table 1: Atmospheric parameters and abundances of the two program stars.

Object	$T_{\text{eff}}$	$\log \left(g\right)$	${\text{varv}}_{\text{rad}}$	$\text{varv}\sin \left(i\right)$	$\zeta$	$\xi$		$\log \left(n\left(x\right)\right)$
	(K)	(cgs)	(km s-1)					He	C	N	O	Ne	Mg	Al	Si	S	Ar	Fe
PG 1610$+$062	$\mathrm{14\hspace{0.17em}800}$	$4.054$	$157.4$a𝑎aa𝑎a$a$The radial velocity of PG 1610$+$062 is given here as the average over the four results from the ESI spectra. This value is consistent with radial velocities measured in low-resolution spectra taken at more than a dozen different epochs.	$15.5$	$0.0$	$2.40$		$-0.92$	$-3.34$	$-3.86$	$-3.09$	$-3.87$	$-4.51$	$-5.67$	$-4.24$	$-4.66$	$-5.32$	$-4.36$
Stat.	${}_{-80}{}^{}{}_{}{}^{+80}$	${}_{-0.023}{}^{}{}_{}{}^{+0.022}$	$7.7$a𝑎aa𝑎a$a$The radial velocity of PG 1610$+$062 is given here as the average over the four results from the ESI spectra. This value is consistent with radial velocities measured in low-resolution spectra taken at more than a dozen different epochs.	${}_{-1.5}{}^{}{}_{}{}^{+1.4}$	${}_{-0.0}{}^{}{}_{}{}^{+15.0}$	${}_{-0.26}{}^{}{}_{}{}^{+0.27}$		${}_{-0.03}{}^{}{}_{}{}^{+0.04}$	${}_{-0.07}{}^{}{}_{}{}^{+0.07}$	${}_{-0.12}{}^{}{}_{}{}^{+0.12}$	${}_{-0.05}{}^{}{}_{}{}^{+0.05}$	${}_{-0.04}{}^{}{}_{}{}^{+0.04}$	${}_{-0.06}{}^{}{}_{}{}^{+0.06}$	${}_{-0.07}{}^{}{}_{}{}^{+0.08}$	${}_{-0.07}{}^{}{}_{}{}^{+0.06}$	${}_{-0.04}{}^{}{}_{}{}^{+0.05}$	${}_{-0.33}{}^{}{}_{}{}^{+0.22}$	${}_{-0.05}{}^{}{}_{}{}^{+0.06}$
Sys.	${}_{-300}{}^{}{}_{}{}^{+300}$	${}_{-0.100}{}^{}{}_{}{}^{+0.100}$	…a𝑎aa𝑎a$a$The radial velocity of PG 1610$+$062 is given here as the average over the four results from the ESI spectra. This value is consistent with radial velocities measured in low-resolution spectra taken at more than a dozen different epochs.	