Фюъѓьхэђ тчџђ шч ъ§јр яюшёъютющ ьрјшэћ. Рф№хё ю№шушэрыќэюую фюъѓьхэђр : http://www.sao.ru/Doc-en/Science/Public/Conf/magstars-2010/p399.pdf
Фрђр шчьхэхэшџ: Thu Aug 11 13:04:20 2011
Фрђр шэфхъёш№ютрэшџ: Mon Feb 4 16:06:30 2013
Ъюфш№ютър:

Magnetic Stars, 2011, pp. 399 409

Variations of Flux Intensity in Large Features and Sp ectral Lines of 2 CVn in Ultraviolet
Sokolov N. A.
Central Astronomical Observatory at Pulkovo, St. Petersburg, Russia

Abstract. Variation of flux intensity at the cores of large features and spectral lines of the classical magnetic CP star 2 CVn in the ultraviolet spectral region A from 1150 to 3200 А is investigated. This study is based on the archival International Ultraviolet Explorer data obtained at different phases of the rotational cycle. The shapes of two light curves at 1375, and 1415 А at the core of a A large feature at 1400 А significantly differ. The light curve at 1375 А shows A A a similar shape as in the "pseudocontinuum". The same behaviour of the flux is found at the cores of Si II resonance lines at 1260 64 and 1485 А while A, А shows the phase displacement of the flux minimum. the light curve at 1415 A The phase displacement is also present at the cores of the Si II resonance lines at 1304 09 and 1485 А Minimal values of the amplitude of flux variations are A. reached at the cores of large features at 1560 and 1770 А and at the cores of A strong Si II resonance lines. The flux at the cores of a large feature at 1770 А A and Fe II resonance line at 1725 31 А does not vary within measurement errors. A The investigation of the flux variability in the wings of Ly line indicates that the fluxes, which formed in the inner layers of the atmosphere are redistributed into the outer layers of the atmosphere of 2 CVn. Key words: ables: other stars: chemically peculiar stars: individual: 2 CVn stars: vari-

1

Intro duction

The magnetic Chemically Peculiar (mCP) star 2 CVn (HD 112413, HR 4915) displays strong line profile variations in the visual spectral region, attributable to the nonuniform chemical abundance distribution on the stellar surface, particularly of the lines of Fe, Cr and Ti (Khokhlova & Pavlova, 1984), Eu, Cr and Si (Goncharskii et al., 1983) and O (Gonzalez & Artru, 1994). Ryabchikova et al. (1999) report the first identification of the Eu III 6666.347 line in the spectrum of 2 CVn. Later Kochukhov et al. (2002) used a new magnetic Doppler Imaging code to reconstruct the magnetic field geometry and surface abundance distributions for six chemical elements of this star. The study in the ultraviolet has been restricted to a much lower resolution, usually without resolving the individual stellar lines (e. g., Leckrone & Snijders, 1979). Hensberge et al. (1986) identified the ions Mg II, Si II, Cr II, Mn II, Fe II, Fe III, Co II, Ni II, Yb II and W II in the mid ultraviolet spectrum of 2 CVn, which has been obtained with the BUSS (BalloonBorne Ultraviolet Spectrograph). Highdispersion IUE observations of the two resonance lines of Mg II, at 1650
Based on INES data from the IUE satellite

