2XMMi J225036.9+573154 – a new eclipsing AM Her binary discovered using XMM–Newton

We report the discovery of an eclipsing polar, 2XMMi J225036.9+573154, using XMM–Newton . It was discovered by searching the light curves in the 2XMMi catalogue for objects showing X-ray variability. Its X-ray light curve shows a total eclipse of the white dwarf by the secondary star every 174 min. An extended pre-eclipse absorption dip is observed in soft X-rays at  φ= 0.8–0.9  , with evidence for a further dip in the soft X-ray light curve at  φ∼ 0.4  . Further, X-rays are seen from all orbital phases (apart from the eclipse) which make it unusual amongst eclipsing polars. We have identified the optical counterpart, which is faint  ( r = 21)  , and shows a deep eclipse (>3.5 mag in white light). Its X-ray spectrum does not show a distinct soft X-ray component which is seen in many, but not all, polars. Its optical spectrum shows Hα in emission for a fraction of the orbital period.     

3.8-μm photometry during the secondary eclipse of the extrasolar planet HD 209458b

We report infrared photometry of the extrasolar planet HD 209458b during the time of secondary eclipse (planet passing behind the star). Observations were acquired during two secondary eclipses at the NASA Infrared Telescope Facility (IRTF) in 2003 September. We used a circular variable filter (1.5 per cent bandpass) centred at 3.8 μm to isolate the predicted flux peak of the planet at this wavelength. Residual telluric absorption and instrument variations were removed by offsetting the telescope to nearby bright comparison stars at a high temporal cadence. Our results give a secondary eclipse depth of 0.0013 ± 0.0011, not yet sufficient precision to detect the eclipse, whose expected depth is  ∼0.002 –0.003  . We here elucidate the current observational limitations to this technique, and discuss the approach needed to achieve detections of hot Jupiter secondary eclipses at 3.8 μm from the ground.     

XMM–Newton observations of the eclipsing polar V2301 Oph

We present XMM–Newton observations of the eclipsing polar V2301 Oph which cover nearly 2.5 binary orbital cycles and two eclipses. This polar is believed to have the lowest magnetic field strength (7 MG) of any known polar. We find evidence for structure in the X-ray eclipse profile which shows a 'standstill' feature lasting  26 ± 4  s. This allows us to place an upper limit on the mass of the white dwarf of  ∼1.2 M_⊙  . We find no evidence for quasi-periodic oscillations (QPOs) in the frequency range 0.02–10 Hz. This coupled with the absence of QPOs in RXTE data suggests that, if present, any oscillations in the shock front have a minimal effect on the resultant X-ray flux. We find no evidence for a distinct soft X-ray component in its spectrum – it therefore joins another seven systems which do not show this component. We suggest that those systems which are asynchronous, have low mass-transfer rates or have accretion occurring over a relatively large fraction of the white dwarf are more likely to show this effect. We find that the specific mass-transfer rate has to be close to 0.1 g cm⁻² s⁻¹ to predict masses which are consistent with that derived from our eclipse analysis. This may be due to the fact that the low magnetic field strength allows accretion to take place along a wide range of azimuth.     

Prediction for the He iλ10830 Å absorption wing in the coming event of η Carinae

We propose an explanation for the puzzling appearance of a wide blue absorption wing in the He  i  λ10830 Å  P Cygni profile of the massive binary star η Car several months before periastron passage. Our basic assumption is that the colliding winds region is responsible for the blue wing absorption. By fitting observations, we find that the maximum outflow velocity of this absorbing material is  ∼2300 km s⁻¹  . We also assume that the secondary star is towards the observer at periastron passage. With a toy model, we achieve two significant results. (1) We show that the semimajor axis orientation we use can account for the appearance and evolution of the wide blue wing under our basic assumption. (2) We predict that the Doppler shift (the edge of the absorption profile) will reach a maximum 0–3 weeks before periastron passage, and not necessarily exactly at periastron passage or after periastron passage.     

