A decelerating jet observed by the EVN and VLBA in the X-ray transient XTE J1752-223
J. Yang, C. Brocksopp, S. Corbel, Z. Paragi, T. Tzioumis, R.P. Fender
aa r X i v : . [ a s t r o - ph . H E ] S e p Mon. Not. R. Astron. Soc. , 1–5 (2010) Printed 3 November 2018 (MN L A TEX style file v2.2)
A decelerating jet observed by the EVN and VLBAin the X-ray transient XTE J1752 − J. Yang ⋆ , C. Brocksopp , S. Corbel , Z. Paragi , , T. Tzioumis and R.P. Fender Joint Institute for VLBI in Europe, Postbus 2, 7990 AA Dwingeloo, The Netherlands Mullard Space Science Laboratory, University College London, Holmbury St Mary, Dorking, Surrey RH5 6NT, UK Universit´e Paris 7 Denis Diderot and Service d’Astrophysique, UMR AIM, CEA Saclay, F-91191 Gif-sur-Yvette, France MTA Research Group for Physical Geodesy and Geodynamics, POB 91, H-1521 Budapest, Hungary Australia Telescope National Facility, CSIRO, P.O. Box 76, Epping, NSW 1710, Australia. School of Physics and Astronomy, University of Southampton, Highfield, Southampton, SO17 1BJ, UK
Accepted 2010 September 7. Received 2010 August 27; in original form 2010 July 23
ABSTRACT
The recently discovered Galactic X-ray transient XTE J1752 −
223 entered its firstknown outburst in 2010, emitting from the X-ray to the radio regimes. Its general X-ray properties were consistent with those of a black hole candidate in various spectralstates, when ejection of jet components is expected. To verify this, we carried out verylong baseline interferometry (VLBI) observations. The measurements were carried outwith the European VLBI Network (EVN) and the Very Long Baseline Array (VLBA)at four epochs in 2010 February. The images at the first three epochs show a movingjet component that is significantly decelerated by the last epoch, when a new jetcomponent appears that is likely to be associated with the receding jet side. Theoverall picture is consistent with an initially mildly relativistic jet, interacting withthe interstellar medium or with swept-up material along the jet. The brightening of thereceding ejecta at the final epoch can be well explained by initial Doppler deboostingof the emission in the decelerating jet.
Key words: stars: individual: XTE J 1752 −
223 – stars: variable: others – ISM: jetsand outflows – radio continuum: stars – X-rays: binaries.
It is clear from the literature of the past three decadesthat, for almost every black hole X-ray transient (XRT)observed at radio wavelengths, a radio counterpart hasbeen discovered (Fender 2006). In a small number ofsources, the ejecta have been resolved and monitoredas they travel away from the central source. Thus, ithas been possible on rare occasions to measure propermotions, sometimes at apparently super-luminal veloc-ities: e.g. GRS 1915+105 (Mirabel & Rodr´ıguez 1994;Fender et al. 1999), GRO J1655 −
40 (Tingay et al. 1995;Hjellming & Rupen 1995), GX 339 − −
564 (Corbel et al. 2002; Kaaret et al.2003) and XTE J1748 −
288 (Hjellming et al. 1998;Miller-Jones et al. 2008). Jet deceleration has beenalso investigated in GRS 1915+105 although no conclusiveevidence was found (Miller-Jones et al. 2007). ⋆ E-mail: [email protected]
The X-ray transient XTE J1752 −
223 was discovered bythe
Rossi X-ray Timing Explorer ( RXTE ) on 2009 October23 (Markwardt et al. 2009) at the start of its first knownX-ray outburst. It showed a long and gradual rise at X-rayenergies, whilst remaining spectrally hard. The X-ray sourcelater evolved, became softer (Homan 2010; Brocksopp et al.2010a) and entered a spectral state commonly associatedwith jet ejection events (Fender et al. 2004, 2009). Theoutburst has been well monitored by the
Monitor of All-sky X-ray Image ( MAXI , Nakahira et al. 2010),
RXTE (Shaposhnikov et al. 2010), and
Swift (Curran et al. 2010)at X-ray energies. All these X-ray observations show thatXTE J1752 −
223 is likely to be a black hole transient.Following the activation of XTE J1752 − ∼ c (cid:13) J. Yang, C. et al.
