EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 604 (August 2017) A&A, 604 (2017) L4 Full HTML
Free Access
Issue
A&A
Volume 604, August 2017
Article Number L4
Number of page(s) 5
Section Letters
DOI https://doi.org/10.1051/0004-6361/201731430
Published online 31 July 2017

© ESO, 2017

1. Introduction

The detection of a possible chemical signature of planet formation in the photospheres of planet host stars is a challenge for current studies. Several authors have attempted to detect this signature by studying the condensation temperature (Tc) trend in planet host stars (e.g., Gonzalez 1997; Smith et al. 2001; Gratton et al. 2001). In particular, Meléndez et al. (2009) showed that the atmosphere of the Sun is deficient in refractory1 elements when compared to the average abundances of 11 solar twins, also showing a clear Tc trend. They proposed that the missing refractories are probably locked in terrestrial planets and rocky material that orbits the Sun. This idea was then supported by Ramírez et al. (2009), who studied a sample of 64 solar twins and analogs with and without planets. On the other hand, other authors suggest that the Tc trends depend on the galactic chemical evolution (GCE), the stellar age, or probably the galactic birth place of the stars (e.g., Adibekyan et al. 2014, 2016), which competes with the proposed chemical signature of planet formation (e.g. González Hernández et al. 2013).

Most multiple and binary stars are believed to have formed coevally from a common molecular cloud. Wide binaries are particularly valuable because both components can be presumed to have the same age and initial composition, greatly diminishing the mentioned age and GCE effects. Several studies have shown that there may be small differences in the chemical composition of their components, possibly due to the planet formation process (Gratton et al. 2001; Desidera et al. 2004, 2006). The similarity between the two components of a binary system enabled achieving a higher precision in a differential study (e.g., Saffe et al. 2015, 2016). To date, more than 2700 planetary systems have been reported2, but only a handful of these systems are binaries with similar components. Three of these remarkable systems were studied in detail: 16 Cyg, HAT-P-1, and HD 80606 (Laws & Gonzalez 2001; Ramírez et al. 2011; Schuler et al. 2011; Tucci Maia et al. 2014; Liu et al. 2014; Saffe et al. 2015). There are also binary systems in which circumstellar planets orbit both stars of the system (e.g., Mack et al. 2014). These studies show that a Tc trend is probably present between the stars of 16 Cyg, but not in the cases of HAT-P-1 or HD 80606. Notably, the binary system ζ2 Ret (where one component is orbited by a dust disk with no planet detected) also shows a Tc trend between their stars, supporting the chemical signature of planet formation (Saffe et al. 2016). This clearly shows that additional systems need to be studied through detailed abundance analyses in order to reach more significant conclusions.

As a result of the planetary transit survey HATNet, Kovács et al. (2007) discovered a giant planet of 0.68 MJup orbiting the star HAT-P-4 (=BD +36 2593) at a distance of 0.04 AU. They estimated a density of 0.41 g cm-3, which is one of the lowest-density hot Jupiters known. Then, Mugrauer et al. (2014) showed that HAT-P-4 forms a wide binary system separated by 91.8 arcsec from its companion (TYC 2569-744-1, hereafter component B), and they also showed that both stars present very similar spectra (G0V + G2V). The estimated separation is 28 446 AU, which was considered by Mugrauer et al. (2014) as a true binary system based on their common proper motion and similar radial velocity. To date, there is no planet detected around the B star, which is also included in the field G191 (FOV = 8 × 8 deg) of the HATNet survey (Kovács et al. 2007). As we show in the next sections, the stellar parameters of both stars are very similar, making this system an ideal case to study the possible chemical signature of planet formation.

Different authors have shown a possible excess of Li  depletion in stars with planets compared to stars without planets (King et al. 1997; Israelian et al. 2004, 2009; Delgado-Mena et al. 2014; Gonzalez 2014, 2015). The Li  depletion was first attributed to the presence of planets, which might possibly increase the angular momentum of the star (e.g., during planet migration) and increase its convective mixing (Israelian et al. 2009). On the other hand, some works showed that the Li  depletion is related to a bias in age, mass, and metallicity and is not due to the presence of planets or the planet formation process (Luck & Heiter 2006; Baumann et al. 2010; Ramírez et al. 2012; Carlos et al. 2016). The study of similar stars in binary systems can help understand the origin of Li  depletion. In the 16 Cyg binary system, the B component hosts a planet of 1.5 MJup (Cochran et al. 1997) and presents a Li  content that is aboutfour times lower than that of the A star (King et al. 1997), which supports the model of Li  depletion by the presence of planets. However, Ramírez et al. (2011) showed that the slightly different masses of stars A and B could explain their different Li  contents, that is, 16 Cyg can in principle support both scenarios. Notably, HAT-P-4 is in some way complementary to 16 Cyg: in this case, the planet host star presents the higher Li  content, which is contrary to the model of Li  depletion due to the presence of planets. Then, it is worthwhile exploring both the Tc trends and the Li  depletion in this remarkable system.

