Binary Stars in the Orion OB1 Association, Subgroup a
E.A. Semenko, I.I. Romanyuk, I.A. Yakunin, D.O. Kudryavtsev, A. V. Moiseeva
aa r X i v : . [ a s t r o - ph . S R ] M a y Binary Stars in the Orion OB1 Association, Subgroup a E. A. Semenko, I. I. Romanyuk, I. A. Yakunin, D. O. Kudryavtsev, andA. V. Moiseeva
Special Astrophysical Observatory of the Russian Academy of Sciences, NizhnyArkhyz, Russia; [email protected]
Abstract.
A detailed spectroscopic survey of chemically peculiar (CP) stars in theOrion OB1 association is the most comprehensive observational program in the field ofstellar magnetism that has been carried out at SAO so far. As at the end of 2018, wehave completed surveying CP stars in the oldest subgroup a of the association. In thispaper we give a short overview of CP members of the subgroup showing both direct andindirect signatures of multiplicity. Among the overall 11 stars which had been classifiedas peculiar, we found several good candidates for further detailed study.
1. Association Orion OB1
In attempting to explore how the stars evolve, we come to the necessity of surveyingthem in groups like stellar clusters or associations. The main reason for that is in com-mon genesis of the stars and thus in high accuracy of an age evaluation. In the case ofhot stars with abnormal chemical composition, which are usually referenced as chem-ically peculiar or CP stars and form about 15% fraction of all the stars of comparabletemperature (Renson & Manfroid 2009), the ability to observe the same mass starsof equal age with and without peculiarities becomes extremely important as the globalprocesses are slow. Atomic di ff usion is an example of such processes. It takes place inthe atmospheres of stars within the wide range of spectral classes where atmosphericlayers are stable. The magnetic field is a substantial stabilizing factor, which is believedto be responsible for the phenomenon of magnetic CP stars. These stars permanentlyhost strong magnetic fields that globally cover the stellar surface, and hence they arethe perfect test beds to study the influence of the magnetic field on atomic di ff usion.In open clusters, Ap / Bp-stars form a small fraction of stellar population. Inspect-ing the WEBDA database , one can note that only few chemically peculiar stars nor-mally occur in a typical cluster. Stellar associations are more populated than individualclusters and the absolute number of stars of interest could reach a dozen or more. Thisfact explains the importance of the associations for statistical studies. Among the near-est stellar associations, the one in Orion is especially attractive. The celestial positionof the Orion OB1 association makes its area completely available for telescopes in bothhemispheres. Moreover, as all the parts of Orion OB1 are located in 300-450 pc fromthe Sun, the vast majority of B-stars there appears brighter than 10 magnitudes andtherefore are available for the moderate aperture telescopes. https://webda.physics.muni.cz a , which is alsothe closest one among four subgroups selected by Blaauw (1964)—the mean distanceis order of 360 pc. From 311 members of the Orion OB1a mentioned by Brown et al.(1994) 24 stars have signatures of chemical peculiarities (Romanyuk et al. 2013). Aswe are interested only in stars hosting the magnetic field, we excluded from further con-sideration the non-magnetic members. Among them there are six Am, two HgMn stars,and one emission-line B-star. The rest fifteen stars were selected for observing with the6-m telescope. A spectropolarimetric survey of CP stars in Orion has been started in2010. By the end of the observational campaign 2017 / a . A paper summarizingthe observational results, occurrence of magnetic field and spatial distribution of stars isunder preparation. In this study, we are examining the radial velocities of CP stars withrespect to the possible presence of unknown binary and multiple stars in our sample.
2. Observational Material
An observational program of CP stars in subgroup a with the Russian 6-m telescope atSAO was accomplished in 2018. This program was carried out in spectropolarimetricmode using the Main Stellar Spectrograph (MSS, Panchuk et al. 2014) installed at theNasmyth focus of the telescope. The working spectral range, with rare exceptions, waslimited to 440–497 nm. In this instance the mean resolving power of spectra R wasabout 15 000. The observational technique and data reduction had been demonstratedin multiple papers (e.g. Kudryavtsev et al. 2006; Romanyuk et al. 2017; Semenko et al.2017).The total number of spectra were measured in this study is indicated for eachobject in column n of a summary Table 1.
