Extreme r-process enhanced stars at high metallicity in Fornax
DDraft version February 18, 2021
Typeset using L A TEX twocolumn style in AASTeX63
Extreme r-process enhanced stars at high metallicity in Fornax ∗ M. Reichert,
1, 2
C. J. Hansen,
3, 4 and A. Arcones
1, 2, 5 Technische Universit¨at Darmstadt, Institut f¨ur Kernphysik, Schlossgartenstr. 2, 64289 Darmstadt, Germany Helmholtz Forschungsakademie Hessen f¨ur FAIR, GSI Helmholtzzentrum f¨ur Schwerionenforschung, 64291 Darmstadt, Germany Max-Planck-Institut f¨ur Astronomie, K¨onigstuhl 17, D-69117 Heidelberg, Germany Copenhagen University, Dark Cosmology Centre, The Niels Bohr Institute, Vibenshuset, Lyngbyvej 2, DK-2100 Copenhagen, Denmark GSI Helmholtzzentrum f¨ur Schwerionenforschung GmbH, Planckstr. 1, D-64291 Darmstadt, Germany (Received December 18, 2020; Revised —; Accepted —)
Submitted to APJABSTRACTWe present and discuss three extremely rapid neutron-capture (r-)process enhanced stars locatedin the massive dwarf spheriodal galaxy Fornax. These stars are very unique with an extreme Euenrichment at high metallicities. They have the largest Eu abundances ever observed in a dwarfgalaxy opening new opportunities to further understand the origin of heavy elements formed by ther-process. We derive stellar abundances of Co, Zr, La, Ce, Pr, Nd, Er, and Lu using 1-dimensional,local thermodynamic equilibrium (LTE) codes and model atmospheres in conjunction with state-of-the art yield predictions. We derive Zr in the largest sample of stars (105) known to date in a dwarfgalaxy. Accurate stellar abundances combined with a careful assessment of the yield predictions haverevealed three metal-rich stars in Fornax showing a pure r-process pattern. We define a new class ofstars, namely Eu-stars, as r-II stars (i.e., [Eu/Fe] >
1) at high metallicities (i.e., [Fe / H] (cid:38) − . < Keywords: stars: chemically peculiar — galaxies: dwarf — galaxies: individual — nuclear reactions,nucleosynthesis, abundances INTRODUCTIONThe creation of heavy elements poses a number ofopen and interesting questions going from a small scalesto the largest scales including formation of stars or evengalaxies. Recent studies of Milky Way (MW) halo starshave shown that stars rich in neutron-capture elementsmay have originated in now dissolved (ultra faint) dwarfgalaxies (Roederer et al. 2018a). Limited to smaller
Corresponding author: M. [email protected] ∗ Based on data obtained from the ESO reduced archive. ProgramIDs: 080.B-0784(A), 171.B-0588(A), 71.B-0641(A). sample sizes, the results seem to indicate that stars withthe strongest rapid neutron-capture process (r-process)enhancement have formed ex situ and later been ac-creted into the MW. This makes dwarf galaxies excellentstudy cases for understanding the r-process.Fornax (Fnx) is one of the most massive dwarf spheri-odals (dSph) in the Local Group (LG). It shows a uniquestar formation history with a sudden increase of star for-mation only ∼ >
1, Beers & a r X i v : . [ a s t r o - ph . GA ] F e b Reichert et al.
Christlieb 2005) at high metallicities. Moreover, we re-port a large study of the neutron-capture element Zr inthe dSph Fornax. Our study presents the first detec-tion of the heavy element Lu in a dSph. In contrastto most r-process enriched stars, the three Eu-stars inFornax are metal-rich, and we use them to understandthe astrophysical source of the r-process by comparingto various yield predictions.Stars with peculiar enhanced r-process abundanceshave rarely been observed at high metallicities. Mostof them are clearly members of a dSph galaxy, e.g.,Ursa Minor (Cos 82, Shetrone et al. 2001; Aoki et al.2007; Sadakane et al. 2004 or Sculptor (SCMS 982,Geisler et al. 2005; Sk´ulad´ottir et al. 2020). We notethat there are also neutron-capture enhanced stars atsimilar metallicities reported already by Letarte et al.(2010) in Fornax, which was not found in Letarte et al.(2018) and Reichert et al. (2020). Also MW field stars,for example, 2MASS 18174532-3353235 (Johnson et al.2013) a possible bulge contender, shows a high neutron-capture enhancement at a high metallicity. Moreover,some halo stars show similar abundance enhancementsand are thought to be accreted into the MW halo, as in-dicated by a low [ α/ Fe] ratio (J1124+4535, Xing et al.2019), while for others, their origin is, however, not soclear. HD 222925 is located in the halo, but due to itshighly eccentric and retrograde orbit it has been sug-gested to be accreted from a satellite galaxy (Roedereret al. 2018a). By looking at the alpha-abundances andfitting a knee to the stellar population, the dSph masscan be assessed. Following the fit of the α -knee of Re-ichert et al. (2020, Eq. 6), the host environment musthave at least the size of Sagittarius or Fornax and anaccretion scenario may therefore be possible (see alsoRoederer et al. 2018b, for a discussion on the origin ofHD 222925).The astrophysical sites of the r-process are still underdiscussion, even if recent kilonova (Abbott et al. 2017;Watson et al. 2019) has shown that neutron star merg-ers (NSM) can produce heavy elements. Galactic chemi-cal evolution (GCE) models (e.g., Matteucci et al. 2014;Cˆot´e et al. 2019; Kobayashi et al. 2020) suggest that anadditional site may be active in our galaxy. An excit-ing possibility are magneto-rotational supernovae (MR-SNe; Nishimura et al. 2015, 2017; M¨osta et al. 2018;Reichert et al. 2021). Stars at intermediate metallicitiesare highly mixed and remain poorly studied in the GCEscheme. Also collapsars (Siegel et al. 2019; Miller et al.2019) have been discussed as a potential site. Our re-sults indicate that both sites, NSM and MR-SNe, canexplain the Eu-stars and probe a late star formationburst in Fornax. Table 1.
Different FLAMES/GIRAFFE setups showingminimum and maximum wavelength coverage and resolvingpower ( R ). Grating λ min [˚A] λ max [˚A] R HR10 5339 5619 19800HR13 6120 6405 22500HR14 6308 6701 17740LR8 8206 9400 6500
This paper is organised as follows. In Sect. 2, wepresent the observations, sample, and stellar parame-ters. The abundances are shown in Sect. 3 including thelargest Zr sample in a dwarf galaxy and heavy neutron-capture elements with focus on Eu and the first Lu de-tection in a dwarf galaxy. In Sect. 4, we discuss in detailthe possible origin of the Eu-stars, which is associatedwith one r-process event and requires late star forma-tion. Finally, we conclude in Sect. 5. OBSERVATIONS, SAMPLE, AND STELLARPARAMETERS2.1.
