Free Access
Issue
A&A
Volume 524, December 2010
Article Number A8
Number of page(s) 4
Section Stellar structure and evolution
DOI https://doi.org/10.1051/0004-6361/201014889
Published online 19 November 2010

© ESO, 2010

1. Introduction

The star RZ Psc (Sp = K0 IV, Herbig 1960) is one of the most unusual variable stars. According to the properties of its light curve, it belongs to the family of young stars of the UX Ori type, the photometric and polarimetric activity of which is caused by the variable circumstellar (CS) extinction (Grinin et al. 1991) According to the current models (Dullemond et al. 2003), the region of the CS disk close to the dust evaporation zone, where the main part of the near infrared excess is formed (Natta et al. 2001), is responsible for this variability. However, RZ Psc shows none of the classical signs of youth: it has neither infrared (JHK) excess (Glass & Penston 1974)1, nor emission in the Hα line (Kaminskii et al. 2000). The star lies at the Galactic latitude b ≃  − 35°, where there are neither young stars, nor star-formation regions. The brightness minima of RZ Psc are also quite unusual: with large amplitudes (up to ΔV  ≈  2.5 mag), they are very brief (1 − 2 days; Zaitseva 1985; Kardopolov et al. 1980; Pugach 1981; Wenzel 1989). Similar minima are sometimes also observed in another UXOrs but, in the case of RZ Psc, they are typical. Several attempts were undertaken to find the period between the star eclipses, but all were unsuccessful (Wenzel 1989).

The above combination of the conflicting observational properties poses many questions about the evolutionary status of RZ Psc and the origin of the eclipses. To clarify these questions, we analyze the Li 6708 Å spectral line, which is commonly used for age estimations of solar-type stars (see e.g., Soderblom et al. 1993; Sestito & Randich 2005; da Silva et al. 2009). In addition, we use the proper motion of RZ Psc from the Tycho-2 Catalog by Hog et al. (2000) to roughly estimate the time at which the star escaped from its birthplace near to the Galactic plane. Both estimates allow us to classify RZ Psc as a post-UXOr.

2. The “lithium” age of RZ Psc

Although RZ Psc has already been investigated for a long time, most papers have been devoted to the study of its variability. There is only one paper (Kaminskii et al. 2000) in which the quantitative analysis of the star spectrum has been made. In that paper, however, there is no information about the Li 6708 Å line. To study this line, the spectrum of RZ Psc was obtained at the Terskol Observatory (Russia). The observations were made on 2009 Nov. 9, with the echelle spectrograph SPECPHOT (spectral resolving power R = 13500) at the 2 m telescope. The spectrum was reduced with the software package DECH by Galazutdinov (1992), which provides all standard tasks of CCD image and spectra processing. The wavelength calibration was made with an Fe-Ar comparison lamp spectrum.

A portion of the reduced spectrum around the Li 6708 Å line is displayed in Fig. 1. One can see that this line is present in the spectrum of RZ Psc and has a moderate depth, with an equivalent width of EW(Li) = 0.202  ±  0.010 Å. For comparison, in Fig. 1 we show two versions of a synthetic spectrum calculated for the same spectral range with the numerical code from Piskunov (1992) and the VALD data base by Piskunov et al. (1995), Kupka et al. (1999). The spectra were broadened by rotation with vsini = 23 km s-1 (Kaminskii et al. 2000). In both cases we used the Kurucz (1979) model of atmosphere with Tef = 5250 K, log g = 4.0, and vturb = 2 km s-1. In one of these spectra, the Li abundance is the same as in the Sun. In the other one, it is 100 times higher. We see that in the first case, the Li 6708 Å line is practically invisible. In the second case, the lithium line in the synthetic spectrum almost coincides with the observed one. Taking into account the possible uncertainties in the estimation of the spectral type of the star, we have calculated several additional synthetic spectra, changing the model parameters in the ranges Tef = 5250 ± 250 K and log g = 4.0 ± 0.5, and these calculations confirmed the result obtained above.

thumbnail Fig. 1

The spectrum of RZ Psc around the Li 6708 Å line (thin line). The synthetic spectra with the Li excess (100 times larger than the solar abundance) (thick gray line) and without excess (dashed line) are shown (see text for details).

