EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 696 (April 2025) A&A, 696 (2025) A123 Full HTML
Open Access
Issue
A&A
Volume 696, April 2025
Article Number A123
Number of page(s) 11
Section Stellar atmospheres
DOI https://doi.org/10.1051/0004-6361/202553969
Published online 11 April 2025

© The Authors 2025

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.

1 Introduction

The λ Boo stars are considered one of the long-standing puzzles in astrophysics. They are a rare class of stars, making up ∼2% of the population of stars of spectral type A (Gray & Corbally 1998; Paunzen et al. 2001b). The main characteristic of λ Boo stars is a notable surface depletion of most Fe-peak elements together with near-solar abundances of lighter elements C, N, O, and S (e.g., Kamp et al. 2001; Andrievsky et al. 2002; Heiter et al. 2002; Alacoria et al. 2022). Determining the origin of the λ Boo peculiarity remains a challenge despite recent efforts (see, e.g., Jura 2015; Kama et al. 2015; Murphy & Paunzen 2017; Kunitomo et al. 2018; Saffe et al. 2021; Alacoria et al. 2022). The scenarios trying to explain the notable abundance pattern point to diverse ideas, including accretion of gas from a circumstellar disk (Venn & Lambert 1990), the ingestion of comets and volatile-rich objects (Gray & Corbally 2002), the interaction of the star with the interstellar medium (ISM) or with a diffuse interstellar cloud (Cowley et al. 1982; Kamp & Paunzen 2002; Martinez-Galarza et al. 2009), and the ablation of volatile gases from a near hot-Jupiter planet (Jura 2015). However, none of the suggested scenarios seems to explain the origin of the λ Boo stars (see, e.g., the recent discussion in Murphy & Paunzen 2017; Alacoria et al. 2022).

The detection of λ Boo stars as members of binary systems is considered an important finding; binary systems are the laboratories for testing the physical conditions under which these stars formed. However, currently few λ Boo stars are clearly identified as members of these systems. For example, Paunzen et al. (2012a,b) identified about a dozen λ Boo stars as members of early-type binary systems, and proposed a method for testing the accretion scenario. They suggested that two early-type stars passing through a diffuse cloud should display, in principle, the same superficial peculiarity (see also Alacoria et al. 2022). The detection of a binary or multiple system including a λ Boo star and a late-type companion is also valuable. In this case, the chemical composition of the late-type component could be considered as a proxy of the original composition from which both stars formed, and is crucial for testing the formation models of λ Boo stars (e.g., Alacoria et al. 2022). If the compositions of the λ Boo star and the late-type star differ significantly, this would be a solid indication that the λ Boo star was originally born with a different composition. In addition, the differential pattern between these stars could help precisely quantify the λ Boo phenomena, that is, to measure element by element the effect produced by the λ Boo peculiarity. However, to our knowledge, only one multiple system including a late-type component is reported in the literature: the remarkable triple system HD 15165 (Alacoria et al. 2022). The examples mentioned show that the detection of multiple systems including a λ Boo component could provide benchmark laboratories, an important tool for the study of the origin of λ Boo stars.

Table 1

Visual binary systems that include a λ Boo star component (the first star listed), identified with Gaia eDR3 data.

Progress in understanding the λ Boo stars has been hindered by a somewhat heterogeneous literature (see, e.g., Sect. 1.2 in Murphy et al. 2015). This motivated us to reevaluate the membership of previously reported λ Boo stars (see Murphy et al. 2015; Gray et al. 2017; Murphy et al. 2020a), using mainly classification spectroscopy. Together, these three works comprise a complete and homogeneous sample of predominantly southern λ Boo stars. On the other hand, El-Badry et al. (2021) recently obtained an extensive catalog of 1.3 (1.1) million spatially resolved binaries with a bound probability >90% (>99%) within ∼1 kpc of the Sun, using Gaia eDR3 data. We caution that 15% of A-type stars have companions with periods of 100–1500 days (Murphy et al. 2018), which would be undetected in the El-Badry et al. (2021) study. Then, with the aim of detecting λ Boo stars in multiple systems, we cross-matched the compilation of 118 λ Boo stars with this recent catalog of resolved binaries. The results of this experiment are presented in Table 1, showing a list of λ Boo stars together with their corresponding binary companions. The columns present the binary number, star name, V magnitude, coordinates J2016.0 (α and δ), proper motions (μα and μδ), parallax π, and separation (in seconds and au). As previously explained, this important group of newly identified binary systems could be used in further studies of λ Boo stars. Most of the λ Boo stars in Table 1 have late-type companions. The binaries include separations ranging between ∼150 au and ∼57 000 au. We note that the list includes one binary system (number 7, HD 198160/HD 198161) that has a previously known λ Boo star candidate (Gray 1988; Stürenburg 1993; Alacoria et al. 2022).

Table 2

Magnitudes and astrometric data for the stars studied in this work.

We present in this work the first detailed analysis of binary systems that include a candidate λ Boo star and a late-type companion. We focused on three relatively bright binary systems included in Table 1, obtaining fundamental parameters and abundances for their components. The separation of the stars allowed us to obtain clean individual spectra that are free from a possible contamination from its stellar companion. This will allow us to determine, for the first time, a solid proxy for the starting chemical composition of the λ Boo stars, that is, the chemical composition from which the λ Boo stars were born. This is a critical constraint for any model that explains the origin of λ Boo stars. In addition, the stars analyzed together with those presented in Table 1, are important laboratories for further studies of λ Boo stars. The new binary systems reported in Table 1 allow us to approximately double the number of λ Boo stars currently known in multiple systems. The sample could also help determine if the presence of a stellar companion could play a role in the development of the peculiarity.

