MAXI J1744-294: Galactic Center X-ray Binary

• MAXI J1744-294 is a transient X-ray source in the Galactic center, typically classified as a black-hole low-mass X-ray binary candidate with recurring outbursts.
• Multi-mission observations using MAXI, Swift, NICER, NuSTAR, XRISM, and IXPE revealed a disk-dominated soft state with a detailed Fe line spectrum and low X-ray polarization for geometric constraints.
• Its unique position in a crowded, high-absorption region makes MAXI J1744-294 a valuable probe for interstellar medium studies and the dynamics of disk winds and jets.

MAXI J1744-294 is a transient X-ray source in the Galactic-center region, generally treated as a black hole low-mass X-ray binary candidate and identified in later work with Swift J174540.2-290037. Since its 2025 outburst, it has become important in two distinct but connected senses: as an accretion source observed predominantly in a bright soft state with a disk-dominated X-ray spectrum, and as a luminous backlight for studies of the interstellar and magneto-ionic medium along one of the most obscured sightlines in the Galaxy (Mandel et al., 17 Sep 2025, Gatuzz et al., 28 Mar 2026).

1. Discovery, nomenclature, and Galactic-center location

MAXI/GSC reported the source on 2025 January 2 as a bright Galactic-center transient. Early fluxes were quoted as $\sim 133$ mCrab in 2–20 keV and $\approx 100$ mCrab in 10–20 keV, with the 10–20 keV flux later rising to $\approx 250$ mCrab over January 13–15 (Majumder et al., 4 Jun 2025, Marra et al., 20 Jun 2025). Swift/XRT and subsequent Chandra observations localized the transient close to Sgr A*, and Chandra/ACIS-S on 2025 March 9 yielded a sub-arcsecond position at RA $17{:}45{:}40.476$, Dec $-29{:}00{:}46.10$ (J2000), placing the source approximately $18''$–$19''$ from Sgr A* and therefore within the projected central parsec (Marra et al., 20 Jun 2025, Mandel et al., 17 Sep 2025).

Later Galactic-center studies adopt the name MAXI J1744-294, or the shorthand M1744, to avoid confusion with other Swift transients in the field. Those same studies identify the source with the 2016 transient Swift J174540.2-290037, so the 2025 event is treated as a recurring outburst rather than an entirely new source class (Parra et al., 28 Mar 2026, Parra et al., 28 Mar 2026). Radio polarimetry subsequently strengthened the Galactic-center association: the measured rotation measure was found to be consistent with that of the Galactic-center magnetar PSR J1745-2900, and this was interpreted as the first direct evidence that MAXI J1744 lies within the Galactic-center region, is bound to Sgr A*, and is part of the nuclear star cluster (Michail et al., 8 Apr 2026).

Its position is astrophysically consequential. The source lies in an exceptionally crowded environment containing Sgr A*, the supernova remnant Sgr A East, the Galactic-center diffuse X-ray emission, and the nearby neutron-star low-mass X-ray binary AX J1745.6-2901. It is also seen through a very large absorbing column and a strong dust-scattering halo, conditions that complicate continuum and line spectroscopy but make the source unusually valuable as a probe of the Galactic-center interstellar medium (Parra et al., 28 Mar 2026, Gatuzz et al., 28 Mar 2026).

2. Multi-mission observational record

The 2025 outburst triggered an unusually broad observing campaign. X-ray coverage included MAXI, Swift/XRT, NICER, NuSTAR, XMM-Newton, XRISM/Resolve and Xtend, Chandra/HETG, and IXPE; radio coverage included MeerKAT and the VLA; near-infrared follow-up included Keck/NIRC2 (Mandel et al., 17 Sep 2025, Parra et al., 28 Mar 2026). XRISM observed the field as a Director’s Discretionary Time target on 2025 March 3 for a net Resolve exposure of $71\,\mathrm{ks}$, while IXPE observed the source on 2025 April 5–8 with an effective exposure of about $150\,\mathrm{ks}$ in the 2–8 keV band (Parra et al., 28 Mar 2026, Marra et al., 20 Jun 2025).

The crowded-field problem became a defining methodological issue. Resolve’s $6\times 6$ pixel array and broad point-spread function imply severe spatial-spectral mixing between MAXI J1744-294, AX J1745.6-2901, and diffuse Galactic-center components. The XRISM analysis therefore developed region-based deblending strategies, response calculations for individual spatial components, and background templates tied to a 2024 pre-outburst XRISM observation of nearly the same field (Parra et al., 28 Mar 2026). This decontamination framework underpins both the later line-spectroscopy results for the source itself and the interstellar-absorption analysis along the same sightline (Parra et al., 28 Mar 2026, Gatuzz et al., 28 Mar 2026).

Outside the X-ray band, the source was detected by MeerKAT and the VLA in radio, with Keck near-infrared imaging failing to reveal a clear counterpart at the source position. The radio detection established an accreting compact object and later enabled polarimetric measurements of the local and Galactic-center magneto-ionic environment, whereas the infrared non-detection remained consistent with a heavily obscured low-mass X-ray binary (Marra et al., 20 Jun 2025, Mandel et al., 17 Sep 2025).

3. X-ray continuum, spectral state, and timing behavior

Early follow-up with NuSTAR and NICER showed a thermal disk component with $\approx 100$0, a steep power-law tail with $\approx 100$1, an iron line near $\approx 100$2, and a 2–10 keV luminosity $\approx 100$3 assuming $\approx 100$4. On that basis the source was provisionally classified as a black hole X-ray binary transient in the high-soft state, with an optically thick, geometrically thin disk dominating the flux and only a weak steep coronal tail (Majumder et al., 4 Jun 2025).

The IXPE-plus-Swift analysis during April 2025 recovered a canonical soft-state continuum in the form

$\approx 100$5

with $\approx 100$6, $\approx 100$7 keV, and $\approx 100$8. In that fit, the absorbed 2–8 keV flux was $\approx 100$9, the unabsorbed 2–8 keV flux was $\approx 250$0, and the unabsorbed 2–8 keV luminosity at 8 kpc was $\approx 250$1. The 2–5 keV band was disk-dominated at about $\approx 250$2 of the total flux, while the 5–8 keV band was already a mixed disk-plus-tail regime (Marra et al., 20 Jun 2025).

Broadband multiwavelength modeling showed that MAXI J1744-294 remained in the bright/soft state through the first months of 2025 and then hardened in April as the flux began to decline. In the NuSTAR/XMM-Newton/Swift analysis, the inner disk temperature increased from about $\approx 250$3–$\approx 250$4 in February and March to about $\approx 250$5–$\approx 250$6 in April, while the Comptonized component hardened from $\approx 250$7 to $\approx 250$8, consistent with growing coronal importance and increasing disk truncation during a soft-to-intermediate evolution (Mandel et al., 17 Sep 2025).

Timing behavior remained comparatively quiescent. The IXPE study reported no dedicated power-density spectrum for MAXI J1744-294, no type-B or type-C QPOs, and no quoted LFQPO parameters at the IXPE epoch; within that work the source was treated as a soft, relatively stable transient with no detected LFQPOs (Majumder et al., 4 Jun 2025). The IXPE/NICER study found a flat NICER power density spectrum over $\approx 250$9–$17{:}45{:}40.476$0 Hz after Poisson-noise subtraction, with fractional rms $17{:}45{:}40.476$1, again consistent with a stable soft state (Marra et al., 20 Jun 2025). NuSTAR timing analysis later reported featureless power spectra, no significant red noise, no coherent pulsations, and no type I X-ray bursts or eclipses/dips across the outburst coverage (Mandel et al., 17 Sep 2025). These properties are not by themselves a proof of black-hole accretion, but they are routinely treated as part of the cumulative case for a black hole candidate.

4. Polarimetry and geometric constraints

X-ray polarimetry yielded a null detection. In the IXPE sample study of eleven black hole X-ray binaries, MAXI J1744-294 was one of the few systems with no significant 2–8 keV polarization detection. Using the model-independent PCUBE analysis on IXPE level-2 events, the measured values were

$17{:}45{:}40.476$2

with $17{:}45{:}40.476$3, $17{:}45{:}40.476$4, $17{:}45{:}40.476$5, and a detection significance of about $17{:}45{:}40.476$6. Because $17{:}45{:}40.476$7, the result was classified as a null detection (Majumder et al., 4 Jun 2025).

A dedicated IXPE analysis tightened the limit by fitting weighted Stokes $17{:}45{:}40.476$8 spectra jointly with Swift/XRT. That study found no significant polarization in any band and derived $17{:}45{:}40.476$9 upper limits of $-29{:}00{:}46.10$0 in 2–8 keV, $-29{:}00{:}46.10$1 in 2–5 keV, and $-29{:}00{:}46.10$2 in 5–8 keV (Marra et al., 20 Jun 2025). Within the broader IXPE black-hole sample, this placed MAXI J1744-294 at the low-polarization end of the empirical trend in which disk-dominated soft states show very low polarization degrees, whereas harder, more Comptonization-dominated states reach several per cent to tens of per cent (Majumder et al., 4 Jun 2025).

Relativistic disk-polarization modeling then used this low X-ray polarization to constrain inclination. With the kynbbrr framework, a standard Novikov-Thorne disk, and two limiting disk-albedo assumptions, the $-29{:}00{:}46.10$3 upper limit implied

$-29{:}00{:}46.10$4

depending on black-hole spin and disk-atmosphere albedo. In the $-29{:}00{:}46.10$5 limit the bounds were $-29{:}00{:}46.10$6, $-29{:}00{:}46.10$7, and $-29{:}00{:}46.10$8 for $-29{:}00{:}46.10$9, $18''$0, and $18''$1, whereas in the $18''$2 limit they were $18''$3, $18''$4, and $18''$5 for the same three spins (Marra et al., 20 Jun 2025). These limits are independent of continuum-fitting and are consistent with a low-to-intermediate inclination system.

Radio polarimetry later supplied a different geometric diagnostic. VLA observations over four epochs in 2025 April detected variable linear polarization at 33 and 43 GHz and inferred a common Faraday rotation screen with

$18''$6

the third largest RM detected within the Galaxy (Michail et al., 8 Apr 2026). On 2025 April 6 a secondary polarized component required an additional local screen with $18''$7. Under the assumption that the secondary component primarily cools by synchrotron radiation, the implied local magnetic field strength was $18''$8–$18''$9 G, and in the context of a jetted X-ray binary this was interpreted as a short-lived knot in a putative jet (Michail et al., 8 Apr 2026). The standard RM relation,

$19''$0

then linked the source not only to a local jet environment but also to the larger Galactic-center Faraday screen (Michail et al., 8 Apr 2026).

5. High-resolution spectroscopy and the line-of-sight medium

High-resolution XRISM spectroscopy transformed MAXI J1744-294 from a generic soft-state transient into a source of line-diagnostic interest. After detailed deblending of the crowded Resolve field, the intrinsic spectrum was found to include a narrow static component of highly ionized Fe emission, likely from a photoionized inner disk atmosphere, together with a weak narrow Fe I K$19''$1 line at 6.4 keV and several narrow features at atypical energies between 6.7 and 7.1 keV (Parra et al., 28 Mar 2026).

The static component was characterized by line widths $19''$2–$19''$3 and by a photoionized plasma with $19''$4, where

$19''$5

The unusual narrow features at approximately 6.74, 6.78, 6.83, and 7.00 keV lacked satisfactory rest-frame identifications and were therefore modeled as highly ionized blueshifted Fe lines with outflow velocities from $19''$6 to $19''$7 (Parra et al., 28 Mar 2026). The line phenomenology was interpreted in terms of multiple layers of photoionized or collisional plasma, and the favored physical possibilities were a multi-phase jet, a multi-phase disk wind, or a hybrid wind-jet configuration. The comparison to SS 433 was made at the level of complex multi-velocity Fe emission rather than identical jet kinematics (Parra et al., 28 Mar 2026).