400

SOKOLOV

and 1942 А were investigated by Leckrone (1984) for two normal stars, six HgMn stars and for A, the magnetic variable 2 CVn. The last star possesses moderately strong Hg II resonance lines, confirming its classification as Hgrich, but its magnesium anomaly is less pronounced than that of the MgMn stars. Fuhrmann (1989) investigated the highresolution spectra of the CP star HR 465. For comparison purposes, the spectra of some other stars, including 2 CVn, were discussed as well. The author have noted that the C II resonance doublet at 1334.5 1335.7 А is comparatively weak A 2 CVn. in the spectrum of To quantify the ultraviolet variations of the C IV doublet at 1548, 1550 А Shore et al. (1987) A, have formed the photometric line index, expressed in magnitudes. Later, Sokolov (2000, 2006) has introduced analogous photometric indices in order to derive the variations of the total absorption in the broad features at the farUV spectral region. Unfortunately, these indices depend strongly upon the intensity stability of the near continuum. Based on the fact that the IUE Newly Extracted Spectra (INES) data are presented in absolute units, it is possible to investigate the flux variations at the cores of the large features and spectral lines. Recently, such an attempt has been made by Sokolov (2010) for an mCP star 56 Ari. Another mCP star for which there are enough IUE data in order to investigate the flux variability at the cores of the large features and spectral lines is 2 CVn. This is done in the present paper.

2
2.1

Observational Data and Analysis
IUE Sp ectra

The IUE spectra used in this study are lowresolution echelle spectra obtained with a resolution of about 6 А Additionally, the "rebinned" spectra from highdispersion images of 2 CVn were A. used as well. In all cases, the spectra were obtained through a large aperture (9.5 т 22 ). Finally, we analyzed 22 SWP, 10 LWR and 6 LWP spectra, distributed quite smoothly over the period of rotation. The description of these IUE spectra in detail is done by Sokolov (2011).

2.2

Data Analysis

To analyse the IUE spectra of 2 CVn we used the linearized leastsquares method. An attempt was made to describe the light curves in a quantitative way by adjusting a Fourier series. The method was already applied to the IUE data of 2 CVn and has shown very good descriptions of the monochromatic light curves in the "pseudocontinuum" (Sokolov, 2011). Thus, the observations were also fitted by a simple cosine wave: F (, T ) = A0 () + A1 () cos( T + i ()), (1)

where F (, T ) is a flux for the given , T = T - T0 and = 2 /P . The T0 and P are zero epoch and rotational period of the Farnsworth (1932) ephemeris, respectively. The coefficients A0 () of the fitted curves define the average distribution of energy over the cycle of variability, while the coefficients A1 () define the semiamplitude of the flux variations for the given .

3

Identification of Large Features and Sp ectral Lines in the Sp ectrum of 2 CVn
main absorber with the strong resonance lines of Sokolov, 2011). It should be noted that the the resonance doublet and the auto ionising & Lanz (1987), strong lines in the spectrum of

In the farUV spectral region silicon appears as the at 126064, 130409, 1485, 152633 А (see Fig. 2 A blend at 1304 09 А has two ma jor contributors: A multiplets (Artru & Lanz, 1987). According to Artru

VARIATIONS OF FLUX INTENSITY IN LARGE FEATURES ...

401

-1 A Figure 1: Average distributions of energy in 10-10 erg s-1 cm-2 А of 2 CVn (the solid line) and 56 Ari (the dashed line). The prominent spectral lines and features are shown by vertical and horizontal lines, respectively (see text).