This work is organized as follows. In Sect. 2, we describe the observations and data reduction. In Sect. 3, we present the stellar parameters and chemical abundance analysis. In Sect. 4, we show the results and discussion. Finally, in Sect. 5 we highlight our main conclusions.

2 Observations

We present in Table 2 the visual magnitude V, coordinates, proper motions, parallax, and signal-to-noise per pixel (@5000 Å) for the stars studied in this work. The spectral data of the binary system HD 87304 + CD-33 6615B were acquired through the Gemini High-resolution Optical SpecTrograph (GHOST), which is attached to the 8.1 m Gemini South telescope at Cerro Pachón, Chile. GHOST is illuminated via 1.2” integral field units that provide the input light apertures. The spectral coverage of GHOST between 360–900 nm is appropriate for deriving stellar parameters and chemical abundances using several features. It provides a high resolving power R ∼ 50 000 in the standard resolution mode1. The read mode was set to medium, as recommended for relatively bright targets. The observations were taken on October 10, 2024, and October 24, 2024 (PI: Carlos Saffe, Program ID: GS-2024B-Q-403), using the same spectrograph configuration for both stars. The exposure times were 120 sec and 1100 sec (for HD 87304 and CD-33 6615B), obtaining a final signal-to-noise ratio (S/N) in the range ∼220–275 per pixel measured at ∼5000 Å for both stars. The spectra were reduced using the GHOST data reduction pipeline v1.1.0, which works under DRAGONS2. This is a platform for the reduction and processing of astronomical data.

The spectra of the binary systems HD 98069 + UCAC4 431-054639 and HD 153747 + TYC 7869-2003-1 were obtained at the Complejo Astrónomico El Leoncito (CASLEO) during March 15–16, 2024, June 11–12, 2024, and July 11–12, 2024 (PI: José Alacoria, Program ID: JS-2024A-06). We used the Jorge Sahade 2.15 m telescope equipped with a REOSC echelle spectrograph3 and a SOPHIA 2048 × 2048 (152-VS-X eXcelon) CCD detector. The REOSC spectrograph uses gratings as cross-dispersers. We used a grating with 300 lines mm−1 (disperser # 270), which provides a resolving power of ∼13 500 or more, covering appropriately a spectral range of λ λ 3700–7500. Five individual spectra for each object were obtained and then combined, thus reaching an average S/N per pixel of ∼235 measured at ∼5000 Å. We took the stellar spectra for each target followed by a ThAr lamp in order to derive an appropriate pixel versus wavelength solution. The data were reduced with the Image Reduction and Analysis Facility (IRAF) following the standard recipe for echelle spectra (i.e., bias and flat corrections, order-by-order normalization, scattered light correction). The continuum normalization and other operations (such as Doppler correction and combining spectra) were performed using IRAF.

3 Stellar parameters and abundance analysis

The stellar parameters were determined as homogeneously as possible for the stars in our sample, including late-type and early-type stars. We used the Virtual Observatory Sed Analyzer4 (VOSA, Bayo et al. 2008) and the spectral energy distributions (SEDs) constructed from photometric data to obtain the stellar effective temperatures (Teff) of the objects in our sample via atmospheric model fitting. The observed SEDs were unreddened by VOSA using the extinction maps of Schlegel et al. (1998) and following the procedure of Bilir et al. (2008) to derive Av. We used a grid of Kurucz-NEWODF models by Kurucz (1993) covering Teff between 3500 K and 13000 K with a step of 250 K.

Table 3

Fundamental parameters derived for the stars in this work.

Then, we performed a Bayesian estimation of surface gravities (log g) using Gaia eDR3 parallaxes with the PARAM 1.3 interface5 (da Silva et al. 2006). The temperatures and gravities derived for the stars in our sample are presented in Table 3.

The projected rotational velocities (v sin i) were first estimated by fitting the line Mg II 4481.23 Å and were then refined by fitting Fe I and Fe II lines in the spectra. Synthetic spectra were calculated using the program SYNTHE (Kurucz & Avrett 1981) together with ATLAS12 (Kurucz 1993) model atmospheres, and were then convolved with a rotational profile (using the Kurucz’s command rotate) and with an instrumental profile for each spectrograph (using the command broaden). The resulting v sin i values are shown in the 6th column of Table 3, covering between 10.3 ± 0.2 km s−1 and 140.5 ± 3.6 km s−1 for the stars in our sample. The microturbulence velocity (vmicro) was estimated as a function of Teff following the formula of Gebran et al. (2014), which is valid for ∼6000 K < Teff < ∼ 10 000 K. We adopted an uncertainty of ∼25% for vmicro, as suggested by Gebran et al. (2014), and then this uncertainty was taken into account in the abundance error calculation.

