Hard state neutron star and black hole X-ray binaries in the radio:X-ray luminosity plane

Abstract

Motivated by the large body of literature around the phenomenological properties of accreting black hole (BH) and neutron star (NS) X-ray binaries in the radio:X-ray luminosity plane, we carry out a comparative regression analysis on 36 BHs and 41 NSs in hard X-ray states, with data over 7 dex in X-ray luminosity for both. The BHs follow a radio to X-ray (logarithmic) luminosity relation with slope β = 0.59±0.02, consistent with the NSs' slope (β = 0.44 +0.05 −0.04 ) within 2.5σ. The best-fitting intercept for the BHs significantly exceeds that for the NSs, cementing BHs as more radio loud, by a factor ∼22. This discrepancy can not be fully accounted for by the mass or bolometric correction gap, nor by the NS boundary layer contribution to the X-rays, and is likely to reflect physical differences in the accretion flow efficiency, or the jet powering mechanism. Once importance sampling is implemented to account for the different luminosity distributions, the slopes of the non-pulsating and pulsating NS subsamples are formally inconsistent (> 3σ), unless the transitional millisecond pulsars (whose incoherent radio emission mechanism is not firmly established) are excluded from the analysis. We confirm the lack of a robust partitioning of the BH data set into separate luminosity tracks.

Radio and X-ray monitoring of the accreting millisecond X-ray pulsar IGR J17591−2342 in outburst

“…Previous studies have modelled the radio-X-ray luminosity scaling in the simplistic form LR ∝ LX β . Recent work by Gallo et al (2018) finds β = 0.44 +0.05 −0.04 and β = 0.59±0.02 for the NS-and BH-LMXB populations, respectively. Gallo et al (2018) also found that NS-LMXBs are, on average, ∼ 20 times less radio luminous compared to BH-LMXBs at the same LX.…”

“…In order to quantify the differences between IGR J17591−2342 and other LMXBs, we performed a power-law fit to the radio and X-ray luminosities for different classes of LMXBs, as well as individual sources Figure 2, which plots the measurements for the first 3 epochs for a range of distances) and each observation (Table 1) is labelled with its corresponding ID number. Data points are taken from Bahramian et al (2018) for BHs hard; Migliari & Fender (2006); Tetarenko et al (2016); Gusinskaia et al (2017Gusinskaia et al ( , 2019 for NSs hard; Papitto et al (2013) and updated X-ray value from Coti Zelati et al (2019) for M28I, where strictly simultaneous data was used and a time-resolved X-ray spectral analysis was performed; Hill et al (2011) Gallo et al (2018). The yellow-green dotted line represents the result of our L R ∝ L X β fit for IGR J17591−2342.…”

“…when the number of available measurements is sufficient (> 8 LR-LX points). Using the Python implementation of the Kelly (2007) linear regression algorithm LINMIX_ERR 7 we perform a linear fit in logarithmic space, in the form used by Gallo et al (2014Gallo et al ( , 2018:…”

“…NS-LMXBs have been observed during their bright (Xray luminosity LX ∼ 10 36−38 erg s −1 ) accretion outbursts (e.g., Migliari & Fender 2006), and as they fade back to quiescence. This has allowed the LR-LX relationship in the hard X-ray state to be studied for both individual systems (e.g., Migliari et al 2003;Tudose et al 2009;Tetarenko et al 2016) as well as statistically for the population as a whole (e.g., Gallo et al 2018). Previous studies have modelled the radio-X-ray luminosity scaling in the simplistic form LR ∝ LX β .…”

The connection between the UV/optical and X-ray emission in the neutron star low-mass X-ray binary Aql X-1

“…Firstly, it appears that the radio/X-ray correlation in Aql X-1 has a much lower β jet value (Tudose et al 2009;Tetarenko et al 2018;Gusinskaia et al 2019). Moreover, a recent radio/X-ray study, using the largest sample of NS LMXBs to date, also found a lower correlation index for the population as a whole (β jet = 0.44; Gallo et al 2018). Finally, detailed multi-wavelength studies of Aql X-1 and other systems suggest that the jet of NS LMXBs is not likely to contribute significantly to the emission at NIR/UV/optical wavelengths, either due to NSs having weaker jets than BHs, or due to the jet breaking at a lower frequency (e.g.…”