А CP stars appear from C II at 1334 A and Al II at 1671 А which is close to a strong C I at 1657 А A, A line. Fe II form a depression at 1725 31 А and Cr II produces a line at 1434 А Although in the A A. case of 2 CVn the line from C II at 1334 А is not detectable at this resolution. Moreover, the C II A resonance doublet at 1334.5 1335.7 А is comparatively weak at the highresolution spectrum of A this star (Fuhrmann, 1989). In the previous study of the star 56 Ari we identified which elements are responsible for the depressions of flux centred at 2140, 2250, 2540, 2607, 2624 and 2747 А (Sokolov, 2010). The A synthetic spectrum calculation showed that Fe II appears as the main absorber for these depressions. On the other hand, the same synthetic spectrum calculation showed that mainly Fe, Cr and Ni are responsible for the flux depression at 2140 А A comparison of the average energy distribution A. 2 CVn with the average energy distribution of 56 Ari showed that the depressions of the flux of centred at 2140, 2250, 2540, 2607, 2624 and 2747 А are also present in the spectrum of 2 CVn. A It is well known that the Mg II resonance lines at 2795, 2798 and 2803 А are responsible for the A А This depression is not detectable in the lowresolution mode depression of the flux at 2800 A. in the spectrum of 2 CVn. Prominent depressions of the flux in the nearUV spectral region are indicated in Fig. 1 with their identifications. In order to compare the mean energy distributions of 2 CVn and 56 Ari, the fluxes of 56 Ari were increased by the factor of ten in Fig. 1. Four large features at 1400, 1560, 1770 and 2350 2400 А which are strongly enhanced in A, the spectrum of CP stars, are well seen in spectrum of 2 CVn. Lanz et al. (1996) have given strong arguments supporting the idea that the intense autoionisation resonances of Si II could explain the features at 1400 and 1560 А in the spectrum of CP stars. On the other hand, they were unable to A identify the depression of the flux at 1770 А Another element may cause this strong depression. A. A comparison of the IUE high resolution spectrum of 56 Ari with the full synthetic spectrum, as well as those, including the lines from only one element has shown that metal is responsible for the depression of the flux at 1770 А (see Fig. 9 of Sokolov, 2006). It is necessary to note that A

402

SOKOLOV

А Figure 2: The spectral region of Fe II doublet at 1725 31 A of 2 CVn. The rectangle shows the selected spectral region for the calculation of the average intensity at the core of this doublet. this depression is considerably increased in the spectrum of 2 CVn than that in the spectrum of 56 Ari. The large features at 2350 2400 А which are strongly enhanced in the spectrum of A, 2 CVn. The lines of the metal p eak elements have a CP stars, are well seen in the spectrum of particularly important contribution to the opacity at 2350 2400 А Adelman et al. (1993) have A. given strong arguments supporting the idea that a large number of metal lines can explain the feature at 2350 2400 А Many CP stars have here a very pronounced flux depression compared A. to normal stars (Stepien & Czechowski, 1993). Д

4

Average Flux Determination at the Cores of Large Features and Sp ectral Lines in the Sp ectrum of 2 CVn

To measure the absorption at the cores of large features, the spectra were processed using the spectral reduction software spe developed by S. Sergeev at the Crimean Astrophysical Observatory (CrAO). The program allows measuring the average flux intensity and the corresponding error in any selected rectangular spectral region. For the large features at 1560 and 1770 А the spectral A А wide, while for the largest feature at 1400 А two spectral regions were selected regions are 10 A A and centred at 1375 and 1415 А On the other hand, for the large feature at 2350 2400 А the A. A spectral region was used with 50 А wide and centred at 2375 А A A. The resonance doublets of Si II and Fe II lines are seen as depressions in the IUE lowresolution spectrum of 2 CVn, as illustrated by Fig. 2. Thus, the average intensity of the flux at the cores of the spectral lines was computed by averaging five nearest fluxes for a given : F () = 1 5
5

F ( - step З (i - 3)),
i=1

(2)

where step is equal to 1.676 А and 2.669 А for a SWP camera, and for LWR and LWP cameras, A A respectively. As far as the errors in F () are concerned, we computed them by taking into account the errors in the fluxes as presented in the INES Catalog, according to the standard error propagation theory. In order to check the reliability of the average flux intensity at the cores of spectral lines, the spectra were also processed using the spe program. Experience shows that within the measurement errors the average flux intensity computed by the spe program, and using equation (2) are the same.

VARIATIONS OF FLUX INTENSITY IN LARGE FEATURES ...