We applied an iterative procedure to determine the chemical abundances for the stars in our sample. As a first step, we computed an ATLAS12 (Kurucz 1993) model atmosphere, adopting initially solar abundances from Asplund et al. (2009). The corresponding abundances were then obtained by fitting the observed spectra with the program SYNTHE (Kurucz & Avrett 1981). With the new abundance values, we derived a new model atmosphere and started the process again. In each step, the opacities were calculated for an arbitrary composition and vmicro using the opacity sampling (OS) method, as in previous works (Saffe et al. 2020, 2021, 2022; Alacoria et al. 2022). In this way, the parameters and abundances were consistently derived using specific opacities rather than solar-scaled values, until the same input and output abundances were reached (for more details, see Saffe et al. 2021). Possible differences originating from the use of solar-scaled opacities instead of an arbitrary composition were recently estimated for solar-type stars (Saffe et al. 2018, 2019). These differences could become particularly important when modeling chemically peculiar stars, where solar-scaled models could result in a very different atmospheric structure (e.g., Piskunov & Kupka 2001) and reach abundance differences up to 0.25 dex (Khan & Shulyak 2007).

We derived the chemical abundances for 31 different species, including Li I, C I, O I, Na I, Mg I, Mg II, Al I, Si I, Si II, S II, Ca I, Ca II, Sc I, Sc II, Ti I, Ti II, V I, Cr I, Cr II, Mn I, Fe I, Fe II, Co I, Ni I, Cu I, Zn I, Sr II, Y II, Ba II, La II, and Ce II. The atomic line list and laboratory data used in this work are the same as described in Saffe et al. (2021). Figure 1 presents an example of observed, synthetic, and difference spectra (black, blue dotted, and magenta lines) for some stars in our sample. There is a good agreement between the results of the modeling and the observations for the lines of different chemical species.

The uncertainty in the abundance values was estimated considering different sources. We estimated the measurement error (e1) from the line-to-line dispersion as σ/n$\sigma/\sqrt{n}$, where σ is the standard deviation and n is the number of lines. For elements with only one line, we adopted for σ the standard deviation of the iron lines. Then we determined the contribution to the abundance error due to the uncertainty in stellar parameters. We modified Teff, log g, and vmicro by their uncertainties and recalculated the abundances, obtaining the corresponding differences e2, e3, and e4; we adopted a minimum of 0.01 dex for these errors. Finally, the total error (etot) was estimated as the quadratic sum of e1, e2, e3, and e4. The abundances with their total error etot, the individual errors e1 to e4, and the number of lines n, are presented in Tables A.1 to A.6.

thumbnail Fig. 1

Observed, synthetic, and difference spectra (black, blue dotted, and magenta lines) for some stars in our sample.

3.1 NLTE effects

In the case of λ Boo stars, light-element non-local thermodynamic equilibrium (NLTE) abundances are particularly important. For example, an average O I correction of −0.5 dex was derived by Paunzen et al. (1999) for a sample of λ Boo stars, while for C I they estimated an average correction of −0.1 dex. Rentzsch-Holm (1996) derived neutral carbon NLTE abundance corrections by using a multilevel model atom for stars with Teff between 7000 K and 12 000 K, log g between 3.5 and 4.5 dex, and metallicities from −0.5 dex to +1.0 dex. She showed that C I NLTE effects are negative (calculated as NLTE-LTE) and depend basically on equivalent width (Weq). Near ∼7000 K the three lower levels of CI are always in LTE; however, increasing the Teff values increases the underpopulation of these levels with respect to LTE by UV photoionization. Thus, we estimated NLTE abundance corrections of C I for the early-type stars in our sample by interpolating in their Figs. 7 and 8 as a function of Teff, Weq, and metallicity. We applied a similar correction in previous works (Alacoria et al. 2022), which allowed the comparison of the abundance values.

The NLTE abundance corrections for O I were derived by Sitnova et al. (2013), who used a multilevel model atom with 51 levels. The authors showed that NLTE effects lead to a strengthening of the O I lines, producing a negative NLTE correction. They calculated NLTE corrections for a grid of model atmospheres, including stars with spectral types from A to K (Teff between 10 000 and 5000 K). We estimated NLTE abundance corrections of O I (IR triplet 7771 Å) for the stars in this work, interpolating based on Table 11 of Sitnova et al. (2013), as a function of Teff. We note that other O I lines present corrections lower than ∼−0.02 dex (see, e.g., Table 5 of Sitnova et al. 2013).

thumbnail Fig. 2

Comparison of [Fe/H] values derived in this work with those from literature.

3.2 Comparisons with the literature

We present in Fig. 2 a comparison of the [Fe/H] values derived in this work with those taken from the literature for the stars UCAC4 431-054639 (Steinmetz et al. 2020), HD 87304 (Gray et al. 2017), and HD 153747 (Paunzen et al. 2002b). In general, there is a good agreement with the literature. The two stars with the lowest metallicity (HD 87304 and HD 153747) seem to present literature values that are slightly higher than those in the present work. However, the values could still be considered similar within their error bars.

thumbnail Fig. 3

Observed spectra (black line) and synthetic spectra (blue dotted line) near the Li line 6707.8 Å in the star CD-33 6615B. The synthetic lines are indicated showing the wavelength, atomic number, and intensity.

4 Discussion

We discuss in this section the chemical composition of the binary systems, which include a candidate λ Boo star and a late-type companion. The chemical patterns of the early-type stars are compared to an average pattern of λ Boo stars. We caution that deriving an average pattern for λ Boo stars is not an easy task, due to the relatively low number of stars homogeneously analyzed (see, e.g., Sect. 4.1 of Alacoria et al. 2022). For this work we used the same average chemical pattern as Alacoria et al. (2022). Basically, we adopted the data derived by Heiter et al. (2002), who homogeneously determined abundances for a number of λ Boo stars, and then we excluded from the average those stars without CNO values, similarly to Alacoria et al. (2022).

