arXiv subjects

W. H. G. Lewin

Publications and source records attributed to W. H. G. Lewin.

At least 73 records · Page 4Linked to original sources

arXiv2000

Relativistic Iron Emission and Disk Reflection in Galactic Microquasar XTE J1748-288

We report evidence for an Fe K-alpha fluorescence line feature and disk reflection in the Very High, High, and Low State X-ray spectra of the galactic microquasar XTE J1748-288 during its June~1998 outburst. Spectral analyses are made on data gathered throughout the outburst by the Rossi X-ray Timing Explorer Proportional Counter Array. Gaussian line, relativistic disk emission line, and ionized disk reflection models are fit to the data. In the Very High State the line profile appears strongly redshifted, consistent with disk emission from the innermost stable orbits around a maximally rotating Kerr black hole. In the High State the line profile is less redshifted and increasingly prominent. The Low State line profile is very strong (~0.5 keV equivalent width) and centered at 6.7 +/- 0.10 keV; disk line emission model fits indicate that the inner edge of the disk fluctuates between ~20 and ~100 gravitational radii in this state. The disk reflection fraction is traced through the outburst; reflection from an ionized disk is preferred in the VHS and HS, and reflection from a relatively neutral disk is preferred in the LS. We discuss the implications of our findings for the binary system dynamics and accretion flow geometry in XTE J1748-288.

astro-ph↗

arXiv1999

The Timing Evolution of 4U 1630-47 During its 1998 Outburst

We report on the evolution of the timing of 4U1630-47 during its 1998 outburst using data obtained with the Rossi X-ray Timing Explorer (RXTE). The count rate and position in hardness-intensity, color-color diagrams and simple spectral fits are used to track the concurrent spectral changes. The source showed seven distinct types of timing behavior, most of which show differences with the canonical black hole spectral/timing states. In marked contrast to previous outbursts, we find quasi-periodic oscillation (QPO) signals during nearly all stages of the outburst with frequencies between 0.06Hz and 14Hz and a remarkable variety of other characteristics. In particular we find large (up to 23% rms) amplitude QPO on the early rise. Later, slow 0.1Hz semi- regular short (~ 5 sec), 9 to 16% deep dips dominate the light curve. At this time there are two QPOs, one stable near 13.5Hz and the other whose frequency drops from 6--8Hz to ~4.5 Hz during the dips. BeppoSAX observations during the very late declining phase show 4U1630-47 in a low state.

astro-ph↗

arXiv1999

Identification of a Likely Radio Counterpart of the Rapid Burster

We have identified a likely radio counterpart to the low-mass X-ray binary MXB 1730-335 (the Rapid Burster). The counterpart has shown 8.4 GHz radio on/off behavior correlated with the X-ray on/off behavior as observed by the RXTE/ASM during six VLA observations. The probability of an unrelated, randomly varying background source duplicating this behavior is 1-3% depending on the correlation time scale. The location of the radio source is RA 17h 33m 24.61s; Dec -33d 23' 19.8" (J2000), +/- 0.1". We do not detect 8.4 GHz radio emission coincident with type II (accretion-driven) X-ray bursts. The ratio of radio to X-ray emission during such bursts is constrained to be below the ratio observed during X-ray persistent emission at the 2.9-sigma level. Synchrotron bubble models of the radio emission can provide a reasonable fit to the full data set, collected over several outbursts, assuming that the radio evolution is the same from outburst to outburst, but given the physical constraints the emission is more likely to be due to ~hour-long radio flares such as have been observed from the X-ray binary GRS 1915+105.

astro-ph↗

arXiv1999

On the nature of XTE J0421+560/CI Cam

We present the results of the analysis of RXTE, BATSE and optical/infrared data of the 1998 outburst of the X-ray transient system XTE J0421+560 (CI Cam). The X-ray outburst shows a very fast decay (initial e-folding time ~0.5 days, slowing down to ~2.3 days). The X-ray spectrum in the 2-25 keV band is complex, softening considerably during decay and with strongly variable intrinsic absorption. A strong iron emission line is observed. No fast time variability is detected (<0.5 % rms in the 1-4096 Hz band at the outburst peak). The analysis of the optical/IR data suggests that the secondary is a B[e] star surrounded by cool dust and places the system at a distance of >~ 2 kpc. At this distance the peak 2-25 keV luminosity is ~4 x 10^37 erg/s. We compare the properties of this peculiar system with those of the Be/NS LMC transient A 0538-66 and suggest that CI Cam is of similar nature. The presence of strong radio emission during outburst indicates that the compact object is likely to be a black hole or a weakly magnetized neutron star.