Observations
We present four Fornax stars including three withanomalous high Eu ([Eu / Fe] < . . The quality ofthe available observations allows for an accurate and pre-cise determination of abundances (typically to within ± . and perform thedata processing (sky correction, radial velocity shifts,co-adding, and normalisation of the spectra) as in Re-ichert et al. (2020). 2.2. Sample
These stars are clearly members of Fornax as previousstudies have already assigned them Fornax membershipbased on radial velocities, and distance to the centre ofFornax (see Fig. 1 and Battaglia et al. 2006; Lemasleet al. 2014). Also the proper motions of
Gaia indicatetheir membership (see Table 2).The [ α /Fe] ratio of the stars also fits with the generaltrend of Fornax (Fig. 2) and similar to many other r- [Eu / Fe] = log ( N Eu /N Fe ) − log ( N Eu /N Fe ) (cid:12) , with the number ofeuropium and iron atoms per cm , N Eu and N Fe . https://archive.eso.org/wdb/wdb/adp/phase3 main/form xtreme r-process enhanced stars at high metallicity in Fornax RA [deg] D e c [ d e g ] Battaglia et al. (2006)Lemasle et al. (2014)Hendricks et al. (2014) Letarte et al. (2010)Kirby et al. (2010)Fnx-mem0546 Fnx-mem0556Fnx-mem0595Fnx-mem0607
Figure 1.
Coordinates of sample stars together with point-ings of Battaglia et al. (2006), Lemasle et al. (2014), Hen-dricks et al. (2014), Letarte et al. (2010), and Kirby et al.(2010). The large oval shows the nominal tidal radius andthe small dashed oval the core radius (Battaglia et al. 2006).The black cross marks the center of the galaxy as given byBattaglia et al. (2006). The orange 2D histogram shows theamount of stars observed by
Gaia (Gaia Collaboration et al.2018). Cyan crosses indicate the globular clusters of Fornax(Mackey & Gilmore 2003).
Table 2.
Proper motions in right ascension direction µ δ ,and in declination direction µ α cos δ taken from Gaia (GaiaCollaboration et al. 2018). Furthermore, we list the meanvalues of Fornax (McConnachie & Venn 2020).Object µ δ [mas yr − ] µ α cos δ [mas yr − ]Fnx-mem0546 0 . ± . − . ± . . ± . − . ± . . ± . − . ± . . ± . − . ± . process enriched metal-rich stars, a clear contributionfrom type Ia SNe is visible.2.3. Stellar parameters
The stars were previously analysed by Lemasle et al.(2014), Reichert et al. (2020), and Battaglia et al.(2006). Table 3 shows their determined effective tem-perature, surface gravity, and metallicity (for Battagliaet al. 2006 the metalicities are from the calcium triplet).We note that the star Fnx-mem0546 was excluded fromLemasle et al. (2014) due to convergence problems whendetermining the stellar parameters. In general, effective [Fe/H] [ M g / F e ] Fnx-mem0607Fnx-mem0595Fnx-mem0546 Fnx-mem0556HD 222925J1124+4535 SCMS 982Cos 82Ret IIFnxMW
Figure 2. [Mg/Fe] versus metallicity is shown for thefour Fornax stars that we investigate here (see also Re-ichert et al. 2020): Fnx-mem0607 (red star), Fnx-mem0595(yellow square), Fnx-mem0546 (yellow diamond), and Fnx-mem0556 (yellow circle). For comparison we show other For-nax stars (green stars, Reichert et al. 2020) and MW stars(grey dots, Reddy et al. 2003; Cayrel et al. 2004; Reddyet al. 2006; Ishigaki et al. 2013; Fulbright 2000; Nissen et al.1997; Prochaska et al. 2000; Stephens & Boesgaard 2002;Ivans et al. 2003; McWilliam et al. 1995; Ryan et al. 1996;Gratton & Sneden 1988; Edvardsson et al. 1993; Roedereret al. 2016; Ezzeddine et al. 2020). Moreover, we compareto other typical r-process enhanced stars: Cos 82 (blue di-amond, Shetrone et al. 2001), SCMS 982 (black diamond,Geisler et al. 2005), HD 222925 (magenta diamond, Roed-erer et al. 2018b), J1124+4535 (blue diamond, Xing et al.2019), and Reticulum II (teal diamonds, Ji et al. 2016b). Inall following figures we keep the same symbols and colors forthe various stars. temperatures in Reichert et al. (2020) are hotter than inLemasle et al. (2014) because of different methods in de-riving them (spectroscopic versus photometric, see Ap-pendix B of Reichert et al. 2020 for a more detailed dis-cussion). Here, we adopt stellar parameters of Reichertet al. (2020) with effective temperatures and metal-licities determined with the automated code SP Ace(Boeche & Grebel 2016), surface gravities from photom-etry, and the microturbulence derived with an empiricalrelation from Kirby et al. (2010). The strong neutron-capture element enhancement is also present when usingthe stellar parameters of Lemasle et al. (2014), hence thechoice of stellar parameters does not change our conclu-sions. ABUNDANCESWe combine the previously determined abundances(Reichert et al. 2020) with eight newly analyzed el-ements: Co, Zr, La, Ce, Pr, Nd, Er, and Lu (Ta-ble 4). Hyperfinesplitting is included for the redder
Reichert et al.
Table 3.
Stellar parameters of different studiesReichert et al. (2020) Lemasle et al. (2014) Battaglia et al. (2006)Identifier T eff log g [Fe/H] T eff log g [Fe/H] [Fe/H]Fnx-mem0546 4367 ±
110 0 . ± . − . ± .
08 - - - − . ± . ±
58 0 . ± . − . ± .
09 3971 ±
150 0 . ± . − . ± . − . ± . ±
67 0 . ± . − . ± .
13 3889 ±
150 0 . ± . − . ± . − . ± . ±
63 0 . ± . − . ± .
09 3916 ±
150 0 . ± . − . ± .
02 -
Table 4.
Lines analyzed in addition to those used in Re-ichert et al. (2020).Element Wavelength EP log gf LiteratureCo I 5483 .
34 1.709 -1.500 1Y II 5509 .
89 0.992 -1.015 2Zr I 6127 .
46 0.154 -1.060 3Zr I 6134 .
62 0.000 -1.280 3Zr I 6143 .
20 0.071 -1.100 3La II 6320 .
43 0.173 -1.610 4La II 6390 .
46 0.321 -1.410 5Ce II 5512 .
08 1.007 -0.390 6Pr II 5509 .
15 0.482 -1.168 7Pr II 6165 .
94 0.923 -0.205 7Nd II 5356 .
97 1.263 -0.280 8Nd II 5361 .
17 0.559 -1.480 8Nd II 5361 .
47 0.680 -0.370 7Er II 5414 .
60 0.000 -2.499 7Lu II 6221 .
87 1.540 -0.760 9
References —(1) Lawler et al. (2015); (2) Hannaford et al.(1982); (3) Biemont et al. (1981); (4) Corliss & Bozman(1962); (5) Lawler et al. (2001); (6) Lawler et al. (2009); (7)Meggers et al. (1975); (8) Den Hartog et al. (2003); (9) denHartog et al. (1998). line of La, Co, and Lu (Lawler & Dakin 1989; Kurucz2011). We derive all additional abundances by syn-thesizing and fitting theoretical spectra with the localthermodynamic equilibrium (LTE) spectrum synthesiscode MOOG (version of 2014, Sneden 1973) using 1-dimensional model atmospheres from Kurucz (1970) ,and solar abundances from Asplund et al. (2009). Inaddition, we have tried to extract carbon from severalmolecular absorption lines but those were too weak inthe covered spectral region. Therefore, we can excludea strong carbon enhancement.All derived abundances, including those already ob-tained by Reichert et al. (2020), are presented in Ta-ble 5 for the four Fornax stars. The star Fnx-mem0607 http://kurucz.harvard.edu/grids.html is used as a reference because it has typical abun-dances of neutron-capture elements in Fornax stars. Thestars Fnx-mem0556, and Fnx-mem0595 have the high-est europium abundance ever observed to our knowl-edge with log (cid:15) (Eu) = 0 .