${}_{-0.5}{}^{}{}_{}{}^{+0.6}$	${}_{-0.0}{}^{}{}_{}{}^{+0.1}$	${}_{-0.17}{}^{}{}_{}{}^{+0.16}$		${}_{-0.14}{}^{}{}_{}{}^{+0.12}$	${}_{-0.07}{}^{}{}_{}{}^{+0.08}$	${}_{-0.04}{}^{}{}_{}{}^{+0.04}$	${}_{-0.06}{}^{}{}_{}{}^{+0.06}$	${}_{-0.04}{}^{}{}_{}{}^{+0.05}$	${}_{-0.07}{}^{}{}_{}{}^{+0.07}$	${}_{-0.04}{}^{}{}_{}{}^{+0.05}$	${}_{-0.08}{}^{}{}_{}{}^{+0.07}$	${}_{-0.05}{}^{}{}_{}{}^{+0.06}$	${}_{-0.06}{}^{}{}_{}{}^{+0.10}$	${}_{-0.14}{}^{}{}_{}{}^{+0.13}$
HD 137366	$\mathrm{14\hspace{0.17em}930}$	$3.803$	$-15.5$	$11.3$	$2.5$	$1.91$		$-0.97$	$-3.52$	$-4.15$	$-3.16$	$-3.99$	$-4.66$	$-5.83$	$-4.35$	$-4.87$	$-5.55$	$-4.51$
Stat.	${}_{-20}{}^{}{}_{}{}^{+10}$	${}_{-0.002}{}^{}{}_{}{}^{+0.002}$	${}_{-0.1}{}^{}{}_{}{}^{+0.1}$	${}_{-0.1}{}^{}{}_{}{}^{+0.1}$	${}_{-0.6}{}^{}{}_{}{}^{+0.4}$	${}_{-0.07}{}^{}{}_{}{}^{+0.07}$		${}_{-0.01}{}^{}{}_{}{}^{+0.01}$	${}_{-0.02}{}^{}{}_{}{}^{+0.02}$	${}_{-0.02}{}^{}{}_{}{}^{+0.02}$	${}_{-0.01}{}^{}{}_{}{}^{+0.02}$	${}_{-0.02}{}^{}{}_{}{}^{+0.01}$	${}_{-0.02}{}^{}{}_{}{}^{+0.02}$	${}_{-0.03}{}^{}{}_{}{}^{+0.02}$	${}_{-0.02}{}^{}{}_{}{}^{+0.02}$	${}_{-0.01}{}^{}{}_{}{}^{+0.01}$	${}_{-0.04}{}^{}{}_{}{}^{+0.04}$	${}_{-0.02}{}^{}{}_{}{}^{+0.01}$
Sys.	${}_{-300}{}^{}{}_{}{}^{+300}$	${}_{-0.100}{}^{}{}_{}{}^{+0.100}$	${}_{-0.1}{}^{}{}_{}{}^{+0.1}$	${}_{-0.3}{}^{}{}_{}{}^{+0.2}$	${}_{-0.1}{}^{}{}_{}{}^{+0.1}$	${}_{-0.66}{}^{}{}_{}{}^{+0.42}$		${}_{-0.13}{}^{}{}_{}{}^{+0.12}$	${}_{-0.09}{}^{}{}_{}{}^{+0.09}$	${}_{-0.09}{}^{}{}_{}{}^{+0.06}$	${}_{-0.03}{}^{}{}_{}{}^{+0.02}$	${}_{-0.03}{}^{}{}_{}{}^{+0.03}$	${}_{-0.06}{}^{}{}_{}{}^{+0.05}$	${}_{-0.06}{}^{}{}_{}{}^{+0.05}$	${}_{-0.05}{}^{}{}_{}{}^{+0.05}$	${}_{-0.04}{}^{}{}_{}{}^{+0.05}$	${}_{-0.07}{}^{}{}_{}{}^{+0.05}$	${}_{-0.11}{}^{}{}_{}{}^{+0.07}$