astro-ph↗

arXiv1999

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 February 1996 and May 1998, 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 days. 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 has been 218 days, in contrast with an average ~180 day recurrence period observed during 1976-1983.

astro-ph↗

arXiv1999

Hard X-ray lags in GRO J1719-24

We have used the Fourier cross spectra of GRO J1719-24, as obtained with BATSE, to estimate the phase lags between the X-ray flux variations in the 20--50 and 50--100 keV energy bands as a function of Fourier frequency in the interval 0.002--0.488 Hz. Our analysis covers the entire ~80 day X-ray outburst of this black-hole candidate, following the first X-ray detection on 1993 September 25. The X-ray variations in the 50--100 keV band lag those in the 20--50 keV energy band by an approximately constant phase difference of 0.072 +/- 0.010 rad in the frequency interval 0.02--0.20 Hz. The peak phase lags in the interval 0.02--0.20 Hz are about twice those of Cyg X-1 and GRO J0422+32.These results are consistent with models for Comptonization regions composed of extended non-uniform clouds around the central source.

astro-ph↗

arXiv1999

The Intensity Distribution of Faint Gamma-ray Bursts Detected with BATSE

We have recently completed a search of 6 years of archival BATSE data for gamma-ray bursts (GRBs) that were too faint to activate the real-time burst detection system running onboard the spacecraft. These "non-triggered" bursts can be combined with the "triggered" bursts detected onboard to produce a GRB intensity distribution that reaches peak fluxes a factor of 2 lower than could be studied previously. The value of the V/Vmax statistic (in Euclidean space) for the bursts we detect is 0.177 +/- 0.006. This surprisingly low value is obtained because we detected very few bursts on the 4.096 s and 8.192 s time scales (where most bursts have their highest signal-to-noise ratio) that were not already detected on the 1.024 s time scale. If allowance is made for a power-law distribution of intrinsic peak luminosities, the extended peak flux distribution is consistent with models in which the redshift distribution of the gamma-ray burst rate approximately traces the star formation history of the Universe. We argue that this class of models is preferred over those in which the burst rate is independent of redshift. We use the peak flux distribution to derive a limit of 10% (99% confidence) on the fraction of the total burst rate that could be contributed by a spatially homogeneous (in Euclidean space) subpopulation of burst sources, such as type Ib/c supernovae. These results lend support to the conclusions of previous studies predicting that relatively few faint "classical" GRBs will be found below the BATSE onboard detection threshold.

astro-ph↗

arXiv1998

Hard X-ray variability of the black-hole candidate GRO J0422+32 during its 1992 outburst

We have studied the hard X-ray variability of the soft X-ray transient GRO J0422+32 with BATSE in the 20-100 keV energy band. Our analysis covers 180 days following the first X-ray detection of the source on 1992 August 5, fully covering its primary and secondary X-ray outburst. We computed power density spectra (PDSs) in the 20-50, 50-100, and 20-100 keV energy bands, in the frequency interval 0.002-0.488 Hz. The PDSs of GRO J0422+32 are approximately flat up to a break frequency, and decay as a power law above, with index ~1. The canonical anticorrelation between the break frequency and the power density at the break, observed in Cyg X-1 and other BHCs in the low state, is not observed in the PDSs of GRO J0422+32. We compare our results with those of similar variability studies of Cyg X-1. The relation between the spectral slope and the amplitude of the X-ray variations of GRO J0422+32 is similar to that of Cyg X-1; however, the relation between the hard X-ray flux and the amplitude of its variation is opposite to what has been found in Cyg X-1. Phase lags between the X-ray flux variations of GRO J0422+32 at high and low photon energies, could only be derived during the first 30 days of its outburst. During this period, the variations in the 50-100 keV lag those in the 20-50 keV energy band by an approximately constant phase difference of 0.039(3) rad in the frequency interval 0.02-0.20 Hz.

astro-ph↗

arXiv1998

ASCA Observation of the Galactic Jet Source XTE J0421+560 (CI Cam) in Outburst

We observed the newly discovered Galactic jet source XTE J0421+560 (= CI Cam) with ASCA from 1998 April 3.3 to April 4.1 (UT), three days after the beginning of the outburst. The X-ray intensity in the 1-10 keV band gradually decreased with an e-folding time of about 30 hours; the decline was accompanied by spectral softening. Two flare-like intensity enhancements were detected below 1 keV. We could fit the average spectrum above 0.8 keV with a two-temperature model (5.7 and 1.1 keV) of thermal emission from an optically thin ionization-equilibrium plasma. The broad iron-K profile, however, requires an extra emission line at 6.4 keV, or Doppler broadening (or both). The former can be explained in terms of reflection from cold matter, while the latter can be attributed to emission from the twin jets. In both cases, the time evolution of the emission measure and of the temperature are difficult to explain by emission from a single plasma, suggesting that heat input and/or injection of material was occurring during the outburst.

