Preamble

Vol. 66 No. 4

July, 2025

Acta Astronomica Sinica Vol. 66 No. 4 Jul., 2025
doi: 10.15940/j.cnki.0001-5245.2025.04.008

Study of Outburst Evolution in Black Hole X-ray Binaries with Insight-HXMT
MA Rui-can¹,²†
(1 Key Laboratory of Particle Astrophysics, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing 100049)
(2 Dongguan Neutron Science Center, Institute of High Energy Physics, Chinese Academy of Sciences, Dongguan 523808)

Abstract

Thanks to its broadband coverage (1–250 keV) and large effective area, the Insight-HXMT (Insight Hard X-ray Modulation Telescope) satellite has achieved a series of important scientific results during its operational lifetime. Insight-HXMT has monitored the outburst evolution of a sample of black hole X-ray binaries, providing crucial observational data for studying accretion properties during these events. The outbursts from these black hole transients include not only typical full outbursts that evolve through hard, intermediate, and soft states, but also "failed state transition outbursts" that only reach the hard or intermediate states. The broadband data from Insight-HXMT not only offer important opportunities for deeply understanding the properties of accretion disks, coronae, and jets in black hole X-ray binaries, but also provide key observational evidence for investigating outburst mechanisms and accretion radiation physics. Using Insight-HXMT data, this review focuses on the evolution and properties of black hole X-ray binary outbursts.

Keywords: accretion, accretion disks, X-rays, binaries, stars: black holes

1. Introduction

Black hole X-ray binaries consist of a black hole and a companion star. When material from the companion is accreted by the black hole, a large amount of gravitational potential energy is released, producing intense X-ray radiation, sometimes accompanied by disk winds and relativistic jets. This enables direct investigation of complex accretion radiation physics processes in regions near the black hole, including the dynamics of black hole accretion disks and jet formation mechanisms. However, due to current observational limitations, direct imaging of black hole X-ray binaries is not possible, and we must rely primarily on spectral and timing characteristics to understand the geometry and properties of the accretion flow.

Spectral analysis reveals that X-ray radiation is mainly composed of emission from the accretion disk \cite{1}, a high-energy corona formed through Comptonization by hot electrons (with radiation extending up to 100 keV \cite{2}), and a reflection component (including iron lines at 6.4–7 keV and a Compton hump at 20–30 keV) \cite{3}. In terms of timing properties, variations on different timescales are observed, with one particularly important phenomenon being quasi-periodic oscillations (QPOs) \cite{4,5}.

The Insight-HXMT (Insight Hard X-ray Modulation Telescope) satellite is China's first space X-ray astronomy satellite, and its successful launch and operation mark an important milestone for China in the field of space astronomy \cite{6}. Insight-HXMT covers an energy range of 1–250 keV and offers advantages of broadband coverage, a large field of view, and large effective area, without suffering from pile-up effects when observing bright sources. These features provide unique capabilities for studying short-timescale variability and broadband spectral properties of high-energy objects such as black holes and neutron stars. During its nearly seven years of on-orbit operation, Insight-HXMT has achieved a series of important scientific results \cite{7,8,9}. Its high-cadence observations and broadband coverage (1–250 keV) give it unique advantages in detecting and studying transient outbursts, such as those from black hole X-ray binaries, providing valuable data support for revealing the radiation mechanisms of these systems and offering crucial experimental evidence for the establishment and verification of theoretical models.

This review systematically examines the properties of black hole X-ray binary outburst evolution and summarizes the important scientific results from Insight-HXMT regarding this topic.

2. Outburst Evolution Process of Black Hole X-ray Binaries

Most black hole X-ray binaries are transients, remaining in quiescence for months or even decades before experiencing sudden increases in accretion rate that produce X-ray outbursts lasting weeks to months. Based on their X-ray properties, these "typical outbursts" can be divided into different spectral states, with their evolution tracing a counterclockwise "q"-shaped track in the hardness-intensity diagram (HID) \cite{10,11,12}.

At the beginning of an outburst, the source transitions from quiescence to the (low) hard state ((L)HS), where the spectrum is dominated by a non-thermal component with a power-law index typically between 1.5–2.1 \cite{4,13}. In the hard state, the accretion disk is generally believed to be truncated at tens to hundreds of gravitational radii \cite{14,15,16}. However, some studies have shown that broad iron lines detected in the hard state \cite{17} suggest the disk remains at or near the innermost stable circular orbit \cite{18}. The power density spectrum shows prominent QPOs, usually type-C QPOs and strong broadband noise features \cite{19}. Persistent, compact jets are also generally present in the hard state \cite{20}. As the source brightness increases, the inner radius of the accretion disk decreases \cite{21} and the disk temperature rises, with thermal disk emission becoming comparable to non-thermal coronal emission \cite{22}, indicating the source has entered the intermediate state (IMS) \cite{1,12}.