2. Observations and data reduction

Observations of HAT-P-4 binary system were acquired through the Gemini Remote Access to CFHT ESPaDOnS Spectrograph (GRACES). This device takes advantage of the high-resolution ESPaDOnS3 spectrograph, located at the Canada-France-Hawaii Telescope (CFHT) and fed by an optical fiber connected to the 8.1 m Gemini North telescope at Maunakea, Hawaii. We used the 1-fiber object-only observing mode, which provides an average resolving power of ~67 500 between 4500 and 8500 Å 4. The stellar spectra were obtained under a fast-turnaround (FT) observing mode requested to the Gemini Observatory (program ID: GN-2016A-FT-25, PI: C. Saffe). The observations were taken on June 3, 2016, with the B star observed immediately after the A star, using the same spectrograph configuration for both stars. The exposure times were 2 × 17 min on each target, obtaining a final signal-to-noise ratio (S/N) of ~400 measured at ~6000 Å in the combined spectra. The final spectral coverage is 4050−10 000 Å. The Moon was also observed with the same spectrograph setup, achieving a similar S/N to acquire the solar spectrum as initial reference. GRACES spectra were reduced using the code OPERA5 (Martioli et al. 2012). More recent documentation on OPERA can be found at the ESPECTRO project webpage6.

3. Stellar parameters and abundance analysis

Fundamental parameters (Teff, log g, [Fe/H], and vturb) were derived following a similar procedure as in our previous work (Saffe et al. 2015, 2016). We measured the equivalent widths (EW) of the Fe i and Fe ii lines using the IRAF task splot, and then continued with other chemical species. Both the lines list and relevant laboratory data were taken from the literature (Liu et al. 2014; Meléndez et al. 2014; Bedell et al. 2014). We imposed an excitation and ionization balance of Fe  lines using the differential version of the FUNDPAR program (Saffe 2011; Saffe et al. 2015). This code made use of ATLAS9 model atmospheres (Kurucz 1993) together with the 2014 version of the MOOG program (Sneden 1973). Stellar parameters were determined using the Sun as reference, that is, (A–Sun) and (B–Sun), by adopting (5777 K, 4.44 dex, 0.00 dex, 0.91 km s-1) for the Sun7. Then, we recalculated the parameters using the A star as reference, that is, (B–A), by adopting for the A star the parameters derived for the case (A–Sun) (see Table 1). The errors in the stellar parameters were derived following the procedure detailed in Saffe et al. (2015), which takes into account the individual and the mutual covariance terms of the error propagation. Within the errors, the parameters of the B star agree using the Sun or the A star as reference. We note that the A star is ~0.1 dex more metal rich than the B star. Figure 1 shows abundance vs. excitation potential (top panel) and abundance vs. EWr (bottom panel), both for the case (B–A).

Table 1

Stellar parameters derived for each star.

thumbnail Fig. 1

Differential abundance vs. excitation potential (upper panel) and vs. reduced EW (lower panel) for the case (B–A). Filled and hollow circles correspond to Fe i and Fe ii.

The next step was the derivation of abundances for all remaining chemical elements. The hyperfine structure splitting (HFS) was considered for V i, Mn i, Co i, Cu i, and Ba ii using the HFS constants of Kurucz & Bell (1995) and performing spectral synthesis for these species. We also derived the Li i abundance using spectral synthesis with the resonance line 6707.80 Å, which includes the doublet 6707.76 Å, 6707.91 Å and HFS components. We corrected the Na i abundance for non-local thermodynamic equilibrium (NLTE) effects by interpolating in the data of Shi et al. (2004) and adopting Na(NLTE) – Na(LTE) ~−0.07 dex for each star. We also applied NLTE corrections to O i (−0.18 dex and −0.17 dex for the A and B stars) by interpolating in the data of Ramírez et al. (2007). These corrections are relative to the Sun, which implies that NLTE effects for the case (B–A) are not significant given the high similarity between the stars A and B.