403

4.1

Variations of Large Features

Left part of Figure 3 exhibits variations of average intensity of the fluxes at the cores of large features versus the rotational phase. First of all, the shapes of two light curves at the core of a large feature at 1400 А significantly differ. The minimum of the light curve at 1375 А is reached A A at phase 0.05 while the minimum of the light curve at 1415 А is at phase 0.26. However, the A amplitudes of two light curves at 1375 and 1415 А are approximately the same and are equal to A -1 0.41 and 0.32 З 10-10 erg s-1 cm-2 А , respectively. According to the model computation of Lanz A et al. (1996), two features at 1400 and 1560 А are connected with the intense autoionisation A resonance of Si II. A qualitative comparison of the light curves at 1375 and 1560 А shows a good A agreement. The minimum of the light curve at 1375 А is reached at phase 0.05 while the minimum A of the light curve at 1560 А is at phase 0.10. Also, the amplitudes of two light curves at 1375 A -1 and 1560 А are approximately the same and are equal to 0.41 and 0.38 З 10-10 erg s-1 cm-2 А , as A A illustrated by Fig. 3. The identifications of two features at 1770 and 2375 А are connected with a large concenA tration of metal lines (see Sect. 3). The flux variations at the cores of the features at 1770 and 2375 А also significantly differ. Thus, the amplitude of the light curve at the core of the feature A -1 А A at 1770 A is equal to 0.11 З 10-10 erg s-1 cm-2 А . Practically, the flux at the core of this feature does not vary within the measurement errors. On the other hand, the amplitude of the light curve -1 A at the core of the feature at 2375 А is equal to 0.43 З 10-10 erg s-1 cm-2 А . However, the shapes A of light curves are approximately the same, as illustrated by Fig. 3. One can see that the behavior of the flux is different at the cores of large features, even if the same element is responsible for the features. This puzzling situation was one of the incentives for studying the variations of the flux at the cores of the Si II resonance lines and the depressions for which a concentration of Fe II lines is responsible.

4.2

Variations of Si I I Resonance Lines

Even in normal stars Si II appears as the main absorber with the strongest resonance lines recognizable at 1260 64, 1304 09, 1485, 1526 33 А (Artru & Lanz, 1987). Right part of Figure 3 A exhibits the flux variations at the cores of these depressions versus the rotational phase. Note that the vertical scales are the same for each part of the figure. One can see from Fig. 3 that the light curves have similar shapes in the cores of resonance lines at 1260 64 and 1526 33 А A. The minima of the light curves at 1260 64 and 1526 33 А are reached at phases 0.05 and A 0.02, respectively. A similar behaviour is revealed by the nearest monochromatic light curves in the "pseudocontinuum" (see Fig. 3 of Sokolov, 2011). Also, the amplitudes of the light curves at the cores of depressions at 1260 64 and 1526 33 А are in good agreement and are equal to 0.49 and A -1 0.63 З 10-10 erg s-1 cm-2 А , respectively. It should be noted that the light curve at 1375 А ws A Asho a similar shape. On the other hand, the minima of the light curves at 1304 09 and 1485 А are A reached at phases 0.12 and 0.30, as illustrated by Fig. 3. However, the minima of the monochromatic light curves are reached at phase 0.0 in the nearest "pseudocontinuum". Moreover, the amplitudes of the light curves are different at 1304 09 and 1485 А and amount to 0.38 and A -1 0.64 З 10-10 erg s-1 cm-2 А , respectively. It should be noted that the light curve at 1415 А eals A Arev a similar shape. The monochromatic light curves in the nearUV spectral region with > 2505 А A in the "pseudocontinuum" also demonstrate such behaviour. The nearUV spectral region is quite important in order to investigate the variability of the flux at the cores of Fe II lines.

404

SOKOLOV

Figure 3: Left: Phase diagrams of the light curves in the cores of large features in -1 10-10 erg s-1 cm-2 А A for 2 CVn. The upper panel represents two phase diagrams centred at 1375 А (crosses) and at 1415 А (circles). 1 error bars accompany each data point. Solid lines A A denote the fit according to equation (2). Right: Phase diagrams of the light curves at the cores of Si II resonance lines.

VARIATIONS OF FLUX INTENSITY IN LARGE FEATURES ...