98 (Table 5). Within uncer-tainties, this is comparable only to SCMS 982 in Sculp-tor (log (cid:15) (Eu) = 0 . ± .
18, Geisler et al. 2005). In thefollowing, we will refer to Eu-stars, defining them asr-II stars (i.e., [Eu/Fe] >
1) at high metallicities (i.e.,[Fe/H] (cid:38) − . xtreme r-process enhanced stars at high metallicity in Fornax Table 5.
Derived absolute abundances (including also those already presented in Reichert et al. (2020, R20)).Element Fnx-mem0546 Fnx-mem0556 Fnx-mem0595 Fnx-mem0607 ReferenceMg I 6 . ± .
27 6 . ± .
24 6 . ± .
11 6 . ± .
17 R20Sc I 1 . ± .
26 1 . ± .
10 1 . ± .
29 1 . ± .
11 R20Ti II 3 . ± .
31 4 . ± .
25 3 . ± .
28 4 . ± .
11 R20Cr I 3 . ± .
39 5 . ± .
26 4 . ± .
18 4 . ± .
44 R20Mn I 3 . ± .
37 3 . ± .
14 4 . ± .
23 4 . ± .
28 R20Fe I 6 . ± .
08 6 . ± .
09 6 . ± .
13 6 . ± .
09 R20Co I 3 . ± .
29 3 . ± .
26 3 . ± .
24 3 . ± .
19 This studyNi I 4 . ± .
20 5 . ± .
21 5 . ± .
23 5 . ± .
23 R20Y II 0 . ± .
25 1 . ± .
29 1 . ± .
24 - R20Zr I 2 . ± .
30 2 . ± .
13 2 . ± .
18 1 . ± .
20 This studyBa II 1 . ± .
12 1 . ± .
30 1 . ± .
14 1 . ± .
12 R20La II 1 . ± .
28 1 . ± .
26 1 . ± .
13 0 . ± .
09 This studyCe II 1 . ± .
25 1 . ± .
24 1 . ± .
24 - This studyPr II 1 . ± .
25 1 . ± .
34 1 . ± .
23 0 . ± .
15 This studyNd II 1 . ± .
19 1 . ± .
29 1 . ± .
30 0 . ± .
22 This studyEu II 0 . ± .
19 0 . ± .
13 0 . ± .
12 0 . ± .
34 R20Er II - 1 . ± .
25 1 . ± .
28 - This studyLu II 0 . ± .
23 0 . ± .
26 0 . ± . < .
23 This study[Ba / Eu] − .
05 ( r ) − .
88 ( r ) − .
85 ( r ) − .
19 ( r + s ) -[Ba / La] − .
33 ( r ) − .
80 ( r ) − .
33 ( r ) − .
03 ( r + s ) - Largest Zr sample in a dwarf galaxy
We have derived (neutral) zirconium abundances forall Fornax stars presented in Reichert et al. (2020). Formany stars it is the first Zr determination, making itthe largest sample of Zr abundances in a dwarf galaxyto date. Figure 4 shows the [Zr / Fe] in Fornax includ-ing the neutron-capture enhanced stars. The averageZr abundance for all Fornax stars is slightly sub solarand the reference star (Fnx-mem0607) is located slightlybelow this general trend. There may be corrections tothe abundances due to the LTE assumption which tendsto affect for neutral atoms when they are the minorityspecies as it is the case for Zr (c.f., to the derived abun-dances and discussions of Andrievsky et al. 2017).The three Eu-enhanced Fornax stars (Fnx-mem0546,Fnx-mem0556, and Fnx-mem0595) are clearly enhancedalso in Zr. This agrees with these stars having an r-process pattern typical of other r-I or r-II stars, see forexample the Reticulum II stars (teal diamonds) in Fig. 4and the discussion in Sect. 4.1. Despite the high metal-licities and Zr being dominated by the s-process in thesolar system ( ∼ Heavy neutron-capture elements
The three Eu-enhanced Fornax stars are enriched ineuropium with ∼ (cid:15) (Eu) = 0 .
98, Fnx-mem0595and Fnx-mem0556 are the most europium-enriched starsobserved until today, together with SCMS 982 that isenriched by an s- or i- process (see Sk´ulad´ottir et al.2020, for a discussion). Even though rare, there areseveral stars with enhanced europium at low metallic-ities. However, such stars are even more rare at highmetallicities. In Fig. 5, we show some of those Eu-stars:Cos 82 (Shetrone et al. 2001; Aoki et al. 2007; Sadakaneet al. 2004), J1124+4535 (Xing et al. 2019), HD 222925(Roederer et al. 2018b), and SCMS 982 (Geisler et al.2005; Sk´ulad´ottir et al. 2020). With the exception ofSCMS 982, these stars show a dominant contributionfrom the r-process as indicated by their [Ba / Eu] ratioand discussed below (Fig. 7).The combination of metal-rich stars together withthe extreme enhancements of neutron capture elementsgives us the unique possibility to derive abundances ofneutron capture elements that are otherwise challeng-ing to detect. Therefore, we are able to derive lutetiumabundances (Fig. 6).Due to the heavy blending of the absorption line, weclaim a detection if more than three points deviate bymore than three sigma (dashed line in Fig. 6) from theflux when assuming no enhancement. Even though un-certain due to irregular noise next to the absorption
Reichert et al.
Wavelength [Å] R e l a t i v e F l u x SCMS 982Fnx-mem0607Fnx-mem0595Fnx-mem0556Fnx-mem0546CNNi I Eu II Fe IFe I/Ni I
Wavelength [Å] R e l a t i v e F l u x SCMS 982Fnx-mem0607Fnx-mem0595Fnx-mem0556Fnx-mem0546Ba IIFe I Fe I Zr I Zr I Fe I/Fe II
Figure 3.
Spectra (including an offset for better visibil-ity) for Fnx-mem0546, Fnx-mem0556, Fnx-mem0595, Fnx-mem0607, and SCMS 982 for regions around the Eu absorp-tion line (top panel) and around a Ba II line (bottom panel). [ Z r / F e ] Fnx-mem0546Fnx-mem0556Fnx-mem0595 Fnx-mem0607J1124+4535HD 222925 Cos 82SCMS 982Ret II FnxMW
Figure 4. [Zr/Fe] versus metallicity following the same no-tation and references as in Fig. 2. line, we clearly detect lutetium in Fnx-mem0546, butalso in Fnx-mem0556 and Fnx-mem0595. Because the l o g ( E u ) Fnx-mem0556Fnx-mem0546Fnx-mem0595 Fnx-mem0607HD 222925J1124+4535 Cos 82SCMS 982Ret IIFnxMW [Fe/H] [ E u / F e ] Figure 5.