Table 2. Table 3: Stellar parameters derived from photometry.

Parameter	Value
PG 1610$+$062:
Effective temperature $T_{\text{eff}}$	${\mathrm{14\hspace{0.17em}800}}_{-1100}^{+2500}$ K
Surface gravity $\log \left(g\left(\text{cm}{\text{s}}^{-2}\right)\right)$	${3.2}_{-1.2}^{+1.7}$
Angular diameter $\Theta$	$\left(8.6\pm 0.5\right)\times {10}^{-12}$ rad
Color excess $E\left(B-V\right)$	$\le 0.09$ mag
HD 137366:
Effective temperature $T_{\text{eff}}$	$\mathrm{15\hspace{0.17em}000}\pm 900$ K
Surface gravity $\log \left(g\left(\text{cm}{\text{s}}^{-2}\right)\right)$	$3.6\pm 0.9$
Angular diameter $\Theta$	$\left(5.91\pm 0.19\right)\times {10}^{-10}$ rad
Color excess $E\left(B-V\right)$	$0.056\pm 0.018$ mag

Table 3. Table 5: Ejection velocities (relative to the rotating Galactic disk) and heliocentric distances of candidate MS stars that were possibly ejected from the Galactic disk beyond the velocity limit of classical mechanisms.

Star	${\text{varv}}_{\text{ej}}$ (km s-1)	Distance (kpc)	Reference
HVS 5	${640}_{-40}^{+50}$	${31.2}_{-2.5}^{+3.2}$	(1), (2)
B711	${600}_{-50}^{+90}$	${28.5}_{-2.2}^{+3.1}$	(1), (2)
B434	$590\pm 20$	${40.5}_{-3.7}^{+4.7}$	(1), (2)
LAMOST-HVS1	${568}_{-17}^{+19}$	${19.1}_{-3.8}^{+5.1}$	(3)
PG 1610$+$062	$\mathrm{𝟓𝟓𝟎}\pm \mathrm{𝟐𝟎}$	${17.3}_{-1.0}^{+1.2}$	This work
HVS 7	$530\pm 30$	${48.2}_{-3.7}^{+4.3}$	(1), (2)
HVS 12	${510}_{-30}^{+40}$	${51.7}_{-6.1}^{+9.0}$	(1), (2)
LAMOST-HVS4	${480}_{-10}^{+13}$	$27.9\pm 1.5$	(4)a𝑎aa𝑎a$a$Assuming a MS nature, Li et al. (2018) give a Galactic rest-frame velocity at a disk intersection of $697\pm 12$ km s-1, which transforms to the given ejection velocity.
EC 19596$-$5356	${475}_{-83}^{+74}$	${13.81}_{-3.63}^{+4.80}$	(5)
HIP 56322	${471}_{-99}^{+189}$	${6.09}_{-1.92}^{+3.17}$	(5)
HIP 105912	${457}_{-133}^{+130}$	${4.17}_{-1.14}^{+1.70}$	(5)
HVS 8	${450}_{-30}^{+40}$	${37.2}_{-3.6}^{+4.4}$	(1), (2)
B733	$450\pm 10$	${9.9}_{-0.9}^{+0.7}$	(1), (2)
BD -2 3766	${425}_{-109}^{+151}$	${4.22}_{-1.10}^{+1.50}$	(5)
B485	${420}_{-10}^{+20}$	${33.3}_{-1.7}^{+3.7}$	(1), (2)
PHL 346	${418}_{-47}^{+49}$	${8.55}_{-1.33}^{+1.61}$	(5)
PB 5418	${415}_{-100}^{+141}$	${6.09}_{-1.49}^{+2.03}$	(5)
PG 1332$+$137	${413}_{-77}^{+38}$	${6.54}_{-1.70}^{+2.13}$	(5)
HIP 114569	${408}_{-71}^{+89}$	${1.60}_{-0.31}^{+0.40}$	(5)
PHL 2018	${399}_{-66}^{+68}$	${6.93}_{-1.77}^{+2.39}$	(5)
PG 1209$+$263	${390}_{-100}^{+293}$	${30.93}_{-7.28}^{+7.94}$	(5)
HD 271791	${390}_{-30}^{+70}$	$21\pm 4$	This workb𝑏bb𝑏b$b$Based on the distance and radial velocity from Heber et al. (2008) and on proper motions from Gaia DR2.
PG 0914$+$001	${369}_{-157}^{+240}$	${20.62}_{-5.28}^{+6.74}$	(5)

Table 4. Table A.12: Observed oscillation parameters of PG 1610 + + 062 derived from the two ATLAS light curves.