The intermediate state can be further divided into the hard intermediate state (HIMS) and soft intermediate state (SIMS), with the SIMS having a softer spectrum than the HIMS. The strong broadband noise and type-C QPOs characteristic of the hard and HIMS are replaced by relatively weak broadband noise and type-B QPOs \cite{23}. In the intermediate states, compact jets are typically suppressed while highly relativistic, intermittent jets appear \cite{24}. When the source enters the (high) soft state ((H)SS), the spectrum becomes dominated by thermal disk emission, and the power-law index of the non-thermal component increases to values typically greater than 2.1 \cite{1}. Sometimes weak type-A QPOs and power-law noise are observed in the soft state \cite{25}. In this state, jets are generally undetectable, and the disk inner radius is believed to reach the innermost stable circular orbit. As the outburst evolves, the source's flux decreases, returning from the intermediate state back to the hard state and eventually to quiescence, completing a full typical outburst.

Studying these typical outbursts of black hole transients helps us understand the accretion processes and underlying physics of black hole X-ray binaries. However, some black hole transient outbursts never reach the soft state, with spectra remaining dominated by a power-law component. These are called "failed state transition outbursts" \cite{26,27,28,29}. Figure 1 [FIGURE:1] shows the HID tracks for different types of outbursts. Failed state transition outbursts are not rare, accounting for approximately 38% of all outbursts \cite{27}. The cause of failed state transition outbursts may be that the accretion rate is too low, below 0.1 times the Eddington luminosity, preventing the disk inner radius from reaching the innermost stable circular orbit and the corona from cooling and collapsing \cite{27,28}. Another type of "failed outburst" may lack a clear initial hard state, or the initial hard state may be too weak or too brief to be detected above the sensitivity threshold of current satellites. This latter "failed outburst" phenomenon is quite rare, having been observed in only a handful of sources to date, and will not be discussed in detail here. Interested readers can refer to \cite{30,31,32} for more information. Thanks to Insight-HXMT's broadband coverage and high-cadence monitoring capabilities, its data play an important role in studying the outburst processes of black hole X-ray binaries, including both "classic outbursts" and various types of "failed outbursts," not only expanding the outburst sample but also providing important references for understanding the evolution of accretion geometry and radiation mechanisms in black hole transients.

3. Observational Cases of Black Hole X-ray Binary Outburst Evolution with Insight-HXMT

Since its launch in 2017, Insight-HXMT has accumulated a wealth of valuable observational data during its seven years of stable operation, leveraging its advantages of broadband coverage (1–250 keV) and large effective area. The accumulated observation time for black hole X-ray binaries such as MAXI J1820+070, MAXI J1348–630, and Swift J1727.8–1613 exceeds 24 Ms. These data provide crucial support for spectral and timing analyses, opening new research opportunities for deeply understanding the properties of accretion disks, coronae, and jets in black hole X-ray binaries, while offering important evidence for investigating outburst mechanisms and accretion radiation physics. This section presents a systematic overview of the important scientific results achieved by Insight-HXMT in studying black hole X-ray binary outburst evolution, focusing on representative cases with complete outburst observations.

3.1 MAXI J1820+070 Outburst in 2018

MAXI J1820+070 is a black hole X-ray binary discovered in March 2018 by MAXI (Monitor of All-sky X-ray Image). Insight-HXMT conducted continuous broadband (1–250 keV) monitoring throughout its entire outburst, yielding a series of important scientific results. For example, You et al. \cite{9} performed spectral analysis using Insight-HXMT data to study the dynamic evolution of the corona and accretion disk during the hard state. The authors proposed that as the corona contracts toward the black hole, its outflow velocity also increases. The dynamic evolution of the disk-corona geometry is shown in Figure 2 [FIGURE:2]. This result is significant for understanding the physical properties of coronae and jet formation mechanisms in black hole X-ray binaries. Additionally, Ma et al. \cite{8} detected QPOs up to 200 keV throughout the outburst, providing key observational constraints for jet physics.

