Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://mavr.sao.ru/hq/lizm/conferences/pdf/2003/2003_p191.pdf Äàòà èçìåíåíèÿ: Wed Apr 21 16:48:55 2010 Äàòà èíäåêñèðîâàíèÿ: Sat Jun 26 16:24:14 2010 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: partial solar eclipse

Magnetic stars, 2004, 191-197

Spectroscopic study of the Am SB2 eclipsing binary HR 6611
z Mikul´ Z.1 , Zverko J.2 , Romanyuk I. I.3 , Zinovsky J.2 , Elkin V. G.3 , Kudryavtsev D. O. asek ´
1 2 3

3

Institute of Theoretical Physics and Astrophysics, Masaryk University, Kotl´ a 2, 611 37 Brno, Czech Republic arsk´ Astronomical Institute, Slovak Academy of Sciences, 059 60 Tatransk´ Lomnica, Slovakia a Special Astrophysical Observatory of Russian AS, Nizhnij Arkhyz 369 167, Russia

Abstract. Detailed sp ectroscopic study confirmed a mild Am-p eculiarity of b oth comp onents of the double-lined sp ectroscopic eclipsing binary HR 6611 and revealed a slightly asynchronous rotation of the primary one. The analysis of Zeeman sp ectra of the binary and 4 non-magnetic stars, do es not exclude an effective magnetic field on HR 6611A, varying cyclically with a semi-amplitude of 0.2 kG, but definitely not with the orbital p erio d 3.895 d. Key words: binaries: eclipsing ­ stars: individual: HR 6611 ­ stars: abundances ­ stars: magnetic fields

1

Intro duction, motivation

HR 6611 (HD 161321, BD+143329, ADS 10749, V624 Her, HIP 86809) was recognized as a double-lined spectroscopic binary by Petrie (1928). Zissel (1972), based on his own photoelectric observations in V color, discovered shallow eclipses and derived light elements with the orbital period P = 3.895 d. He also recomputed the spectroscopic orbit and determined masses and radii of the components. He inspected original spectra as well new ones and noted the spectra of both the components of the binary to be Am. Popper (1984) using an additional set of high-dispersion spectrograms reanalyzed Zissel's (1972) curves and yielded improved parameters of the components. He claimed: the more massive component of V624 Her appears to be more evolved in the mass-radius, temperature-gravity, and HR diagram than any Am star that has al l properties wel l determined. Bertaud & Floquet (1974) summarized spectral types as Am, A3­A7, A3m and A3pm and Ribas et al. (1998) give A3m + A7 V. Cowley et al. (1969) alleged that the metallicity of the primary component was not particularly well developed, while Lacy et al. (2002), comparing WW Cam with HR 6611, even noted that HR 6611 had no observational determination of metallicity but abundances could not be far from solar. The flagrant disagreement in the evaluation of the degree of the chemical peculiarity of both components shows that the detailed abundance study of the binary is very appealing. Babcock (1958) listed HR 6611 among 66 stars that probably but not definitely show the Zeeman effect. Recently, Elkin et al. (2002) investigated the fine structure of the famous 5200 ° depression with the A 1 m telescope of the Special Astrophysical Observatory. An unidentified feature 5150 ° has proven to be A a relatively reliable indicator of the magnetic field presence. The detail was in the spectrum of HR 6611 relatively strong indicating the possible presence of the field. These two reasons induced us to start obtaining high-quality spectra of the star with the Zeeman analyzer attached to the 6 m telescope of the Special Astrophysical Observatory of the Russian Academy of Sciences, Nizhnij Arkhyz.
c Sp ecial Astrophysical Observatory of the Russian AS, 2004