3. Techniques of Measuring
Although the region of Orion has been intensively studied for a long time, still there isa lack of information about a multiplicity of Orion’s stars. As the mean distance to theassociation exceeds 350 pc, techniques like speckle-interferometry or adaptive opticslose their e ffi ciency in searches for the new multiple systems. Observing the changesin the stellar radial velocity one can detect the presence of components unseen underother circumstances. Such research was not in our plans, but we decided to checkthe possible presence of yet unknown spectroscopic binaries in our sample using theavailable spectra.The radial velocity can be measured from a displacement of strong spectral lines.In spotted stars, hydrogen lines give more adequate estimates as this element is not suf-fering from an inhomogeneous surface distribution which is typical for iron, chromium,silicon or magnesium. To measure the radial velocity from hydrogen lines, we fitted thecore of the H β line with a Gaussian function.For the stars with a significant number of lines in their spectra, we extracted themean LSD-profiles (e.g. Kochukhov et al. 2010). Measuring the average I Stokes pro-files gives the information about stellar radial velocity and its rotational broadening.All values of the mean projected rotational velocity v sin i in Table 1 were estimatedfrom LSD-profiles unless other was specified. Line masks for LSD were constructedinary Stars inthe Orion OB1Association, Subgroup a Table 1. Summary of results. New binary and possible binary stars are marked byasterisk in a column “ RV ”. Column “ n ” shows the number of measured spectra.HD m V , mag Sp. type π , mas v sin i RV , km s − n ±
10, 176 ± . ± .
4, 26 . ± . ∗ ± . ± . . ± . ± . ± . . ± . ± . ± . ± . ± . ± . ± . ± . ∗ . ± . ± . ± . ± . ∗ ± . ± . ±
11 (syn) 26 . ± .
4. Binary Stars
Looking into the output of NASA / ADS database, one can note that the interest to binaryand multiple systems in Orion is mostly concerned around young stars and protostellarobjects. Along with excess of papers devoted to individual stars the number of obser-vational surveys remains very small. Generally, a study by Morrell & Levato (1991)is the only example of such surveys. Among the 96 most probable members of OrionOB1 that had been studied by the authors, five stars were common in our sample.
Known Systems
For two stars — HD 35008 and HD 35456 — Morrell & Levato (1991) had reportedvariable radial velocities. Hence, in our study these stars are considered as presumablybinary systems. The same authors also recognized HD 35575 as a single-lined binarysystem and found its orbital elements.Our observations of HD 35008 in 2013, 2014, and 2017 showed constant radialvelocity RV = . − , though with large scattering. The latter may be causedby the real changes of the radial velocity, but again this scattering might result from Semenko etalsubstantial spectral variability of the star (Fig. 1). After accumulation of all availablemeasurements of RV for this star, we were failed in constructing of its spectroscopicorbit. N o r m a li ze d i n t e n s it y JD2456319.456JD2456995.579JD2458116.278
Figure 1. Spectral variability of He and Mg lines in HD 35008. Spectra shifted ina vertical direction for demonstration purposes.
HD 35456 in our observations had demonstrated a constant radial velocity RV = . − with a small scatter. This star had been observed with spectropo-larimetry at SAO for five years since 2010. In Morrell & Levato’s paper (1991), theradial velocity of HD 35456 was declared as variable based on 12 measurements. Inparticular, during four days RV grew up from 4 to 24 km s − . HD 35456 is a known bi-nary system with a close three magnitudes fainter companion at 0.7 arcsec away (Balegaet al. 2012).Conclusion about multiplicity of HD 35575 made by Morrell & Levato relies gen-erally on their own observations which had been made during a week in 1980. Proposedorbital solution suggests very short 2.24 days period and an eccentricity e = . Table 2. Radial velocity of HD 35575.HJD, 2450000 + RV , km s − . ± . . . . . . . The last star in the list of known binary and multiple stars is HD 35502. Thismassive magnetic CP star is a member of a triple system. An exhaustive study of thissystem was done by Sikora et al. (2016).inary Stars inthe Orion OB1Association, Subgroup a . . . . . . − R V , k m s − . . . . . . − R V ( O − C ) , k m s − Figure 2. Radial velocity of HD 35575 phased according to ephemerides of binarysystem published by Morrell & Levato (1991). Measurements from Table 2 aredisplayed as filled dots.
New Binary and Possible Binary Stars
Variable radial velocity or a presence of two sets of lines in a composite spectrumwere the reason to classify the stars HD 33917, HD 35730, and HD 36549 as the newspectroscopic binaries.HD33917. The mean profiles extracted from intensity spectra of HD 33917 havea complex triangular shape. From the best fitting solution, we have concluded thatHD 33917 is a possible SB2 system consisting of two stars which though have the simi-lar radial velocities at the moments of observation. This pattern is repeating in all obser-vations. A sample of the mean LSD-profile is presented in Fig. 3. In observations, theradial velocity of broader component remains nearly constant with small changes (seeTable 3). The measured projected rotational velocity averaged over four spectra in thiscase is h v sin i i = . ± . − . A narrow-lined component rotates slower, itsprojected velocity h v sin i i = . ± . − . Significant spectral variability and avariable radial velocity may lead to the higher error in results. Table 3. Radial velocity of HD 33917.HJD, 2450000 + RV , km s − RV , km s − . ± . . ± . . . . . . . . . . . . . Trying to find the other indicators of stellar multiplicity, we have analyzed photo-metric series collected in the All-Sky Automated Survey (ASAS, Pojmanski (1997)). Semenko etal −300 −200 −100 0 100 200 300Velocity, km s −1 M ea n L S D I ObservedFit
Figure 3. Mean LSD intensity profile extracted from a spectrum of HD 33917.