Furthermore, Peng et al. \cite{33} conducted a detailed analysis of MAXI J1820+070 using Insight-HXMT data, revealing time lags between soft and hard X-ray photons during three phases of the outburst. The results show that during the first two phases, low-energy photons (below 140 keV) lag behind high-energy photons (140–170 keV), while during the decay phase the opposite occurs, with high-energy photons lagging behind low-energy photons. These lag timescales are on the order of several days. After accounting for these energy-dependent time delays, the typical "q"-shaped track in the hardness-intensity diagram can be corrected to a linear relationship. Based on timing analysis results, the authors further explored the possible evolution of the corona: from a radial corona covering the accretion disk at the outburst onset, gradually evolving into a vertically collapsing jet-like corona, which first expands vertically during the decay phase before eventually covering the disk again. This work provides new insights into the evolution of disk-corona structures in black hole X-ray binary systems.

You et al. \cite{34} also investigated the formation mechanism of magnetically arrested disks (MADs) in X-ray binaries by combining Insight-HXMT X-ray data with optical observations from AAVSO (American Association of Variable Star Observers) and radio data from AMI-LA (Arcminute Microkelvin Imager-Large Arrays). The results show that radio and optical fluxes lag behind X-ray flux by approximately 8 days and 17 days, respectively. The authors suggest that these delays may result from magnetic field amplification in the accretion disk by an expanding corona during the X-ray outburst, leading to MAD formation. They further propose that thermal-viscous instability in the outer disk may be the key factor causing the optical band delay. This study not only reveals the connection between accretion disk magnetic field evolution and multi-wavelength radiation but also provides important observational evidence for understanding the formation and dynamics of large-scale magnetic fields in black hole accretion disks.

3.2 MAXI J1348–630 Outburst in 2019

Zhang et al. \cite{35} performed a detailed spectral analysis of the black hole X-ray binary MAXI J1348–630 during its 2019 outburst using data from Insight-HXMT and Swift's X-ray Telescope (XRT). The source evolved from the hard state through intermediate and soft states, then back through intermediate and hard states, completing a full outburst that followed the classic outburst track of black hole transients and displayed a "q"-shaped pattern in the HID. Throughout the outburst, the spectrum could be well fitted by a multi-temperature disk plus power-law model. In the SIMS and soft state, the disk luminosity and color temperature followed the standard relationship, indicating that the inner disk radius was stable (reaching the innermost stable circular orbit). However, during other phases of the outburst, MAXI J1348–630 exhibited unusual behavior inconsistent with typical black hole transient evolution: during the initial hard state, the accretion disk had a smaller inner radius and higher color temperature than during the soft state (see Figure 3 [FIGURE:3] for the evolution of disk-corona parameters). This peculiar disk behavior can be partially explained by self-consistent Comptonization models such as the \textsc{simplcut} model, which accounts for inverse Compton scattering of disk photons by a high-energy corona. However, even after considering coronal scattering effects, the unusual behavior persists. To explain the anomalous trend of increasing inner radius with decreasing color temperature, the authors propose that the hardening factor in the early outburst is larger than typical values (approximately 1.7). Further analysis shows that this evolutionary trend between disk inner radius and temperature indeed requires variation of the hardening factor, evolving from approximately 3.5 in the hard state to about 1.7 in the HIMS. The authors interpret this evolution of the hardening factor as revealing a genuine and possibly rare accretion disk evolution process: during the early outburst, the inner disk is in a condensation process from optically thin material and has not yet reached sufficient optical depth, making its spectrum incompatible with standard optically thick disk modeling. As the outburst evolves and the source transitions from the hard state to the HIMS, the accretion disk density continuously increases and reaches equilibrium. These findings are important for understanding the outburst evolution and accretion disk dynamics in black hole X-ray binaries.

Weng et al. \cite{36} conducted detailed timing analysis of the 2019 outburst of MAXI J1348–630, revealing that radiation time lags between the disk and corona produce hysteresis effects and X-ray delays. The authors present a physical picture of the entire outburst: hard X-ray emission from the corona almost immediately causes optical brightening of the outer disk; subsequently, enhanced accretion in the outer disk propagates inward, leading to a delayed soft X-ray outburst on the viscous timescale (approximately 8–12 days). This disk-corona model successfully reproduces the observed "q"-shaped hardness-intensity diagram. The study also utilized broadband (1–150 keV) X-ray and ultraviolet observations from Insight-HXMT and Swift to discover that soft X-ray radiation from the accretion disk lags behind hard X-ray radiation from the corona. This finding is crucial for understanding energy transfer and interaction mechanisms between the disk and corona. After correcting for this time delay, the authors reinterpreted the hardness-intensity diagram, showing a linear correlation between thermal disk flux and non-thermal power-law flux, in stark contrast to the previous non-linear "q"-shaped track. The authors further discuss that this time delay mechanism may be universally applicable to other black hole X-ray binary systems. These results not only challenge the previous view that accretion flow and radiation properties are uniquely determined by mass accretion rate but also provide new perspectives for future observational and theoretical studies.