192

´ ´ MIKULASEK, ZVERKO, ROMANYUK, ZIZNOVSKY ELKIN, KUDRYAVTSEV

2

Observations

Our spectroscopic investigation of HR 6611 was based on the analysis of 23 Zeeman spectra obtained with CCD cameras in the Main Stellar Spectrograph of the 6 m telescope from February 2002 to July 2003. The CCD1 camera (1) 1160 â 1040, pixel size 16 µm, gave spectra with R 20 000 and S/N better than 100 within (4445­4605) ° The CCD2 camera (2) 2050 â 2050, pixel size 16 µm, gave spectra R 20 000 and S/N better A. than 100 within (4405 - 4645) ° The observations were carried out with the achromatic circular polarization A. analyzer (Na jdenov & Chuntonov 1976). The technique of the magnetic observations was described in details in Romanyuk et al. (1998). The ESO MIDAS (Kudryavtsev 2000) and NICE software package (Knyazev & Shergin 1995) were used to reduce the spectra. The Zeeman spectra of comparison non-magnetic standards Procyon, Arcturus, Mirfak ( Persei) and HD 158974 were exposed immediately before or after the exposure of the binary. Table 2 summarizes the spectroscopic material used.

3
3.1

Sp ectral analysis
Parameters of the system

The eclipsing binary HR 6611 is a well detached system of two practically spherical stars orbiting around their center of gravity on nearly circular orbits. According to Popper (1984) the inclination of the orbital plane is i = (79.4 ± 0.2) , the masses and radii of the primary (A) and secondary (B) components are: MA = (2.27 ± 0.014) M , MB = (1.87 ± 0.013) M , and RA = (3.03 ± 0.03) R , RB = (2.21 ± 0.03) R . The eclipses are partial and very shallow. For the orbital phase we adopted Zissel's (1972) light elements of the primary minimum: J Dmin = 2 440 321.005 + 3.894 977E .

3.2

Atmospheres

As a base for the choice of stellar atmosphere models of components and computing of synthetic spectra we adopted the physical parameters of both the components derived by Popper (1984). Combining spectrophotometric and photometric methods developed for eclipsing binaries, Popper (1984) derived the effective temperatures and surface gravities (in c.g.s.) Teff = 8150 K, log g = 3.83 and Teff = 7950 K, log g = 4.02 for the components A and B, respectively, and the luminosity ratio: LB /LA = 0.48. The line list was extracted from the VALD2 database (Kupka et al. 1999, Ryabchikova et al. 1999). We interpolated the models of stellar atmospheres in the Kurucz tables (Kurucz 1993) for the parameters given above and used SYNSPEC code (Hubeny 1987, Krti a 1998) to compute theoretical spectra for each ´ ck component. The value of the microturbulence giving the best agreement for both strong and weak spectral lines, vturb = 4.5 km s-1 , is fairly high and the same for both components.

3.3

V sin i and the character of rotation of comp onents

The value of pro jected equatorial rotational velocity V sin i of each component was estimated by comparing the model profiles with the observed ones of a well defined Fe ii 4508.2 ° line. A The best fit computed and observed line profiles in disentangled spectra was reached for V A sin i = 35 km s-1 for the A and VB sin i = 29 km s-1 for the B components. The uncertainty of these values was not larger than 1 km s-1 . The corresponding values of V sin i for synchronously rotating components with rotational axes oriented perpendicularly to the orbital plane are, however, 38.7 km s-1 and 28.2 km s-1 for the A and B components, respectively. While the less massive component is well synchronized, the rotation of the more massive component seems to be apparently slower than the synchronized one. The surprising asynchronism could be removed only by admitting a non-perpendicularity of the rotational axis or due to overestimation of the star's radius by Popper by some 10%. Both the admissions appear to be unrealistic, and so we conclude that very probably the rotational period of the A component is P rotA 4.3 days and it is larger than the orbital period 3.895 d. The asynchronous rotation of the primary component could be naturally explained by the fact that the primary component has recently entered into the epoch of fast reconstruction of the inner and outer parts of its interior, which necessarily results in changes of the rotational period. As the binary is not very close, the