The mass of the white dwarf in the recurrent nova U Scorpii

We present spectroscopy of the eclipsing recurrent nova U Sco. The radial velocity semi-amplitude of the primary star was found to be     from the motion of the wings of the He  ii λ 4686-Å emission line. By detecting weak absorption features from the secondary star, we find its radial velocity semi-amplitude to be     . From these parameters, we obtain a mass of     for the white dwarf primary star and a mass of     for the secondary star. The radius of the secondary is calculated to be     , confirming that it is evolved. The inclination of the system is calculated to be     , consistent with the deep eclipse seen in the light-curves. The helium emission lines are double-peaked, with the blueshifted regions of the disc being eclipsed prior to the redshifted regions, clearly indicating the presence of an accretion disc. The high mass of the white dwarf is consistent with the thermonuclear runaway model of recurrent nova outbursts, and confirms that U Sco is the best Type Ia supernova progenitor currently known. We predict that U Sco is likely to explode within ∼700 000 yr.     

The evolution of Rapid Burster outbursts

We describe the evolutionary progression of an outburst of the Rapid Burster. Four outbursts have been observed with the Rossi X-Ray Timing Explorer between 1996 February and 1998 May, and our observations are consistent with a standard evolution over the course of each. An outburst can be divided into two distinct phases. Phase I is dominated by type I bursts, with a strong persistent emission component; it lasts for 15–20 d. Phase II is characterized by type II bursts, which occur in a variety of patterns. The light curves of time-averaged luminosity for the outbursts show some evidence for reflares, similar to those seen in soft X-ray transients. The average recurrence time for Rapid Burster outbursts during this period was 218 d, in contrast to an average ∼180‐d recurrence period observed during 1976–1983.     

A TiO study of the dwarf nova IP Pegasi

We present red spectra in the region ∼ λ 7000–8300 Å of the eclipsing dwarf nova IP Peg, with simultaneous narrow-band photometry centred at 7322 Å. We show that by placing a second star on the slit we can correct for the telluric absorption bands which have hitherto made the TiO features from the secondary star unusable. We use these TiO features to carry out a radial velocity study of the secondary star, and find this gives an improvement in the signal-to-noise ratio of a factor of 2 compared with using the Na  i doublet. In contrast with previous results, we find no apparent ellipticity in the radial velocity curve. As a result we revise the semi-amplitude to K ₂=331.3±5.8 km s⁻¹, and thus the primary and secondary star masses to 1.05^-0.07_+0.14 M⊙ and 0.33^-0.05_+0.14 M⊙ respectively. Although this is the lowest mass yet derived for the secondary star, it is still overmassive for its observed spectral type. However, the revised mass and radius bring IP Peg into line with other cataclysmic variables in the mass–radius–period relationships.
By fitting the phase-resolved spectra around the TiO bands to a mean spectrum, we attempt to isolate the light curve of the secondary star. The resulting light curve has marked deviations from the expected ellipsoidal shape. The largest difference is at phase 0.5, and can be explained as an eclipse of the secondary star by the disc, indicating that the disc is optically thick when viewed at high inclination angles.     

Episodic absorption in the outflow of the old nova V603 Aquilae

We report on the time-dependent behaviour of ultraviolet spectral lines in Hubble Space Telescope Goddard High-Resolution Spectrograph data of the classical nova V603 Aql. In particular, episodic blueshifted absorption (extending to ∼−2500 km s⁻¹) is present, with a variability time-scale down to ∼1 min. The data provide a rare opportunity to study the rapid evolution of absorption structures that may be associated with accretion-disc winds in cataclysmic variables. At least three absorption events are recorded (at blueward velocities only) over ∼5 h, each lasting ∼10–15 min. The derived velocity, acceleration and optical depth properties provide an empirical picture of stochastically variable structures in the outflow, with no evidence for short-term (less than ∼1 h) cyclic or modulated behaviour in the overall absorption properties. In contrast, the emission components of the ultraviolet resonance lines are very stable in velocity and strength in this low-inclination system. On at least two occasions there is an intriguing short-term 'flare' in the ultraviolet continuum flux (of up to ∼40 per cent). Though there is no clear one-to-one relation in these data between the continuum fluctuations and the occurrence of the absorption events, the time-scales for the two variable phenomena are essentially the same. The irregular absorption episodes in the ultraviolet data of V603 Aql presently defy a clear physical interpretation. Their overall characteristics are discussed in the context of instabilities in radiation-pressure-driven disc winds.     