Absolute Eu (upper panel) and relative [Eu/Fe]abundances (lower panel) versus metallicity following thesame notation and references as in Fig. 2. continuum shows some artifacts next to the absorptionline in Fnx-mem0595, we assign an additional error of0 . ORIGIN OF EU-STARSIn this section we probe the origin of the Eu-stars byfirst finding the predominant process that enriched thesestars. Following, we assess which possible sites couldhost this process. We make use of yield predictions andabundance ratios for this purpose, but we caution theuse of a single element ratio to assign a dominant pro-cess contribution. We point towards using such ratiostogether with observationally derived abundance pat-terns, as these present a more complete chemical traceof the true stellar enrichment.4.1. r-, i-, or s-process?
At high metallicity, the neutron-capture elementsstem mainly from the s- and r-process, possibly alsofrom the i-process. In order to check whether the Eu-stars are enriched by an r-process event or whether theyhave a strong contribution from the s-process, we firstcheck the [Ba/Eu] ratio shown in Fig. 7 and Table 5. xtreme r-process enhanced stars at high metallicity in Fornax n o r m . f l u x Lu II
Fe I Fe I Ni I V I
Fnx-mem0546 [Lu/H] = -9.00[Lu/H] = 0.12
Fnx-mem0556 Lu II
Fe I Fe I Ni I V IY I [Lu/H] = -9.00[Lu/H] = 0.510.20.00.2 r e s i d u a l n o r m . f l u x Fnx-mem0595 Lu II
Fe I Fe I [Lu/H] = -9.00[Lu/H] = 0.08
Fnx-mem0607 Lu II
Fe I Fe I Ni I V IY I [Lu/H] = -9.00[Lu/H] = -0.776218 6220 6222 62240.20.00.2 r e s i d u a l Wavelength [Å]
Figure 6.
Normalized flux together with a synthetic spectrum. The synthetic spectrum is shown for a negligible amount ofLu (black line) and for an enhanced Lu abundance (red line). A red band indicates a deviation of the Lu abundance in therange of a typical error ( ± . For the three stars, the ratio is well below zero and thisis often used to define neutron-capture-rich, r-processenhanced stars ([Ba/Eu] <
0, Beers & Christlieb 2005).A stronger constraint on the heavy-element enhance-ment is the estimate of the pure process trace, which istypically assessed through the [Ba/Eu] being less than ∼ − . ∼ . Reichert et al. N S N S ( B ) N S N S ( R ) N S B H ( R ) D i s k D i s k D i s k M R - S N ( W ) M R - S N ( O ) [ B a / E u ] FnxFnx-mem0546 Fnx-mem0556Fnx-mem0595 Fnx-mem0607Cos 82 SCMS 982J1124+4535 HD 222925Ret II
Figure 7. [Ba/Eu] versus metallicity (right panel) following the same notation and references as in Fig. 2. Theoreticalpredictions are shown for s-process (horizontal lines) and r-process (left panel). The s-process yields are from F.R.U.I.T.Y.database (Cristallo et al. 2011) and are indicated by horizontal lines starting at the metallicity of the model. These models arefor three AGB masses: 1 . M (cid:12) (solid lines), 2 . M (cid:12) (dashed lines), and 5 . M (cid:12) (dotted lines). The red, thin lines indicate ametallicity of Z = 0 . Z = 0 . α ,Fe]=0.5. The r-process ratios are shown withbars that cover the calculate range taking into account uncertainties due to the various astrophysical conditions and differentnuclear physics theoretical models. The ratio is shown for the early, dynamical ejecta of neutron star merger (NSNS) andneutron star black hole merger (NSBH) as well as for their disk ejecta (Disk 1, Disk 2, and Disk 3). Results are also presentedfor MR-SNe based on two simulations. For more details and references see Sect. 4.3 and Cˆot´e et al. (2020); Eichler et al. (2019);Reichert et al. (2021). theoretical predictions of the r-process within uncertain-ties. However, those uncertainties are rather large andthus it is not possible to conclude which scenario con-tributed most to the observed abundances.Now we explore another possibility to check whetherthe three Eu-star’s abundances are mainly due to ther-process. We compare their abundance patterns tothat of HD 222925 (Roederer et al. 2018b), a starthat has been identified as r-process enhanced. Fig-ure 8 shows this comparison and demonstrates that twoof the Eu-stars (Fnx-mem0556 and Fnx-mem0595) fitwith an excellent reduced χ < l o g Fnx-mem0546 (3.27)Fnx-mem0556 (0.83)Fnx-mem0595 (0.67)35 40 45 50 55 60 65 70 75
Atomic Number, Z l o g Figure 8.
Abundances of Fnx-mem0546 (diamonds), Fnx-mem0556 (circles), and Fnx-mem0595 (squares) scaled to fitthe r-process pattern of HD 222925 (black line, with errorsindicated as grey bands, Roederer et al. 2018b). In thelegend we give the reduced χ of the fit for Z >
50. Bottompanel shows the relative differences between the Fornax starsand HD 222925. xtreme r-process enhanced stars at high metallicity in Fornax
One r-process event
After showing that the Eu-stars are r-process en-hanced, we estimate how many r-process events are nec-essary to enrich these stars and calculate the ejectedeuropium per event. The three Eu-stars were born in agas cloud/region with a total mass M gas . Magg et al.(2020) provide an approximation for the dilution masswhere the new elements from an r-process event(s) aremixed: M gas = 1 . · M (cid:12) E . n − . , (1)here n is the ambient density and E is the explosionenergy of the event. Magg et al. (2020) uses n = 1corresponding to the environment where Pop III starsformed, however this has an small impact on the finaldilution mass due to the small exponent. The explo-sion energy ( E ) of the event can be assumed to bebetween 10 and 10 erg (c.f., 10 erg for GW190817;Kathirgamaraju et al. 2019 and the explosion energy of10 erg of hypernovae; Nomoto et al. 2006). The dilu-tion mass ( M gas ) therefore lies between 10 to 10 M (cid:12) ,which is at least three orders of magnitudes smallerthan the total stellar mass of Fornax M Fornax ∼ M (cid:12) ((cid:32)Lokas 2009).Before the r-process event(s), the mass of Eu in thegas was M pre − eventEu and the mass injected by the eventwas M r − eventEu . We can estimate M pre − eventEu using theabundance of our reference stars, Fnx-mem0607, in thefollowing definition:log (cid:15) (Eu) = log M pre − eventEu A Eu · M gas + 12 , (2)where A Eu = 152 is the average mass number of eu-ropium. Therefore, the Eu in the r-process stars is:log (cid:15) (Eu) = log M pre − eventEu + M r − eventEu A Eu · M gas + 12 , (3)and from this expression we obtain an europiummass per event M r − eventEu between ∼ . · − and ∼ · − M (cid:12) . This is similar to the values reported byJi et al. (2016a) to explain the r-process enhanced starsin Reticulum II ( ∼ . · − − · − M (cid:12) ). Therefore,it is likely that Fnx-mem0556, Fnx-mem0595, and Fnx-mem0546 also got enriched by a single r-process event.Moreover, the stars probably formed only a few Myr af-ter the event. Otherwise, the r-process material wouldhave been mixed into a larger amount of gas (van deVoort et al. 2019).Our simple estimate has several input parameters(e.g., n , E ), therefore we show in Fig. 9 an overview Figure 9.