SPECTROSCOPIC STUDY OF THE AM SB2 ECLIPSING BINARY HR 6611

193

1.05 1 0.95 0.9 Relative intensity 0.85 0.8 0.75 0.7 0.65 0.6 0.55 4507 4507.5 4508 4508.5 Wavelength [A] 4509 4509.5 Fe II 4508.2

Figure 1: Comparison of the observed and computed profiles of the Fe ii 4508.2 ° line. V sin i: ful l line ­ A 39 km s-1 ; broken line ­ 35 km s-1 ; dotted line ­ 29 km s-1 ; asterisks ­ observed. tidal interaction between the components is not effective enough to reestablish the perfect synchronization of the system.

3.4

Abundances

The synthetic spectra of each component were added up for each orbital phase using the luminosity ratio mentioned above and the ephemeris given in Zissel's paper (1974). By fitting the computed and observed lines and repeating the procedure we received abundances of chemical elements in each of the component. In the component A, the underabundant elements are Si and Sc (by a factor of 0.01), C (0.05), O (0.15) and Ca (0.5), while overabundant elements are: Cr and Fe (2), Mg (3), Ti (4), Na, Ni and Zr (5), Ba (14) and Y (20). In the component B, the underabundant elements are: Sc (0.03), Ca (0.07), C (0.3), Mg (0.4), Si (0.5) and Ti (0.7). The overabundant elements in the B component are: Cr and Fe (2), Ni and Ba (6) and Na (8). The value derived for Ti in the B component indicates the inaccuracy of determination rather than the real underabundance. In general, the abundances derived agree well with the data available in the literature (Conti 1970) and thus the Am classification of both the binary components is confirmed.

4
4.1

Search for magnetic field
Measuring Zeeman shifts

The synthetic spectrum procedure allowed us to resolve and identify particular spectral lines of both components in the spectrum of the binary. Having the lines identified, we selected those of them that were not seriously blended with other lines in the same component neither were contaminated by lines of the other component pro jected onto the same wavelength. Unfortunately, there are not too many lines complying with these criteria. Moreover, the spectra of both components are practically identical due to the close atmospheric parameters, and that one taken near the primary minimum ( = 0) does not contain any uncontaminated line. The selected spectral lines used for measuring of Zeeman splits and their Land´ factors are listed in e Table 1. The Land´ factors are from Beckers (1969). e We used 4 to 18 (typically 13) lines in the component A and 2 to 13 (typically 7) lines in the component B.


194

´ ´ MIKULASEK, ZVERKO, ROMANYUK, ZIZNOVSKY ELKIN, KUDRYAVTSEV

Table 1: List of spectral lines and their Land´ factors. e 4415.123 4427.310 4447.717 4450.482 4464.450 4466.552 4472.929 4491.405 4501.273 4508.288 4515.339 4520.245 4522.634 4541.524 ion+mult. Fe ii (41) Fe ii (2) Fe ii (68) Ti ii (19) Ti ii (40) Fe i (350) Fe ii (37) Fe ii (37) Ti ii (31) Fe ii (38) Fe ii (37) Fe ii (37) Fe ii (38) Fe ii (38) geff 1.167 1.500 2.000 1.029 0.333 1.167 1.500 0.400 0.929 0.500 1.029 1.500 0.900 0.800 4554.029 4555.893 4558.650 4563.761 4565.740 4571.968 4576.340 4583.837 4588.199 4604.982 4620.521 4629.339 4634.070 ion+mult. Ba ii (1) Fe ii (37) Cr ii (44) Ti ii (50) Cr ii (39) Ti ii (82) Fe ii (38) Fe ii (38) Cr ii (44) Ni i (98) Fe ii (38) Fe ii (37) Cr ii (44) geff 1.167 1.238 1.167 0.833 0.600 0.944 1.200 1.167 1.071 1.855 1.333 1.333 0.500