4. Results and discussion

Condensation temperatures were taken from the 50% Tc values derived by Lodders (2003) for a solar system gas with [Fe/H] = 0. As suggested by the referee, it would be helpful to calculate other Tc sequences for different metallicity values. We corrected for GCE effects for the case (star – Sun), but not for the case (B–A) by adopting the GCE fitting trends of González Hernández et al. (2013), that is, we followed the same procedure as Saffe et al. (2015). Figure 2 presents the corrected abundance values vs. Tc for (A–Sun)8. Table 2 presents the slopes and uncertainties of the linear fits. The positive slopes for the case (star – Sun) indicate a higher content of refractories (those with Tc> 900 K) relative to volatiles (Tc< 900 K).

Table 2

Derived slopes (abundance vs. Tc) and their uncertainties.

thumbnail Fig. 2

Differential abundances vs. Tc for (A–Sun). The long dashed line is a weighted linear fit to the differential abundance values, while the continuous line shows the solar-twins trend of Meléndez et al. (2009).

The differential abundances for (B–A) are presented in Fig. 3. The continuous line in this figure presents the solar-twins trend of Meléndez et al. (2009; vertically shifted). Long dashed lines are weighted linear fits to all elements and to refractory elements, showing similar negative slopes (see also Table 2). Some species such as Sr i and Ce ii seem to possibly drive the trends, but we obtained very similar slopes after excluding these species. The average refractory and volatile [X/H] abundance values for (B–A) are −0.105± 0.007 dex and –0.065 ± 0.015 dex, which together with the negative slopes of Fig. 3 points toward a higher content of refractories in the A star than in its stellar companion.

thumbnail Fig. 3

Differential abundances (B–A) vs. Tc. Long dashed lines are weighted linear fits to all species and to refractory species. The solar-twins trend of Meléndez et al. (2009) is shown with a continuous line.

4.1. Do the HAT-P-4 stars have a peculiar composition?

We detected a difference of ~0.1 dex in metallicity between the stars A and B, which is one of the highest differences found in similar systems (e.g., Desidera et al. 2004, 2006). In this section we discuss the possible chemical pattern of the two stars. λ Boötis stars show moderate surface underabundances of most Fe-peak elements, but solar abundances of C, N, O, and S (e.g., Paunzen 2004), in contrast to the metal-rich content of stars A and B.

δ Scuti stars are pulsating variables with ~A6-F6 spectral types and masses of between 1.4−3.0 M⊙ (see, e.g., the catalogs of Rodríguez et al. 2000; Liakos & Niarchos 2017). We determined the stellar masses through a Bayesian method using the PARSEC9 isochrones (Bressan et al. 2012), obtaining 1.24 ± 0.06 M⊙ and 1.17 ± 0.05 M⊙ for stars A and B. The temperature and mass of A and B are lower than those of all δ Scuti stars of these catalogs. One of the coolest δ Scuti stars with an abundance determination is CP Boo (6320 K, Galeev et al. 2012), which is ~300 K hotter than the A star. However, the A star is enhanced (≳0.20 dex) in Na i, Mg i, Al i, Sc ii, V i, and strongly enhanced (>0.40 dex) in Sr ii and Y ii compared to CP Boo. Moreover, stars A and B lie beyond the instability strip boundaries (e.g., Figs. 6 and 7 of Liakos & Niarchos 2017), and no stellar pulsations have been reported.

Blue stragglers (BS) are stars that are significantly bluer than the main-sequence turnoff of the cluster (or population) to which they belong. There are binaries where one component is a BS, and it is impossible to fit both components with a single isochrone (Desidera et al. 2007). However, in our system the ages of the two stars agree within the errors (2.7 ± 1.3 Gyr and 2.9 ± 1.8 Gyr), also when they are derived with the PARSEC isochrones (Bressan et al. 2012). BS stars present significant rotational velocities, intense activity, and very low Li  content (e.g., Fuhrmann & Bernkopf 1999; Schirbel et al. 2015; Ryan et al. 2001). None of these characteristics are seen in star A or star B.