Absolute abundances of europium versus dilu-tion mass. The amount of europium needed to enhance astar to a given absolute abundance value is color coded.The orange box indicates the range, predicted from Fnx-mem0556, where the left and right vertical lines are ob-tained from different explosion energies and the horizontallines from the absolute europium abundances including theerror. The range of estimated Eu masses to enhance Reticu-lum II (Ji et al. 2016a) is given by blue lines. The maximumEu mass ejected from NSMs based on simulation is shown asa black line (8 · − M (cid:12) ), while the maximum ejected eu-ropium mass from MR-SNe simulations is traced by red line(1 . · − M (cid:12) ). covering different possible values. In this figure, theabsolute europium abundance of Fnx-mem0556 are in-dicated by horizontal orange lines, the estimated dilu-tion mass by vertical orange lines, and the colors cor-respond to the Eu mass ejected by the r-process eventusing Eq. 3. The latter changes smoothly over severalorders. However, the amount of Eu needed to explainthe Eu-stars ( M Eu ∼ − − − M (cid:12) , Fig. 9) agreeswith having only one r-process event.4.3. Chasing the r-process site in Fornax
Which r-process event is responsible for the enhancedEu in the three Fornax stars? The possible candidatesare compact binary mergers (two neutron stars or a neu-tron star and a black hole) and MR-SNe including theearly explosion phase and the late evolution as collapsar.In Fig. 7, we have compared the [Ba/Eu] of the Eu-stars to various theoretical predictions from compactbinary mergers (dynamical and disk ejecta) and fromMR-SNe. We test the dynamical ejecta of mergers oftwo neutron stars (NSNS (B), Bovard et al. 2017) and(NSNS (R), Korobkin et al. 2012), and a neutron starand a black hole (NSBH (R), Korobkin et al. 2012);ejecta from the accretion disk formed after merger sur-rounding the compact object (Disk 1, Disk 2, and Disk3, Wu et al. 2016); ejecta from magneto-rotational su-pernovae MR-SN (W) (Winteler et al. 2012) and MR-SN0
Reichert et al. (R) (Reichert et al. 2021). A similar set of models wasused also by Cˆot´e et al. (2020) to compare to meteoriteratios of r-process isotopes. However, there are too largetheoretical uncertainties in the nuclear physics input aswell as the variation in the astrophysical conditions andthis makes it impossible to conclude from the [Ba/Eu]ratio alone which scenario has contributed to the en-hanced Eu.The mass of Eu estimated above ( M Eu ∼ − − · − M (cid:12) ) could be also used as a constraint on the r-process site. We compare this amount of Eu to the massthat is ejected in NSM and MR-SNe, as obtained in hy-drodynamic simulations. However, these simulations arestill uncertain due to various aspects (e.g., resolution,magnetic fields, neutrino matter interactions, high den-sity equation of state) and predictions about the amountof Eu ejected are only approximate.In compact binary mergers, the masses of ejected eu-ropium in the dynamic ejecta depends on the simulationand can differ by orders of magnitudes (see, e.g., Cˆot´eet al. 2018 for an overview of different ejecta massesof the dynamical ejecta). Just to name a few exam-ples, the europium mass based on the simulation of Bo-vard et al. (2017) is M Eu ∼ − − − M (cid:12) . Yieldsof other simulations are slightly higher with M Eu ∼ − − · − M (cid:12) (Korobkin et al. 2012) and similar M Eu ∼ · − − · − M (cid:12) (Goriely et al. 2011). Inthe case of a NSBH merger, the ejected europium massof the dynamic ejecta can reach M Eu ∼ · − M (cid:12) (Ko-robkin et al. 2012). In addition, the disk ejecta will alsocontribute to europium ( M Eu ∼ · − − · − M (cid:12) ,Wu et al. 2016). However, it is still under discussionhow much each ejected component contributes and howneutron-rich the components are (see, e.g., Shibata &Hotokezaka 2019, for a recent review).In GW170817, Watson et al. (2019) directly observedSr and predicted a mass M (Sr) (cid:38) · − M (cid:12) . Whenassuming that the ejecta is scaled like the solar r-processresidual (Sneden et al. 2008) the ejected europium massis M Eu ∼ · − M (cid:12) . We note that the lighter heavyelements like strontium usually scatter with respect tothe heavier elements. Because of this and other involveduncertainties, this value should be treated as a rough es-timate only. Nevertheless, it agrees well with the massesof M Eu ∼ · − − . · − M (cid:12) obtained in Cˆot´e et al.(2018) for GW170817.The amount of Eu produced in MR-SNe based on re-cent simulations is (cid:46) · − M (cid:12) (Winteler et al. 2012), (cid:46) . · − M (cid:12) (Nishimura et al. 2015), (cid:46) · − M (cid:12) (Nishimura et al. 2017), and (cid:46) · − M (cid:12) (Reichertet al. 2021). These values are on the lower end of ourassumed uncertainties (Fig. 9), but they could still be responsible for the enrichment. After some MR-SNe, ablack hole forms and is surrounded by an accretion disk.The disk conditions may be similar to those found in ac-cretion disks after neutron star mergers and may favouran r-process. Siegel et al. (2019) have found that Eumay be produced in collapsars, however, more detailedinvestigations are necessary to understand whether theconditions are appropriate for an r-process, see Milleret al. (2019).In Fig. 9, we show the maximum europium massejected by NSM (black line, 8 · − M (cid:12) ) and by MR-SNe(red line, 1 . · − M (cid:12) ). Moreover, we include the esti-mated europium mass necessary to explain r-II stars inReticulum II (Ji et al. 2016a). All of these estimates areclose to the mass that is necessary to enhance the Eu-stars, as derived from the observed stellar abundances.In addition to Eu, one could use α − elements, e.g.,Mg, to check whether MR-SNe are the r-process site re-sponsible for more element abundances of the Eu-starsin Fornax. If an abnormal supernova produces r-processand huge amount of Mg, one would possibly see a signa-ture of this in the Mg abundances of those stars. How-ever, in Fig. 2, the three stars present normal Mg abun-dances (see also Table 5). This may be an indication ofa neutron star merger producing the observed Eu or ofMR-SNe ejecting normal amount of Mg as any other su-pernova in Fornax. Therefore, we cannot use Mg as anindicator for rare supernovae and their r-process yields.In summary, based on the abundance ratio of [Ba/Eu],on the estimated amount of Eu, and on the Mg abun-dance, we cannot determine the r-process site enrichingthe Eu-stars in Fnx. Based on Fig. 9, NSMs seem apromising site to explain the r-material in the Eu-stars,however, their time delay coupled with recent star for-mation in Fnx may pose an issue that is easier to over-come if MR-SNe would be the r-site.4.4. Time scale - late r-process event linked to a starformation burst in Fornax
The strong r-process enhancement, despite the re-maining unknown formation site, combined with thetypical low α − content indicates, together with the Gaia proper motions, that these rare Eu-stars are indeed For-nax members (see Sect. 2.2).From the time scale perspective, neutron star mergersremain viable sources and could explain the observedabundances. Neutron star mergers are expected to con-tribute with some delay and thus such an event couldaccount for producing r-process material late or at highmetallicities. However, this would imply that there issome star formation just after the neutron stars mergeand eject r-process material. Similarly linked to the late xtreme r-process enhanced stars at high metallicity in Fornax . ± .