4.1 Binary system HD 87304+C D-33 6615B

The star HD 87304 was classified by Gray et al. (2017) as A8 V kA2.5mA2.5 λ Boo. However, the authors caution that the identification as a member of the λ Boo class should be followed up with high-resolution abundance studies to confirm a λ Boo abundance pattern, as we did for the present work. The candidate λ Boo star HD 87304 is accompanied by CD-33 6615B, a late-type star separated by 6.97 arcsec or 3160.67 au (El-Badry et al. 2021). El-Badry et al. (2021) estimated empirically a probability that each pair is a chance alignment (R in their paper and R_chance_align in their catalog). The authors consider “high bound probability” or “high-confidence” pairs those with R < 0.1, corresponding to >90% probability of being bound. In particular, this pair presents R = 1.24 10−4, and is considered a high bound probability pair. The separation allowed us to analyze the two stars independently without contamination from the companion, which is different from other λ Boo stars (see, e.g., Paunzen et al. 2012a,b). To our knowledge, there is no detailed abundance determination for any component of this binary system.

We note that the Li abundance is significantly supersolar in CD-33 6615B ([Li/H] = 1.72 ± 0.22 dex). We present in Fig. 3 a spectral region near the Li line 6707.8 Å in this star. The observed and synthetic spectra are shown with black and blue dotted lines. For the synthetic lines, the plot indicates the wavelength, atomic number, and intensity (between 1 and 0). The lithium line 6707.8 Å was detected in CD-33 6615B, but not in their companion HD 87304 (which is a λ Boo star, as we discuss below). Interestingly, for the case of the triple system HD 15615, the Li line was detected in the early-type star HD 15164 (see Fig. 5 in Alacoria et al. 2022), but not in their λ Boo companion HD 15165. However, we caution that the lithium abundance is sensitive to different effects probably not related to the λ Boo peculiarity. For example, there is a known correlation between lithium content and age. Moreover, its abundance depends on other factors such as metallicity in solar-type stars (see, for example, Fig. 3 of Martos et al. 2023) or even the possible engulfment of a rocky planet (Saffe et al. 2017; Soares et al. 2025). We therefore consider that the lithium content deserves to be further explored with a larger sample of stars.

We present in Fig. 4 the chemical pattern of the stars CD33 6615B and HD 87304 (left and right panels), compared to an average pattern of λ Boo stars (blue). For each star, we present two panels, corresponding to elements with atomic number z < 32 and z > 32. The error bars of the λ Boo pattern show the standard deviation derived from different stars, while the error bars for our stars correspond to the total error, etot. Figure 4 shows that the different chemical compositions of the two stars is striking. On the one hand, CD-33 6615B presents a mostly solar chemical pattern (e.g., [Fe/H] = −0.05 ± 0.15 dex), with some elements above solar values (Y, Ba and La). In particular, the light elements C, O, and S present solar abundances.

On the other hand, HD 87304 presents a chemical pattern that agrees with those of λ Boo stars (see Fig. 4). For instance, C and O present solar or slightly subsolar values ([C/H] = −0.15 ± 0.16 dex, [O/H] = −0.24 ± 0.27 dex), which is similar to other λ Boo stars. Other metals, such as Ca, Ti, and Fe, present abundances ∼1 dex below solar values. The abundance values confirm the bona fide λ Boo nature of this object. Interestingly, although C and O are almost solar in HD 87304, they seem to be slightly lower in CD-33 6615B ([C/H] = −0.03 ± 0.17 dex, [O/H] = −0.01 ± 0.17 dex). This would perhaps suggest that the λ Boo phenomena also slightly modifies the light element abundances (in addition to a stronger effect on heavier species); however, we caution that NLTE effects could play a role in the light elements.

Other metals present a significant difference between the two stars (e.g., Fe differs by Δ[Fe/H] ∼ 1.12 ± 0.21 dex). This is one of the largest differences in metallicity found between two stars in a binary system (see, for example, Saffe et al. 2017, 2022).

In summary, this binary system is composed of a late-type star with a mostly solar-like composition, and an early-type object with a λ Boo chemical pattern. This pair shows that λ Boo stars are born with a very different composition, approximately solar, and reinforces the idea that λ Boo stars are Population I objects. A similar result was obtained by studying the triple system HD 15165 (Alacoria et al. 2022), the only system reported to date that includes a late-type star with solar-like composition (HD 15165C) and a λ Boo companion (HD 15165). This is the most likely result of the analysis; however, we caution that other explanations are also possible. We come back to this point in Sect. 4.4.

thumbnail Fig. 4

Chemical pattern of the stars CD-33 6615B and HD 87304 (left and right panels), compared to an average pattern of λ Boo stars (blue).

4.2 Binary system HD 98069 + UCAC4 431-054639

The star HD 98069 was classified as A9 V kA2mA2 (λ Boo) by Murphy et al. (2020a). Interestingly, the K2 (Kepler-2) light curve revealed its δ Scuti nature, with eight pulsation peaks exceeding 1 mmag and seven peaks exceeding 0.05 mmag (Murphy et al. 2020a,b). This candidate λ Boo star (HD 98069) is accompanied by UCAC4 431-054639, a late-type star separated by 55.24 arcsec or 16148.71 au (El-Badry et al. 2021). This binary presents a chance alignment probability of R = 1.24 × 10−4, and is considered a high bound probability pair (El-Badry et al. 2021). The separation allows us to analyze the two stars independently without contamination from the companion. To our knowledge, there is no detailed abundance determination for HD 98069.