There are no firm reasons to identify any component of the binary system as peculiar. A possible different accretion would therefore be the most plausible explanation for the chemical differences observed. We estimated the rocky mass of depleted material in the A star following Chambers (2010). Adopting a convection zone similar to the current Sun (Mcz = 0.023 M⊙), we obtain Mrock ~ 10 M⊕. However, adopting a higher Mcz value (e.g., 0.050 M⊙) at the time of planet formation, we derive Mrock ~ 20 M⊕. Cody & Sasselov (2005) found that for stars with 1.1 M⊙, a polluted convection zone is slightly smaller (~1%) than its unpolluted counterpart, but the trend is reversed for stars with M ≤ M⊙. More recently, Van Saders & Pinsonneault (2012) found that the acoustic depth to the base of the CZ varies at the 0.5−1% per 0.1 dex level in [Z/X]. A change of 1% in Mcz translates into ~0.5 M⊕ of derived accreted material. Then, the estimate of at least 10 M⊕ of accreted material should be considered as a first-order approximation.

4.2. Hot-Jupiter planet and the Li content

The formation of hot-Jupiter planets is mainly explained by the model of core accretion (Pollack et al. 1996) and by gravitational instability (Boss 2000). In both cases, some type of migration from the original location is required to explain the current orbit at ~0.04 AU, with the possible accretion of inner planets into the star (e.g., Mustill et al. 2015). Another possibility is the in situ formation of the hot Jupiter (e.g., Bodenheimer et al. 2000), without migration from greater distances. However, the abundances of the C/O and O/H ratios suggest that some hot Jupiters originate beyond the snow line (Brewer et al. 2017). With the current data, it is difficult to distinguish between the different formation scenarios. Additional planets were searched for in the A star by transits (Smith et al. 2009; Ballard et al. 2011) and radial velocity (Knutson et al. 2014), without success. A scenario with a possible migration and accretion can therefore not be discarded.

Figure 4 shows that the line Li i 6707.8 Å is stronger in the A star (blue dotted line) than in the B star (black continuous line), with abundances of 1.47 ± 0.05 and 1.17 ± 0.04 dex. In this binary system, the star that hosts the planet also presents the higher Li  content, in contrast to the model that explains the Li  depletion by the presence of planets. A similar rotational velocity (7.0 ± 0.9 and 6.1 ± 1.0 km s-1 estimated from our spectra) suggests that the rotational mixing is not the cause of the different Li  content. The sligthly different masses of stars A and B suggest a difference of 0.2−0.3 dex in the Li  abundances from fitting general literature trends (e.g., Fig. 3 of Delgado-Mena et al. 2015; and Fig. 9 of Ramírez et al. 2012). This can explain (at least in part) the higher Li  content, in agreement with the scenario proposed by Ramírez et al. (2011) for 16 Cyg. Another possible scenario is the accretion of material in the A star, which could increase its Li  abundance and agree with its higher metal content.

thumbnail Fig. 4

Stellar spectra near the Li line 6707.8 Å for the A (blue dotted line) and B (black continuous line) stars.

4.3. Relative Teff difference

In this section we discuss the relative Teff difference between stars A and B. Balmer lines could be used as Teff proxies for solar-type stars (e.g., Barklem et al. 2002). Figure 5 presents the region near the Hα line for the stars A (blue dotted line) and B (black continuous line) almost superimposed, together with a synthetic spectrum for comparison (red dashed line) calculated with Teff = Teff(A)−150 K = 5886 K. This suggest that stars A and B present a very similar Teff. We also used the photometric calibration of Casagrande et al. (2010) with the colors (B−V)T and (J−Ks) from the Tycho and 2MASS catalogs. We estimate a difference of +95 K and −5 K for (B–A). However, we note that the errors reported for VT and BT vary between 0.07−0.13 mag, while the errors in JHK are notably lower at 0.01−0.03 mag.

We also explored how the relative Teff difference could change the abundance difference. For the case (B–A) we varied Teff by −50, −100, and −150 K for the reference A star and recomputed new solutions for (B–A). In this way, we derived relative [Fe /H ] differences between B and A of +0.11, +0.04, and −0.01 dex. This shows that a Teff difference near ~150 K would be required to remove the abundance offset between stars A and B. However, Fig. 5 shows that a synthetic spectra calculated for Teff = Teff(A)−150 K = 5886 K does not fit the observed Hα profiles and does not respect the excitation and ionization equilibrium of the Fe  lines (Fig. 6). We therefore find no clear reason to assume such a difference in Teff between stars A and B.

thumbnail Fig. 5

Spectral region near the Hα line for stars A and B (blue dotted and black continuous line). A synthetic spectrum (red dashed line) with Teff(A)−150 K = 5886 K is also shown for comparison.

thumbnail Fig. 6

Differential abundance vs. excitation potential (upper panel) and vs. reduced EW (lower panel) for the A star, adopting Teff(A)–150 K = 5886 K. Filled and empty circles correspond to Fe i and Fe ii.