86 Gyr andFnx-mem0595 an age of 5 . ± .
78 Gyr. Fornax shows acomplex star formation history with several outbreaks.A significant number of stars were formed at early times,i.e., more than 10 Gyr ago. Moreover, there was a veryrecent period of star formation 3 − CONCLUSIONSWe study neutron-capture elements in Fornax starsincluding three stars at high metallicities with extremeenhancements of heavy r-process elements. We definethese new type of stars as Eu-stars, they are r-II stars([Eu/Fe] >
1) at high metallicity ([Fe/H] (cid:38) − . (cid:15) (Eu) =0 . ± .
12, which is the highest europium abundanceever observed. Thus, Eu is approximately three timesmore abundant than in our Sun while their iron abun-dance is a factor of seven smaller. In addition to an en-richment in heavy neutron-capture elements, Zr is alsoenhanced in the three stars. In order to compare toother Fornax stars, we have derived Zr abundances for105 stars. This is the largest Zr sample in a dSph todate and offers a new chance to explore the galacticchemical evolution of Fornax. Moreover, we have de-rived lutetium abundances for the first time for stars ina dSph galaxy.We have demonstrated that the enhancements inneutron-capture elements is due to the r-process as in- dicated by the [Ba/Eu] ratio and typical r-process pat-tern. Moreover, we give an estimate of the amount ofEu necessary to explain these r-process stars, namely M (Eu) ∼ · − − · − M (cid:12) . This agrees with anindividual r-process event being enough to explain theobserved Eu abundances. Based on the elemental ratiosand the europium mass ejected, we try to identify the r-process site by comparing to theoretical yield predictionsfrom neutron star mergers and magneto-rotational su-pernovae. However, the uncertainties in the astrophysi-cal conditions and the nuclear physics input prevent usof making any firm conclusion about the site. Thereis a clear need of improved hydrodynamic simulationswith detailed microphysics as well as more theoreticaland experimental information of the extreme neutron-rich nuclei involved in the r-process.The r-process event responsible for the Eu-stars wasoccurring during a star formation episode. If the eventwas a NSM, this could come from neutron stars bornin early supernovae. The delay of the merger was co-inciding with the star formation event where the Eu-stars were born shortly after. Despite NSM yield a verypromising range of Eu abundances, the time scale is abit more tricky in this scenario. Another possibilityis that during the star formation event, massive starsformed and at least one died fast as a MR-SN ejectingthe r-process material necessary to explain the observedabundances. Therefore, an active star forming environ-ment simultaneously with the r-process event is in anycase required for the formation of stars with such anenhanced europium content. The age of the stars ap-proximately traces the time when the r-process eventoccurred. Their age is estimated to be around 4 Gyr(Lemasle et al. 2014), which coincides with a sudden in-crease of star formation in Fornax (Coleman & de Jong2008; de Boer et al. 2012). We conclude that the ex-istence of these Eu-stars proves that the r-process canefficiently form r-II stars across a broad range of dwarfgalaxies - from the faintest low-mass ones to the mostmassive dSph galaxies. It is not unique to ultra faintdwarf galaxies as suggested before. Moreover, we em-phasize that gas dilution and star formation time scalesmust be considered in the search for the r-process sites.Future observations are critical to find more Eu-starsthat are key to understand the origin of heavy elementsproduced by the r-process.ACKNOWLEDGMENTSThe authors thank M. Eichler, M. Hanke, A. Koch,and ´A. Sk´ulad´ottir for valuable discussions.2 Reichert et al.
MR and AA were supported by the ERC Start-ing Grant EUROPIUM-677912, Deutsche Forschungs-gemeinschaft through SFB 1245, and Helmholtz Forschungsakademie Hessen f¨ur FAIR. CJH acknowl-edges support from the Max Planck Society.This work has benefited from the COST Action“ChETEC” (CA16117) supported by COST (EuropeanCooperation in Science and Technology).APPENDIX A. ZIRCONIUM ABUNDANCESWe list all derived zirconium abundances together with metallicities from Reichert et al. (2020) in Table 6.REFERENCES
Abbott, B. P., Abbott, R., Abbott, T. D., et al. 2017, Phys.Rev. Lett., 119, 161101,doi: 10.1103/PhysRevLett.119.161101Abohalima, A., & Frebel, A. 2018, ApJS, 238, 36,doi: 10.3847/1538-4365/aadfe9Andrievsky, S. M., Korotin, S. A., Hill, V., & Zhukova,A. V. 2017, Odessa Astronomical Publications, 30, 54,doi: 10.18524/1810-4215.2017.30.117154Aoki, W., Honda, S., Sadakane, K., & Arimoto, N. 2007,PASJ, 59, L15, doi: 10.1093/pasj/59.3.L15Arlandini, C., K¨appeler, F., Wisshak, K., et al. 1999, ApJ,525, 886, doi: 10.1086/307938Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009,ARA&A, 47, 481,doi: 10.1146/annurev.astro.46.060407.145222Battaglia, G., Tolstoy, E., Helmi, A., et al. 2006, A&A, 459,423, doi: 10.1051/0004-6361:20065720Beers, T. C., & Christlieb, N. 2005, ARA&A, 43, 531,doi: 10.1146/annurev.astro.42.053102.134057Biemont, E., Grevesse, N., Hannaford, P., & Lowe, R. M.1981, ApJ, 248, 867, doi: 10.1086/159213Bisterzo, S., Travaglio, C., Gallino, R., Wiescher, M., &K¨appeler, F. 2014, ApJ, 787, 10,doi: 10.1088/0004-637X/787/1/10Boeche, C., & Grebel, E. K. 2016, A&A, 587, A2,doi: 10.1051/0004-6361/201526758Bouchard, A., Carignan, C., & Staveley-Smith, L. 2006, AJ,131, 2913, doi: 10.1086/503629Bovard, L., Martin, D., Guercilena, F., et al. 2017, PhRvD,96, 124005, doi: 10.1103/PhysRevD.96.124005Cayrel, R., Depagne, E., Spite, M., et al. 2004, A&A, 416,1117, doi: 10.1051/0004-6361:20034074Coleman, M., Da Costa, G. S., Bland-Hawthorn, J., et al.2004, AJ, 127, 832, doi: 10.1086/381298Coleman, M. G., & de Jong, J. T. A. 2008, ApJ, 685, 933,doi: 10.1086/589992 Corliss, C. H., & Bozman, W. R. 1962, Experimentaltransition probabilities for spectral lines of seventyelements; derived from the NBS Tables of spectral-lineintensitiesCˆot´e, B., Fryer, C. L., Belczynski, K., et al. 2018, ApJ, 855,99, doi: 10.3847/1538-4357/aaad67Cˆot´e, B., Eichler, M., Arcones, A., et al. 2019, ApJ, 875,106, doi: 10.3847/1538-4357/ab10dbCˆot´e, B., Eichler, M., Yag¨ue, A., et al. 2020, arXiv e-prints,arXiv:2006.04833. https://arxiv.org/abs/2006.04833Cristallo, S., Abia, C., Straniero, O., & Piersanti, L. 2015,ApJ, 801, 53, doi: 10.1088/0004-637X/801/1/53Cristallo, S., Piersanti, L., Straniero, O., et al. 2011, ApJS,197, 17, doi: 10.1088/0067-0049/197/2/17de Boer, T. J. L., Tolstoy, E., Hill, V., et al. 2012, A&A,539, A103, doi: 10.1051/0004-6361/201118378den Hartog, E. A., Curry, J. J., Wickliffe, M. E., & Lawler,J. E. 1998, SoPh, 178, 239, doi: 10.1023/A:1005088315480Den Hartog, E. A., Lawler, J. E., Sneden, C., & J., C. J.2003, The Astrophysical Journal Supplement Series, 148,543, doi: 10.1086/376940Denissenkov, P. A., Herwig, F., Woodward, P., et al. 2019,MNRAS, 488, 4258, doi: 10.1093/mnras/stz1921Edvardsson, B., Andersen, J., Gustafsson, B., et al. 1993,A&AS, 102, 603Eichler, M., Sayar, W., Arcones, A., & Rauscher, T. 2019,ApJ, 879, 47, doi: 10.3847/1538-4357/ab24cfEzzeddine, R., Rasmussen, K., Frebel, A., et al. 2020, ApJ,898, 150, doi: 10.3847/1538-4357/ab9d1aFulbright, J. P. 2000, AJ, 120, 1841, doi: 10.1086/301548Gaia Collaboration, Brown, A. G. A., Vallenari, A., et al.2018, A&A, 616, A1, doi: 10.1051/0004-6361/201833051Geisler, D., Smith, V. V., Wallerstein, G., Gonzalez, G., &Charbonnel, C. 2005, AJ, 129, 1428, doi: 10.1086/427540Goriely, S., Bauswein, A., & Janka, H.-T. 2011, ApJL, 738,L32, doi: 10.1088/2041-8205/738/2/L32 xtreme r-process enhanced stars at high metallicity in Fornax Table 6.