Parameter	Value
Frequency ${\nu }_{\text{osc}}$	$0.23052\pm 0.00022$ d-1
Period $P_{\text{osc}}$	$4.338\pm 0.005$ d
Reference epoch $T_{\text{ref}}$ (fixed)	$\mathrm{57\hspace{0.17em}230.0}$ MJD
Phase ${\varphi }_{\text{ref}}$ at epoch $T_{\text{ref}}$	$0.30\pm 0.08$
$o$ mean magnitude	$15.746\pm 0.008$ mag
$o$ semiamplitude	$21\pm 11$ mmag
$c$ mean magnitude	$15.457\pm 0.006$ mag
$c$ semiamplitude	$35\pm 7$ mmag

Equations

$${\text{mag}}_j(t)={\overline{\text{mag}}}_j+A_j\cos \left(2\pi \left[(t-T_{\text{ref}}){\nu }_{\text{osc}}+{\varphi }_{\text{ref}}\right]\right).$$

Full text

1 Introduction

Young stars are expected to be found close to their birthplaces, namely the star-forming regions in the Galactic disk. Finding them far away in the Galactic halo implies that they have been forced to leave their primal environment. Two mechanisms are usually discussed in the literature to explain the presence of these so-called runaway stars (see, e.g., Hoogerwerf et al. 2001 and references therein). In the binary-supernova scenario (Blaauw 1961) the massive primary star of a binary explodes as a core-collapse supernova and the secondary component is released at almost orbital velocity. In the dynamic scenario (Poveda et al. 1967) the runaway stars are formed via gravitational interactions in young and dense stellar clusters, for instance close binary-binary encounters, where the least massive star is usually set free. With typical ejection velocities below a few hundred km s*-1*, both of these disk runaway scenarios are by far less powerful than the Hills mechanism (Hills 1988), which describes the disruption of a binary system during a close flyby of the supermassive black hole at the Galactic center (GC). Due to the strong tidal forces, one component is captured while the other is able to leave the site at very high velocity (up to thousands of km s*-1*). To highlight their unique origin (and to follow the nomenclature by Vickers et al. 2015), stars stemming from this particular mechanism are referred to as Hills stars in this work. Apart from their formation channels, ejected stars may also be classified according to whether they are gravitationally bound to or unbound from the Milky Way. Stars exceeding their local escape velocity from the Galaxy are commonly called hypervelocity stars (HVSs), the first of which were discovered in 2005 (Brown et al. 2005, Hirsch et al. 2005, Edelmann et al. 2005). A dedicated spectroscopic survey covering 29% of the sky revealed 21 candidate HVSs, all of which are late B-type stars that are unbound from the Milky Way if they are main-sequence (MS) stars, and thus at distances of 50–120 kpc (Brown et al. 2014). Until recently, the Hills mechanism was widely assumed to be the only ejection scenario that is capable of producing MS HVSs (Brown 2015). However, high-precision astrometry from Gaia’s second data release (DR; Gaia Collaboration et al. 2018) shows that some of the candidate HVSs no longer qualify as Hills stars because the GC can be most likely ruled out as their spatial origin (Irrgang et al. 2018a). Because the ejection velocities of those dismissed Hills stars are higher than the upper limits for the two “classical” disk ejection scenarios mentioned above, a powerful yet neglected or unknown mechanism (e.g., dynamical interactions with massive stars or intermediate-mass black holes) must be at work (Irrgang et al. 2018a). To gain deeper insights, more stars ejected by this mechanism have to be studied in detail. Here, we investigate PG 1610$+$062, a blue star at high Galactic latitude ($b=+37.80\degr$). It was first discovered during the Palomar-Green survey (Green et al. 1986) where it was classified as a horizontal branch B star. Apart from a re-classification as MS B-type star by Geier et al. (2015), no attempt has been made since then to study this object in more detail. The star attracted our attention in the course of the MUCHFUSS project (Geier et al. 2011) because a set of low-resolution spectra indicated that its radial velocity might be variable (Geier et al. 2015). Unlike the faint stars of the HVS sample, which have visual magnitudes between 17.5 and 20 mag (Brown et al. 2014), PG 1610$+$062 is bright enough ($V=15.6$ mag) for a high-precision quantitative spectroscopic (Sect. 2) and photometric (Sect. 3) analysis. A kinematic investigation (Sect. 4) yields an ejection velocity of $550\pm 40$ km s*-1*, granting PG 1610$+$062 a place in the top five of the most extreme disk runaway MS stars known to date (Sect. 5).

2 Spectroscopic analysis