The singularity at the center of a black hole is, by definition, beyond observation. But everything that happens on the way in — the spiraling gas, the intense radiation, the inexplicable jets of plasma punching out perpendicular to the disk at near-light speed — is not. That boundary region, where ordinary matter makes its last stand before crossing the event horizon, is now the most productive frontier in high-energy astrophysics, and instruments like NASA's IXPE (Imaging X-ray Polarimetry Explorer) and the European Space Agency's XMM-Newton are rewriting what physicists thought they understood about how black holes feed and grow.
The basic picture has been in place since the 1970s: infalling gas forms an accretion disk, friction and magnetic turbulence heat the disk to millions of degrees, and the resulting X-ray emission is bright enough to be seen across the observable universe. But "basic picture" is doing a lot of work in that sentence. The geometry of the inner disk, the role of the corona — a diffuse, extremely hot plasma sitting above the disk — and the mechanism by which some systems launch collimated jets while others don't have remained stubbornly unclear. Recent observations are starting to close those gaps, not by overturning the established framework but by revealing the fine structure within it.
What accretion actually looks like up close
The accretion disk around a stellar-mass black hole in a binary system is small enough — a few hundred kilometers across at the innermost stable circular orbit — that no telescope can image it directly. What observers have instead is spectroscopy: the spectrum of X-rays emitted by hot plasma encodes information about temperature, density, velocity, and, crucially, how close the emitting material is to the event horizon. The innermost stable circular orbit scales with black hole spin; a maximally spinning Kerr black hole has an ISCO just 1.2 Schwarzschild radii from the center, while a non-spinning Schwarzschild black hole pushes it out to 3 radii. Measuring the relativistic broadening of iron K-alpha emission lines — iron fluorescence at 6.4 keV, smeared and skewed by extreme gravity and Doppler shifts — gives a spin measurement accurate enough to distinguish between competing models of disk structure.
IXPE, launched in December 2021, adds a dimension that spectroscopy alone cannot: polarization. The orientation and degree of X-ray polarization from an accreting system carries information about the geometry of the corona and the scattering history of photons before they escape. In observations of Cygnus X-1, one of the most studied black hole binaries in the sky, IXPE data published in 2022 revealed that the X-ray polarization angle is aligned with the radio jet direction — a result that had been predicted by some models but never confirmed. The corona, the data suggest, is not a spherical cloud but a vertically extended, elongated structure sitting above the disk along the jet axis. That geometry would naturally produce polarized emission aligned with the jet, and it would help explain how the corona couples magnetically to the outflow.
That magnetic coupling is where theorists have focused enormous effort. The Blandford-Znajek mechanism, proposed in 1977, describes how a spinning black hole can extract rotational energy electromagnetically, threading magnetic field lines through the ergosphere and driving them into the surrounding plasma. The result is a Poynting-flux-dominated outflow — a jet powered not by thermal pressure but by the spin energy of the black hole itself. Simulations using general relativistic magnetohydrodynamics (GRMHD) have reproduced this behavior in striking detail, and the Event Horizon Telescope's 2019 image of M87's central region showed the expected magnetically dominated jet base. But simulations are not observations, and the question of whether the Blandford-Znajek process dominates in real systems — as opposed to the Blandford-Payne mechanism, which extracts energy from the disk rather than the hole — has remained open.
Jets, state transitions, and the missing link
Black hole binaries are extraordinarily useful laboratories precisely because they are not static. A system like GX 339-4 or MAXI J1820+070 can transition from a quiescent, undetectable state to a luminous outburst and back within weeks to months, cycling through distinct spectral states that correspond to different disk geometries and jet behaviors. In the hard state, the disk is truncated — receding far from the ISCO — the corona is large, and a steady, compact radio jet is present. As the system brightens and the disk moves inward, transitioning toward the soft state, the jet disappears. What destroys the jet, and why, is one of the field's central puzzles.
MAXI J1820+070, which erupted in 2018, became one of the best-studied transient black hole binaries ever observed because it was caught early and monitored continuously across X-ray, optical, and radio wavelengths. NICER (Neutron star Interior Composition Explorer), NASA's other current X-ray timing instrument aboard the ISS, tracked the inner disk radius shrinking in real time as the system evolved from hard to soft state — a direct demonstration of the disk moving inward as accretion rate rises. When IXPE observed it during a later outburst in 2023, the polarization data showed a rotated polarization angle relative to what hard-state sources typically exhibit, suggesting the corona geometry changes dramatically during state transitions. The jet collapse may be tied to the corona collapsing — as the disk fills in toward the ISCO, it may disrupt the vertical magnetic field structure that sustains both the corona and the outflow.
Supermassive black holes — millions to billions of solar masses, sitting at galactic centers — operate on the same physical principles but at scales where the jet's effect on the surrounding environment becomes cosmological. The jets of M87, driven by a 6.5-billion-solar-mass black hole, extend 5,000 light-years from the nucleus in optical emission and much farther in radio. They are energetic enough to heat the intracluster medium of the Virgo Cluster, suppressing star formation across a region far larger than the host galaxy. This "AGN feedback" is now a standard ingredient in cosmological simulations of galaxy formation, but the energy budget — how much of the accreted mass-energy actually goes into the jet versus radiation versus outflows — depends on parameters that are still poorly constrained.
What the next generation will resolve
ESA's Athena mission, scheduled for launch in the mid-2030s, will carry a microcalorimeter spectrometer with an energy resolution roughly fifty times better than current CCD-based instruments. That resolution will allow detailed mapping of the warm-hot circumgalactic medium around active galaxies, and it will make iron-line spin measurements routine rather than heroic. The proposed AXIS mission (Advanced X-ray Imaging Satellite), currently under NASA review, would combine Chandra-class angular resolution with a much larger collecting area, enabling snapshot imaging of AGN jets and their interaction with surrounding gas at cosmological distances. Meanwhile, the planned STROBE-X mission would apply NICER-style timing precision to a much larger detector area, making it possible to track disk dynamics on millisecond timescales — the light-crossing time of the innermost disk around a stellar-mass black hole.
The picture that is emerging from current and planned instruments is one of tight coupling: between the disk and the corona, between the corona and the jet, between the jet and the galactic environment. A black hole does not simply sit and eat. It processes infalling material into radiation and outflows with an efficiency that dwarfs nuclear fusion, and it feeds energy back into its environment in ways that shape galaxy evolution over billions of years. The details of that coupling — which magnetic field configuration launches the jet, why some transitions kill it, how spin interacts with accretion rate to set the efficiency — are where the science is happening right now, one carefully polarized X-ray photon at a time.
