Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Return of the Clocked Burster: Exceptionally Short Recurrence Time in GS 1826-238

Tomoshi Takeda, Toru Tamagawa, Teruaki Enoto, Wataru Iwakiri, Akira Dohi, Tatehiro Mihara, Hiromitsu Takahashi, Chin-Ping Hu, Amira Aoyama, Naoyuki Ota, Satoko Iwata, Takuya Takahashi, Kaede Yamasaki, Takayuki Kita, Soma Tsuchiya, Yosuke Nakano, Mayu Ichibakase, Nobuya Nishimura

TL;DR

GS 1826$-$238, the archetypal clocked X-ray burst source, exhibits an exceptionally short recurrence time of $t_{ m rec} = 1.603 \pm 0.040$ hr during a 2025 hard-state episode observed by the NinjaSat CubeSat, challenging the canonical $t_{ m rec} \propto F_{ m bol}^{-\eta}$ relation with $\eta \approx 1.05$. The study combines time-averaged persistent emission modeling with time-resolved burst spectroscopy across 19 bursts, finding no PRE and a reduced burst-emitting area during this epoch; the hard-state spectrum is well described by a Comptonization model, and the source transitions to a soft state thereafter. A key interpretation is ignition confined to a smaller NS surface area, potentially aided by crustal heating, which would raise the local accretion rate and trigger earlier ignition; the authors also assess whether a soft spectral component could reconcile the results with prior trends. These findings place new constraints on ignition physics, accretion geometry, and crustal evolution in neutron-star LMXBs, highlighting the role of multi-dimensional accretion and thermal history in burst variability.

Abstract

We report the discovery of an exceptionally short burst recurrence time in the well-known clocked burster GS 1826$-$238, observed with the CubeSat X-ray observatory NinjaSat. In 2025 May, GS 1826$-$238 underwent a soft-to-hard state transition for the first time in 10 years. On June 23, NinjaSat began monitoring GS 1826$-$238 in the hard state and continued until it returned to a steady soft state. During this period, we detected 19 X-ray bursts: 14 during the hard state, 4 in the transitional state, and 1 in the soft state. In the hard state, we identified a new clocked bursting epoch, during which the burst recurrence time remained highly stable and unprecedentedly short among the clocked bursting phases of GS 1826$-$238, with $t_{\rm rec} = 1.603 \pm 0.040$ hr ($1σ$ error). Previous observations showed that the burst recurrence time in GS 1826$-$238 decreased with increasing mass accretion rate, reached its minimum value of $t_{\rm rec} \sim 3$ hr, and then increased again. The observed 1.6 hr recurrence time is therefore exceptionally short, indicating anomalous ignition conditions. We propose that this phenomenon reflects fuel accumulation over a smaller fraction of the neutron star surface, resulting in a higher local accretion rate compared to earlier epochs. This scenario naturally accounts for the exceptionally short recurrence time, together with the observed reductions during bursts in blackbody normalization (proportional to the emitting area) and fluence. We also discuss possible contributions from residual heat in the neutron star crust or the presence of an additional soft spectral component.

Return of the Clocked Burster: Exceptionally Short Recurrence Time in GS 1826-238

TL;DR

GS 1826238, the archetypal clocked X-ray burst source, exhibits an exceptionally short recurrence time of hr during a 2025 hard-state episode observed by the NinjaSat CubeSat, challenging the canonical relation with . The study combines time-averaged persistent emission modeling with time-resolved burst spectroscopy across 19 bursts, finding no PRE and a reduced burst-emitting area during this epoch; the hard-state spectrum is well described by a Comptonization model, and the source transitions to a soft state thereafter. A key interpretation is ignition confined to a smaller NS surface area, potentially aided by crustal heating, which would raise the local accretion rate and trigger earlier ignition; the authors also assess whether a soft spectral component could reconcile the results with prior trends. These findings place new constraints on ignition physics, accretion geometry, and crustal evolution in neutron-star LMXBs, highlighting the role of multi-dimensional accretion and thermal history in burst variability.

Abstract

We report the discovery of an exceptionally short burst recurrence time in the well-known clocked burster GS 1826238, observed with the CubeSat X-ray observatory NinjaSat. In 2025 May, GS 1826238 underwent a soft-to-hard state transition for the first time in 10 years. On June 23, NinjaSat began monitoring GS 1826238 in the hard state and continued until it returned to a steady soft state. During this period, we detected 19 X-ray bursts: 14 during the hard state, 4 in the transitional state, and 1 in the soft state. In the hard state, we identified a new clocked bursting epoch, during which the burst recurrence time remained highly stable and unprecedentedly short among the clocked bursting phases of GS 1826238, with hr ( error). Previous observations showed that the burst recurrence time in GS 1826238 decreased with increasing mass accretion rate, reached its minimum value of hr, and then increased again. The observed 1.6 hr recurrence time is therefore exceptionally short, indicating anomalous ignition conditions. We propose that this phenomenon reflects fuel accumulation over a smaller fraction of the neutron star surface, resulting in a higher local accretion rate compared to earlier epochs. This scenario naturally accounts for the exceptionally short recurrence time, together with the observed reductions during bursts in blackbody normalization (proportional to the emitting area) and fluence. We also discuss possible contributions from residual heat in the neutron star crust or the presence of an additional soft spectral component.