We present in Fig. 5 the chemical pattern of the stars UCAC4 431-054639 and HD 98069 (left and right panels), compared to an average pattern of λ Boo stars. The symbols of Fig. 5 are similar to those used in Fig. 4. It is clear from Fig. 5 that the two stars present a significantly different chemical pattern, similar to the previous binary system. On the one hand, UCAC4 431-054639 presents mostly a solar chemical pattern within ± ∼ 0.20 dex, with some species showing slightly subsolar (Mg, Sc, Fe) or slightly supersolar (Ca, Y) abundance values. For example, for iron we obtained [Fe/H] = −0.16 ± 0.17 dex, while light-element abundances (CNOS) were not derived.

On the other hand, the chemical pattern of HD 98069 is very different from its stellar companion (see Fig. 5, right panel). Carbon displays an abundance close to solar, or slightly subsolar ([C/H] = −0.21 ± 0.05). Most species are strongly depleted compared to the solar values by ∼1 dex or more (Al, Ca, Ti, Cr, Fe) except perhaps Si, which is less depleted ([Si/H]= −0.37 ± 0.22 dex). However, we caution that the Si abundance was derived using only one line. Thus, the general pattern of HD 98069 seems to agree with the values of λ Boo stars. The metallicity values of the two stars in this binary system differ by Δ[Fe/H] ∼ 0.89 ± 0.21 dex, which is also significant, although a less extreme difference than the previous binary system.

In summary, this binary system is composed of a late-type star with a mostly solar-like composition, and an early-type object with a λ Boo-like chemical pattern.

thumbnail Fig. 5

Chemical pattern of the stars UCAC4 431-054639 and HD 98069 (left and right panels), compared to an average pattern of λ Boo stars (blue).

4.3 Binary system HD 153747 + TYC 7869-2003-1

The star HD 153747 was classified by Paunzen et al. (2001b) as hA7mA0 V λ Boo, and later as A7 V kA0mA0 λ Boo by Murphy et al. (2015), who discussed the membership of this object in the λ Boo class. This object is also reported as λ Boo in other literature works (Gray et al. 2017; Murphy et al. 2020a). Desikachary & McInally (2014) reported multiperiodicity (periods between 0.96 and 1.2 h) in the light curve of HD 153747, while Paunzen et al. (2001b) noted their δ Sct variability. For this star, Paunzen et al. (2002b) reported a depleted metal content of [Z/H] = −0.86 ± 0.20 dex, estimated using Δ m2 from the Geneva seven-color system as well as Δ m1 from the uvby β photometric system. This candidate λ Boo star (HD 153747) is accompanied by TYC 7869-2003-1, a late-type star separated by 322.64 arcsec or 57131.24 au (El-Badry et al. 2021). This pair presents a chance alignment probability of R = 0.0885, somewhat higher than previous binary systems, although it is still considered a high bound probability pair (El-Badry et al. 2021). The separation allows us to analyze the two stars independently without contamination from its companion. To our knowledge, there is no detailed (spectroscopic) abundance determination for any component of this binary system.

We present in Fig. 6 the chemical pattern of the stars TYC 7869-2003-1 and HD 153747 (left and right panels), compared to an average pattern of λ Boo stars. The symbols of Fig. 6 are similar to those used in Figs. 4 and 5. The two stars present a significantly different chemical pattern, similar to previous binary systems. TYC 7869-2003-1 displays mostly a solar chemical pattern, while HD 153747 presents a chemical pattern which closely follows the average pattern of λ Boo stars. HD 153747 displays nearly solar C and O abundances ([C/H] = 0.09 ± 0.12 dex, [O/H] = −0.18 ± 0.16 dex), while other metals show strong depletions of ∼1 dex (e.g., [Fe/H]= −0.94 ± 0.16 dex). The metallicities of the two stars in this binary system differ by Δ[Fe/H] ∼ 0.94 ± 0.22 dex.

In summary, this binary system is composed of a solar-composition late-type star and a λ Boo early-type star.

4.4 The relevance of λ Boo stars with late-type companions

We discuss in this section the importance of finding λ Boo stars accompanied by late-type stars. For the first time, we studied binary systems composed of a late-type object and a candidate λ Boo star. Different authors note that the identification of λ Boo stars (starting with candidates suggested by spectral classification) should be followed up with a high-resolution abundance analysis to confirm their λ Boo nature (see, e.g., Andrievsky et al. 2002; Heiter et al. 2002; Murphy et al. 2015; Gray et al. 2017; Alacoria et al. 2022). For this work we applied a similar procedure, and confirmed the λ Boo membership of three early-type stars (HD 87304, HD 98069, and HD 153747), as we can see in Figs. 4, 5 and 6. These three stars could then be considered bona fide λ Boo stars.

We also showed that the late-type companions that belong to these binary systems mostly present a solar-like composition. This composition could be used as a proxy for the initial composition of the material from which the λ Boo star formed (under the hypothesis that they were born from the same molecular cloud). To our knowledge, the starting solar-like composition was only shown in the triple system HD 15165 (Alacoria et al. 2022), giving us the opportunity to strengthen this result in the present work. This is an important constraint for any model attempting to explain the λ Boo phenomenon. Most scenarios trying to explain λ Boo stars usually assume a starting solar-like composition (e.g., Martinez-Galarza et al. 2009; Jura 2015). The present work provides three numerical examples of possible starting and ending compositions, which could be further explored to test and numerically constrain the formation models of λ Boo stars.

