Discovery of a New Spectral Transition in Swift J0243.6+6124 in the Sub-Eddington Regime

TL;DR

This study analyzes Swift J0243.6+6124 in its sub-Eddington accretion regime using broadband data from Insight-HXMT and NICER to map its spectral evolution across multiple outbursts. It identifies a new sub-Eddington transition at $L_t \approx 4.5\times10^{37}$ erg s$^{-1}$, accompanied by a turnover in the blackbody normalization, adding to the source’s known transitions and underscoring the complexity of its emission. The authors interpret the transition within a multipolar magnetic-field framework, with weak and strong magnetic poles dominating at different accretion rates, yielding an effective dipole field of $\sim 6.6\times10^{12}$ G while permitting local surface fields to exceed $10^{13}$ G. Methodologically, they perform broadband spectral fits with a physical model, quantify $L_t$ via a broken-linear radius–luminosity relation, and discuss cross-calibration–induced luminosity offsets, providing constraints on the magnetic topology of this extreme XRP.

Abstract

We conduct a detailed spectral analysis of the Galactic ultraluminous X-ray pulsar Swift J0243.6+6124 in its sub-Eddington regime, using Insight-HXMT and NICER observations during multiple outbursts including the 2018 giant outburst. We discover a new transition at $L_{\rm t} \approx 4.5 \times 10^{37}\ {\rm erg\ s^{-1}}$, accompanied by systematic evolution of spectral parameters, in particular a significant turnover in the blackbody normalization. This transition luminosity in the sub-Eddington regime represents the fifth transition identified so far in Swift J0243.6+6124, further highlighting the complexity of its accretion-powered emission. We interpret the transition in terms of a multipolar magnetic-field configuration, where weak ($\sim 2.8 \times 10^{12}\ {\rm G}$) and strong ($\sim 1.6 \times 10^{13}\ {\rm G}$) magnetic poles dominate the emission at different accretion rates. On the magnetospheric scale, this configuration is equivalent to an effective dipole field of $\sim 6.6 \times 10^{12}\ {\rm G}$, while allowing the local surface field to exceed $10^{13}\ {\rm G}$.

Figures (4)

• Figure 1: Bolometric luminosity evolution during the decay phases of three outbursts. The luminosity is derived from Insight-HXMT observations in the 2–150 keV band. The red, blue, and green curves correspond to the outbursts in 2018 (MJD 58125–58170), 2019 (MJD 58493–58523), and 2023 (MJD 60152–60181), respectively. The purple dashed line marks the newly identified transition luminosity(with the uncertainty interval at the 90% confidence level marked by the purple bar), while the black dashed line represents the previously reported critical luminosity $L_{\rm 1}$ of the source2020ApJ...902...18K. The x-axis shows the elapsed days since the first data point used for each outburst.
• Figure 2: Parameter evolution versus luminosity derived from spectral fits performed with three instrument combinations. