Support ATel At Patreon

[ Previous | Next | ADS ]

31.4 day periodicity in Cyg X-3 ASM-data: evidence for a tertiary body or precessing disc ?

ATel #3325; Osmi Vilhu (Univ. Helsinki, Finland), Pasi Hakala (Univ. Turku, Finland), Diana Hannikainen (Aalto University Metsähovi Radio Observatory, Finland)
on 5 May 2011; 08:00 UT
Credential Certification: Osmi Vilhu (osmi.vilhu@helsinki.fi)

Subjects: X-ray, Binary, Black Hole, Neutron Star, Star, Variables

Referred to by ATel #: 3437

The X-ray/gamma-ray activity of the microquasar Cygnus X-3 has been reported in several recent ATELs: #2644 (INTEGRAL), #2645 and #3239 (AGILE), #2646 and #3233 (Fermi LAT). Here we report the finding of a periodic change in the X-ray light curve minimum time (Observed-Computed, O-C) using RXTE/ASM soft X-ray data. The ASM data consisted of 97450 dwell datapoints taken during 5374 days (up to October 2010). We analysed these data using the Singh et al. (2002) parabolic ephemeris. Search for observed times of minima (corresponding to minimum countrates) were limited to orbital phases between -0.1 and 0.1 resulting in 7018 binary orbits (P_orb=4.8 h). These were compared with the minimum times computed from the ephemeris resulting in (O-C). This showed a strong periodic spike at 31.364 day with PSD = 35 (using the IDL-code dynpsrch.pro, see http://www.helsinki.fi/~vilhu/DYNPSRCH.PS ). If the search was done over 10 orbit bins 2141 data points with somewhat less prominent dynamical spectrum resulted (see http://www.helsinki.fi/~vilhu/DYNPSRCH_10PER.PS ). The soft X-ray light curve across the orbital phase is somewhat asymmetric in the sense that the declining side (around phase 0.75) is steeper than the rising one (phase 0.25). We quantified the asymmetry from dwell data by (F075-F025)/(F075+F025) where F075 and F025 are mean count rates between orbital phases 0.6-0.9 and 0.1-0.4, respectively, in those orbits where data are enough (5492 data bins). We folded both (O-C) and asymmetry with the 31.364 d period and obtained fits for the modulation ( using A(0) + A(1)*SIN(2Pi*phase+A(2) ). Here the phase was computed with zero-point MJD 50088.0 ( see http://www.helsinki.fi/~vilhu/OMC_AND_ASYMMETRY.PS and http://www.helsinki.fi/~vilhu/OMC_AND_ASYMMETRY_10PER.PS ): O-C : A(0)=-1.15+-0.20 min; A(1)=2.44+-0.28 min; A(2)=-0.17+-0.11 ; O-C (10 per) : A(0)=-1.13+-0.37 min; A(1)=2.43+-0.53 min; A(2)= 0.16+-0.21 ; Asymmetry: A(0)= 0.02+-0.002 ; A(1)=0.016+-0.003 ; A(2)= 0.087+-0.19 . If the (O-C)-modulation above is applied to the average light curve (normalising each orbit first by the daily mean, http://www.helsinki.fi/~vilhu/ASM_MEAN.PS ) then this size of asymmetry- modulation follows, giving strong support to the reality of the (O-C)-modulation with amplitude of 2.4 minutes. We assume that this modulation is due to a third body orbiting around the binary system. The light-time effect would change the minimum time (O - C) in a periodic manner like in many short period eclipsing binaries having multiple components (Pribulla et al., 2005). Assuming circular orbits this light-time effect would fix the tertiary body mass provided that the orbit inclination and Cyg X-3 mass are known. Since these are not known, a wide variety of possibilities exist (see http://www.helsinki.fi/~vilhu/3RDBODY.PS ): Mcyg=2Msun: M3rd=6+-1 (inc=90deg), M3rd=30+-7 (inc=30deg); Mcyg=10Msun:M3rd=12+-2 (inc=90deg), M3rd=40+-10 (inc=30deg). Hence, it is not excluded that the tertiary body is a Wolf Rayet star. This type of star is widely assumed to be the companion star in the Cyg X-3 binary system (compact star + WR star). CygX-3 could even be a LMXRB with a low-mass star orbiting around a neutron star or black hole and the observed WR-signatures are from the tertiary body. However, we note that the 31.4 day periodicity may be similar to the 3rd periods found in several X-ray binaries and possibly due to (warped) disc precession. A paper explaining the details and discussing physical models is in preparation.

atel