In addition, the solar-like composition of the late-type stars is in agreement with the idea that λ Boo stars are Population I objects. For example, Paunzen et al. (2014) concluded that it is possible to distinguish λ Boo stars from intermediate Population II stars on the basis of elemental abundances, though not in terms of kinematics. The authors suggest the use of binary systems to strengthen the conclusion that λ Boo stars are a distinct population from the Population II group. The results of the present work support the idea that λ Boo stars belong to Population I, in agreement with Paunzen et al. (2014). This is the most likely result of the analysis; however, we caution that other explanations are also possible. For example, we mentioned that 15% of A-type stars have companions with periods of 100–1500 days (Murphy et al. 2018) which would be undetected in the El-Badry et al. (2021) study. Furthermore, 21% of those companions are white dwarfs (WDs). Perhaps the λ Boo star once had a tight orbit with a higher-mass companion that became a red giant, transferred mass, and is now an undetected WD, such as those suggested by Murphy et al. (2018). For example, Van Winckel et al. (1995) presented five extremely Fe-deficient post-AGB binary systems with orbital periods from one to a few years, suggesting that mass transfer had occurred in these systems. Paunzen et al. (1998) suggested that a similar mass-transfer might also be an additional mechanism to be considered for the λ Boo phenomenon. Then, in our binary systems the cool star could have been captured by the hypothetical tight binary and could be of any age.

In general, the present results are in agreement with those previously obtained in the triple system HD 15165 (Alacoria et al. 2022), in which a λ Boo star (HD 15165) is accompanied by a late-type star with solar-like composition (HD 15165C). To our knowledge, λ Boo stars were detected in binary (and multiple) systems accompanied by early-type stars (e.g., Paunzen et al. 2012a,b; Alacoria et al. 2022), by late-type companions (this work and Alacoria et al. 2022), and even in one case by a Brown Dwarf (ζ Del, detected by Saffe et al. 2021). However, no bona fide λ Boo stars have been detected in open clusters (e.g., Paunzen et al. 2001a,b; Gray & Corbally 2002), including searches in different intermediate-age open clusters. Other chemically peculiar stars (such as Am or Ap) were detected in the same clusters (e.g., Gray & Corbally 2002). It would be valuable to detect λ Boo stars in open clusters, which could also be used to study their origin.

The work of Jura (2015) mentioned the possibility that λ Boo stars could originate by accreting the winds from late-type stellar companions (their Sect. 4.1). Jura (2015) considered that only a small fraction of the material from the late-type stars could reach the λ Boo star, and then ruled out this scenario (instead, the author proposed winds from hot-Jupiter planets as a plausible mechanism). In the present work, we analyzed binary systems including λ Boo stars and late-type companions. However, the three binary systems analyzed here present wide separations (from 3160 au to 57 131 au; see Table 1), making the suggested mechanism unlikely. On the other hand, there are no planets reported orbiting the λ Boo stars studied in this work.

thumbnail Fig. 6

Chemical pattern of the stars TYC 7869-2003-1 and HD 153747 (left and right panels), compared to an average pattern of λ Boo stars (blue).

5 Concluding remarks

In the present work, we cross-matched a homogeneous list of candidate λ Boo stars with a recent catalog of resolved binaries detected with Gaia eDR3, with the aim of finding λ Boo stars as members of multiple systems. Then we performed a detailed abundance determination of three of these binary systems. The main results of this work are as follows:

  • We presented a group of 19 newly identified binary systems that contains a candidate λ Boo star (see Table 1). The new systems approximately double the number of λ Boo stars currently known in multiple systems. This important group could be used in further studies of λ Boo stars.

  • For the first time, we performed a detailed abundance analysis of three binary systems made up of a candidate λ Boo star and a late-type companion. We confirmed the true λ Boo nature of the three early-type stars, and obtained a mostly solar-like composition for the late-type components. In particular, the binary system HD 87304 + CD-33 6615B presents a mutual metallicity difference of Δ[Fe/H] ∼ 1.12 ± 0.21 dex, one of the largest differences found in a binary system.

  • Adopting as a proxy the chemical composition of the late-type stars, we showed that the three λ Boo stars were initially born with a solar-like composition. This is an important constraint for any scenario trying to explain the origin of λ Boo stars. The present work provides three numerical examples of possible starting and ending compositions to test the models.

  • Finally, the solar-like composition of the late-type stars supports the idea that λ Boo stars are Population I objects. This is in agreement with the suggestions of previous works (Paunzen et al. 2014; Alacoria et al. 2022). However, we caution that other explanations are also possible.

The present work shows the importance of finding λ Boo stars that belong to multiple systems. The stars used in this work correspond mostly to southern λ Boo stars. We encourage performing a similar work on additional binary systems, and expanding the list of homogeneous λ Boo candidates to the northern hemisphere.

Data availability

The reduced spectra are available at the CDS via anonymous ftp to cdsarc.cds.unistra.fr (130.79.128.5) or via https://cdsarc.cds.unistra.fr/viz-bin/cat/J/A+A/696/A123

Acknowledgements

We thank the referee for constructive comments that improved the paper. The authors thank Dr. R. Kurucz for making their codes available to us. CS acknowledge financial support from CONICET (Argentina) through grant PIP 11220210100048CO and the National University of San Juan (Argentina) through grant CICITCA 21/E1235. IRAF is distributed by the National Optical Astronomical Observatories, which is operated by the Association of Universities for Research in Astronomy, Inc., under a cooperative agreement with the National Science Foundation. Based on data acquired at Complejo Astronómico El Leoncito, operated under agreement between the Consejo Nacional de Investigaciones Científicas y Técnicas de la República Argentina and the National Universities of La Plata, Córdoba and San Juan.