Thus, the spectral analysis shows that the atmosphere of RZ Psc has a huge excess of lithium (about 100 times the solar abundance) and, according to this criterion, RZ Psc is not yet a main sequence star. The considerably high (for this spectral type) value of vsini agrees with this conclusion (see the discussion by Bouvier 2008). On the other hand, in the atmospheres of young solar-type stars (T Tauri stars) the excess of lithium is usually much greater than in our case (see Fig. 2a). For example, V718 Per (about the same spectral type as RZ Psc) has a lithium excess equals to about 3.2 − 3.5 dex (Grinin et al. 2008), which is indicative of a primordial Li abundance (Pavlenko & Magazzu 1996). From this point of view, RZ Psc is not any more a young star. To estimate its age, we used the rough calibration of Li excess as a function of age for clusters of different ages (King 1993; Soderblom et al. 1990; 1993; Sestito & Randich 2005). Our analysis of these data shows (see Fig. 2) that the excess of Li in the atmosphere of RZ Psc corresponds approximately to a stellar age between the age of the Pleiades (about 70 Myr) and the Orion (about 10 Myr) clusters. It is admittedly a rough estimate, taking into account the high dispersion of Li abundance versus age in both clusters. Nevertheless, this estimate qualitatively agrees with the absence of IR excess in RZ Psc and its isolated position far from the star-forming regions.

thumbnail Fig. 2

Comparison of EW (Li, 6708 Å) in the spectrum of RZ Psc (circle) with the data for the stars in the clusters of different age (triangles): Orion (10 Myr) from King (1993); Pleiades (70 Myr) and Hyades (600 Myr) from Soderblom et al. (1990; 1993).

3. The kinematic age of RZ Psc

Another possibility to estimate the age of RZ Psc is the proper motion of the star. According to the Tycho-2 catalog by Hog et al. (2000), the proper motion of RZ Psc is quite large: pmRA = 25.4  ±  2 mas yr-1, pmDE =  −11.9  ±  2.1 mas yr-1. The corresponding values of the proper motion in the galactic coordinate system are: pml = 7.89  ±  0.60 deg per Myr, pmb =  −2.76  ±  0.49 deg per Myr.

Assuming that RZ Psc has not moved away very far from its birthplace near to the Galaxy plane (GP), one can calculate the vertical component of its motion W using the approximation by Perrot & Grenier (2003) for the gravitation potential of the Galactic disk. In this approximation the Galactic disk is locally described by an infinite plane. The corresponding vertical acceleration is given by g(z)=4πGρdzd(1ez/zd),\begin{equation} g(z) = -4\pi\textmd{G}\rho_{\rm d} z_{\rm d} (1 - {\rm e}^{-z/z_{\rm d}}), \label{1} \end{equation}(1)where G is the gravitational constant, ρd = 7.6  ×  10-2   M   pc-3 is a volume density at z = 0 (Crézé et al. 1998), zd = 260  ±  60 pc is an exponential scale height (Ojha et al. 1996).

Substituting g(z) in the right part of equation of motion in z-direction and replacing dW/dt by WdW/dz, after integration on z we obtain the vertical component of the velocity W(x)=±W02vd2f(x),\begin{equation} W(x) = \pm \sqrt{W_0^2 - v_{\rm d}^2\,f(x)}, \label{2} \end{equation}(2)where x = z/zd, f(x)=x1+ex,\begin{equation} f(x) = x - 1 + {\rm e}^{-x}, \label{3} \end{equation}(3)vd=2zd(2πGρd)1/2,\begin{equation} v_{\rm d} = 2z_{\rm d}(2\pi G\rho_{\rm d})^{1/2}, \label{4} \end{equation}(4)W0 is the initial velocity at x = 0. Signum “+” in Eq. (2) corresponds to the motion toward the southern Galactic pole.