VLBI Network) experiment and three follow-up VLBA ex-periments at 5 GHz in 2010 February. In this paper, wepresent the results of these VLBI (Very Long Baseline In-terferometry) observations.
The performed VLBI experiments are summarised in Ta-ble 1. The low declination and potentially weak flux ofXTE J1752 −
223 were possible problems for VLBI obser-vations. Therefore, to quickly resolve these concerns, weinitiated an e-VLBI experiment with the western Euro-pean telescopes (Szomoru 2008). The EVN experiment used1024 Mbps data rate (16 channels, 16 MHz per chan-nel, 2 bit sampling, dual polarisation). The real-time cor-relation was done with the Earth orientation parameters(EOP) predicted from the EOP model of one day earlier.We applied 2-second integration time, and 16 frequencypoints per channel. The participating stations were Medic-ina, Yebes, Torun, Onsala, and Westerbork. In the EVNexperiment, we used the following J2000 coordinate: RA =17 h m . s − ◦ ′ . ′′
782 (positional uncer-tainty: σ = 0 . ′′ . ◦
8) phase-referencing source: PMN J1755 − h m . s − ◦ ′ . ′′ σ ∼
15 mas). As these sources have low elevation ( < ◦ )in Europe, we used a short cycle time: 160 seconds on thetarget and 80 seconds on the reference source. We also ob-served a strong and compact source (NRAO 530) as thefringe finder and bandpass calibrator.We successfully detected a radio source, consistent withan ejection from the black hole candidate (Brocksopp et al.2010b), during the EVN experiment and then performedthree follow-up VLBA experiments. We used the samephase-referencing source, cycle time, and observing fre-quency. The recording data rate was 512 Mbps (16 channels,8 MHz per channel, 2 bit sampling, dual polarisation). Therewere eight VLBA telescopes available at the first epoch, nineat other two epochs. The data were correlated with the sameparameters as the EVN experiment.We also performed an EVN experiment in March andtwo VLBA experiments in April to study other ejectionevents and attempt to detect the core. These additionalresults will be presented by Brocksopp et al. in a generalpaper. We used the NRAO software
AIPS (Astronomical ImageProcessing System) to perform the initial calibrations. The a-priori amplitude calibration was done with measured sys-tem temperatures and antenna gain curves. We correctedthe EOP model for the VLBA data before any phase cali-brations. We follow the same procedure for both the EVNand VLBA data reduction. (1) We corrected the parallacticangle. (2) We ran the global fringe-fitting for NRAO 530with half-minute solution interval and then solved for theinstrumental bandpass. (3) With the bandpass solutions,
Table 1.
The summary of the image parameters of Fig. 1.
Exp. Date Array N ant T obs S peak σ rms b maj b min φ padd/mm/yy (hour) (mJy/b) (mJy/b) (mas) (mas) ( ◦ )RY001 11/02/10 EVN 5 1.2 2.32 0.21 14.5 6.1 − we re-ran the fringe-fitting to solve for the instrumentalphase and delay using two-minute data of NRAO 530. (4)We ran the fringe-fitting to solve for the phase, the fringerate, and the delay for the calibrators with a solution inter-val of the scan length ( ∼ − <
10 mJy) most likely due to scat-ter broadening. In view of this problem, we removed both Scand Mk. The phase wrapped slowly (fringe rate < − −
223 by linear interpolation. (5) The data wereaveraged in each IF and then split into single-source filesafter all the corrections were applied.The reference source PMN J1755 − Difmap (Shepherd et al. 1994). The source was well repre-sented by a circular Gaussian model with a size of 4 . ∼
200 mJy at 5 GHz. Finally, weself-calibrated the u – v data with the model and applied theamplitude and phase solutions to both sources in AIPS. − The imaging results for the X-ray transient XTE J1752 − h m . s − ◦ ′ . ′′ − inthe dirty map at the first epoch, when natural weightingis used. After removing component A with a circular Gaus-sian model, we notice that there may be at least one more jetcomponent. One candidate is located at angular separation18.7 mas, position angle − . ◦