Time-series Paschen-β spectroscopy of SU Aurigae

We present time-series echelle spectra of the Paβ line of the T Tauri star SU Aur, observed over three consecutive nights. The line shows strong variability (∼10 per cent) over the velocity range 100–420 km s⁻¹ in the red broad absorption component, and weaker variability (∼2 per cent) over the velocity range  −200–0 km s⁻¹  in the blue wing. The variability in the velocity range  −200–0 km s⁻¹  is correlated with that in  200–400 km s⁻¹  , and the variability in these velocity ranges anticorrelates with that in  0–100 km s⁻¹  . The mean spectrum from the second night shows the suggestion of a blueshifted absorption component at about  −150 km s⁻¹  , similar to that found in the Hα and Hβ lines. We find the position of the subpeak in the red absorption component changes steadily with time, and its motion modulates at half the rotational period. We also find that the modulation of the line equivalent width is possibly associated with a half and a third of the rotational period, which is consistent with the surface Doppler images of SU Aur. Radiative transfer models of a rotationally modulated Paβ line, produced in the shock-heated magnetospheric accretion flow, are also presented. Models with a magnetic dipole offset reproduce the overall characteristics of the observed line variability, including the line equivalent width and the motion of the subpeak in the red absorption trough.     

A ROSAT observation of the dipping, flaring and eclipsing polar MN Hya (RX J0929.1–2404)

We present the results of a 22.5 ks pointed ROSAT PSPC observation of the 3.4-h period eclipsing polar MN Hya (RX J0929.1−2404). The X-ray light curve exhibits a 'double-humped' shape, with a secondary minimum occuring at φ∼ 0.45, a morphology consistent with two-pole accretion. Strong aperiodic flaring activity, with flux enhancements of ∼ 6 × the quiescent level, is also observed. A pre-eclipse 'dip' occurs in the phase interval φ= 0.87–0.95 with the X-rays becoming harder, indicative of photoelectric absorption by the pre-shock flow. There is also evidence of a secondary spectrally hard 'dip' near φ = 0.45–0.55, which might be associated with a second accretion stream flowing to the other magnetic pole.   The X-ray spectrum is best represented by a combination of a ∼50 eV blackbody and a thermal bremsstrahlung component of kT 1.6 keV, with a total absorption column of N _H  = 2.9 × 10²⁰ cm⁻².   The primary maximum (φ∼ 0.65) has a slightly larger column and normalization compared to the secondary maximum. Although there are few photons, the dip spectrum is very flat in comparison to other phases, and is best represented by a single bremsstrahlung component. This is indicative of the spectral hardening seen in the light curves attributed to photoabsorption. The ratio of unabsorbed bremsstrahlung and blackbody luminosities is ∼ 0.1 for the best-fitting average spectral models. This implies a magnetic field strength  30 MG on the basis of the empirical L _hard/ L _soft −  B relationships, although consideration of the cyclotron flux and aspect effects could allow for an even higher field (55 MG).     