3.3 4U 1543–47 Outburst in 2021

After 21 years in quiescence, the black hole X-ray binary 4U 1543–47 underwent a new outburst in 2021, reaching a peak flux of 8 Crab. Thanks to Insight-HXMT's immunity to detector pile-up effects at high flux levels, the outburst was observed with high cadence. Zhao et al. \cite{37} conducted an in-depth study of the accretion dynamics and geometry during this outburst using combined observations from Insight-HXMT, NuSTAR (Nuclear Spectroscopic Telescope Array), and Swift. The best-fit model and residuals for Insight-HXMT low-energy (black) and medium-energy (red) data are shown in Figure 5 [FIGURE:5]. Through spectral analysis, the authors found significant reflection components when the source was in the soft state, indicated by the purple line in Figure 5. They propose that this reflection may result from photons from the inner disk region being bent by strong gravitational fields and re-illuminating the disk surface, producing a self-irradiation effect. Notably, the best-fit parameters indicate that this reflection component contributes more than 50% of the total flux. Using general relativistic ray-tracing simulations, the authors found that when the accretion rate approaches or exceeds the Eddington limit, the accretion disk develops a geometrically thick, funnel-like structure that successfully reproduces the observed results. For 4U 1543–47, particularly when the disk surface inclination exceeds 45°, the observed self-irradiation intensity can be reasonably explained. This study not only reveals the geometric structure evolution of accretion disks at high accretion rates but also provides new observational evidence and theoretical insights for understanding radiation processes in strong gravitational field environments.

3.4 Swift J1727.8–1613 Outburst in 2023

Swift J1727.8–1613 is a new X-ray transient first detected by Swift in August 2023, with subsequent multi-wavelength observations confirming it as a black hole X-ray binary. During this outburst, the source reached a peak brightness of 7 Crab, and Insight-HXMT conducted high-cadence monitoring. Liu et al. \cite{38} and Peng et al. \cite{39} performed detailed X-ray spectral analysis using quasi-simultaneous observations from Insight-HXMT, NuSTAR, and NICER (Neutron star Interior Composition Explorer). Spectral fitting results show that, unlike typical black hole X-ray binary spectra, this source exhibits a significant additional hard component. Based on this finding, Peng et al. \cite{39} estimated the source's spin parameter and measured an accretion disk inclination of approximately 40° after accounting for this component. Peng et al. \cite{39} propose that this additional hard component may be related to relativistic jets, the jet base beneath slower jets, or the corona, providing new perspectives for understanding the high-energy radiation mechanism of this source.

3.5 Failed Transition Outburst of H 1743–322 in 2018

In addition to complete outbursts, some black hole X-ray binaries exhibit failed transition outbursts that only evolve to the hard or intermediate state. Wang et al. \cite{40} studied the 2018 failed transition outburst of the black hole transient H 1743–322 using observations from Insight-HXMT, NICER, and NuSTAR, covering a broad X-ray band from 1 to 120 keV. Through spectral and timing analysis of the entire outburst, the authors found that although the source showed slight spectral softening at peak flux, it remained in the hard state throughout the outburst, making it a failed transition outburst that never entered the soft state. The HID is shown in Figure 6 [FIGURE:6]. The authors also studied the evolution of type-C QPO properties, finding that the QPO centroid frequency increased from approximately 0.1 Hz to about 0.4 Hz during the outburst rise and decreased during the decay. Additionally, they found a positive correlation between the X-ray flux of the continuum spectrum, the photon index, and the QPO rms amplitude. The QPO amplitude in the soft X-ray band (12%–16%) was slightly higher than in the hard X-ray band (8%–10%). By comparing the spectral and timing properties of failed transition outbursts with successful outbursts, the authors found that although this failed transition outburst did not reach the accretion rate threshold for transitioning to the soft state, both types of outbursts follow the same initial evolutionary track. This work enriches the sample of black hole transient outbursts and is important for further understanding their outburst mechanisms.

4. Summary and Outlook

Black hole X-ray binary systems are important objects of study in high-energy astrophysics, and their complex outburst evolution processes have long been a focus of astronomers. By studying the outburst evolution of these transient systems, we can gain deep insights into the accretion geometry, its evolution, and the physical laws of accretion radiation. The operation of Insight-HXMT has provided valuable observational data for this field, particularly through its broadband X-ray monitoring capabilities and high-cadence observations, enabling more in-depth studies of short-timescale variability and spectral properties of black hole X-ray binaries. From typical full outbursts to "failed state transition outbursts," Insight-HXMT data have played an important role in studying outburst mechanisms, constructing accretion radiation models, and constraining fundamental source properties. Furthermore, with the upcoming launch of China's next-generation space science mission—the enhanced X-ray Timing and Polarimetry mission (eXTP)—we anticipate revealing more secrets about black hole X-ray binary systems and advancing the field of high-energy astrophysics.

