A&A 376, 885-891 (2001)
I. A. Usenko 1,3 - V. V. Kovtyukh 1,3 - V. G. Klochkova 2,4 - V. E. Panchuk 2,4
1 - Astronomical Observatory of Odessa State University, Odessa 65014, Ukraine
2 - Special Astrophysical Observatory, Russian Academy of Sciences, Nizhny Arkhyz,Stavropol Territory 369167, Russia
3 - Isaac Newton Institute of Chile, Odessa Branch, Ukraine
4 - Isaac Newton Institute of Chile, SAO RAS Branch, Russia
Received 27 February 2001 / Accepted 24 July 2001
The small-amplitude Cepheid V1726 Cyg and two members of open cluster Platais 1 (Platais 1 star No. 1 (1921) and Platais 1 star No. 111 (1600)) were investigated, using high-resolution CCD spectra. The following results were obtained: 1) All objects have the same metallicities, close to that of the Sun (for V1726 Cyg weighted average [Fe/H]=+0.05, for Platais 1 star No. 1 (1921) [Fe/H]=+0.13); 2) values of and for the B-stars are in excellent agreement with those determined from (B-V) colour indices using a , ) calibration; 3) the elemental abundances indicate that V1726 Cyg is in the post first dredge-up stage with an age near , and is crossing the Cepheid instability strip for the third time. Mean values of K and permit us to refine its colour excess to , which for a distance of pc corresponds to . The Cepheid could therefore be pulsating in the fundamental tone, although pulsation in the first overtone is not excluded; 4) Platais 1 Star No. 1 (1921) is a slowly rotating HgMn-star with a high helium content, while Platais 1 star No. 111 (1600) is a rapidly rotating main-sequence star with a helium content comparable to that of the Sun; 5) the age of the open cluster is estimated to be about yr.
Key words: galaxy: open clusters: and associations - stars: abundances - stars: Cepheids - stars: general
In this paper we continue to present the results of spectroscopic investigations of Cepheids and main-sequence (MS) stars in selected Galactic open clusters that was began in Paper I (Usenko et al. 2001). Examined here are three objects belonging to the insufficiently explored open cluster Platais 1 (Platais 1986): the small-amplitude Cepheid (DCEPS) V1726 Cyg and two B-stars.
As in the case of SU Cas, V1726 Cyg is suspected to be an overtone pulsator, according to Turner et al. (1994). Of the other cluster stars, one of the B-stars (star No. 111) is a typical MS object, while the other (star No. 1) is located near the turn-off point for the open cluster. In Paper I we identified in the Cas OB2 association a similar star HD 17327a, located near the turn-off point that is a mercury-manganese star with a very high helium content and a low rotation velocity. Lying closer to the MS, HD 17443 has a helium content close to that of the Sun, along with a high rotational velocity. The question arises: is it a unique case, or can we observe similar stars in other open clusters and associations?
This very poorly studied small cluster was investigated photometrically in detail for the first time by Platais (1986), owing to its spatial proximity to the small-amplitude Cepheid (DCEPS) V1726 Cyg. Subsequent observations by Turner et al. (1994) have shown that V1726 Cyg is an outlying member of the open cluster. Photoelectric photometry of V1726 Cyg obtained during the last 15 years (Berdnikov 1986, 1986; Pastukhova & Shugarov 1994) did not indicate any obvious evolutionary effects on its pulsation period (Berdnikov & Pastukhova 1994), but more recent studies (Berdnikov et al. 2000; Turner et al. 1999) do provide evidence for an increasing pulsational period in this DCEPS. Nevertheless, the value of derived for V1726 Cyg, from the interstellar reddening and distance of Platais 1, implies that the star is pulsating in the first overtone mode. On the other hand, according to the colour coefficient in the PLC relation, it is assumed that V1726 Cyg pulsates in the fundamental mode (Turner et al. 1994).
Since Platais 1 contains very faint objects, in addition to V1726 Cyg, we selected two B-stars for observations at the suggestion of Dr. D. G. Turner. Star No. 111 from Turner et al. (1994) (or Platais 1600) is a MS object, not far from the open cluster's turn-off point, while star No. 1 (or Platais 1921) is situated very close to the turn-off point. Turner et al. (1994) suggested that it is a B9 IIIp shell star. Photometric information about the stars is given in Table 1.
|V1726 Cyg||9.01||0.89||0.58||F6 Ib|
|No. 1 (1921)||11.15||0.30||0.19||B9 IIIp|
|No. 111 (1600)||12.54||-0.12||-||B9.5 V|
High-resolution spectra of the stars were obtained with an échelle spectrometer PFES (Panchuk et al. 1998), installed on the 6 m telescope of the Special Astrophysical Observatory of the Russian Academy of Sciences (Russia, Northern Caucasus). The spectra contained 23 orders, with an average resolving power 000, and a signal-to-noise ratio -100. Information concerning the program stars and their CCD spectra is given in Table 2.
|No. 1 (1921)||s20702||1003.304||4420-7767||30|
|No. 111 (1600)||s20817||1004.461||4420-7767||54|
By means of the MIDAS software, we extracted the spectra from CCD frames, subtracted dark frames, removed cosmic ray hits, and performed wavelength calibration. All the equivalent widths values were measured using the DECH20 code (Galazutdinov 1992).