Equation (2) describes the oscillations of a star in the gravitation field of the disk after ejection from the disk plane with the initial velocity W0: at some point xm the velocity W drops up to zero, whereupon the star goes back, intersects the disk plane, and continues the motion at the opposite side of the disk plane. With Eq. (2) we can calculate the time of the star motion from x1 to x2 (x2  ≤  xm): t(x1,x2)=tdx1x2dxa2f(x),\begin{equation} t(x_1,x_2) = t_{\rm d} \int_{x_1}^{x_2}\frac{{\rm d}x}{\sqrt{a^2 - f(x)}}, \label{5} \end{equation}(5)Here a = W0/vd, td = zd/vd  ≈  12 Myr. The time t(0,xm) = p/4, where p is the period of oscillations of a star in the gravity field of the disk. If the current value W(x) > 0 (the star moves away the Galactic plane), then the kinematic age tk = t(0,x). In the opposite case tk = p/4 + t(x,xm).

Accordingly, we need two parameters to calculate tk: a) the current distance z of the star from the disk plane; and b) the corresponding velocity W(z). Both parameters depend on the distance to the star D: i) z = Dsinb − z, where z is a distance of the Sun from the Galactic plane (in pursuance Joshi 2007 we adopt z = 25 pc); ii) W(z) = Vrsinb + D pmb + W, where pmb is the proper motion of RZ Psc in ster/s, W =  − 7 km s-1 is the velocity of the Sun toward the nordic Galactic pole (Perrot & Grenier 2003), Vr =  − 11.5  ±  1.5 km s-1 – the radial velocity of the star (Kaminski et al. 2000). For the estimate of D, we used the photometric data for the out-of-eclipse state of RZ Psc (V ≈ 11.5m, Zaitseva 1985) and adopted Tef = 5250 K, log g = 4.0 (see above), and the stellar radius R = 1.5   R (which is compatible with log g = 4.0 at M ≈ 1   M). The interstellar extinction toward RZ Psc AV  ≈  0 (Kaminski et al. 2000). Taking this into account we obtain D  ≈  240 pc2. Using this value at the calculations of z, zm and W(z) we obtain tk = 37.6  ±  5.4 Myr. Here, the uncertainty is due to the uncertainty of pmb ( − 2.76  ±  0.49 deg/Myr). About the same uncertainty appears at the variations of D within 240  ±  50 pc: $t_{\rm k} = 37.6^{-7.2}_{+6.9}$ Myr. Taking this into account we can give an approximate estimate of the kinematic age of the star: tk  ≃  30 − 40 Myr3.

Thus, both estimates of the age of RZ Psc, based on the lithium excess and the proper motion of the star give values that essentially exceed the characteristic time of dissipation of protoplanetary disks (~10 Myr; Strom et al. 1993). This allows us to classify RZ Psc as an intermediate object between MS stars with debris disks and UXOrs.

4. Discussion and conclusion

The question arises: if RZ Psc is a post UXOr, what is, in this case, the reason for the quick and sporadic eclipses of the star? The apparent answer would be a screening of the star from time to time by material remnant from the CS disk: planetesimals, comets, and rocks orbiting the star and dissipating in its nearest vicinity.

The idea of comet-like activity in the neighborhoods of young stars as a possible source of the variable CS extinction is not new. It was claimed many years ago by Gahm & Greenberg (1983). However, in the case of young stars, there is a power alternative source of the variable extinction: the inhomogeneous matter of CS disks. This source dominates in UX Ori stars. Therefore, the case of RZ Psc is interesting and important because there are no alternative explanations for the rapid and sporadic eclipses of the star.

Using the flux variation rate (about 0.1 mag per hour) estimated by Zaitseva (1985) from intense photometric monitoring of two deep minima of RZ Psc, one can estimate the tangential velocity vt of the opaque screen (the dust cloud) intersecting the line-of-sight and its approximate distance to the star. Assuming an stellar radius of R = 1.5 R and stellar mass of M = 1 M, we obtained vt  ≈  40 km s-1 and a distance  ≈ 0.6 AU. In order to screen the star from an observer during two days (the duration of eclipses), the clouds have to be quite compact (about 0.05 AU).