Radiation mechanisms and geometry of Cygnus X-1 in the soft state

We present X-ray/ γ -ray spectra of Cyg X-1 observed during the transition from the hard to the soft state and in the soft state by ASCA , RXTE and CGRO /OSSE in 1996 May and June. The spectra consist of a dominant soft component below ∼2 keV and a power-law-like continuum extending to at least ∼800 keV. We interpret them as emission from an optically thick, cold accretion disc and from an optically thin, non-thermal corona above the disc. A fraction f ≳0.5 of total available power is dissipated in the corona.
We model the soft component by multicolour blackbody disc emission taking into account the torque-free inner-boundary condition. If the disc extends down to the minimum stable orbit, the ASCA RXTE data yield the most probable black hole mass of M _X≈10 M_⊙ and an accretion rate,     , locating Cyg X-1 in the soft state in the upper part of the stable, gas-pressure-dominated, accretion-disc solution branch.
The spectrum of the corona is well modelled by repeated Compton scattering of seed photons from the disc off electrons with a hybrid, thermal/non-thermal distribution. The electron distribution can be characterized by a Maxwellian with an equilibrium temperature of kT _e∼30–50 keV, a Thomson optical depth of τ ∼0.3 and a quasi-power-law tail. The compactness of the corona is 2≲ℓ_h≲7, and a presence of a significant population of electron–positron pairs is ruled out.
We find strong signatures of Compton reflection from a cold and ionized medium, presumably an accretion disc, with an apparent reflector solid angle, Ω/2π∼0.5–0.7. The reflected continuum is accompanied by a broad iron K α line.     

Spectroscopy and orbital periods of the old novae V533 Herculis, V446 Herculis and X Serpentis

We report spectroscopic orbital periods of 0.147 d (=3.53 h) for V533 Her, 0.207 d (=4.97 h) for V446 Her and 1.478 d for X Ser. V533 Her (Nova Herculis 1963) shows absorption features in its He  i and Balmer lines which appear only in a limited range of orbital phase, suggesting that it is a low-inclination SW Sextantis star. V446 Her is unusual in that it has started normal dwarf nova eruptions after a nova outburst, but we find nothing else unusual about it – in particular, a distance estimate based on its dwarf nova outbursts agrees nicely with another based on the rate of decline of its nova eruption, both giving d ∼1 kpc. In X Ser, unlike in other old novae with long periods, no spectral features of the secondary star are visible. This and its outburst magnitude both suggest that it is quite distant and luminous, and at least 1 kpc from the Galactic plane.     

uvbyβ photometry of the short-period binary VV Ursae Majoris

We present well-sampled uvby light curves, supplemented by a few β filter measurements, of the Algol binary VV UMa. The light curves are analysed using two different codes to derive the orbital and absolute stellar parameters of this binary. We find reasonably good fits to the light curves and determine the stellar effective temperatures T _eff,1≃9000–9600 K , and T _eff,2≃5300–5600 K with a mass ratio q ≃0.35 . From the light-curve fits we discard the possibility of an anomalous gravity-darkening exponent for the secondary star of this system, as previously suggested.
We find evidence of short-term, small-amplitude variations in the brightness of the system. Two periodicities of about 1.10 and 0.51 h seem to be present in the data for at least two different nights, even within the secondary eclipse. This suggests that VV UMa may be a new Algol binary with a low-amplitude variable primary star, but new data collected during longer observing runs are necessary to confirm the pulsating nature of the brightness variations.     

HST and ground-based eclipse observations of V2051 Ophiuchi: binary parameters

We report on high-speed eclipse photometry of the dwarf nova V2051 Oph while it was in a low brightness state, at B  ≃ 16.2 mag. In comparison with the average IUE spectra, the ultraviolet continuum and emission lines appear reduced by factors of, respectively, ≃ 4 and ≃ 5. Flickering activity is mostly suppressed and the light curve shows the eclipse of a compact white dwarf at the disc centre which contributes ≃ 60 per cent of the total light at 3900–4300 Å. We use measurements of contact phases in the eclipse light curve to derive the binary geometry and to estimate masses and relevant dimensions. We find a mass ratio of q  = 0.19 ± 0.03 and an inclination of i  = 83 ± 2°. The masses of the component stars are M ₁ = 0.78 ± 0.06 M⊙ and M ₂ = 0.15 ± 0.03 M⊙. Our photometric model predicts K ₁ = 83 ± 12 km s⁻¹ and K ₂ = 436 ± 11 km s⁻¹. The predicted value of K₁ is in accordance with the velocity amplitude obtained from the emission lines after a correction for asymmetric line emission in the disc is made. The secondary of V2051 Oph is significantly more massive than the secondaries of the other ultrashort period dwarf novae. V2051 Oph is probably a relatively young system, with a secondary star that has not had enough time to evolve out of thermal equilibrium.     