Similarly to Paper I, prior to the calculations we made a visual inspection of the program B-stars spectra, because star No. 1 was suspected to be a peculiar shell star, and No. 111 has a high projected rotational velocity. The H and H line profiles for star No. 1 are shown in Figs. 1 and 2, respectively. As seen in the figures, there are no emission features in the H and H line wings. Hence, star No. 1 is not a shell star.
|Figure 1: The H line profile for star No. 1 (1921). The best fit between the synthetic and observed profiles is presented.|
|Open with DEXTER|
|Figure 2: The H line profile for star No. 1 (1921). The best fit between the synthetic and observed profiles is presented.|
|Open with DEXTER|
On the other hand, star No. 111 is an usual MS object with the high projected rotational velocity (see Figs. 3 and 4).
|Figure 3: The same as Fig. 1 for star No. 111 (1600).|
|Open with DEXTER|
As noted from the figures, prior to making equivalent width measurements, we estimated values of for each B-star. For that we used a synthetic spectrum technique (STARSP code of Tsymbal 1996) software and SYNSPEC code (Hubeny et al. 1994), fitting a synthetic spectrum to the observed one. The results are given in Table 3. In our analysis, we did not use lines with equivalent widths greater than 150 mÅ.
The internal accuracy of the equivalent widths is of the order of 5-10%. This estimate is based upon a comparison of the values derived from the lines present in two overlapping spectral orders.
The method of analysis implemented in this work was described by us in detail in Paper I. We used atmosphere models, interpolated from the Kurucz (1992) grid along with the WIDTH9 code for calculating the chemical composition.
For V1726 Cyg, as for SU Cas in Paper I, we used so-called "solar" values derived by us using unblended solar lines (from the solar spectrum by Kurucz et al. 1984). The corresponding solar atmosphere model was recalculated with km s-1 from Kurucz's grid, using the WIDTH9 code. For the B-stars we used the oscillator strengths from the Kurucz (1995) database (CD-ROM 15, 18). Solar elemental abundances were taken from Grevesse & Noels (1993) (see Paper I).
|Figure 4: The same as Fig. 2 for star No. 111 (1600).|
|Open with DEXTER|
As in Paper I, program objects with different spectral types and luminosity classes require the determination of atmospheric parameters by different methods. For V1726 Cyg we obtained values for the effective temperature ( ), surface gravity (), and microturbulent velocity () as follows.
1) was found using the ratio of the depths of spectral lines by the method of Kovtyukh & Gorlova (2000) with an accuracy of 50-80 K.
2) The surface gravity was determined by forcing the Fe I and Fe II to produce the same abundance (within an accuracy of about 0.20 dex).
3) was obtained by forcing the abundances from the Fe II lines to be independent of the equivalent widths (with an accuracy of about 0.30 km s-1).
For the B-stars we used:
1) , calibrations (Bessel et al. 1998).
2) Comparisons of the observed H and H line profiles with synthetically generated ones.
The (B-V) and EB-V data were taken from Turner et al. (1994).
All data for the preliminary and values determination for B-stars are given in Table 3.
|Star||(B-V)0||(U-B)0||H||H||(B-V)0, (U-B)0||H||H||(km s-1)|
|No. 1 (1921)||9900||9900||10000||10000||3.30||3.50||3.50||10|
|No. 111 (1600)||12250||14000||13000||13000||4.00||4.00||4.00||150|
For the determination in the case of star No. 1, we used the same method, as adopted for V1726 Cyg. For star No. 111 we adopted a value of of 3 km s-1, which seems more appropriate for late B-stars. All of the adopted atmospheric parameters are listed in Table 4.