4.4. Possible interpretations of the data

We propose that at the time of planet formation, the A star locked the orbiting refractory material in planetesimals and rocky planets, and formed the gas giant in the external disk. This was followed by the accretion of most of the refractories (possibly due to migration of the gas giant) onto the A star. This scenario explains the higher metallicity of the A star (no chemically peculiar patterns are seen), the higher refractory abundances, and the higher Li  content. It is also supported by the slightly higher mass of the A star (corresponding to a lower convective mass and lower mixing of the accreted material) and by the non-detection of additional inner planets. A very similar scenario was recently proposed for the solar-twin star HIP 68468 (Meléndez et al. 2017), where the authors suggested that the possible inward migration of a close-in giant planet likely produces the engulfment of the inner planetary material by the star. This would explain the high Li  content (0.6 dex greater than expected for its age), a refractory enhancement compared to the Sun, and a Tc trend. We also note that chemical tagging studies usually claim to be able to distinguish groups of stars with common origin within ~0.04 dex of precision (e.g., Hogg et al. 2016). An intrinsic difference of ~0.1 dex in metallicity would therefore imply a great challenge for these works.

1

Refractory and volatile species are those with Tc> 900 K and Tc< 900 K.

3

Echelle SpectroPolarimetric Device for the Observation of Stars.

5

Open source Pipeline for ESPaDOnS Reduction and Analysis.

7

We estimated a vturb of 0.91 km s-1 for the Sun by requiring zero slope between Fe i abundances and EWr.

8

For the B star, the values are similar to those of Fig. 2.

Acknowledgments

The authors thank R. Kurucz and C. Sneden for making their codes available to us. E.J., R.P., M.F., and M.J.A. acknowledge the financial support from CONICET in the forms of Post-Doctoral Fellowships. We also thank the referee Ivan Ramírez for constructive comments that greatly improved the paper.