The further study of RZ Psc and the search for stars with similar properties can give valuable information about the disk properties at the transitional phase of their evolution. It is important, first of all, to observe the mid-IR excess of this star at λ  ≥  5 μm. There are indications that this excess exists, including the observations of the highly polarized radiation in the deep minima (Kiselev et al. 1991; Shakhovskoi et al. 2003) and the so-called blueing effect (Zaitseva 1985; Kardopolov et al. 1980; Pugach 1981). Both these effects are caused in UX Ori stars by the scattered radiation of CS dust (Grinin 1988). This radiation is the relatively stabilized agent (different in different stars), which provides the limitation of the amplitudes of the algol-type minima. For RZ Psc (ΔV)max  ≈  2.5, which corresponds to the intensity of scattered radiation Isc ≈ 0.1I (which is comparable with the level of Isc in the other UXORs (Grinin et al. 1991). The absence of near IR excess in RZ Psc means that in this case the main part of CS dust lies far from the star, in the circumbinary disk. This dust is probably formed (as in MS stars with debris disks) by collisions of large particles, rocks, and planetesimals (see e.g. Wyatt 1996) and is heated by stellar radiation. The energy distribution of RZ Psc at mid-IR wavelengths will enable us to estimate the characteristic size of the region occupied by CS dust. The other important problem is to find the distance from the star. It allows us to estimate more precisely the time when RZ Psc has left its birthplace.


1

RZ Psc is absent in the IRAS and Spitzer data bases.

2

At this distance, the proper motion of RZ Psc would correspond to a tangential velocity  ≃ 32 km s-1.

3

Note that the kinematic age of RZ Psc is comparable with the age of the Gould Belt, a nearby starburst region where many stars have formed over 30 to 40 Myr ago (see Perrot & Grenier 2003, and references there). This allows us to assume that RZ Psc is a runaway member of this cluster.

Acknowledgments

We thank the staff of the Terskol Observatory for the observations and the referee for the useful comments that improved the paper. This research has been supported in part by the program of the Presidium of RAS “Formation and Evolution of Stars and Galaxies” and grant N.Sh.- 3645.2010.2.