High-speed energy-resolved STJ photometry of the eclipsing binary UZ For

We present high-time-resolution optical photometry of the eclipsing binary UZ For using a superconducting tunnel junction (STJ) device, a photon-counting array detector with intrinsic energy resolution. Three eclipses of the ∼18-mag 126.5-min orbital binary were observed using a 6×6 array of tantalum STJs at the 4.2-m William Herschel Telescope on La Palma. The detector presently provides individual photon arrival-time accuracy to about 5 μs, and a wavelength resolution of about 60 nm at 500 nm, with each array element capable of counting up to ∼5000 photon s⁻¹. The data allow us to place accurate constraints on the accretion geometry from our time- and spectrally resolved monitoring, especially of the eclipse ingress and egress. We find that there are two small accretion regions, located close to the poles of the white dwarf. The positions of these are accurately constrained, and show little movement from eclipse to eclipse, even over a number of years. The colour of the emission from the two regions appears similar, although their X-ray properties are known to be significantly different: we argue that the usual accretion shock may be absent at the non-X-ray-emitting region, and instead the flow here interacts directly with the white dwarf surface; alternatively, a special grazing occultation of this region is required. There is no evidence for any quasi-periodic oscillations on time-scales of the order of seconds, consistent with relatively stable cyclotron cooling in each accretion region.     

The X-ray eclipse of OY Car resolved with XMM–Newton: X-ray emission from the polar regions of the white dwarf

We present the XMM–Newton X-ray eclipse light curve of the dwarf nova OY Car. The eclipse ingress and egress are well resolved for the first time in any dwarf nova placing strong constraints on the size and the location of the X-ray emitting region. We find good fits to a simple linear eclipse model, giving ingress/egress durations of  30 ± 3 s (Δφ_orb= 0.0054 ± 0.0005)  . Remarkably, this is shorter than the ingress/egress duration of the sharp eclipse in the optical, as measured by Wood et al. (1989) and ascribed to the white dwarf  (43 ± 2 s)  . We also find that the X-ray eclipse is narrower than the optical eclipse by  14 ± 2 s  , which is precisely the difference required to align the second and third contact points of the X-ray and optical eclipses. We discuss these results and conclude that X-ray emission in OY Car arises most likely from the polar regions of the white dwarf.
Our data were originally reported by Ramsay et al. (2001b) , but they did not make a quantitative measurement of eclipse parameters. We have also corrected important timing anomalies present in the data available at that time.     

Photometric variability of WC9 stars

Do some Wolf–Rayet stars owe their strong winds to something else besides radiation pressure? The answer to this question is still not entirely obvious, especially in certain Wolf–Rayet subclasses, mainly WN8 and WC9. Both of these types of Wolf–Rayet stars are thought to be highly variable, as suggested by observations, possibly due to pulsations. However, only the WN8 stars have so far been vigorously and systematically investigated for variability. We present here the results of a systematic survey during three consecutive weeks of 19 Galactic WC9 stars and one WC8 star for photometric variability in two optical bands, V and I . Of particular interest are the correlated variations in brightness and colour index in the context of carbon dust formation, which occurs frequently in WC9 and some WC8 stars. In the most variable case, WR76, we used this information to derive a typical dust grain size of  ∼ 0.1 μm  . However, most photometric variations occur at surprisingly low levels, and in fact almost half of our sample shows no significant variability at all above the instrumental level (  σ∼ 0.005– 0.01  mag).     

