Sergey V. Marchenko

The Luminous Eclipsing SMC OB + WN Binary HD 5980 before and during the Recent LBV-like Outburst: An Extreme Case of Colliding Winds

The 1994 LBV-(luminous blue variable)like outburst of one of the luminous, hot components of the binary HD 5980 made it the brightest star in the Small Magellanic Cloud for an interval of 5 months. The most intriguing question to arise from this event is the following: Why did the HD 5980 spectrum change from an H-poor WN3 with veiled OB absorption lines about 20 yr ago to an H-rich WN11 without central absorption lines during the outburst? In an attempt to answer this apparent enigma, we present and analyze new phase-dependent spectroscopic, polarimetric, and light-curve observations. Together with other published data, these new observations allow us to improve the orbital parameters considerably, except for the radial velocity amplitudes and hence the masses, which are only roughly constrained. Especially important in HD 5980 is the strong collision of the two nearly equal pre-outburst winds. The emission-line spectrum generated by the collision tends to mask the underlying line spectra of both components when the system is relatively quiescent. We argue that the pre-erupting system consists of a very luminous but moderately massive H-rich O type supergiant, possibly with emission lines, and a low-mass, H-poor, relatively faint WN companion, whose lines are mostly drowned out by wind collision emission, the spectrum of which largely imitates that of a WNE star. It was the O supergiant that erupted in a normal way as an H-rich, visually bright WN11 star. In this way, the need for peculiar evolutionary scenarios (e.g., rapid evolution from a faint, low-mass, H-poor WNE star to a luminous, H-rich WNL star) is avoided.

The Hipparcos distance determination of the Wolf-Rayet system gamma(2) Velorum (WC8+O) and its ramifications

Hipparcos parallax measurements give a distance to the Wolf-Rayet WC8 + O spectroscopic binary gamma(2) Vel of d = 258(-31)(+41) pc and a distance to the O4I(n)f star zeta Pup of d = 429(-77)(+120) pc. Adopting for gamma(2) Vel an interstellar extinction of A(v) = 0.06 mag, this implies an absolute magnitude M-v = -5.4 mag for the WC8 + O binary system. Given that the binary components have a magnitude difference Delta m = 1.4 mag, we derive M-v(WC8) = -3.7 and M-v(O) = -5.0 mag. The latter indicates an 08.5III rather than an O9I companion, as was adopted during the last 25 years. Apparently gamma(2) Vel is not a member of, but a foreground object before the open cluster Cr173 and the association Vel OB2. Given a re-assessment of the distance of the Gum Nebula, gamma(2) Vel is still one of its ionizing sources, while zeta Pup appears to be located at the back of the Gum Nebula. Consequences of the Hipparcos distance determination of gamma(2) Vel for its mass, mass loss rate, luminosities at various wavelengths, and, briefly, its association with the Gum Nebula, are discussed

delta Ceti Is Not Monoperiodic: Seismic Modeling of a beta Cephei Star from MOST Space-based Photometry

The β Cephei star δ Ceti was considered one of the few monoperiodic variables in its class. Despite (or perhaps because of) its apparently simple oscillation spectrum, it has been challenging and controversial to identify this stars pulsation mode and constrain its physical parameters seismically. Broadband time-resolved photometry of δ Ceti spanning 18.7 days with a duty cycle of about 65% obtained by the Microvariability and Oscillations of Stars (MOST) satellite—the first scientific observations ever obtained by MOST—reveals that the star is actually multiperiodic. Besides the well-known dominant frequency of f1 = 6.205886 day-1, we have discovered in the MOST data its first harmonic 2f1 and three other frequencies (f2 = 3.737, f3 = 3.673, and f4 = 0.318 day-1), all detected with a signal-to-noise ratio > 4. In retrospect, f2 was also present in archival spectral line-profile data but at lower S/N. We present seismic models whose modes match exactly the frequencies f1 and f2. Only one model falls within the common part of the error boxes of the stars observed surface gravity and effective temperature from photometry and spectroscopy. In this model, f1 is the radial (l = 0) first overtone, and f2 is the g2 (l = 2, m = 0) mode. This model has a mass of 10.2 ± 0.2 M☉ and an age of 17.9 ± 0.3 Myr, making δ Ceti an evolved β Cephei star. If f2 and f3 are rotationally split components of the same g2 mode, then the stars equatorial rotation velocity is either 27.6 km s-1 or half this value. Given its v sin i of about 1 km s-1, this implies that we are seeing δ Ceti nearly pole-on.