Acknowledgments

We thank the reviewers for their valuable suggestions, which have significantly improved the quality of this article. We also thank the Insight-HXMT core science team and ground system team for their support and assistance.

References

\cite{2001, 132: 377}
\cite{Homan J, Belloni T. Ap\&SS, 2005, 300: 107}
\cite{Remillard R A, McClintock J E. 2006, ARA\&A, 44: 49}
\cite{McClintock J E, Haswell C A, Garcia M R, et al. ApJ, 2001, 555: 477}
\cite{Done C, Gierliński M, Kubota A. A\&ARv, 2007, 15: 1}
\cite{Done C, Diaz Trigo M. MNRAS, 2010, 407: 2287}
\cite{Miller J M, Homan J, Steeghs D, et al. ApJ, 2006, 653: 525}
\cite{Esin A A, McClintock J E, Narayan R. ApJ, 1997, 489: 865}
\cite{Remillard R A, Sobczak G J, Muno M P, et al. ApJ, 2002, 564: 962}
\cite{Fender R P. MNRAS, 2001, 322: 31}
\cite{Gierliński M, Done C, Page K. MNRAS, 2008, 388: 753}
\cite{Cúneo V A, Alabarta K, Zhang L, et al. MNRAS, 2020, 496: 1001}
\cite{Casella P, Belloni T, Stella L. ApJ, 2005, 629: 403}
\cite{Fender R P, Belloni T M, Gallo E. MNRAS, 2004, 355: 1105}
\cite{Motta S E. AN, 2016, 337: 398}
\cite{Brocksopp C, Bandyopadhyay R M, Fender R P. NewA, 2004, 9: 249}
\cite{Tetarenko B E, Sivakoff G R, Heinke C O, et al. ApJS, 2016, 222: 15}
\cite{Capitanio F, Belloni T, Del Santo M, et al. MNRAS, 2009, 398: 1194}
\cite{Alabarta K, Altamirano D, Méndez M, et al. MNRAS, 2021, 507: 5507}
\cite{Ma R C, Soria R, Tao L, et al. MNRAS, 2022, 514: 5667}
\cite{McClintock J E, Remillard R A. Black Hole Binaries// Compact Stellar X-ray Sources, Cambridge: Cambridge University Press, 2006, 157-213}
\cite{Sunyaev R A, Titarchuk L G. A\&A, 1980, 86: 121}
\cite{Gilfanov M. LNP, 2010, 794: 17}
\cite{van der Klis M, Jansen F, van Paradijs J, et al. Nature, 1985, 316: 225}
\cite{van der Klis M. ARA\&A, 1989, 27: 517}
\cite{Zhang S N, Li T, Lu F, et al. SCPMA, 2020, 63: 249502}
\cite{Li C K, Lin L, Xiong S L, et al. NatAs, 2021, 5: 378}
\cite{Ma X, Tao L, Zhang S N, et al. NatAs, 2021, 5: 94}
\cite{Soria R, Ma R, Tao L, et al. MNRAS, 2022, 515: 3105}
\cite{Peng J Q, Zhang S, Wang P J, et al. ApJ, 2023, 955: 96}
\cite{Peng J Q, Zhang S, Chen Y P, et al. MNRAS, 2023, 518: 2521}
\cite{You B, Cao X, Yan Z, et al. Science, 2023, 381: 961}
\cite{Zhang W, Tao L, Soria R, et al. ApJ, 2022, 927: 210}
\cite{Weng S S, Cai Z Y, Zhang S N, et al. ApJ, 2021, 915: 61}
\cite{Zhao S J, Tao L, Li P P, et al. A\&A, 2024, 685: A42}
\cite{Liu H X, Xu Y J, Zhang S N, et al., arXiv: 2406.03834}
\cite{Peng J Q, Zhang S, Shui Q C, et al. ApJ, 2024, 960: 142}
\cite{You B, Tuo Y, Li C, et al. NatCo, 2021, 12: 1025}
\cite{Méndez M, van der Klis M. ApJ, 1997, 479: 926}
\cite{Wang P J, Kong L D, Chen Y P, et al. MNRAS, 2022, 512: 4541}
\cite{Homan J, Wijnands R, van der Klis M, et al. ApJS, 2001, 132: 377}