0; the other at angular sepa-ration 70.6 mas and position angle − . ◦
3. Both candidateshave a peak brightness ∼ .
91 mJy beam − ( ∼ σ rms ) usingnatural weighting. In Fig. 1a, the two candidates show thesecond positive contours. If either candidate is removed bycircular Gaussian model fitting, the other also becomes faint.If we reduce the contribution of the long baselines, both be- c (cid:13) , 1–5 decelerating jet in the X-ray transient XTE J1752 − (b) 18 Feb, VLBA (d) 26 Feb, VLBA(c) 23 Feb, VLBA(a) 11 Feb, EVN A A AA B R e l a t i v e d i s t an c e ( m a s ) MJD 55238.4 (day)
Figure 1.
The decelerating jet of the X-ray transient XTE J1752 − σ off-sourcenoise level and increase by a factor of -1.4, -1, 1, 1.4, 2, 2.8. The related map parameters are listed in Tab. 1. The right panel plots thefitting results (Tab. 3) of the proper motion data (Tab. 2) of component A using the models with (solid curve) and without decelerationrate (dotted line). The reference time MJD 55238.4 corresponds to the first VLBI observations. come brighter and show a small peak ( ∼ u – v coverageduring the 1.2-hour observations, neither components can beunambiguously identified as a true jet component. However,there is evidence for the extended emission for the sourceas the total restored flux density is much lower than that( ∼
16 mJy at 5.5 GHz) measured by the ATCA (Brocksoppet al. in prep.).In the follow-up VLBA observations, the higher reso-lution and sensitivity are achieved by more telescopes andlonger observing time. To image the extended source, weused natural weighting again. Because of the resolved struc-ture and the decaying peak flux density, the source is onlyseen clearly in the dirty map with the synthesised beam(16 . × . − . ◦
6) at the first VLBAepoch. However, the large-scale beam pattern around thefaint source could also be easily identified at the later twoepochs. If we taper the long baselines, use the short baselinesonly, or increase the image pixel size, the source becomes sig-nificantly brighter in the dirty map at the later two epochs.None of the suspected ejecta candidates in the EVN imageare further seen after 7 days in the later VLBA images. Be-cause the diffuse emission can not be well restored by cleancomponents, Gaussian models were used in making all theVLBI images of Fig. 1. Due to the limited SNR (6 – 12),circular rather than elliptical Gaussian model fitting wasadopted to reduce the number of free parameters.Table 2 lists the best-fit parameters of the circularGaussian model. To show the motion of component A, wetake the position of component A measured at the first epochas the reference origin. The random position error was es-timated by √ b maj b min , where b maj and b min are the size ofthe major and minor axes of the used restoring beam, SNRis the signal to noise ratio ( S peak σ rms ) listed in Column (7) ofTable 2. Note that the rather large systematic position error Table 2.
The circular Gaussian model fitting results of the de-tected jet components in the X-ray transient XTE J1752 − Comp. MJD Separation Position Angle Size Flux SNR(day) (mas) ( ◦ ) (mas) (mJy)A 55238.4 0 ± . . ± . − . ± .
90 13.8 2.20 10.7A 55250.6 85 . ± . − . ± .
63 19.0 2.32 8.8A 55253.6 100 . ± . − . ± .
70 13.9 1.05 6.1B 55253.6 387 . ± . . ± .