Appendix A Chemical abundances

In this section we present the chemical abundances and their corresponding errors. The total error etot was derived as the quadratic sum of the line-to-line dispersion e1 (estimated as σ/n$\sigma/\sqrt{n}$, where σ is the standard deviation) and the error in the abundances (e2, e3, and e4) when varying Teff, log g, and vmicro by their corresponding uncertainties6. For chemical species with only one line, we adopted as σ the standard deviation of iron lines. Abundance tables show the average abundance and the total error etot, together with the errors e1 to e4.

Table A.1

Chemical abundances for the star HD 87304.

Table A.2

Chemical abundances for the star CD-33 6615B.

Table A.3

Chemical abundances for the star HD 98069.

Table A.4

Chemical abundances for the star UCAC4 431-054639.

Table A.5

Chemical abundances for the star HD 153747.

Table A.6

Chemical abundances for the star TYC 7869-2003-1.

References

  1. Alacoria, J., Saffe, C., Jaque Arancibia, M., et al. 2022, A&A, 660, A98 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  2. Andrievsky, S., Chernyshova, I., Paunzen, E., et al. 2002, A&A, 396, 641 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  3. Asplund, M., Grevesse, N., Sauval, A., & Scott, P. 2009, ARA&A, 47, 481 [Google Scholar]
  4. Bayo, A., Rodrigo, C., Barrado y Navascués, D., et al. 2008, A&A, 492, 277 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. Bilir, S., Soydugan, E., Soydugan, F., et al. 2008, AN, 329, 835 [NASA ADS] [Google Scholar]
  6. Cowley, C. R., Sears, R. L., Aikman, G., & Sadakane, K. 1982, AJ, 254, 191 [NASA ADS] [Google Scholar]
  7. da Silva, L., Girardi, L., Pasquini, L., et al. 2006, A&A, 458, 609 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  8. Desikachary, K., McInally, C. J. 1979, MNRAS, 188, 67 [Google Scholar]
  9. El-Badry, K., Rix, H.-W., Heintz, T. 2021, MNRAS 506, 2269 [NASA ADS] [CrossRef] [Google Scholar]
  10. Gebran, M., Monier, R., Royer, F., et al. 2014, Putting A Stars into Context: Evolution, Environment, and Related Stars, Proceedings of the international conference held on June 3–7, 2013 at Moscow M.V. Lomonosov State University in Moscow, Russia, eds. G. Mathys, E. Griffin, O. Kochukhov, R. Monier, & G. Wahlgren (Moscow: “Pero”), 193 [Google Scholar]
  11. Gray, R. O. 1988, AJ, 95, 220 [NASA ADS] [CrossRef] [Google Scholar]
  12. Gray, R. O., & Corbally, C. J., 1998, AJ, 116, 2530 [NASA ADS] [CrossRef] [Google Scholar]
  13. Gray, R. O., & Corbally, C. J. 2002, AJ, 124, 989 [NASA ADS] [CrossRef] [Google Scholar]
  14. Gray, R., Riggs, Q., Koen, C., et al. 2017, AJ, 154, 31 [NASA ADS] [CrossRef] [Google Scholar]
  15. Heiter, U. 2002, A&A 381, 959 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  16. Jura, M. 2015, AJ, 150, 166 [NASA ADS] [CrossRef] [Google Scholar]
  17. Kama, M., Folsom, C. P., & Pinilla, P. 2015, A&A, 582, L10 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  18. Kamp, I., & Paunzen, E. 2002, MNRAS 335, L45 [NASA ADS] [CrossRef] [Google Scholar]
  19. Kamp, I., Iliev, I. Kh., Paunzen, E. et al. 2001, A&A, 375, 899 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Khan, S., & Shulyak, D. 2007, A&A, 469, 1083 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  21. Kunitomo, M., Guillot, T., Ida, S., & Takeuchi, T. 2018, A&A, 618, A132 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  22. Kurucz, R. L. 1993, ATLAS9 Stellar Atmosphere Programs and 2 km/s grid, Kurucz CD-ROM 13 (Cambridge, MA: Smithsonian Astrophysical Obs.) [Google Scholar]
  23. Kurucz, R. L., & Avrett, E. H. 1981, SAO Special Report No. 391 [Google Scholar]
  24. Martinez-Galarza, J., Kamp, I., Su, K. Y., et al. 2009, ApJ, 694, 165 [NASA ADS] [CrossRef] [Google Scholar]
  25. Martos, G., Meléndez, J., Rathsam, A., et al. 2023, MNRAS, 522, 3217 [NASA ADS] [CrossRef] [Google Scholar]
  26. Murphy, S. J., & Paunzen, E. 2017, MNRAS, 466, 546 [NASA ADS] [CrossRef] [Google Scholar]
  27. Murphy, S., Corbally, C., Gray, R., et al. 2015, PASA, 32, e036 [NASA ADS] [CrossRef] [Google Scholar]
  28. Murphy, S., Moe, M., Kurtz, D., et al. 2018, MNRAS, 474, 4322 [NASA ADS] [CrossRef] [Google Scholar]
  29. Murphy, S., Gray, R., Corbally, C., et al. 2020a, MNRAS, 499, 2701 [NASA ADS] [CrossRef] [Google Scholar]
  30. Murphy, S., Paunzen, E., Bedding, T. R., et al. 2020b, MNRAS, 495, 1888 [NASA ADS] [CrossRef] [Google Scholar]
  31. Paunzen, E. 2001, A&A, 373, 633 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  32. Paunzen, E., Heiter, U., Handler, G., et al. 1998, A&A, 329, 155 [NASA ADS] [Google Scholar]
  33. Paunzen, E., Kamp, I., Iliev, I. Kh., et al. 1999, A&A, 345, 597 [NASA ADS] [Google Scholar]
  34. Paunzen, E., Duffee, B., Heiter, U., et al. 2001, A&A, 373, 625 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  35. Paunzen, E., Handler, G., Weiss, W. W., et al. 2002, A&A, 392, 515 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  36. Paunzen, E., Fraga, L., Heiter, U., et al., 2012a, From interacting binaries to exoplanets: essential modeling tools, IAU Proceeding Symposium 282, eds. M. Richards, & I. Hubeny, https://doi.org/10.1017/S1743921311027773 [Google Scholar]
  37. Paunzen, E., Heiter, U., Fraga, L. et al., 2012b, MNRAS, 419, 3604 [NASA ADS] [CrossRef] [Google Scholar]
  38. Paunzen, E., Iliev, I., Fossati, L., et al. 2014, A&A, 567, A67 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  39. Piskunov, N., & Kupka, F. 2001, AJ, 547, 1040 [NASA ADS] [CrossRef] [Google Scholar]
  40. Rentzsch-Holm, I. 1996, A&A, 312, 966 [NASA ADS] [Google Scholar]
  41. Saffe, C., Jofré, E., Martioli, E., et al. 2017, A&A, 604, L4 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  42. Saffe, C., Flores, M., Miquelarena, P., et al. 2018, A&A, 620, A54 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  43. Saffe, C., Jofré, E., Miquelarena, P. et al. 2019, A&A, 625, A39 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  44. Saffe, C., Miquelarena, P., Alacoria, J., et al. 2020, A&A, 641, A145 [EDP Sciences] [Google Scholar]
  45. Saffe, C., Miquelarena, P., Alacoria, J., et al. 2021, A&A, 647, A49 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  46. Saffe, C., Alacoria, J., Miquelarena, P., et al. 2022, A&A, 668, A157 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  47. Saffe, C., Miquelarena, P., Alacoria, J., et al. 2024, A&A, 682, L23 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  48. Schlegel, D., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525 [Google Scholar]
  49. Sitnova, T., Mashonkina, L., & Ryabchikova, T. 2013, Astron. Lett., 39, 126 [NASA ADS] [CrossRef] [Google Scholar]
  50. Soares, B., Adibekyan, V., Mordasini, C., et al. 2025, A&A, 693, A47 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  51. Steinmetz, M., Guiglion, G., McMillian, P. J., et al. 2020, AJ, 160, 83 [NASA ADS] [CrossRef] [Google Scholar]
  52. Stürenburg, S. 1993, A&A, 277, 139 [NASA ADS] [Google Scholar]
  53. Van Winckel, H., Waelkens, C., & Waters, L. B. F. M. 1995, A&A, 293, L25 [NASA ADS] [Google Scholar]
  54. Venn, K., & Lambert, D. 1990, ApJ, 363, 234 [Google Scholar]
3