Metallicities and zirconium abundances of 105 stars in Fornax.ID [Fe/H] log (cid:15) (Zr) ID [Fe/H] log (cid:15) (Zr)[LHT2010] BL147 − . ± .
05 1 . ± .
22 [WMO2009] For-0956 − . ± .
08 1 . ± . − . ± .
05 1 . ± .
11 2MASS J02384113-3444205 − . ± .
05 1 . ± . − . ± .
08 2 . ± .
27 2MASS J02401677-3429346 − . ± .
07 1 . ± . − . ± .
05 1 . ± .
11 [MOW91] 8 − . ± .
10 1 . ± . − . ± .
11 1 . ± .
19 [LDH2014] Fnx-mem0715 − . ± .
13 1 . ± . − . ± .
10 1 . ± .
10 [WMO2009] For-0391 − . ± .
07 1 . ± . − . ± .
09 1 . ± .
13 2MASS J02383503-3441380 − . ± .
06 1 . ± . − . ± .
05 1 . ± .
14 2MASS J02390157-3436488 − . ± .
08 1 . ± . − . ± .
06 1 . ± .
20 2MASS J02395144-3421211 − . ± .
09 1 . ± . − . ± .
09 1 . ± .
16 [LDH2014] Fnx-mem0574 − . ± .
05 1 . ± . − . ± .
10 1 . ± .
14 2MASS J02391102-3428348 − . ± .
10 1 . ± . − . ± .
06 1 . ± .
14 2MASS J02390031-3430302 − . ± .
05 1 . ± . − . ± .
13 2 . ± .
15 [LDH2014] Fnx-mem0638 − . ± .
07 1 . ± . − . ± .
11 1 . ± .
19 [LDH2014] Fnx-mem0631 − . ± .
08 1 . ± . − . ± .
06 1 . ± .
14 2MASS J02401790-3427010 − . ± .
07 1 . ± . − . ± .
08 1 . ± .
13 [WMO2009] For-0365 − . ± .
06 1 . ± . − . ± .
05 1 . ± .
20 2MASS J02401752-3426065 − . ± .
09 1 . ± . − . ± .
08 1 . ± .
15 [KGS2010] For 64059 − . ± .
10 1 . ± . − . ± .
07 1 . ± .
14 [WMO2009] For-0387 − . ± .
11 1 . ± . − . ± .
06 1 . ± .
14 2MASS J02385365-3433048 − . ± .
08 1 . ± . − . ± .
04 1 . ± .
17 [WMO2009] For-0910 − . ± .
05 2 . ± . − . ± .
07 1 . ± .
11 2MASS J02391606-3430135 − . ± .
08 2 . ± . − . ± .
11 1 . ± .
14 2MASS J02390853-3430556 − . ± .
07 1 . ± . − . ± .
07 1 . ± .
19 2MASS J02395427-3435114 − . ± .
11 1 . ± . − . ± .
06 1 . ± .
22 2MASS J02391398-3428364 − . ± .
11 1 . ± . − . ± .
05 1 . ± .
16 [LDH2014] Fnx-mem0682 − . ± .
06 1 . ± . − . ± .
03 1 . ± .
12 [LDH2014] Fnx-mem0717 − . ± .
08 1 . ± . − . ± .
11 1 . ± .
17 2MASS J02391437-3434427 − . ± .
09 2 . ± . − . ± .
12 1 . ± .
20 [LDH2014] Fnx-mem0678 − . ± .
07 1 . ± . − . ± .
07 1 . ± .
12 2MASS J02391783-3430570 − . ± .
10 1 . ± . − . ± .
08 1 . ± .
19 [WMO2009] For-1581 − . ± .
08 1 . ± . − . ± .
05 1 . ± .
17 [LHT2010] BL233 − . ± .
12 1 . ± . − . ± .
09 2 . ± .
17 [WMO2009] For-2026 − . ± .
06 1 . ± . − . ± .
10 1 . ± .
19 2MASS J02393808-3437062 − . ± .
08 1 . ± . − . ± .
06 1 . ± .
11 [WMO2009] For-1120 − . ± .
08 2 . ± . − . ± .
09 1 . ± .
16 2MASS J02394195-3430361 − . ± .
06 2 . ± . − . ± .
04 1 . ± .
21 2MASS J02384018-3439121 − . ± .
07 1 . ± . − . ± .
04 1 . ± .
12 2MASS J02390819-3436537 − . ± .
08 1 . ± . − . ± .
11 1 . ± .
25 2MASS J02392769-3437487 − . ± .
08 2 . ± . − . ± .
10 1 . ± .
13 2MASS J02395604-3424106 − . ± .
06 2 . ± . − . ± .
04 1 . ± .
19 2MASS J02392483-3434383 − . ± .
05 2 . ± . − . ± .
16 2 . ± .
21 WEL 60 − . ± .
09 1 . ± . − . ± .
03 1 . ± .
11 2MASS J02394309-3440186 − . ± .
08 1 . ± . − . ± .
09 1 . ± .
13 2MASS J02393412-3433096 − . ± .
08 2 . ± . − . ± .
07 1 . ± .
24 [LDH2014] Fnx-mem0522 − . ± .
06 1 . ± . − . ± .
09 1 . ± .
19 [LHT2010] BL298 − . ± .
06 1 . ± . − . ± .
04 1 . ± .
20 2MASS J02395732-3431211 − . ± .
07 2 . ± . − . ± .
12 1 . ± .
19 [WMO2009] For-0949 − . ± .
08 1 . ± . − . ± .