References

  1. Bouvier, J. 2008, A&A, 489, 53 [Google Scholar]
  2. Crézé, M., Chereul, E., Bienymé, O., & Pichon, C. 1998, A&A, 329, 920 [NASA ADS] [Google Scholar]
  3. da Silva, L., Torres, C. A. O, de la Reza, R., et al. 2009, A&A, 508, 833 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  4. Dullemond, C. P., van den Ancker, M. E., Acke, B., & van Boekel, R. 2003, ApJ, 594, L47 [NASA ADS] [CrossRef] [Google Scholar]
  5. Gahm, G. H., & Greenberg, J. M. 1983, in Asteroids, Comets, Meteors, ed. C.-I. Lagerkvist, & H. Rickman, Uppsala Univ., 375 [Google Scholar]
  6. Galazutdinov, G. A. 1992, SAO Preprint, 92 [Google Scholar]
  7. Glass, I. S., & Penston, M. V. 1974, MNRAS, 167, 237 [NASA ADS] [CrossRef] [Google Scholar]
  8. Grinin, V. P. 1988, Soviet. Astron. Lett., 14, 27 [Google Scholar]
  9. Grinin, V. P., Kiselev, N. V., Minikulov, N. H., Chernova, G. P., & Voshchinnikov, N. V. 1991, ApSS, 186, 283 [Google Scholar]
  10. Grinin, V., Stempels, H. C., Gahm, G. F., et al. 2008, A&A, 489, 1233 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  11. Herbig, G. H. 1960, ApJ, 131, 632 [NASA ADS] [CrossRef] [Google Scholar]
  12. Hog, E., Fabricius, C., Makarov, V. V., et al. 2000, A&A, 355, L27 [NASA ADS] [Google Scholar]
  13. Joshi, Y. C. 2007, MNRAS, 378, 768 [NASA ADS] [CrossRef] [Google Scholar]
  14. Kaminskii, B. M., Kovalchuk, G. U., & Pugach, A. F. 2000, Astron. Rep., 44, 611 [NASA ADS] [CrossRef] [Google Scholar]
  15. Kardopolov, V. I., Sahanionok, V. V., & Shutemova, N. A. 1980, Perem. Zvezdy, 21, 310 [Google Scholar]
  16. King, J. 1993, AJ, 105, 1087, [NASA ADS] [CrossRef] [Google Scholar]
  17. Kiselev, N. N., Minikulov, N. K., & Chernova, G. P. 1991, Astrophys., 34, 175 [Google Scholar]
  18. Kupka, F., Piskunov, N. E., Ryabchikova, T. A., Stempels, H. C., & Weiss, W. W. 1999, A&A, 138, 119 [Google Scholar]
  19. Kurucz, R. 1979, ApJS, 40, 1 [NASA ADS] [CrossRef] [Google Scholar]
  20. Natta, A., Prusti, T., Neri, R., et al. 2001, A&A, 371, 186 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  21. Ojha, D. K., Bienymé, O., Robin, A. C., et al. 1996, A&A, 311, 456 [NASA ADS] [Google Scholar]
  22. Pavlenko, Ya. V., & Magazzu, A. 1996, A&A, 311, 961 [NASA ADS] [Google Scholar]
  23. Perrot, C. A., & Grenier, I. A. 2003, A&A, 404, 519 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  24. Piskunov, N. N. 1992, in Stellar Magnetizm, ed. Yu. V. Glagolevsky, & I. I. Romanjuk (St. Petersburg: Nauka), 92 [Google Scholar]
  25. Piskunov, N. E., Kupka, F., Ryabchikova, T. A., Weiss, W. W., & Jeffery, C. S. 1995, A&A, 112, 525 [Google Scholar]
  26. Pugach, A. F. 1981, Astrophys., 17, 47 [Google Scholar]
  27. Sestito, P., & Randich, S. 2005, A&A, 442, 615 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  28. Shakhovskoi, D. N., Grinin, V. P., & Rostopchina, A. N. 2003, Astron. Rep., 47, 580 [NASA ADS] [CrossRef] [Google Scholar]
  29. Soderblom, D. R., Oey, M. S., Johnson, D. R. H., & Stone, R. P. S. 1990, AJ, 99, 595 [NASA ADS] [CrossRef] [Google Scholar]
  30. Soderblom, D. R., Jones, B. F., Balachandran, S., et al. 1993, AJ, 106, 1059 [NASA ADS] [CrossRef] [Google Scholar]
  31. Strom, S. E., Edwards, W., & Skrutskie, M. F. 1993, in Protostars and Planets III, ed. E. H. Levy, & J. I. Lunine (Tucson: University of Arizona Press), 837 [Google Scholar]
  32. Wyatt, M. 2008, ARA&A, 46, 339 [NASA ADS] [CrossRef] [Google Scholar]
  33. Wenzel, W. 1989, Inform. Bull. Var. Stars., 3280, 1 [Google Scholar]
  34. Zaitseva, G. V. 1985, Variable Stars, 22, 181 [Google Scholar]

All Figures

thumbnail Fig. 1

The spectrum of RZ Psc around the Li 6708 Å line (thin line). The synthetic spectra with the Li excess (100 times larger than the solar abundance) (thick gray line) and without excess (dashed line) are shown (see text for details).

In the text
thumbnail Fig. 2

Comparison of EW (Li, 6708 Å) in the spectrum of RZ Psc (circle) with the data for the stars in the clusters of different age (triangles): Orion (10 Myr) from King (1993); Pleiades (70 Myr) and Hyades (600 Myr) from Soderblom et al. (1990; 1993).

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.