On loan from the Institute d’Astrophysique de Liege, Belgium.

6

We adopt a minimum of 0.01 dex for the errors e2, e3, and e4.

All Tables

Table 1

Visual binary systems that include a λ Boo star component (the first star listed), identified with Gaia eDR3 data.

In the text
Table 2

Magnitudes and astrometric data for the stars studied in this work.

In the text
Table 3

Fundamental parameters derived for the stars in this work.

In the text
Table A.1

Chemical abundances for the star HD 87304.

In the text
Table A.2

Chemical abundances for the star CD-33 6615B.

In the text
Table A.3

Chemical abundances for the star HD 98069.

In the text
Table A.4

Chemical abundances for the star UCAC4 431-054639.

In the text
Table A.5

Chemical abundances for the star HD 153747.

In the text
Table A.6

Chemical abundances for the star TYC 7869-2003-1.

In the text

All Figures

thumbnail Fig. 1

Observed, synthetic, and difference spectra (black, blue dotted, and magenta lines) for some stars in our sample.
In the text
thumbnail Fig. 2

Comparison of [Fe/H] values derived in this work with those from literature.
In the text
thumbnail Fig. 3

Observed spectra (black line) and synthetic spectra (blue dotted line) near the Li line 6707.8 Å in the star CD-33 6615B. The synthetic lines are indicated showing the wavelength, atomic number, and intensity.
In the text
thumbnail Fig. 4

Chemical pattern of the stars CD-33 6615B and HD 87304 (left and right panels), compared to an average pattern of λ Boo stars (blue).
In the text
thumbnail Fig. 5

Chemical pattern of the stars UCAC4 431-054639 and HD 98069 (left and right panels), compared to an average pattern of λ Boo stars (blue).
In the text
thumbnail Fig. 6

Chemical pattern of the stars TYC 7869-2003-1 and HD 153747 (left and right panels), compared to an average pattern of λ Boo stars (blue).
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.