The central wavelengths of selected lines in the right circularly polarized (R) and left circularly polarized (L) strips of the Zeeman spectra were derived by fitting their profiles with a gaussian. If the Zeeman analyzer operates properly, the shift = R - L will depend only on the product of the longitudinal component of the total magnetic field of the star Beff (effective magnetic field) and the Land´ factor geff specific to the e particular line: = 9.34 · 10
-10

g

eff

2 B

eff

= 0.0193 (/4543)2 g

eff

Beff ,

(1)

° ° where is the wavelength of the line in A and Beff in kG, 4543 A being the center of the spectral region ° studied. As the standard deviation of the shift measurements for the A and B components are 0.007 A and 0.009 ° respectively, we should expect the typical inner uncertainty of the determination of the effective field A, of the A and B components to be 0.10 kG and 0.18 kG, respectively. All measured R­L shifts were corrected for some instrumental shifts which occurred to be both timeand wavelength-dependent. The course of the dependence was determined by means of the R­L shifts of 33 spectrograms of nonmagnetic sharp-lined standards: Arcturus (19), Mirfak ( Persei) (8), Procyon (3) and HD 158974 (3).

4.2

Effective magnetic field of HR 6611 A, B

The search for the effective magnetic field for the A and B components of HR 6611 consists in determination of 23+21=44 values of the effective magnetic field and 15 coefficients describing the behavior of instrumental shifts. This was done simultaneously for all the measured R­L shifts (1080 shifts) by the robust regression (Mikul´ et al. 2003) that effectively eliminates the influence of outliers. asek Using the measurements of the R­L shifts on the 33 comparison Zeeman spectra and the 23 Zeeman spectra of the double-lined spectroscopic binary HR 6611, we find information about the effective magnetic field of both components and its variation, see Table 2. Some values of BeffB of the B component suffer from their uncertainty caused by the small number of the lines measured and the bad phase distribution. The data display an extensive scatter with the center at BeffB = (-0.03 ± 0.15) kG, with a standard deviation of 0.55 kG. The semi-amplitude of sine variations of the effective magnetic field is (0.0 ± 0.25) kG, see Fig. 2. Information on the magnetic field of the primary component is more complete. Twenty-three fairly accurate measurements (the typical inner uncertainty 0.10 kG) are satisfactorily well distributed on the orbital phase diagram, so we can test the presence of potential orbital phase variations of the field (see Fig.3(a)). Fig. 3(a) tellingly displays that the effective magnetic field of the A component very probably does not vary with the orbital period (the amplitude is (0.12 ± 0.16) kG). Nevertheless, the fact that the typical uncertainty of a single determination of the effective magnetic field is more than two times smaller than the standard deviation of them, std(BeffA ) = 0.22 kG, suggests that another period may be in play, namely, the true rotational period. We have attempted to search for possible better periods that may be candidates for


SPECTROSCOPIC STUDY OF THE AM SB2 ECLIPSING BINARY HR 6611

195

2 1.5 Beff in kG 1 0.5 0 -0.5 -1 -1.5 -2 -2.5 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

orbital phase

Figure 2: Dependence of the effective magnetic field of the B component on the orbital phase (P = 3.895 d).

0.6

0.6 P = 4.248 d

Beff in kG

0.4

0.4

0.2

Beff in kG

0.2

0

0

-0.2

-0.2

-0.4

-0.4

-0.6

-0.6

-0.8 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

-0.8 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

orbital phase

rotational phase

(a)

(b)

Figure 3: Dependence of the effective magnetic field of the A component on (a) the orbital phase (P = 3.895 d) and (b) the possible rotational phase (P = 4.248 d, J Dmax = 2 442 755.1).