Paper Structure

This paper contains 12 sections, 4 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Observation and Data Reduction
  3. Results
  4. Spectral State Transition
  5. Type-I X-ray Burst
  6. Persistent Emission
  7. Time-resolved Burst Spectroscopy
  8. Discussion
  9. Exceptionally Short Burst Recurrence Time
  10. Deviation from the Previous $t_{\rm rec}$--$F_{\rm bol}$ Relation
  11. Physical Origin of the Short Recurrence Time
  12. Properties of the persistent and burst emissions observed from GS 1826$-$238

Figures (4)

  • Figure 1: (a) NinjaSat light curves of GS 1826$-$238 in 2--10 keV, 10--20 keV, and their hardness ratio with a bin size of 1.5 hr. The time intervals between 50-s before and 300-s after the burst onset are excluded to show the persistent flux, while these burst onsets are indicated by black vertical lines in each panel. (b) MAXI/GSC light curve in the 2–20 keV energy range with 1-day bins, downloaded from the MAXI on-demand archive Mihara2024. (c) Same as (b), but zoomed in on the latter part. This period is highlighted in panel (b) by the shaded teal region, while the shaded gray region corresponds to the NinjaSat observation interval. (d) Color-Color diagram observed during the NinjaSat observation campaign with a bin size of 6.0 hr. The IDs of the observed bursts are indicated at corresponding data points.
  • Figure 2: Average energy spectra of the persistent emission (blue circles) and burst emission (pink crosses) in the hard state (IDs 1--14 except 10), along with their respective best-fit models (top), and the corresponding data-to-model ratios (bottom). The burst spectrum was extracted using the interval 10--20 s after burst onset and rebinned to maintain a minimum significance of $5\sigma$ per bin for display purposes.
  • Figure 3: Evolution of the blackbody components during the Type-I X-ray burst from time-resolved spectroscopy, including data from the previous three clocked bursting epochs (Galloway2017). (a) bolometric (0.1--200 keV) unabsorbed flux in units of 10$^{-8}$ erg cm$^{-2}$ s$^{-1}$; (b) blackbody temperature, $kT_{\rm bb}$ (keV); (c) blackbody normalization, $K_{\rm bb}$${(R_{\rm km}/D_{10})^2}$, where $R_{\rm km}$ represents the source radius in kilometers and $D_{10}$ indicates the distance to the source in units of 10 kpc; (d) reduced C-stat value for the fit.
  • Figure 4: (a) Variations of (a-1) burst recurrence time, (a-2) burst fluence, (a-3) $\alpha$-value, and (a-4) blackbody normalization $K_{\rm bb}$ as a function of the bolometric persistent flux in the 0.1--1000 keV range. The dotted curve in panel (a-1) represents an empirical fit to the data from Galloway2017 (open black circles), assuming $t_{\rm rec} \propto F_{\rm per}^{-1.05}$Galloway2004. Panels (a-2) and (a-4) include linear fits to the same subset of data. The blue triangles represent the data from the 2003 April epoch reported by Thompson2008ApJ...681..506T, while the red squares show the results from the 2025 June epoch. In both cases, open and filled markers correspond to spectral fits without and with an additional disk blackbody component below 2 keV, respectively (see Section \ref{['sec:persistent']} for details). For the three previous epochs, $K_{\rm bb}$ values are the mean and standard deviation over 20--50 s after each burst onset, based on Figure \ref{['fig:BBcomponents_vs_time']}(c). For the 2025 June epoch, $K_{\rm bb}$ is derived from a spectral fit over the same interval. (b) $F_{\rm bol}t_{\rm rec}$ plotted against $K_{\rm bb}$ during the burst tail. The black dotted line represents the best-fit linear model ($F_{\rm bol}t_{\rm rec} = a K_{\rm bb}$) to all data points excluding the 2025 June epoch results with additional soft component, yielding $a = 0.388 \pm 0.009$$(R_{\rm km}/D_{10})^{-2}$erg cm$^{-2}$ and $\chi^2/{\rm d.o.f} = 5.01/3$.