The Unusual 2001 Periastron Passage in the “Clockwork” Colliding-Wind Binary WR 140

We follow, using both optical spectroscopy and photometry, the ‘‘ textbook ’’ colliding-wind WR+O binary WR 140 through and between the periastron passages of 1993 and 2001. An extensive collection of high-quality spectra allows us to derive precise orbital elements for both components simultaneously. We confirm the extremely high eccentricity of the system, e ¼ 0:881 � 0:005, find an excellent match of the newly derived period to the previous estimates, P ¼ 2899:0 � 1:3 days, and improve the accuracy of the time of periastron passage, T0 ¼ HJD 2; 446; 147:4 � 3:7. Around periastron, at orbital phases � � 0:995 1:015, additional emission components appear on the tops of the broad Wolf-Rayet emission lines of relatively low ionization potential. The phase-dependent behavior of these excess line emissions points to their origin in the wind-wind collision zone, which allows us to place some limits on the orbital inclination of the system, i ¼ 50 � � 15 � , and half-opening angle of the bow shock cone, � ¼ 40 � � 15 � . The relatively sudden appearance and disappearance of the extra emission components probably signify a rapid switch from an adiabatically to a radiatively dominated regime and back again. Multiyear UBV photometry provides one more surprise: in 2001 at � ¼ 0:02 0:06, the system went through a series of rapid, eclipse-like events. Assuming these events to be related to an episode of enhanced dust formation at periastron, we estimate the characteristic size of the dust grains to be a � 0:07 lm. Subject headings: binaries: spectroscopic — stars: early-type — stars: individual (WR 140) — stars: Wolf-Rayet On-line material: machine-readable tables

Massive Binary WR 112 and Properties of Wolf-Rayet Dust

Some hot, massive, Population I Wolf-Rayet (W-R) stars of the carbon subclass are known to be prolific dust producers. How dust can form in such a hostile environment remains a mystery. Here we report the discovery of a relatively cool, extended, multiarc dust envelope around the star WR 112, most likely formed by wind-wind collision in a long-period binary system. We derive the binary orbital parameters, the dust temperature, and the dust mass distributions in the envelope. We find that amorphous carbon is a main constituent of the dust, in agreement with earlier estimates and theoretical predictions. However, the characteristic size of the dust grains is estimated to be ~1 μm, significantly larger than theoretical limits. The dust production rate is 6.1 × 10-7 M☉ yr-1, and the total detectable dust mass is found to be about 2.8 × 10-5 M☉ (for d = 4.15 kpc). We also show that, despite the hostile environment, at least ~20% of the initially formed dust may reach the interstellar medium.

A 2.3 Day Periodic Variability in the Apparently Single Wolf-Rayet Star WR 134: Collapsed Companion or Rotational Modulation?

The apparently single WN 6 type star WR 134 (HD 191765) is distinguished among the Wolf-Rayet star population by its strong, presumably cyclical (≈2.3 day) spectral variations. A true periodicity—which is still very much debated—would render WR 134 a prime candidate for harboring either a collapsed companion or a rotating, large-scale, inhomogeneous outflow. We have carried out an intensive campaign of spectroscopic and photometric monitoring of WR 134 from 1989 to 1997 in an attempt to reveal the true nature of this object. This unprecedentedly large data set allows us to confirm unambiguously the existence of a coherent 2.25±0.05 day periodicity in the line-profile changes of He II λ4686, although the global pattern of variability is different from one epoch to another. This period is only marginally detected in the photometric data set. Assuming the 2.25 day periodic variability to be induced by orbital motion of a collapsed companion, we develop a simple model that aims to investigate (1) the effect of this strongly ionizing, accreting companion on the Wolf-Rayet wind structure, and (2) the expected emergent X-ray luminosity. We argue that the predicted and observed X-ray fluxes can only be matched if the accretion on the collapsed star is significantly inhibited. Additionally, we performed simulations of line-profile variations caused by the orbital revolution of a localized, strongly ionized wind cavity surrounding the X-ray source. A reasonable fit is achieved between the observed and modeled phase-dependent line profiles of He II λ4686. However, the derived size of the photoionized zone substantially exceeds our expectations, given the observed low-level X-ray flux. Alternatively, we explore rotational modulation of a persistent, largely anisotropic outflow as the origin of the observed cyclical variability. Although qualitative, this hypothesis leads to greater consistency with the observations.

Wind inhomogeneities in Wolf-Rayet stars. IV. Using clumps to probe the wind structure in the WC8 star HD 192103