08 1 . ± .
17 [WMO2006] F01-16 − . ± .
06 2 . ± . − . ± .
06 1 . ± .
17 [LHT2010] BL163 − . ± .
07 1 . ± . − . ± .
06 1 . ± .
19 2MASS J02393084-3435451 − . ± .
06 2 . ± . − . ± .
08 1 . ± .
13 [MOW91] 21 − . ± .
11 2 . ± . − . ± .
05 1 . ± . Reichert et al.
Gratton, R. G., & Sneden, C. 1988, A&A, 204, 193Grebel, E. K. 1997, Reviews in Modern Astronomy, 10, 29Hampel, M., Stancliffe, R. J., Lugaro, M., & Meyer, B. S.2016, ApJ, 831, 171, doi: 10.3847/0004-637X/831/2/171Hannaford, P., Lowe, R. M., Grevesse, N., Biemont, E., &Whaling, W. 1982, ApJ, 261, 736, doi: 10.1086/160384Hansen, C. J., El-Souri, M., Monaco, L., et al. 2018, ApJ,855, 83, doi: 10.3847/1538-4357/aa978fHansen, C. J., Primas, F., Hartman, H., et al. 2012, A&A,545, A31, doi: 10.1051/0004-6361/201118643Hendricks, B., Koch, A., Walker, M., et al. 2014, A&A, 572,A82, doi: 10.1051/0004-6361/201424645Ishigaki, M. N., Aoki, W., & Chiba, M. 2013, ApJ, 771, 67,doi: 10.1088/0004-637X/771/1/67Ivans, I. I., Sneden, C., James, C. R., et al. 2003, ApJ, 592,906, doi: 10.1086/375812Ji, A. P., Frebel, A., Chiti, A., & Simon, J. D. 2016a,Nature, 531, 610, doi: 10.1038/nature17425Ji, A. P., Frebel, A., Simon, J. D., & Chiti, A. 2016b, ApJ,830, 93, doi: 10.3847/0004-637X/830/2/93Johnson, C. I., McWilliam, A., & Rich, R. M. 2013, ApJL,775, L27, doi: 10.1088/2041-8205/775/1/L27Kathirgamaraju, A., Giannios, D., & Beniamini, P. 2019,MNRAS, 487, 3914, doi: 10.1093/mnras/stz1564Kirby, E. N., Guhathakurta, P., Simon, J. D., et al. 2010,ApJS, 191, 352, doi: 10.1088/0067-0049/191/2/352Kobayashi, C., Karakas, A. I., & Lugaro, M. 2020, ApJ,900, 179, doi: 10.3847/1538-4357/abae65Koch, A., Reichert, M., Hansen, C. J., et al. 2019, A&A,622, A159, doi: 10.1051/0004-6361/201834241Korobkin, O., Rosswog, S., Arcones, A., & Winteler, C.2012, MNRAS, 426, 1940,doi: 10.1111/j.1365-2966.2012.21859.xKurucz, R. L. 1970, SAO Special Report, 309—. 2011, Canadian Journal of Physics, 89, 417,doi: 10.1139/p10-104Lawler, J. E., Bonvallet, G., & Sneden, C. 2001, ApJ, 556,452, doi: 10.1086/321549Lawler, J. E., & Dakin, J. T. 1989, J. Opt. Soc. Am. B, 6,1457, doi: 10.1364/JOSAB.6.001457Lawler, J. E., Sneden, C., & Cowan, J. J. 2015, ApJS, 220,13, doi: 10.1088/0067-0049/220/1/13Lawler, J. E., Sneden, C., Cowan, J. J., Ivans, I. I., & DenHartog, E. A. 2009, ApJS, 182, 51,doi: 10.1088/0067-0049/182/1/51Lemasle, B., de Boer, T. J. L., Hill, V., et al. 2014, A&A,572, A88, doi: 10.1051/0004-6361/201423919Letarte, B., Hill, V., Tolstoy, E., et al. 2010, A&A, 523,A17, doi: 10.1051/0004-6361/200913413 —. 2018, A&A, 613, C1,doi: 10.1051/0004-6361/200913413e(cid:32)Lokas, E. L. 2009, MNRAS, 394, L102,doi: 10.1111/j.1745-3933.2009.00620.xMackey, A. D., & Gilmore, G. F. 2003, MNRAS, 340, 175,doi: 10.1046/j.1365-8711.2003.06275.xMagg, M., Nordlander, T., Glover, S. C. O., et al. 2020,MNRAS, 498, 3703, doi: 10.1093/mnras/staa2624Matteucci, F., Romano, D., Arcones, A., Korobkin, O., &Rosswog, S. 2014, MNRAS, 438, 2177,doi: 10.1093/mnras/stt2350McConnachie, A. W., & Venn, K. A. 2020, AJ, 160, 124,doi: 10.3847/1538-3881/aba4abMcWilliam, A., Preston, G. W., Sneden, C., & Shectman,S. 1995, AJ, 109, 2736, doi: 10.1086/117485Meggers, W. F., Corliss, C. H., & Scribner, B. F. 1975,Tables of spectral-line intensities. Part I, II - arranged byelements.Miller, J. M., Sprouse, T. M., Fryer, C. L., et al. 2019,arXiv e-prints, arXiv:1912.03378.https://arxiv.org/abs/1912.03378M¨osta, P., Roberts, L. F., Halevi, G., et al. 2018, ApJ, 864,171, doi: 10.3847/1538-4357/aad6ecNishimura, N., Sawai, H., Takiwaki, T., Yamada, S., &Thielemann, F. K. 2017, ApJ, 836, L21,doi: 10.3847/2041-8213/aa5deeNishimura, N., Takiwaki, T., & Thielemann, F.-K. 2015,ApJ, 810, 109, doi: 10.1088/0004-637X/810/2/109Nissen, P. E., Hoeg, E., & Schuster, W. J. 1997, in ESASpecial Publication, Vol. 402, Hipparcos - Venice ’97, ed.R. M. Bonnet, E. Høg, P. L. Bernacca, L. Emiliani,A. Blaauw, C. Turon, J. Kovalevsky, L. Lindegren,H. Hassan, M. Bouffard, B. Strim, D. Heger, M. A. C.Perryman, & L. Woltjer, 225–230Nomoto, K., Tominaga, N., Umeda, H., Kobayashi, C., &Maeda, K. 2006, NuPhA, 777, 424,doi: 10.1016/j.nuclphysa.2006.05.008Pasquini, L., Avila, G., Blecha, A., et al. 2002, TheMessenger, 110, 1Prochaska, J. X., Naumov, S. O., Carney, B. W.,McWilliam, A., & Wolfe, A. M. 2000, AJ, 120, 2513,doi: 10.1086/316818Reddy, B. E., Lambert, D. L., & Allende Prieto, C. 2006,MNRAS, 367, 1329,doi: 10.1111/j.1365-2966.2006.10148.xReddy, B. E., Tomkin, J., Lambert, D. L., & AllendePrieto, C. 2003, MNRAS, 340, 304,doi: 10.1046/j.1365-8711.2003.06305.xReichert, M., Hansen, C. J., Hanke, M., et al. 2020, A&A,641, A127, doi: 10.1051/0004-6361/201936930 xtreme r-process enhanced stars at high metallicity in Fornax15