Hotspots and a clumpy disc: variability of brown dwarfs and stars in the young σ Ori cluster

The properties of accretion discs around stars and brown dwarfs in the σ Ori cluster (age 3 Myr) are studied based on near-infrared (IR) time series photometry supported by mid-IR spectral energy distributions (SEDs). We monitor ∼30 young low-mass sources over eight nights in the J and K band using the duPont telescope at Las Campanas. We find three objects showing variability with J -band amplitudes  ≥0.5 mag  ; five additional objects exhibit low-level variations. All three highly variable sources have been previously identified as highly variable; thus, we establish the long-term nature of their flux changes. The light curves contain periodic components with time-scales of  ∼0.5–8 d  , but have additional irregular variations superimposed – the characteristic behaviour for classical T Tauri stars. Based on the colour variability, we conclude that hotspots are the dominant cause of variations in two objects (#19 and #33), including one likely brown dwarf, with spot temperatures in the range of 6000–7000 K. For the third one (#2), a brown dwarf or very low-mass star, inhomogeneities at the inner edge of the disc are the likely origin of variability. Based on mid-IR data from Spitzer , we confirm that the three highly variable sources are surrounded by circum-(sub)-stellar discs. They show typical SEDs for T Tauri-like objects. Using SED models, we infer an enhanced scaleheight in the disc for the object #2, which favours the detection of disc inhomogeneities in light curves and is thus consistent with the information from variability. In the σ Ori cluster, about every fifth accreting low-mass object shows persistent high-level photometric variability. We demonstrate that estimates for fundamental parameters in such objects can be significantly improved by determining the extent and origin of the variations.     

Opacity effects and shock-in-jet modelling of low-level activity in Cygnus X-3

We present simultaneous dual-frequency radio observations of Cygnus X-3 during a phase of low-level activity. We constrain the minimum variability time-scale to be 20 min at 43 GHz and 30 min at 15 GHz, implying source sizes of 2–4 au. We detect polarized emission at a level of a few per cent at 43 GHz which varies with the total intensity. The delay of ∼10 min between the peaks of the flares at the two frequencies is seen to decrease with time, and we find that synchrotron self-absorption and free–free absorption by entrained thermal material play a larger role in determining the opacity than absorption in the stellar wind of the companion. A shock-in-jet model gives a good fit to the light curves at all frequencies, demonstrating that this mechanism, which has previously been used to explain the brighter, longer lived giant outbursts in this source, is also applicable to these low-level flaring events. Assembling the data from outbursts spanning over two orders of magnitude in flux density shows evidence for a strong correlation between the peak brightness of an event, and the time-scale and frequency at which this is attained. Brighter flares evolve on longer time-scales and peak at lower frequencies. Analysis of the fitted model parameters suggests that brighter outbursts are due to shocks forming further downstream in the jet, with an increased electron normalization and magnetic field strength both playing a role in setting the strength of the outburst.     

Sakurai's object: the ionized nebula at radio wavelengths

Sakurai's object (V4334 Sgr) is a planetary nebula nucleus which is undergoing its final helium shell flash. This is the first of these rare and important events to be observable with non-optical instruments. We report the first radio detection, using a short (2-h) observation with the Very Large Array (VLA) at 4.86 GHz. The radio emission structure is coincident with the 34-arcsec diameter planetary nebula seen in optical emission lines. We find a statistical distance ∼ 3.8 ± 0.6 kpc, with a range of 1.9 <  D  < 5.3 kpc, depending on the planetary nebula (PN) mass. While we have no direct evidence for a new (post-flash) stellar wind, we estimate an upper limit to the mass-loss rate due to any such wind of 1.7 × 10⁻⁷ M⊙ yr⁻¹. The number of emitting knots in the radio-visible nebula indicates an electron density of ∼ 2 × 10⁸ m⁻³ in those knots, and a total emitting ionized mass of ∼ 0.15 M⊙, at an assumed distance of 3.8 kpc. The radio flux density indicates an Hβ flux of ∼ 6 × 10⁻¹⁶ W m⁻², suggesting an extinction E ( B  −  V ) ∼ 1.15, comparable with reddening estimates in the direction of V4334 Sgr.