405

Figure 4: The same as in Fig. 3 for the light curves at the cores of Fe II lines

4.3

Variations of Fe I I Lines

А In the farUV spectral region the Fe II doublet at 1725 31 A gives only one depression of the flux (Artru & Lanz, 1987). A synthetic spectrum calculation shows that Fe II appears as the main absorber of the flux at 2250, 2540, 2607, 2624 and 2747 А (see Sect. 3). Additionally, the flux A А was included in our investigation. On the other hand, two depressions of depression at 2140 A the flux at 2607 and 2624 А have shown the same behaviour. Thus, we have included in our A investigation only the flux depression at 2607 А Figure 4 exhibits the variations of average intensity A. of the flux at the cores of Fe II lines versus the rotational phase. Note that the vertical scales are the same for each part of the figure. As we can see on the graphs of Fig. 4, the variability of the flux at the cores of Fe II lines reveals approximately the same behaviour as the monochromatic light curves with > 2505 А in the "pseudocontinuum" (see. Fig. 3 of Sokolov, 2011). The light curves at the A cores of Fe II lines show a phase displacement of the flux minimum from 0.07 at 2250 А to 0.31 at A А Additionally, the minimum of the light curve at the core of Fe II lines at 2140 А is located 2747 A. A

406

SOKOLOV

A Figure 5: Phase diagrams of the light curves in 10-10 erg s-1 cm-2 А in the wings of Ly line for 2 CVn. Upp er and lower panels show shortwavelength and longwavelength sides of the Ly line, respectively. Solid lines denote the fit according to equation (2). А at phase 0.16. However, the amplitude of the light curve at 2140 A is large enough and is equal -10 erg s-1 cm-2 А-1 . It is due p ossibly to the fact that mainly three chemical elements to 0.69 З 10 A (Fe, Cr, Ni) are responsible for this depression. The monochromatic light curves in the "pseudo continuum" only with > 2505 А demonstrate the same phase displacement of the flux minimum. It A should be noted that the amplitudes of the light curves here are larger than those at the cores of Fe II lines. On the other hand, we have the light curve at the core of Fe II resonance line at 1725 31 А A, -10 erg s-1 cm-2 А-1 . Similar to the situation at the core where the amplitude is equal to 0.13 З 10 A of a large feature at 1770 Аthe flux does not vary at the core of this line either. However, the A variability of the flux in the nearest "pseudocontinuum" is significant. Thus, the amplitudes of the -1 monochromatic light curves at 1690 and 1794 А are equal 0.41 and 0.51 З 10-10 erg s-1 cm-2 А , A A respectively.

-1

4.4

Variations of LymanAlpha Line

Leckrone & Snijders (1979) have compared the Lymanalpha (Ly ) profiles of 2 CVn at two phases, 0.0 and 0.5, using the Copernicus data. The authors have drawn the conclusion that brightness variations at the core of Ly line are anomalous with respect to the adjacent ultraviolet regions. In addition, they noted that the coverage of the complete cycle by a future space instrument will be necessary to establish the specific phasing of these variations. Based on the fact that the star is bright enough (mv = 2.90), the IUE data have allowed to investigate the behavior of the monochromatic light curves in the shortwavelength and longwavelength sides from the core of the Ly line.

VARIATIONS OF FLUX INTENSITY IN LARGE FEATURES ...

407

Several monochromatic light curves in the wings of the Ly line at different wavelengths were formed. Figure 5 exhibits the flux variations in the shortwavelength and longwavelength sides of the Ly line versus the rotational phase. As we can see on the graphs of Fig. 5, the monochromatic light curves have identical shapes in the shortwavelength and longwavelength sides of the Ly line. The light curves at 1187 and 1246 А show the minimum and maximum of the flux at phases A 0.0 and 0.5, respectively. Note that the monochromatic light curves in the "pseudocontinuum" show the same behaviour in the farUV spectral region. These monochromatic light curves are most removed from the line centre of Ly at 1215 А On the other hand, the monochromatic light curves A. at 1202 and 1234 А reveal the flux minimum and maximum at phases 0.5 and 0.0, respectively. A Also, the light curve in the V filter shows the same behaviour in the visual spectral region (Pyper, 1969). At the same time, the variation of the flux at 1190 and 1239 А is practically zero over A the period of rotation. In other words, the socalled "null wavelength regions" are also disposed in these wavelengths. It should be noted that the fluxes at the cores of the Ly line varies with small amplitudes at 1213 А for two CP stars CU Vir and 56 Ari (Sokolov, 2000, 2006). A