In Tables 5 and 6 we list the calculated abundances for V1726 Cyg and star No. 1, respectively. For the rapidly rotating B-star No. 111 (as with HD 17443 from Paper I) we estimated only the helium and magnesium abundances using two strong features He I 4471 Å and Mg II 4481 Å. That was done using spectral synthesis (see Fig. 5). The helium and magnesium abundances turn out to be solar.
|No. 1 (1921)||10000||3.50||3.50|
|No. 111 (1600)||13000||4.00||3.00|
|Figure 5: The fit between the observed and synthetic spectra for star No. 111 (1600) near the He I 4471 Å and Mg II 4481 Å lines.|
|Open with DEXTER|
From an analysis of the results of Tables 5 and 6, we notice that DCEPS V1726 Cyg and B-star No. 1 have chemical compositions very similar to those of DCEPS SU Cas and HD 17327a in the Cas OB2 association (Paper I), respectively. The weighted average metallicity for V1726 Cyg [Fe/H]=+0.05 and [Fe/H]=+0.13 for Star No. 1. These values are close to that of the Sun. Indeed, V1726 Cyg appears to be carbon deficient, with a very noticeable overabundance of nitrogen, and the oxygen content close to solar. Data for so-called "odd elements", sodium and aluminium, suggest a small overabundance for the first element and a solar-like content for the second. As with SU Cas, the content of -elements (except for the magnesium deficit) is close to solar. The Fe-group elements have a slight overabundance, however except for manganese and nickel. The abundances of s-process elements are close to solar, and exhibit some overabundance (see Fig. 6).
|Figure 6: Elemental abundance for V1726 Cyg.|
|Open with DEXTER|
It is interesting that star No. 1 is a mercury-manganese star, much like HD 17327a in Paper I. It has an overabundance of helium, carbon and oxygen, and a deficient of nitrogen. Furthermore, there is a deficit of Mg and Si, and an overabundance of P, Sc, Mn, Y, Zr and Hg. The latter are distinctive features of this type of peculiar B-star (see Fig. 7). Nevertheless, star No. 1 is located in the diagram (Ryabchikova 1997) between the classical HgMn star region and the Searl-Sargent group (see Fig. 8). According to its value of , the object seems to be closer to the Searl-Sargent group, whereas HD 17327a is a classical HgMn star.
|Figure 7: Elemental abundance for star No. 1 (1921).|
|Open with DEXTER|
|Figure 8: (Mn/Fe) vs. (Fe/H) abundances relation. Open squares, - classical (SB1) HgMn stars, filled circles, - Searle-Sargent (SB2) group HgMn stars; the open star indicate HD 17327a, while the filled star, - No. 1 (1921).|
|Open with DEXTER|
In Paper I we noted that the HgMn-star HD 17327a, unlike the main-sequence B star HD 17443, has an overabundance of helium. That fact was interpreted to be the result of the light-induced drift (LID) mechanism (Atutov & Shalagin 1988). For star No. 1 we obtain the same result, while star No. 111 has a helium content close to solar.
As seen from Table 4, our values of and obtained spectroscopically for stars Nos. 1 and 111 give corresponding values for (B-V)0 of 087 and 120, respectively, according to a , ) calibration from Bessell et al. (1998). There is excellent agreement between our results and those of Turner et al. (1994) (see Table 5 from their paper). Their colour excesses 37 and 0 48, respectively), derived for those stars using UBV photometry, are therefore very accurate.
For V1726 Cyg Turner et al. (1994) estimated . That value for the reddening corresponds to an intrinsic colour-index of 02. For the mean cluster distance of pc, the corresponding absolute magnitude for V1726 Cyg is 07 which suggest pulsation is in the first overtone.
To check that suggestion, we estimated a corresponding intrinsic colour for K and ( 11) from the average standard light and colour curves (Berdnikov & Pastukhova 1994), as interpolated from the , ) calibration, cited above. We obtained 485, which corresponds to a colour excess of 34. If the value of 02 from Turner et al. (1994) is correct, than the corresponding intrinsic colour-index 392, bluer by 0 093 than our estimation. Having noted that Turner et al. (1994) explained the same difference of 09 between the average 46 for this DCEPS and those of the most Cepheids with similar periods as a consequence of pulsation in the first overtone, we can assume that V1726 Cyg is pulsating in the fundamental tone.
It is interesting to check our spectroscopic value of
ones obtained from Gray's (1992)
The colour excess estimated by us for V1726 Cyg is therefore closer to that for star No. 1 which is located 1 arcmin east of the variable. In principle all three values are close to the mean reddening of 39 obtained by Turner et al. (1994) for the open cluster Platais 1. On that basis, we may use their mean values of R=AV/EB-V=3.07 and pc.