Observations of HD 33917 had been lasting for eight years and thus allowed to find longtrends in data. The longest reliable period that could be derived from ASAS photometryis 2654 days, which is comparable with the length of time series. The amplitude of vari-ability is an order of 10 mmag. Bearing in mind these parameters, we cannot excludean instrumental origin of long term photometric variability in ASAS data, although thelong period agrees well with the observed amplitude of stellar radial velocity.HD35730. This star had already been observed spectroscopically by Morrell &Levato (1991). Six individual observations night by night did not show any changesin stellar radial velocity, it stayed constant around 29 km s − . We have been observingHD 35730 five times from 2010 till 2017. Our results in Table 4 witness the radialvelocity reversing its sign. Apparently, HD 35730 is a new single-lined spectroscopicbinary. Table 4. Radial velocity of HD 35730.HJD, 2450000 + RV , km s − . ± . − . . − . . . . . . Small amount of acquired measurements does not suggest finding of optimal or-bital solution at this time. However, we can select one possible solution with a period P orb = .
86 days and nearly circular orbit. The complete set of orbital parametersis in Table 5. To obtain this solution we use after averaging the original measurementsfrom a paper of Morrell & Levato (1991). In the future we plan to explore the star withnew measurements.inary Stars inthe Orion OB1Association, Subgroup a Table 5. Orbital parameters of HD 35730 in the case of a long orbital period.Parameter Value P (d) 909 . ± . T p (HJD) 2455868 . ± . e . ± . ω (deg) 225 . ± . γ (km s − ) 24 . ± . K (km s − ) 37 . ± . The radial velocity curve computed for the presented orbital solution is plotted inFig. 4. . . . . . . R V , k m s − . . . . . . Orbital phase − R V ( O − C ) , k m s − Figure 4. Radial velocity of HD 35730 according to a solution with a circular orbitand long period. Di ff erent symbols mean the same as in Fig. 2. HD36549. This star has been observed at SAO for five years. In eight spectra wedid not find any signatures of a magnetic field, but the other attributes of CP stars like aspectral variability and abnormal intensity of lines were clearly seen.On six unevenly spaced nights, the measured radial velocity of HD 36549 rangedfrom − . − (Table 6). At two moments separated by three years, themeasurements gave exactly the same value of RV . All together, these facts let us to putlimits on the orbital parameters of the system. In our vision, HD 36549 is a single-linedbinary system with a nearly circular orbit ( e = P orb whichis a factor of 1095.92 days. At the moment, the available number of measurements isinsu ffi cient to make the final decision about the correct value of P orb . Semenko etal Table 6. Radial velocity of HD 36549.HJD, 2450000 + RV , km s − . ± . . . . . . . . . − . . − . . − . .
5. Rotational Velocities
A significant fraction of stellar population in associations consists of the early B and O-stars which typically rotate fast. However, in our sample the fastest rotators are amonglate B-stars with He-wk or Si anomalies. When the projected velocity v sin i approachesto 200 km s − the modeling becomes needed to measure stellar rotation. This could be atricky problem because the number of lines decreases as the e ff ective temperature goeshigher. That is why we used the di ff erent methods for slowly or moderately rotatingstars and for those with v sin >
200 km s − .The rotational velocity of HD 35008 was found from the best fit of Mg and Helines. At some moment, the lines of He i ii − , HD 294046 rotates slower ( v sin i =
129 km s − ). Theoutstanding object in our sample is HD 35881, a late B-star with the rotational veloc-ity v sin i =
335 km s − . For the rest stars the rotational velocities are well distributedwithin 20–100 km s − with a slow trend to higher values.
6. Conclusions
In the current study we have analyzed a sample of chemically peculiar members of theOrion OB1 association, subgroup a . The conclusion about star’s membership relies ona fundamental paper by Brown et al. (1994). The measured stellar radial velocities gen-erally confirm that the stars belong to Orion OB1 as their values are close to 23km s − ,the mean radial velocity of the association itself. Also, ignoring the di ff erence in RV forsome newly discovered spectroscopic binary systems, their systemic velocity is agreedwell with other association members.Our sample contains only potentially magnetic CP stars, excluding HgMn andAm. The presence of slow rotators without measurable magnetic field in the samplenonetheless make possible the wrong classification and thus the stars HD 35730 andHD 36549 might be in fact HgMn.Before this work was started, only four binary or multiple stars out of 15 in thesample had been known. We have shown that at least three another stars have a variableradial velocity that allows to classify them as new spectroscopic binaries. So far, theinary Stars inthe Orion OB1Association, Subgroup a Acknowledgments.
The authors acknowledge the Russian Foundation for BasicResearch for partial financial support of this study (grant No.18-52-06004 Az_a). Thisresearch has made use of the WEBDA database, operated at the Department of Theo-retical Physics and Astrophysics of the Masaryk University. Observations at the SAORAS telescopes are supported by the Ministry of Science and Higher Education of theRussian Federation.