We present the most intensive, high-quality spectroscopic monitoring of optical Wolf-Rayet emission lines ever obtained. The Wolf-Rayet star HD 192103 (\WR 135; subtype WC8) was observed in the 5650¨5840 regime alternately from both the William Herschel Telescope and the Canada-France- Ae Hawaii Telescope. The —nal data consist of a series of 197 spectra spread over 64 hr, each with a resolv- ing power j/*j ^ 20,000 and a signal-to-noise ratio in the continuum ^450 per 3 pixel resolution element. We clearly and unambiguously identify stochastic, structured patterns of intrinsic variability at the 1%¨2% level of the line —ux in the broad C III j5696 emission line. The j5801/12 doublet emission is also found to be variable at the 0.2%¨0.5% level of the line —ux. We —nd a correlation between the variability patterns observed in C III and C IV, which suggests a signi—cant overlap in the emission volumes of these transitions, although C IV is known to arise somewhat closer to the star. We attempt to reproduce the observed line pro—le variation patterns using a simple phenomenological model, which assumes the wind to be fully clumped. With a minimal set of assumptions, we are able to reproduce both the shape and the variability in the C III j5696 emission pro—le. We show that the variability pattern provides constraints on the radial extent of WR 135ˇs wind where C III is produced, as well as on the local wind acceleration rate. However, our simple clump model does not reproduce the lower variability in the C IV doublet unless we assume the C IV emission to occur in a much larger volume than C III, implying that signi—cant C IV emission occurs farther out in the wind than C III. We suggest that while some C IV emission might occur farther out, possibly because of reionization from shocks, a more likely explanation is that wind clumping signi—cantly increases with distance from the star, leading to larger variability levels in C III, formed farther out than most of C IV. Alternatively, optical depth eUects and/or local ionization gradients within clumps could conspire to attenuate clumping eUects in the C IV emis- sion line while enhancing them in the C III line.

Oscillations in the massive wolf-rayet star WR 123 with the MOST satellite

We present the results of intensive visual-broadband photometric monitoring of the highly variable WN8 Wolf-Rayet star WR 123, obtained by the MOST (Microvariability and Oscillations of STars) satellite. This first Canadian astronomical space telescope observed WR 123 for 38 days nonstop during 2004 June and July. Fourier analysis shows that no periodic signal is stable for more than several days in the low-frequency domain (f < 1 day-1), where most of the stochastic power is contained. Also, no significant variability is seen in the high-frequency domain (10 day-1 < f < 1400 day-1) down to the level of 0.2 mmag, an order of magnitude lower than theoretical predictions for strange-mode pulsations. On the other hand, there seems to be a relatively stable 9.8 hr periodic signal present throughout the whole run. This period is probably too short to represent the axial rotation of the star, unless it is related to multiple substructures equidistantly spread along the stellar equator. It is also too short to be orbital in nature; it is more likely to be related to pulsational instablilities (although with a much longer period than expected), thus finally revealing a possible fundamental driver behind the highly variable wind of this object, and others of similar type.

STELLAR MODEL ANALYSIS OF THE OSCILLATION SPECTRUM OF η BOOTIS OBTAINED FROM MOST

Eight consecutive low-frequency radial p-modes are identified in the G0 IV star η Bootis based on 27 days of ultraprecise rapid photometry obtained by the MOST (Microvariability and Oscillations of Stars) satellite. The MOST data extend smoothly, to lower overtones, the sequence of radial p-modes reported in earlier ground-based spectroscopy by other groups. The sampling is nearly continuous; hence, the ambiguities in p-mode identifications due to aliases, such as the cycle day-1 alias found in ground observations, are not an issue. The lower overtone modes from the MOST data constrain the interior structure of the model of η Boo, giving a best fit on a grid of ~300,000 stellar models for a composition of (X,Z) = (0.71,0.04), a mass of M = 1.71 ± 0.05 M☉, and an age of t = 2.40 ± 0.03 Gyr. The surface temperature and luminosity of this model, which were constrained only by using the oscillation modes, are close (1 σ) to current best estimates of η Boos surface temperature and luminosity. With the interior fit anchored by the lower overtone modes seen by MOST, standard models are not able to fit the higher overtone modes with the same level of accuracy. The discrepancy, model minus observed frequency, increases from 0.5 μHz at 250 μHz to 5 μHz at 1000 μHz and is similar to the discrepancy that exists between the Suns observed p-mode frequencies and the p-mode frequencies of the standard solar model. This discrepancy promises to be a powerful constraint on models of three-dimensional convection.

THE WOLF-RAYET BINARY V444 CYGNI UNDER THE SPECTROSCOPIC MICROSCOPE. II. PHYSICAL PARAMETERS OF THE WOLF-RAYET WIND AND THE ZONE OF WIND COLLISION

New, extensive, high signal-to-noise, phase-dependent optical spectroscopy, along with simultaneous narrowband continuum photometry, leads to restrictions on the electron temperature and density in the wind close to the Wolf-Rayet component of the eclipsing binary V444 Cygni (WN5 + O6 V-III). Detailed study of the phase dependence of the equivalent widths and line profiles of the W-R star reveals significant ionization stratification in the W-R wind. Our previous discussion of the wind-wind collision effects on He I lines is extended to He II lines. We find that (1) the wind-wind collision zone is detached from the surface of the O star; (2) the radiation field of the O star does not inhibit the initial acceleration of the W-R wind; however, it does brake the flow just prior to entrance into the collision zone; (3) the shocked gas experiences rapid, tcool (2-4) × 104 s, and profound cooling via radiative losses, leading to high compression of the postshock gas.