To estimate luminosities and radii of the program objects we can use our spectroscopic with the mean distance from Turner et al. (1994). For the B-stars we used bolometric corrections from Bessel et al. (1998). However, for V1726 Cyg we have one spectrogram only. We therefore estimated the mean value of K with interpolated from the , ) calibration of Bessell et al. (1998), using the averaged standard (B-V)0 colour curve (Berdnikov & Pastukhova 1994) with the mean value of 553. Equation (1) gives a mean value of K, which is very close to the estimate mentioned above.
|V1726 Cyg (F)||-2.99||-2.99||3.10||31.8||4.2|
|V1726 Cyg (1st)||-3.42||-3.42||3.28||38.6||4.8|
|No. 1 (1921)||-0.97||-1.23||2.40||5.3||3.0|
|No. 111 (1600)||+0.08||-0.76||2.21||2.5||3.0|
L and R estimates have been obtained for each star (see Table 7)
using these values for
and AV along with the cluster
distance and the value of
Since V1726 Cyg is an object
in the post red supergiant evolutionary phase, its evolutionary mass has
been obtained from the mass-luminosity relation obtained from
stellar evolutionary models that include core overshooting,
|Figure 9: HR diagram for Platais 1 members. Evolutionary tracks are from Schaller et al. (1992) recalculated for Z=0.02.|
|Open with DEXTER|
It is interesting to estimate ages for the Platais 1 objects. From the Schaller et al. (1992) grids of models we obtain age estimates of , and yrs for V1726 Cyg, star No. 1 and star No. 111, respectively. Turner et al. (1994) estimated the age of this open cluster to be about 1.1- yr, like that of the open cluster NGC 1647. If star No. 1 lies near the turn-off point of the open cluster, then an age of about can be adopted.
We can summarize the results of our detailed high-resolution spectroscopic investigation:
1) All Platais 1 objects studied have metallicities close to that of the Sun.
2) The values of and derived spectroscopically for the two B-stars in this open cluster are in excellent agreement with ones determined from the Turner et al. (1994) (B-V) colour-indices in conjunction with a , ) calibration. That implies that the mean values of 39 and distance pc found by Turner et al. (1994), can be considered to be correct.
3) The DCEPS V1726 Cyg is found to be a Cepheid in the post first dredge-up stage. In spite of having only one spectrogram for the star obtained at phase 11, we estimated a corresponding intrinsic colour-index near 0 485 and obtain a value for the colour excess 33. Because we used an average standard (B-V)0 colour curve, the mean values of K and have been obtained carefully. The derived interstellar extinction of 02 implies an absolute magnitude of 99 for the Cepheid. In contrast with the value of 42 from Turner et al. (1994) (indicating pulsation in the first overtone), our value is in excellent agreement with that for pulsation in the fundamental mode according to Turner's (1992) relation.
It should be noted that Turner et al. (1994) obtained a mean value of 02, using the intrinsic colour calibrations for F-G - supergiants from Fernie (1963), Johnson (1966), Parsons (1971) and Kron (1978). Therefore, either those calibrations are incorrect or V1726 Cyg has a blue companion of approximate spectral type B5-B6 V (note that V1726 Cyg is bluer by 0 09 than most Cepheids with periods near 4 24, according to Turner et al. 1994).
As seen from Fig. 9, the Cepheid V1726 Cyg is located in the HR diagram between portions of the evolutionary tracks for stars with and 4.8 for the case of fundamental mode and first overtone pulsations, respectively. Both positions correspond to the third crossing of the Cepheids instability strip. That feature has a realistic observational confirmation, namely an increasing pulsational period corresponds to movement towards the red edge of Cepheids instability strip (Berdnikov et al. 2000; Turner et al. 1999; Turner et al. 2001). Moreover, Turner et al. (2001) have concluded that it is crossing the Cepheids instability strip for the third time.
Nevertheless, since we obtained only one spectrogram for V1726 Cyg, we cannot contend flatly that this DCEPS is pulsating in the fundamental mode. According to Turner et al. (2001), its light curve's Fourier parameters correspond with those found for Cepheids pulsating in the first overtone. In our opinion, this DCEPS is in need of more detail spectroscopical investigations. It is necessary to obtain more specific information about the mean effective temperature, gravity and radius of the star. Such data could help clarify the identification of the V1726 Cyg pulsational mode.
4) The two B-stars from Platais 1 have approximately equal evolutionary masses of 3 . It is interesting that star No. 111, like HD 17443 in Cas OB2, is a typical MS star with a high rotational velocity and has a helium content comparable to that of the Sun. Star No. 1 appears to be a HgMn-star with an overabundance of He, C, O, P, Sc, Mn, Y, Zr, Hg and a deficiency of N, Al, Mg, and Si. That can be well explained by the LID mechanism operating in the stellar atmosphere. Much like HD 17327a in Cas OB2, the star lies near the turn-off point of the cluster and is a more evolved object than star No. 111. Unlike HD 17327a, however, star No. 1 probably belongs to the Searl-Sargent group.
5) The age of the open cluster was estimated to be about yrs, while that of the Cepheid is yrs.
Authors are thankful to Drs. D. G. Turner and A. Miroshnichenko and Mrs. E. A. Panko for useful comments. I. A. Usenko is acknowledged to ISF for support (ISF research grant YSU 082057).