18 11.9 0.86 6.4 from the reference source will not affect our proper motionmeasurements. The fitted size has the same random error asthe position for each component. Since the measured sizes( > ∼ The angular separation of component A versus time is shownin the right panel of Fig. 1. We take the position and thetime of component A measured at the first epoch as thereference origin. We fit these data points to the followingproper motion model: r = r + µ t − . µt (1)where r is the angular separation; t is the observing time; r and µ are the angular separation and the proper mo-tion at t = 0, ˙ µ is the apparent deceleration rate. The dot-ted straight line and the solid curve represent the uniform c (cid:13) , 1–5 J. Yang, C. et al.
Table 3.
Best-fit parameters using the proper motion modelswith and without the deceleration rate.
Model r µ ˙ µ χ /DoF DoF(mas) (mas day − ) (mas day − )˙ µ = 0 0 . ± .
41 9 . ± .
15 0 . ± .
02 3.9 1˙ µ = 0 2 . ± .
39 6 . ± .
05 0 118.4 2 proper motion model ( ˙ µ = 0) and the proper motion modelwith the deceleration rate ( ˙ µ = 0) respectively. The best-fitparameters are listed in Table 3. The model of ˙ µ = 0 givesan average proper motion of ¯ µ = µ = 6 . ± .
05 mas day − with the reduced χ = 118 .
4. The degree of freedom (DoF)was listed in the last column. The model of ˙ µ = 0 gives µ = 9 . ± .
15 mas day − at MJD 55238.4 and a deceler-ation rate of ˙ µ = 0 . ± .
02 mas day − with the reduced χ = 3 .
9. With the deceleration rate, component A has aproper motion of 4.0 mas day − at the last epoch. It is clearthat the deceleration rate should be included in the propermotion model.Jet deceleration was also found in XTE J1550 −
564 us-ing
Chandra observations (Corbel et al. 2002; Kaaret et al.2003) and XTE J1748 −
288 with VLA observations(Hjellming et al. 1998; Miller-Jones et al. 2008). Comparedwith them, the observed deceleration in XTE J1752 −
223 isfree from the blending of multiple jet components caused bythe low resolution (Hjellming & Rupen 1995). If there is asequence of ejecta which decreased sequentially more rapidlyin flux density with increasing distance from the core, thecluster of components may show a decreasing proper mo-tion. In our case, these VLBI observations have a resolu-tion of <
10 mas and can well identify single ballistic ejectawith a time resolution of less than one day, assuming aninitial proper motion 10 mas day − . XTE J1752 −
223 is thesecond known case of gradual jet deceleration, although ona much smaller scale ( ∼
100 mas) than the first case ofXTE J1550 − −
223 is most likely due to inter-action with the external dense interstellar medium (ISM) orthe residual slowly-moving ejecta from the previous ejectionalong the jet path.
We interpret component A as an approaching jet componentsince it is the only component detected at the four epochsand our VLBI observations were performed just after the as-sociated radio flare reached its peak flux density. The ATCAobservations (Brocksopp et al., in prep.) show that it had adecaying flux density and a pretty stable and steep spectralindex: α = − S ν ∝ ν α ) between 5.5 and 9 GHz, i.e. therewas no indication of another ejection event during our VLBIobservations. Thus, the possibility that the components Aand B detected at the last three epochs are associated with adifferent ejection event can be ruled out. Besides the propermotion, component A shows a hint of expansion. The evo-lution of its size is displayed in Fig. 2. The first three datapoints give an expansion speed of 0 . ± . − withreduced χ = 1 .