5

Discussion

Our results indicate that the variations of the average flux intensity at the cores of large features and spectral lines show different behaviours. First of all, the shapes of two light curves at 1375 and 1415 А at the core of a large feature at 1400 А curves significantly differ. The minimum of the A A light curve at 1375 А is reached at phase 0.05, while the minimum of the light curve at 1415 А is A A located at phase 0.26. However, the difference in wavelength between these spectral regions amounts to 40 А This puzzling situation was one of the incentives for studying the highdispersion spectra of A. 2 CVn in these sp ectral regions. In the top panel of Fig. 6 two highdisp ersion sp ectra are plotted in the spectral region near 1375 А These spectra SWP 27894 and SWP 15828 were obtained at phases A. 0.064 (the dashed line) and 0.524 (the solid line), respectively. From Fig. 6 we see that the fluxes of the spectrum SWP 27894 are systematically lower than the fluxes of the spectrum SWP 15828. In the bottom panel of Fig. 6 two highdispersion spectra are plotted in the spectral region near 1415 А These spectra SWP 04813 and SWP 27880 were obtained at the phases 0.204 (dashed line) A. and 0.702 (full line), respectively. Again, we can see that the fluxes of the spectrum SWP 04813 are lower than the fluxes of the spectrum SWP 27880. The comparisons of the highdispersion spectra in the spectral regions of the broad feature at 1400 А are in agreement with the phase diagrams A obtained from the lowdispersion spectra (see Fig. 3). However, the nature of this behaviour of the fluxes in the spectral regions at 1375 and 1415 А is still unclear. Possibly, the effect of different A species of spectral lines can play some role here. The light curves at the cores of Si II resonance lines at 1260 64 and 1526 33 А show shapes, A similar to the ones revealed in the "pseudocontinuum". At the same time, the light curves at the cores of Si II resonance lines at 1304 09 and 1485 А demonstrate the phase displacement of the A flux minimum. On the other hand, the flux at the cores of a large feature at 1770 А and Fe II A resonance line at 1725 31 А es not vary within the measurement errors. Moreover, the light Ado curves at the cores of Fe II lines show the phase displacement of the flux minimum from 0.07 at 2250 А to 0.31 at 2747 А A A. The vertically dependent abundance stratification in CP stars is suggested by many authors (e. g., Ryabchikova, 2008, and references therein). This effect may influence our results because the effective depth at which the fluxes are formed at < 2000 А can differ from the effective depth at A which the fluxes are formed at > 2505 А We can expect that the flux comes from the upper layers A. of the atmosphere at the cores of some large features and depressions than the flux coming from the nearest "pseudocontinuum" in the spectral region with < 2000 А Therefore, possibly, some large A. features and depressions show the displacement of position of the flux minimum, though the flux

408

SOKOLOV

Figure 6: Two spectral regions of the broad feature at 1400 А for 2 CVn. The highdispersion A spectra in the region of 1375 А (top panel) and the highdispersion spectra in the region of 1415 А A A (bottom panel) (see the text). in the nearest "pseudocontinuum" does not reveal such a displacement. However, the spectrum in the "pseudocontinuum" is also blocked by a great number of spectral lines of various chemical elements. The main thing is that the energy blocking by the silicon boundfree transitions and metal boundbound transitions decreases the flux in the UV spectral region. The blocked flux appears in the visual and red parts of the spectrum. Such an explanation is supported by the antiphase relationship of light curves in the visual and UV spectral regions. Probably, our investigation indicates that not only this mechanism may influence on the redistribution of the flux in atmosphere of 2 CVn. The fluxes, situated in the wings of the Ly line at different distances from its center are hence formed at different depths in the stellar atmosphere. The investigation of the flux variability in the wings of Ly line indicates that the fluxes, which formed in the inner layers of atmosphere are redistributed into the outer layers of the atmosphere of 2 CVn. Therefore, there should be a layer in the atmosphere of the star where the fluxes do not vary over the period of rotation. However, an independent investigation of the Ly line variations in others CP stars is needed in order to confirm our results.