196

´ ´ MIKULASEK, ZVERKO, ROMANYUK, ZIZNOVSKY ELKIN, KUDRYAVTSEV

Table 2: List of Zeeman spectra of HR 6611. Effective magnetic field measured. 1 ­ number of exposition, 2 ­ the number of CCD camera used, 3 ­ J Dhel - 2 450 000 of the center of exposition, 4 ­ orbital phase, 5 ­ effective magnetic field of A component, 6 ­ effective magnetic field of B component, both in kG 1 2227 2522 2628 2701 2702 2926 3120 3121 3217 3218 3614 3614 3616 3709 3709 3710 3710 3711 3711 3713 3713 3714 3714 2 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 333.6160 417.3721 454.4531 457.3948 458.3795 544.3194 660.6354 661.6285 688.5688 689.5323 805.2944 805.3014 807.3243 830.3535 830.3604 831.3181 831.3250 832.4444 832.4514 834.3806 834.3875 835.3333 835.3403 4 0.130 0.621 0.153 0.908 0.161 0.225 0.088 0.343 0.260 0.507 0.228 0.230 0.749 0.661 0.663 0.909 0.911 0.199 0.201 0.695 0.698 0.940 0.942 5 -0.68 ± -0.09 ± +0.01 ± +0.28 ± +0.30 ± -0.45 ± -0.20 ± +0.22 ± -0.32 ± -0.13 ± -0.13 ± -0.04 ± -0.21 ± +0.08 ± +0.01 ± +0.29 ± -0.01 ± +0.20 ± +0.12 ± +0.01 ± -0.03 ± -0.10 ± +0.14 ± 0.09 0.17 0.13 0.12 0.10 0.11 0.12 0.09 0.10 0.08 0.07 0.08 0.10 0.07 0.07 0.08 0.08 0.07 0.08 0.07 0.08 0.06 0.07 6 -1.35 ± 0.15 +1.13 +1.63 +0.53 -1.83 +0.10 +1.27 -0.18 +0.18 -0.05 -0.63 +0.35 -0.21 -0.32 -0.05 +0.18 -0.14 +0.04 -0.22 -0.09 +0.37 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 28 21 18 21 15 25 20

04 05 08 18 19 10 11 02 03 02 03 02 03

12 11 11 10 13 11 11 12 13 10 12 09 09

the rotational period of the primary and have found the best solution for the elements: P = (4.248 ± 0.006) d, J Dmax = (2 442 755.1 ± 0.2) (Fig. 3(b)). The semi-amplitude of variations is (0.19 ± 0.06) kG, the mean value is very close to zero: (-0.02 ± 0.04) kG. Adopting Popper's values of the radius of HR 6611 A: RA = (3.03 ± 0.03) R , and of the inclination angle of its rotational axis i = (79.4 ± 0.2) , we arrived for the possible rotational period P = (4.248 ± 0.006) d, to the VA sin i = (35.5 ± 0.4) km s-1 , which is practically identical with its observed value of 35 km s-1 . Nevertheless, to prove or disprove this potential rotational period in question we need definitely to check it by further observational data taken preferably from another observatory. The problem is of high theoretical importance, as no Am star with a variable effective magnetic field and asynchronous rotation has been known up to now.

5

Conclusions

Am stars are known as non-magnetic CP stars, which has recently been confirmed by Shorlin et al. (2002) in their highly sensitive search for magnetic fields in B, A and F stars and Monin et al. (2002) in their magnetic survey of bright northern main sequence stars, thus negating all the former positive indications above the observational accuracy limits. On the other hand, the classical understanding of non-magnetic, non-variable CP1 (Am) and CP3 (Hg-Mn) stars became questioned with finding by Adelman et al. (2002) a mercury spot on the HgMn star And. Our preliminary spectroscopic study has focused on a bit unusual couple of Am stars creating the SB2 eclipsing binary HR 6611. The exclusiveness of the system refers namely to its primary component which seems to be a star entering the terminal stage of its main sequence life. We confirmed the Am peculiarity of both binary components and revealed the asynchronous rotation of the more massive star, which is most likely a consequence of its proposed fast period of evolution. The measurements of effective magnetic field done by the Zeeman analyzer of the 6 m telescope of SAO AS, Russia, may admit the presence of a variable