1. The expansion speed is much slower thanits proper motion, indicating that its expansion was signifi-cantly confined too. Because component A shows an increas- ing size and a decaying peak brightness, its size estimationat a later stage is limited by the image sensitivity and thelack of short baselines. For this reason, the last data pointwas omitted in the linear fitting. According to the evolutionof component A, a jet component is expected to have morecompact structure at the earlier stage. Compared with thesize (7.9 mas) of component A measured at the first epoch,component B shows a much larger size (11.9 mas). This in-dicates that component B is most likely an evolved compo-nent, which was ejected on the receding jet side at the sametime as component A.According to the expansion speed, component A wasejected 8.7 days earlier, i.e. at MJD 55229 . µ app + ¯ µ rec > . − if they were indeed ejected on the inferred birth date.Since ¯ µ app > ¯ µ rec , there is ¯ µ app > . − ,which is significantly higher than the average proper motion(6.9 mas day − ) measured during our observations. Thus,component A had already been significantly decelerated be-fore our VLBI observations.If the jet expansion is linear and symmetric on bothsides, the ratio of the approaching and receding componentsizes is (e.g. Miller-Jones et al. 2004): R app ( t app ) R rec ( t rec ) = t app t rec = 1 + ¯ β (0 , t app ) cos θ − ¯ β (0 , t rec ) cos θ (2)where t app and t rec are the intrinsic times at which lightleaves the approaching and receding jet components respec-tively and arrive at the telescope at the same observingtime, ¯ β is the average jet speed in units of light speed c and θ is the inclination angle of the jet axis. By a linearextrapolation, the approaching jet component has a size of21.3 mas at the fourth epoch. If there is no deceleration,¯ β (0 , t app ) = ¯ β (0 , t rec ) = β , we can give β cos θ = 0 . c , whichrequires β > . c and θ ◦ . Note that the size of the re-ceding jet component detected for the first time is not likelyto be affected by the over-resolution since it is much youngerthan the approaching jet component ( t rec ∼ . t app ). Giventhe jet deceleration and t rec t app at the same telescopetime, then ¯ β (0 , t rec ) > . c and ¯ β (0 , t app ) . c . Therefore,we can take 0 . c as the lower limit of the jet birth speed inthe case of the jet deceleration.The ratio of the flux density measured at R app = R rec ( t app = t rec , free from its intrinsic luminosityevolution effect) for a pair of discrete jet components(Mirabel & Rodr´ıguez 1999): S app S rec = (cid:18) β cos θ − β cos θ (cid:19) − α (3)The receding jet component had a size of R rec = 11 . t app = t rec , the radio core is inferredto be at the centre: angular separation ∼
174 mas and posi- c (cid:13) , 1–5 decelerating jet in the X-ray transient XTE J1752 − -10 -5 0 5 10 1505101520 S i z e o f c o m ponen t A ( m a s ) MJD 55238.4
Figure 2.
Evolution of the size of component A. The solidstraight line shows the fitting result of the first three data points. tion angle ∼ − ◦ . The radio core was not detected duringany of the VLBI epochs but, since all four observations tookplace during the X-ray soft state, this is to be expected, ac-cording to the unified model of Fender et al. (2004, 2009).The radio position confirms that its optical counterpart is Swift -UVOT source A (Curran et al. 2010). If we take theflux density at the first epoch as the upper limit of S app ,then S app S rec β cos θ . c < ¯ β (0 , t rec ) cos θ , in agree-ment with the jet deceleration scenario on both sides. Theobserved flux density from the receding jet is deboosted by afactor: δ − α rec , where δ rec = (1 − β ) . (1+ β cos θ ) − . Becauseof the jet deceleration, the receding ejecta is less beamedaway from our line of sight, and thus it looks relativelybrighter. Note that the non-detection of the receding jetcomponent at the first epoch may also be because it stayedat an earlier stage ( t rec ∼ . t app if β cos θ = 0 .
3) and itsflux density was still much lower than the peak flux den-sity of the radio flare. The caveat in the above argument isthat component B might have not followed the same lumi-nosity evolution model as component A. It is also possiblethat the brightening of the receding ejecta was due to sud-den jet-cloud interaction, as in the case of XTE J1748 − In this paper, we present the results of the first VLBIobservations of the new Galactic black hole candidateXTE J1752 −
223 during its first known outburst. With EVNand VLBA observations at four epochs in 2010 February,we imaged its radio counterpart at 5 GHz. We detect anejected component at the first three epochs and find thatits proper motion shows significant deviation from the uni-form proper motion model and requires a deceleration rateof 0 . ± .