VARIATIONS OF FLUX INTENSITY IN LARGE FEATURES ...

409

6

Conclusions

The archival IUE spectrophotometric data of 2 CVn have permitted to analyse the light variations at the cores of large features and spectral lines. The variations of flux intensity at the cores of large features and spectral lines revel various behaviours. The effect of different species of spectral lines is observed in different regions at 1375 and 1415 А of the large feature 1400 А The light curve A A. at 1375 А shows a similar shape like in the "pseudocontinuum". The same behaviour of the flux A is observed at the cores of Si II resonance lines at 1260 64 and 1485 А while the light curve at A, А shows the phase displacement of the flux minimum. The phase displacement is also present 1415 A at the cores of Si II resonance lines at 1304 09 and 1485 А The same phase displacement is at A. the cores of the Fe II depressions in the spectral region with > 2505 А However, the variability of A. the flux at the cores of the Fe II depressions in this spectral region is the same as in the "pseudo continuum". The minimal values of the amplitude of the flux variations are reached at the cores of large features at 1560 and 1770 А and at the cores of the strong Si II resonance lines at 1260 64, A 1304 09 and 1485 А The flux at the cores of a large feature at 1770 А and Fe II resonance line A. A А does not vary within the measurement errors. at 1725 31 A The investigation of the flux variability in the wings of the Ly line indicate that the fluxes, which formed in the inner layers of atmosphere are redistributed into the outer layers of the atmosphere of 2 CVn. However, an independent investigation of the Ly line variations in the highresolution mode is needed in order to confirm our results.

References
Adelman S. J., Cowley C. R., Leckrone D. S., Roby S. W., Wahlgren G. M., 1993, ApJ, 419, 276 Artru M.C., Lanz T., 1987, A&A, 182, 273 Farnsworth G., 1932, ApJ, 75, 364 Fuhrmann K., 1989, A&AS, 80, 399 Goncharskii A. V., Ryabchikova T. A., Stepanov V. V., Khokholova V. L., Yagola A. G., 1983, Soviet Astr., 14, 652 Gonzalez J. F., Artru M.C., 1994, A&A, 289, 209 Hensberge H., Van Santvoort J., Van der Hucht K. A., Morgan T. H., 1986, A&A, 158, 113 Khokhlova V. L., Pavlova V. M., 1984, Soviet Astr. Lett., 10, 158 Kochukhov O., Piskunov N., Ilyin I., Tuominen I., 2002, A&A, 389, 420 Lanz T., Artru M.C., Le Dourneuf M., Hubeny T., 1996, A&A, 309, 218 Leckrone D. S., 1984, ApJ, 286, 725 Leckrone D. S., Snijders M. A. J., 1979, ApJS, 39, 549 Pyper D. M., 1969, ApJS, 18, 347 Ryabchikova T., 2008, Contr. of the Astron. Obs. SkalnatД Pleso, 38, 257 e Ryabchikova T., Piskunov N., Savanov I., Kupka F., Malanushenko V., 1999, A&A, 343, 229 Shore S. N., Brown D. N., Sonneborn G., 1987, AJ, 94, 737 Sokolov N. A., 2000, A&A, 353, 707 Sokolov N. A., 2006, MNRAS, 373, 666 Sokolov N. A., 2010, Astrophys. Space Sci., 330, 37 Sokolov N. A., 2011, this proceedings Stepien K., Czechowski W., 1993, A&A, 268, 187 Д