SPECTROSCOPIC STUDY OF THE AM SB2 ECLIPSING BINARY HR 6611

197

effective magnetic field in the A component with an amplitude of about 0.4 kG. Up to now we have not known reliably the rotational period and hence the period of possible magnetic variations, but it is definitely not the orbital one. To test the presence of a measurable magnetic field in the Am component HR 6611A and to find its rotational period, we have to collect more Zeeman spectra, but the effort will surely be paid in return.
Acknowledgements. This work was partly supported by the Grant Agency of the Slovak Academy of Sciences, Vega No.3014 and by the Bilateral Czech-Slovak Co-operation in Research, the pro ject No.2002/131(54). Partial supports were provided by the grant No.205/02/0445 of the Grant Agency of the Czech Republic and of the grant APVT No. 51-802 of the Science and Technology Assistance Agency of the Slovak Republic. We acknowledge the availability of the VALD2 database.

References
Adelman S. J., Gulliver A. F., Kochukov O. P., Ryabchikova T., 2002, Astrophys. J., 575, 449 Babcock H. W., 1958, Astrophys. J. Suppl. Ser., 3, 141 Beckers J. M., 1969, AFCRL-69-0115, Physical Sci. Research Papers, No. 371 Bertaud Ch., Floquet M., 1974, Astron. Astrophys. Suppl. Ser., 16, 71 Conti P., 1970, Publ. Astr. Soc. Pacific, 82, 781 Cowley A., Cowley C., Jaschek M., Jaschek C., 1969, Astron. J., 74, 375 Elkin V. G., Kudryavtsev D. O., Romanyuk I. I., 2002, Pis'ma Astron. Zh., 28, 195 Hubeny I., Lanz T., Jeffery C. S., 1987, in: News Letters on Analysis of Astronomical Spectra, No. 20, ed. C. S. Jeffery, ´ (CCP7; St. Andrews Univ.), 30 Knyazev A. Yu., Shergin V. S., 1995, SAO Technical Report No. 239 Krti a J., 1998, Proc. of the 20th Stellar Conference, eds.: J. Du & M. Zejda, Brno, 73 ck sek Kudryavtsev D. O., 2000, in: Magnetic fields of Chemically Peculiar and Related Stars, eds.: Yu. V. Glagolevskij & I. I. Romanyuk, (Spec. Astrophys. Obs., Moscow), 84 Kupka F., Piskunov N. E., Ryabchikova T. A., Stempels H. C., Weiss W. W., 1999, Astron. Astrophys. Suppl. Ser. 138, 119 Kurucz R. L., 1993, CD ROM13 Lacy C. H. S., Torres G., Claret A., Sabby J. A., 2002, Astron. J., 123, 1013 z Mikul´ Z., Zinovsky J., Zverko J., Polosukhina N. S., 2003, Contr. Astr. Obs. Sk. Pleso, 33, 29 asek ´ Monin D. N., Fabrika S. N., Valyavin G. G., 2002, Astron. Astrophys., 396, 131 Na jdenov I. D., Chuntonov G. A., 1976, Soobshch. Spec. Astrofiz. Obs., 16, 63 Petrie R. M., 1928, Publ. Dominion Astrophys. Obs., 4, 81 Popper D. M., 1984, Astron. J., 89, 1057 Ribas I., Chim´ enez A., Torra J., Jordi C., & Oblak E., 1998, Astron. Astrophys., 330, 600 Romanyuk I. I., Elkin V. G., Kudryavtsev D. O., Landstreet J. D., & Wade G. A., 1998, Bull. Spec. Astrophys. Obs., 45, 93 Ryabchikova T. A., Piskunov N. E., Stempels H. C., Kupka F., & Weiss W. W., 1999, Phys. Scr., 83, 162 Shorlin S. L. S., Wade G. A., Donati J.-F., Landstreet J. D., Petit P., Sigut T. A. S. A., Strasser S., 2002, Astron. Astrophys., 392, 652 Zissel R., 1972, Astron. J., 77, 610