astro-ph↗

arXiv1998

A Search for Non-triggered Gamma Ray Bursts in the BATSE Data Base

We describe a search of archival data from the Burst and Transient Source Experiment (BATSE). The purpose of the search is to find astronomically interesting transients that did not activate the burst detection (or ``trigger'') system onboard the spacecraft. Our search is sensitive to events with peak fluxes (on the 1.024 s time scale) that are lower by a factor of 2 than can be detected with the onboard burst trigger. In a search of 345 days of archival data, we detected 91 events in the 50--300 keV range that resemble classical gamma ray bursts but that did not activate the onboard burst trigger. We also detected 110 low-energy (25--50 keV) events of unknown origins which may include activity from SGR 1806-20 and bursts and flares from X-ray binaries. This paper gives the occurrence times, estimated source directions, durations, peak fluxes, and fluences for the 91 gamma ray burst candidates. The direction and intensity distributions of these bursts imply that the biases inherent in the onboard trigger mechanism have not significantly affected the completeness of the published BATSE gamma ray burst catalogs.

astro-ph↗

arXiv1998

The Beat-Frequency Interpretation of Kilohertz QPOs in Neutron Star Low-Mass X-ray Binaries

Pairs of quasi-periodic oscillations (QPOs) at kilohertz frequencies are a common phenomenon in several neutron-star low-mass X-ray binaries. The frequency separation of the QPO peaks in the pair appears to be constant in many sources and directly related to the neutron star spin frequency. However, in Sco X-1 and possibly in 4U 1608-52, the frequency separation of the QPOs decreases with increasing inferred mass accretion rate. We show that the currently available Rossi X-ray Timing Explorer data are consistent with the hypothesis that the frequency separations in all sources vary by amounts similar to the variation in Sco X-1. We discuss the implications for models of the kilohertz QPOs.

astro-ph↗

arXiv1998

Sidebands Due to Quasi-periodic Oscillations in 4U 1626-67

The low-mass X-ray binary pulsar 4U 1626-67 shows 0.048 Hz quasi-periodic oscillations (QPOs) and red noise variability as well as coherent pulsations at the 0.130 Hz neutron star spin frequency. Power density spectra of observations made with the Rossi X-ray Timing Explorer show significant sidebands separated from the pulsar spin frequency (and its harmonics) by the QPO frequency. These show that the instantaneous amplitude of the coherent pulsations is modulated by the amplitude of the QPOs. This phenomenon is expected in models such as the magnetospheric beat frequency model where the QPOs originate near the polar caps of the neutron star. In the 4--8 keV energy range, however, the lower-frequency sidebands are significantly stronger than their higher-frequency complements; this is inconsistent with the magnetospheric beat frequency model. We suggest that the 0.048 Hz QPOs are instead produced by a structure orbiting the neutron star at the QPO frequency. This structure crosses the line of sight once per orbit and attenuates the pulsar beam, producing the symmetric (amplitude modulation) sidebands. It also reprocesses the pulsar beam at the beat frequencies between the neutron star spin frequency and the QPOs, producing the excess variability observed in the lower-frequency sidebands. Quite independently, we find no evidence that the red noise variability modulates the amplitude of the coherent pulsations. This is also in contrast to the expectations of the magnetospheric beat frequency model and differs from the behavior in some high-mass X-ray binary pulsars.

astro-ph↗

arXiv1997

A varying kHz peak separation in 4U 1608-52

astro-ph↗

arXiv1997

Discovery of a second kHz QPO peak in 4U 1608-52

Using a new technique to improve the sensitivity to weak Quasi-Periodic Oscillations (QPO) we discovered a new QPO peak at about 1100 Hz in the March 1996 outburst observations of 4U 1608-52, simultaneous with the ~ 600-900 Hz peak previously reported from these data. The frequency separation between the upper and the lower QPO peak varied significantly from $232.7 \pm 11.5$ Hz on March 3, to $293.1 \pm 6.6$ Hz on March 6. This is the first case of a variable kHz peak separation in an atoll source. We discuss to what extent this result could be accommodated in beat-frequency models such as proposed for the kHz QPOs. We measured the rms fractional amplitude of both QPOs as a function of energy, and we found that the relation is steeper for the lower than for the upper frequency peak. This is the first source where such difference between the energy spectrum of the two kHz QPOs could be measured.

