We present the radio and X-ray monitoring campaign of the 2019/2020 outburst of MAXI J1348–630, a new black hole X-ray binary (BH XRB) discovered in 2019 January. We observed MAXI J1348–630 for ∼14 months in the radio band with MeerKAT and the Australia Telescope Compact Array, and in the X-rays with MAXI and Swift/XRT. Throughout the outburst, we detected and tracked the evolution of compact and transient jets. Following the main outburst, the system underwent at least four hard-state-only re-flares, during which compact jets were again detected. For the major outburst, we observed the rise, quenching and reactivation of compact jets, as well as two single-sided discrete ejecta travelling away from the BH, launched ∼2 months apart. These ejecta displayed the highest proper motion (≳100 mas d⁻¹) ever measured for an accreting BH binary. From the jet motion, we constrain the ejecta inclination and speed to be ≤46^○ and ≥0.69 c, and the opening angle and transverse expansion speed of the first component to be ≤6^○ and ≤0.05 c. We also infer that the first ejection happened at the hard-to-soft state transition, before a strong radio flare, while the second ejection was launched during a short excursion from the soft to the intermediate state. After travelling with constant speed, the first component underwent a strong deceleration, which was covered with unprecedented detail and suggested that MAXI J1348–630 could be located inside a low-density cavity in the interstellar medium, as already proposed for XTE J1550–564 and H1743–322.

Binary supermassive black hole (SMBH) systems result from galaxy mergers, and will eventually coalesce due to gravitational wave (GW) emission if the binary separation can be reduced to ≲0.1 pc by other mechanisms. Here, we explore a gravitational self-lensing binary SMBH model for the sharp (duration ∼1 h), quasi-regular X-ray flares – dubbed quasi-periodic eruptions – recently observed from two low-mass active galactic nuclei: GSN 069 and RX J1301.9+2747. In our model, the binary is observed ∼edge-on, such that each SMBH gravitationally lenses light from the accretion disc surrounding the other SMBH twice per orbital period. The model can reproduce the flare spacings if the current eccentricity of RX J1301.9+2747 is ϵ₀ ≳ 0.16, implying a merger within ∼1000 yr. However, we cannot reproduce the observed flare profiles with our current calculations. Model flares with the correct amplitude are ∼2/5 the observed duration, and model flares with the correct duration are ∼2/5 the observed amplitude. Our modelling yields three distinct behaviours of self-lensing binary systems that can be searched for in current and future X-ray and optical time-domain surveys: (i) periodic lensing flares, (ii) partial eclipses (caused by occultation of the background mini-disc by the foreground mini-disc), and (iii) partial eclipses with a very sharp in-eclipse lensing flare. Discovery of such features would constitute very strong evidence for the presence of a supermassive binary, and monitoring of the flare spacings will provide a measurement of periastron precession.

We present deep X-ray and radio observations of the fast blue optical transient (FBOT) AT 2020xnd/ZTF 20acigmel at z = 0.2433 from 13 days to 269 days after explosion. AT 2020xnd belongs to the category of optically luminous FBOTs with similarities to the archetypal event AT 2018cow. AT 2020xnd shows luminous radio emission reaching L _ν ≈ 8 × 10²⁹ erg s⁻¹ Hz⁻¹ at 20 GHz and 75 days post-explosion, accompanied by luminous and rapidly fading soft X-ray emission peaking at L _X ≈ 6 × 10⁴² erg s⁻¹. Interpreting the radio emission in the context of synchrotron radiation from the explosion’s shock interaction with the environment, we find that AT 2020xnd launched a high-velocity outflow (v ∼ 0.1c–0.2c) propagating into a dense circumstellar medium (effective M ̇ ≈ 10 − 3 M ⊙ yr⁻¹ for an assumed wind velocity of v _w = 1000 km s⁻¹). Similar to AT 2018cow, the detected X-ray emission is in excess compared to the extrapolated synchrotron spectrum and constitutes a different emission component, possibly powered by accretion onto a newly formed black hole or neutron star. These properties make AT 2020xnd a high-redshift analog to AT 2018cow, and establish AT 2020xnd as the fourth member of the class of optically luminous FBOTs with luminous multiwavelength counterparts.

During a 2018 outburst, the black hole X-ray binary MAXI J1820 + 070 was comprehensively monitored at multiple wavelengths as it underwent a hard to soft state transition. During this transition, a rapid evolution in X-ray timing properties and a short-lived radio flare were observed, both of which were linked to the launching of bi-polar, long-lived relativistic ejecta. We provide a detailed analysis of two Very Long Baseline Array observations, using both time binning and a new dynamic phase centre tracking technique to mitigate the effects of smearing when observing fast-moving ejecta at high angular resolution. We identify a second, earlier ejection, with a lower proper motion of 18.0 ± 1.1 mas d⁻¹. This new jet knot was ejected 4 ± 1 h before the beginning of the rise of the radio flare, and 2 ± 1 h before a switch from type-C to type-B X-ray quasi-periodic oscillations (QPOs). We show that this jet was ejected over a period of ∼6 h and thus its ejection was contemporaneous with the QPO transition. Our new technique locates the original, faster ejection in an observation in which it was previously undetected. With this detection, we revised the fits to the proper motions of the ejecta and calculated a jet inclination angle of (64 ± 5)°, and jet velocities of |$0.97_{-0.09}^{+0.03}c$| for the fast-moving ejecta (Γ > 2.1) and (0.30 ± 0.05)c for the newly identified slow-moving ejection (Γ = 1.05 ± 0.02). We show that the approaching slow-moving component is predominantly responsible for the radio flare, and is likely linked to the switch from type-C to type-B QPOs, while no definitive signature of ejection was identified for the fast-moving ejecta.

The black hole (BH) binary V404 Cyg entered the outburst phase in 2015 June after 26 yr of X-ray quiescence, and with its behaviour broke the outburst evolution pattern typical of most BH binaries. We observed the entire outburst with the Swift satellite and performed time-resolved spectroscopy of its most active phase, obtaining over a thousand spectra with exposures from tens to hundreds of seconds. All the spectra can be fitted with an absorbed power-law model, which most of the time required the presence of a partial covering. A blueshifted iron-Kα line appears in 10 per cent of the spectra together with the signature of high column densities, and about 20 per cent of the spectra seem to show signatures of reflection. None of the spectra showed the unambiguous presence of soft disc–blackbody emission, while the observed bolometric flux exceeded the Eddington value in 3 per cent of the spectra. Our results can be explained assuming that the inner part of the accretion flow is inflated into a slim disc that both hides the innermost (and brightest) regions of the flow, and produces a cold, clumpy, high-density outflow that introduces the high absorption and fast spectral variability observed. We argue that the BH in V404 Cyg might have been accreting erratically or even continuously at Eddington/super-Eddington rates – thus sustaining a surrounding slim disc – while being partly or completely obscured by the inflated disc and its outflow. Hence, the largest flares produced by the source might not be accretion-driven events, but instead the effects of the unveiling of the extremely bright source hidden within the system.