References

  1. Adibekyan, V., González Hernández, J., Delgado-Mena, E., et al. 2014, A&A, 564, A15 [Google Scholar]
  2. Adibekyan, V., Delgado-Mena, E., Figueira, P., et al. 2016, A&A, 592, A87 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  3. Ballard, S., Christiansen, J., Charbonneau, D., et al. 2011, ApJ, 732, 41 [NASA ADS] [CrossRef] [Google Scholar]
  4. Barklem, P. S., Stempels, H. C., Allende Prieto, C., et al. 2002, A&A, 385, 951 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. Baumann, P., Ramírez, I., Meléndez, J., et al. 2010, A&A, 519, A87 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Bedell, M., Meléndez, J., Bean, J., et al. 2014, AJ, 795, 23 [Google Scholar]
  7. Boss, A. 2000, ApJ, 536, 101 [NASA ADS] [CrossRef] [PubMed] [Google Scholar]
  8. Bodenheimer, P., Hubickyj, O., & Lissauer, J. 2000, Icarus, 143, 2 [NASA ADS] [CrossRef] [Google Scholar]
  9. Bressan, A., Marigo, P., Girardi, L., et al. 2012, MNRAS, 427, 127 [NASA ADS] [CrossRef] [Google Scholar]
  10. Brewer, J., Fischer, D., & Madhusudhan, N. 2017, AJ, 153, 83 [NASA ADS] [CrossRef] [Google Scholar]
  11. Carlos, M., Nissen, P., & Meléndez, M. 2016, A&A, 587, A100 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  12. Casagrande, L., Ramírez, I., Meléndez, J., et al. 2010, A&A, 512, A54 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  13. Chambers, J. 2010, AJ, 724, 92 [NASA ADS] [CrossRef] [Google Scholar]
  14. Cochran, W., Hatzes, A., Butler, P., & Marcy, G. 1997, ApJ, 483, 457 [NASA ADS] [CrossRef] [Google Scholar]
  15. Cody, A. M., & Sasselov, D. 2005, AJ, 622, 704 [Google Scholar]
  16. Delgado-Mena, E., Israelian, G., González Hernández, J., et al. 2014, A&A, 562, A92 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  17. Delgado-Mena, E., Bertan de Lis, S., Adibekyan, V., et al. 2015, A&A, 576, A69 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  18. Desidera, S., Gratton, R. G., Scuderi, S., et al. 2004, A&A, 420, 683 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  19. Desidera, S., Gratton, R. G., Lucatello, S., et al. 2006, A&A, 454, 581 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Desidera, S., Gratton, R. G., Lucatello, S., et al. 2007, A&A, 462, 1039 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  21. Fuhrmann, K., & Bernkopf, J. 1999, A&A, 347, 897 [NASA ADS] [Google Scholar]
  22. Galeev, A., Ivanova, D., Shimansky, V., et al. 2012, Astron. Rep., 56, 11, 850 [CrossRef] [Google Scholar]
  23. Gonzalez, G. 1997, MNRAS, 285, 403 [NASA ADS] [CrossRef] [Google Scholar]
  24. Gonzalez, G. 2014, MNRAS, 441, 1201 [NASA ADS] [CrossRef] [Google Scholar]
  25. Gonzalez, G. 2015, MNRAS, 446, 1020 [NASA ADS] [CrossRef] [Google Scholar]
  26. González Hernández, J., Delgado-Mena, E., Sousa, S. G., et al. 2013, A&A, 552, A6 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  27. Gratton, R. G., Bonanno, G., Claudi, R. U., et al. 2001, A&A, 377, 123 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  28. Hogg, D., Casey, A., Ness, M., et al. 2016, ApJ, 833, 262 [NASA ADS] [CrossRef] [Google Scholar]
  29. Israelian, G., Santos, N., Mayor, M., et al. 2004, A&A, 414, 601 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  30. Israelian, G., Delgado-Mena, E., Santos, N., et al. 2009, Nature, 462, 12 [Google Scholar]
  31. King, J., Deliyannis, C., Hiltgen, D., et al. 1997, AJ, 113, 1871 [NASA ADS] [CrossRef] [Google Scholar]
  32. Kovács, G., Bakos, G., Torres, G., et al. 2007, ApJ, 670, L41 [NASA ADS] [CrossRef] [Google Scholar]
  33. Knutson, H., Fulton, B., Montet, B., et al. 2014, ApJ, 785, 126 [NASA ADS] [CrossRef] [Google Scholar]
  34. Kurucz, R. L. 1993, ATLAS9 Stellar Atmosphere Programs and 2 km s grid, Kurucz CD-ROM 13 (Cambridge, MA: Smithsonian Astrophysical Obs.) [Google Scholar]
  35. Kurucz, R., & Bell, B. 1995, Atomic Line Data, Kurucz CD-ROM, 23, (Cambridge, MA: Smithsonian Astrophysical Obs.) [Google Scholar]
  36. Laws, C., & Gonzalez, G. 2001, ApJ, 553, 405 [NASA ADS] [CrossRef] [Google Scholar]
  37. Liakos, A., & Niarchos, P. 2017, MNRAS, 465, 1181 [NASA ADS] [CrossRef] [Google Scholar]
  38. Liu, F., Asplund, M., Ramírez, I., et al. 2014, MNRAS, 442, L51 [NASA ADS] [CrossRef] [Google Scholar]
  39. Lodders, K. 2003, AJ, 591, 1220 [Google Scholar]
  40. Luck, R., & Heiter, U. 2006, AJ, 131, 3069 [NASA ADS] [CrossRef] [Google Scholar]
  41. Mack, C., Schuler, S., Stassun, K., et al. 2014, ApJ, 787, 98 [NASA ADS] [CrossRef] [Google Scholar]
  42. Martioli, E., Teeple, D., Manset, N., et al. 2012, Software and Cyberinfrastructure for Astronomy II, Proc. SPIE, 8451, 21 [Google Scholar]
  43. Meléndez, J., Asplund, M., Gustafsson, B., et al. 2009, AJ, 704, L66 [Google Scholar]
  44. Meléndez, J., Ramírez, I., Karakas, A., et al. 2014, AJ, 791, 14 [Google Scholar]
  45. Meléndez, J., Bedell, M., Bean, J., et al. 2017, A&A, 597, A34 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  46. Mugrauer, M., Ginski, C., & Seeliger, M. 2014, MNRAS, 439, 1063 [NASA ADS] [CrossRef] [Google Scholar]
  47. Mustill, A., Davies, M., & Johansen, A. 2015, ApJ, 808, 14 [NASA ADS] [CrossRef] [Google Scholar]
  48. Pauzen, E. 2004, The A-Star Puzzle, eds. J. Zverko, J. Ziznovsky, S. Adelman, & W. Weiss (Cambridge Univ. Press), IAU Symp., 224, 443 [Google Scholar]
  49. Pollack, J., Hubickyj, O., Bodenheimer, P., et al. 1996, Icarus, 124, 62 [NASA ADS] [CrossRef] [Google Scholar]
  50. Ramírez, I., Allende Prieto, C., & Lambert, D. 2007, A&A, 465, 271 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  51. Ramírez, I., Meléndez, J., & Asplund, M. 2009, A&A, 508, L17 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  52. Ramírez, I., Meléndez, J., Cornejo, D., et al. 2011, ApJ, 740, 76 [NASA ADS] [CrossRef] [Google Scholar]
  53. Ramírez, I., Fish, J., Lambert, D., et al. 2012, ApJ, 756, 46 [NASA ADS] [CrossRef] [Google Scholar]
  54. Rodríguez, E., López-González, M., & López de Coca, P. 2000, A&AS, 144, 469 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  55. Ryan, S., Beers, T., Kajino, T., & Rosolankova, K. 2001, AJ, 547, 231 [Google Scholar]
  56. Saffe, C. 2011, RMxAA, 47, 3 [Google Scholar]
  57. Saffe, C., Flores, M., & Buccino, A. 2015, A&A, 582, A17 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  58. Saffe, C., Flores, M., Jaque Arancibia, M., et al. 2016, A&A, 588, A81 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  59. Schirbel, L., Meléndez, J., Karakas, A., et al. 2015, A&A, 584, A116 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  60. Shi, J. R., Gehren, T., & Zhao, G. 2004, A&A, 423, 683 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  61. Schuler, S., Cunha, K., Smith, V., et al. 2011, ApJ, 737, L32 [NASA ADS] [CrossRef] [Google Scholar]
  62. Smith, V., Cunha, K., & Lazzaro, D. 2001, AJ, 121, 3207 [NASA ADS] [CrossRef] [Google Scholar]
  63. Smith, A. M., Hebb, L., Collier Cameron, A., et al. 2009, MNRAS, 398, 1827 [NASA ADS] [CrossRef] [Google Scholar]
  64. Sneden, C. 1973, ApJ, 184, 839 [NASA ADS] [CrossRef] [Google Scholar]
  65. Tucci Maia, M., Meléndez, J., & Ramírez, I. 2014, ApJ, 790, L25 [NASA ADS] [CrossRef] [Google Scholar]
  66. Van Saders, J., & Pinsonneault, M. 2012, AJ, 746, 16 [Google Scholar]