02 mas day − . In the jet deceleration scenario,it has proper motion decelerating from 9.2 mas day − atthe first epoch to 4.0 mas day − at the last epoch. It alsoshows slow but detectable variation of its transverse sizeindicating that its expansion is also significantly confined.This is the first time that a Galactic jet is found to be grad-ually decelerating on the hundred milliarcsecond scale. Thediscovery provides strong evidence for the existence of sig-nificant interaction around the jet at an early stage of itsevolution. In addition to the approaching jet component, wedetect another jet feature at the last epoch, which is mostlikely associated with the receding ejecta. We infer that thejet deceleration should start at a time much earlier thanour VLBI observations using the birth date (around 2010February 2) from the ATCA radio light curve. Furthermore, we interpret the detection of the receding ejecta as a resultof the reduced Doppler deboosting effect caused by the jetdeceleration on the receding side and give a lower limit of0 . c for the jet birth speed assuming symmetric jet motion.It has been reported by Shaposhnikov et al. (2010) that thedistance, estimated by the spectral-timing correlation scal-ing technique, is around 3.5 kpc. Thus, the proper motionobserved at the first epoch would correspond to an apparentjet speed of ∼ . c , in agreement with our results (but notethat the technique is very model dependent). ACKNOWLEDGMENTS
We thank the EVN PC Chair, T. Venturi and the VLBA Pro-posal Selection Committee for prompt approval of our ToO re-quests. e-VLBI developments in Europe were supported by theEC DG-INFSO funded Communication Network Developmentsproject EXPReS. The National Radio Astronomy Observatoryis a facility of the National Science Foundation operated undercooperative agreement by Associated Universities, Inc. The Eu-ropean VLBI Network is a joint facility of European, Chinese,South African and other radio astronomy institutes funded bytheir national research councils.
REFERENCES
Brocksopp C. et al., 2009, Astronomer’s Telegram, 2278Brocksopp C. et al., 2010, Astronomer’s Telegram, 2400Brocksopp C. et al., 2010, Astronomer’s Telegram, 2438Corbel S. et al., 2002, Sci, 298, 196Corbel S. et al., 2010, Astronomer’s Telegram, 2745Curran P.A. et al., 2010, MNRAS, arXiv:1007.5430Fender R.P. et al., 1999, MNRAS, 304, 865Fender R.P. et al., 2004, MNRAS, 355, 1105Fender R.P., 2006, in Lewin W.H.G., van der Klis M., eds.,Compact Stellar X-ray Sources, Cambridge Univ Press,Cambridge, p. 381Fender R.P. et al., 2009, MNRAS, 396, 1370Hjellming R.M., Rupen M.P., 1995, Nat, 375, 464Hjellming R.M. et al., 1998, BAAS, 30, 1405Homan J., 2010, Astronomer’s Telegram, 2387Kaaret P. et al., 2003, ApJ, 582, 945Markwardt C.B. et al., 2009, Astronomer’s Telegram, 2258Mirabel I.F., Rodr´ıguez L.F., 1994, Nat, 371, 46Mirabel I.F., Rodr´ıguez L.F., 1999, ARA&A, 37, 409Miller-Jones J.C.A. et al., 2004, ApJ, 603, L21Miller-Jones J.C.A. et al., 2007, MNRAS, 375, 1087Miller-Jones J.C.A. et al., 2008, in Bandyopadhyay R.M.et al., eds, A Population Explosion: The Nature and Evo-lution of X-ray Binaries in Diverse Environments, AIPConf. Proceedings, 1010, 50Nakahira S. et al., 2010, PASJ, arXiv:1007.0801Shepherd M.C. et al., 1994, BAAS, 26, 987Shaposhnikov N. et al., 2010, ApJ, arXiv:1008.0597Szomoru A., 2008, in Proceedings of the 9th EVN Sympo-sium on The Role of VLBI in the Golden Age for RadioAstronomy and EVN Users Meeting, Proceedings of Sci-ence, PoS 040Tingay S.J. et al., 1995, Nat, 374, 141 c (cid:13)000