astro-ph↗

arXiv1997

Hard X-ray Lags in Cygnus X-1

We have used the Fourier cross spectra of Cyg X-1, as obtained with BATSE during a period of almost 2000 days, to estimate the phase (or time) lags between X-ray flux variations in the 20-50 keV and the 50-100 keV bands as a function of Fourier frequency, nu. We find that these lag spectra do not depend on source state, as measured by the fractional rms variations of the X-ray flux. For frequencies well below the Nyquist frequency the variations in the 50-100 keV band lag those in the 20-50 keV band by a time interval proportional to nu^{-0.8}. Binning effects cause the observed hard X-ray time lags to decrease to zero at the Nyquist frequency. Some previous results on time lags were affected by these binning effects. At photon energies above 10 keV the time lags are approximately proportional to the logarithm of the ratio the energies of the passbands used.

astro-ph↗

arXiv1997

Kilohertz QPO and Atoll Source States in 4U 0614+09

We report three RXTE/PCA observations of the low-mass X-ray binary 4U 0614+09. They show strong (~ 30% rms) band-limited noise with a cut-off frequency varying between 0.7 and 15 Hz in correlation with the X-ray flux. We observe two non-simultaneous 11-15% (rms) kHz peaks near 728 and 629 Hz in the power spectra of two of our observations when the X-ray flux is ~ 10^{-9} erg cm^{-2} s{-1} (2-10 keV), but find no QPO (< 6% rms) when the X-ray flux is half that. We suggest that count rate may not be a good measure for $\dot M$ even in sources as intrinsically weak as 4U 0614+09, and that QPO frequency and noise cutoff frequency track $\dot M$ more closely than count rate. The QPO increases in rms amplitude from ($11 \pm 1.3$) % at 3 keV to ($37 \pm 12$) % at 23 keV; the fractional amplitude of the band-limited noise is energy-independent. This suggests different sites of origin for these two phenomena. The spectrum of the oscillating flux roughly corresponds to a black body with temperature ($1.56 \pm 0.2$) keV and radius ($ 500 \pm 200$) m (other models fit as well), which might indicate the oscillations originate at a small region on the neutron star surface.

astro-ph↗

arXiv1997

GX 17+2: X-ray spectral and timing behaviour of a bursting Z source

We investigated the properties of the Z pattern of the low-mass X-ray binary GX 17+2 in the X-ray colour-colour diagrams (CDs) and hardness-intensity diagrams (HIDs), and of the power spectra as a function of position in the Z, using all EXOSAT ME data. The Z pattern in the CD does not move to within 2 per cent between observations separated by 1 day to 1.5 years. The power spectra and the characteristics of the Z pattern both support the division of the Z sources into two sub-groups, GX 17+2 being in the group together with Sco X-1 and GX 349+2. It has been noted previously that the maximum frequency of the horizontal branch quasi-periodic oscillations in GX 17+2 is lower than for Cyg X-2, GX 5-1 and GX 340+0. We suggest that an asymmetry in the magnetic field giving rise to polar caps with different emission characteristics may be the origin of this. This may also explain the occurrence of bursts in GX 17+2. We also report the detection of two new bursts in the EXOSAT data of GX 17+2, which we interpret as type I bursts. No regular pulsations are observed during these two bursts nor during the bursts previously detected in GX 17+2 with EXOSAT, with upper limits to the pulsed amplitude of 1.5 per cent for pulsations with a frequency below 512 Hz. Three of the four bursts occurred in the normal branch, while one occurred in the lower part of the flaring branch.

astro-ph↗

arXiv1997

Post-Burst Quasi-Periodic Oscillations from GRO J1744-28 and from the Rapid Burster

The repetitive X-ray bursts from the accretion-powered pulsar GRO J1744-28 show similarities to the type II X-ray bursts from the Rapid Burster. Several authors (notably Lewin et al.) have suggested that the bursts from GRO J1744-28 are type II bursts (which arise from the sudden release of gravitational potential energy). In this paper, we present another similarity between these sources. Rossi X-ray Timing Explorer observations of GRO J1744-28 show that at least 10 out of 94 bursts are followed by quasi-periodic oscillations (QPO) with frequencies of about 0.4 Hz. The period of the oscillations decreases over their 30--80 s lifetime, and they occur during a spectrally hard ``shoulder'' (or ``plateau'') which follows the burst. In one case the QPO show a modulation envelope which resembles simple beating between two narrow-band oscillations at 0.325 and 0.375 Hz. Using EXOSAT observations, Lubin et al. found QPO with frequencies of 0.039 to 0.056 Hz following 10 out of 95 type II bursts from the Rapid Burster. As in GRO J1744-28, the period of these oscillations decreased over their 100 s lifetime, and they occurred only during spectrally hard ``humps'' in the persistent emission. Even though the QPO frequencies differ by a factor of 10, we believe that this is further evidence that a similar accretion disk instability is responsible for the type II bursts from these two sources.

astro-ph↗