All Tables

Table 1

Stellar parameters derived for each star.

In the text
Table 2

Derived slopes (abundance vs. Tc) and their uncertainties.

In the text

All Figures

thumbnail Fig. 1

Differential abundance vs. excitation potential (upper panel) and vs. reduced EW (lower panel) for the case (B–A). Filled and hollow circles correspond to Fe i and Fe ii.
In the text
thumbnail Fig. 2

Differential abundances vs. Tc for (A–Sun). The long dashed line is a weighted linear fit to the differential abundance values, while the continuous line shows the solar-twins trend of Meléndez et al. (2009).
In the text
thumbnail Fig. 3

Differential abundances (B–A) vs. Tc. Long dashed lines are weighted linear fits to all species and to refractory species. The solar-twins trend of Meléndez et al. (2009) is shown with a continuous line.
In the text
thumbnail Fig. 4

Stellar spectra near the Li line 6707.8 Å for the A (blue dotted line) and B (black continuous line) stars.
In the text
thumbnail Fig. 5

Spectral region near the Hα line for stars A and B (blue dotted and black continuous line). A synthetic spectrum (red dashed line) with Teff(A)−150 K = 5886 K is also shown for comparison.
In the text
thumbnail Fig. 6

Differential abundance vs. excitation potential (upper panel) and vs. reduced EW (lower panel) for the A star, adopting Teff(A)–150 K = 5886 K. Filled and empty circles correspond to Fe i and Fe ii.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.