The same sightline also enabled a separate XRISM/Resolve plus Chandra/HETG study of interstellar absorption by sulfur, argon, and calcium. In that work MAXI J1744-294 functioned explicitly as a “background lamp” through the Galactic-center interstellar medium (Gatuzz et al., 28 Mar 2026). Sulfur and argon were found to be predominantly in low-ionization states, with S II and Ar II dominating the cold and warm ISM phases, while calcium was detected mainly in Ca I–III and interpreted as strongly depleted into dust. The source thereby provided the first X-ray constraints on calcium absorption in the interstellar medium (Gatuzz et al., 28 Mar 2026).

From the measured ionic columns, the derived hydrogen columns were

$19''$8

$19''$9

and

$71\,\mathrm{ks}$0

These values agreed within uncertainties and indicated that most of the absorbing column lies in cold and warm phases rather than in hot Galactic-center plasma (Gatuzz et al., 28 Mar 2026). This interstellar result is distinct from continuum-fitting values of $71\,\mathrm{ks}$1–$71\,\mathrm{ks}$2 obtained in source models, but both sets of measurements reinforce the conclusion that MAXI J1744-294 is seen through an exceptionally heavy column (Marra et al., 20 Jun 2025, Gatuzz et al., 28 Mar 2026).

6. System parameters, classification, and astrophysical significance

The source is consistently described as a black-hole candidate in a low-mass X-ray binary, but published compact-object parameters remain model-dependent. One XRISM analysis reported a moderately spinning black hole with spin $71\,\mathrm{ks}$3–$71\,\mathrm{ks}$4, mass $71\,\mathrm{ks}$5–$71\,\mathrm{ks}$6 Solar masses, and disk inclination $71\,\mathrm{ks}$7–$71\,\mathrm{ks}$8, and summarized a most probable mass of $71\,\mathrm{ks}$9 (Chatterjee et al., 28 Jun 2025). The IXPE continuum-fitting analysis, using a fixed $150\,\mathrm{ks}$0 black hole in kerrbb, obtained $150\,\mathrm{ks}$1 and $150\,\mathrm{ks}$2 (Marra et al., 20 Jun 2025). By contrast, later broadband/reflection studies cited in the XRISM crowded-field series treat MAXI J1744-294 as having high spin $150\,\mathrm{ks}$3 and inclination $150\,\mathrm{ks}$4 (Parra et al., 28 Mar 2026, Parra et al., 28 Mar 2026). This indicates that published spin, mass, and inclination estimates have not yet converged.

The classification as a black-hole low-mass X-ray binary nevertheless rests on a broader evidentiary base than any single fit. The source shows a canonical outburst progression dominated by a bright soft state, lacks coherent pulsations and type I X-ray bursts, requires no neutron-star boundary-layer component in broadband X-ray fits, and has luminosity and decay-time behavior treated as more typical of black-hole transients than neutron-star systems (Mandel et al., 17 Sep 2025). The infrared non-detection also remains compatible with a low-mass donor rather than a luminous high-mass companion (Mandel et al., 17 Sep 2025).

Its Galactic-center setting has broader population significance. MAXI J1744-294 has been described as the fourth candidate black-hole transient discovered within a projected distance of one parsec from Sgr A* (Mandel et al., 17 Sep 2025). This aligns with the long-standing suggestion that the central parsec hosts a cusp of black-hole low-mass X-ray binaries, produced or retained by nuclear-cluster dynamics more efficiently than comparable neutron-star systems (Mandel et al., 17 Sep 2025). The radio RM measurement adds an environmental dimension: because the source and the Galactic-center magnetar show similar, very large negative rotation measures, the Galactic-center Faraday screen appears approximately uniform on central-parsec scales, while most of Sgr A*’s still larger RM is likely intrinsic to its own accretion flow rather than to an unrelated foreground structure (Michail et al., 8 Apr 2026).

MAXI J1744-294 therefore occupies an unusual intersection of research domains. It is simultaneously a soft-state black-hole candidate, a crowded-field spectroscopy test case, a probe of disk atmospheres and outflows through narrow Fe emission, a background lamp for multi-element ISM absorption studies, and a radio-polarimetric marker of the Galactic-center magneto-ionic medium. That combination, rather than any single parameter estimate, explains its rapid emergence as a benchmark source in Galactic-center high-energy astrophysics (Parra et al., 28 Mar 2026, Gatuzz et al., 28 Mar 2026, Michail et al., 8 Apr 2026).