1 Introduction

The X-ray transient V4641 Sgr (SAX J1819.3-2525) was discovered by BeppoSAX and RXTE observatories in Feb. 1999 (in't Zand et al. 1999; Markwardt et al. 1999). During most of 1999 the source demonstrated only weak X-ray activity (see e.g. in't Zand et al. 2000), until September 1999 when series of very bright and short X-ray flares occurred within less than $\sim$1.5-2 days. During the brightest flare, detected by the All Sky Monitor aboard RXTE the source reached the level of $\sim$12.2 Crab within a few hours (Smith et al. 1999). Subsequently the X-ray flux from V4641 Sgr rapidly decreased and totally disappeared within less than $\sim$0.3 day. On the decaying part of this giant outburst a pointed TOO RXTE observation of the source has been performed (see e.g. Markwardt et al. 1999).

Analysis of optical images showed that the newly discovered X-ray transient can be associated with a variable star discovered by Goranskij (Goranskij 1978; Goranskij 1990; Samus et al. 1999). Previous unusual activity of this star was observed in 1978 (Goranskij 1978), indicating that V4641 Sgr might be a recurrent transient with the recurrence time of $\sim$20 years.

During the giant outburst in Sep. 1999 strong variability of V4641 Sgr was observed at all wavelengths - radio (Hjellming et al. 1999), optical (Stubbings 1999; Kato et al. 1999), X-rays (1-12 keV) and hard X-rays (20-100 keV) (McCollough et al. 1999). The maximal observed optical brightness was at the level of $m_V\sim8.8$ (Stubbings 1999). Spectroscopic observations in the optical band performed during the outburst indicated a presence of a high velocity wind (Charles et al. 1999).

The VLA images, obtained on Sep. 16.03, soon after the brightest 12.2 Crab flare (occurred on Sep. 15.7, 1999), revealed an elongated extended radio source with a size of $\sim$0.25 $\hbox{$^{\prime\prime}$ }$(e.g. Hjellming et al. 2000). This led to a suggestion that V4641 Sgr might be a new Galactic superluminal source. However, determination of the jet velocity was complicated by quick decay of the radio flux and the absence of any direct observations of the apparent motion, in contrast to the three previously known Galactic superluminal sources GRS 1915+105, GRO J1655-40 and XTE J1748-288.

Optical spectroscopy and photometry of the source performed during the quiescence measured all major parameters of the binary system (Orosz et al. 2001, Table 1), including the mass of the primary $M_{\rm primary}\sim$8.7-11.7 $M_{\odot}$ and of the secondary $M_{\rm secondary}\sim$5.5-8.1 $M_{\odot}$, making V4641 Sgr a firm black hole candidate in a binary system with a high mass companion. The distance to the source was constrained in the range of 7-12 kpc (Orosz et al. 2001; Chaty et al. 2000), which exceeds significantly the initial distance estimates of 0.5 kpc (e.g. Hjellming et al. 2000). Chaty et al. (2000) and Orosz et al. (2001) infer slightly different distance estimates, but this difference do not influence strongly on our results. In our paper we will use the estimate obtained by Orosz et al. (2001).

Based primarily on the optical data: (i) significant brightening of the source during the outburst, by $\Delta m_V\approx 5$, unusual for an HMXB transient and (ii) large peak optical flux, $m_{V,{\rm peak}}\approx 8.5$, Revnivtsev et al. (2002) suggested that in Sep. 1999 V4641 Sgr might have had an episode of a super-Eddington accretion. Super-Eddington accretion rate led to the formation of a massive optically thick and geometrically extended envelope/outflow which enshrouded the central black hole. The envelope, being optically thick in the optical and X-ray bands absorbed/reprocessed the primary emission of the central source and re-emitted bulk of the accretion energy in the UV and EUV bands. Being the direct consequence of near- or super-Eddington accretion onto the black hole, the envelope vanishes during the subsequent evolution of the source when the apparent luminosity drops well below the Eddington value. In this paper we present the results of analysis of the data of Rossi X-ray Timing Explorer data and demonstrate that the X-ray data provides additional strong evidence in support of such picture.

In Sects. 2 and 3 we describe results of analysis of the RXTE data. In Sect. 4 we summarize the overall picture emerging from the RXTE data and discuss the X-ray evidence of the extended optically thick envelope surrounding the X-ray source during the maximum of its activity in Sep. 1999.

 

 

Table 1: The parameters of the binary system V4641 Sgr (SAX J1819.3-2525). From Orosz et al. (2001).

Position
${\rm RA}=18^{\rm h}19^{\rm m}21\hbox{$.\!\!^{\rm s}$ }64$	${\rm Dec} = -25^{\circ}24\hbox{$^\prime$ }25\hbox{$.\!\!^{\prime\prime}$ }6$
l=6.77402	b=-4.78906
Parameter	Value
Orbital period, days	$2.81730\pm0.00001$
Mass function, ($M_{\odot}$)	$2.74\pm0.12$
Black hole mass, ($M_{\odot}$)	9.61+2.08-0.88
Secondary star mass, ($M_{\odot}$)	6.53+1.6-1.03
Total mass, ($M_{\odot}$)	16.19+3.58-1.94
Mass ratio	$1.50\pm0.13$
Orbital separation, ($R_{\odot}$)	21.33+1.25-1.02
Secondary star radius, ($R_{\odot}$)	7.47+0.53-0.47
Secondary star luminosity, ($L_{\odot}$)	610+122-104
Distance, kpc	7.4-12.3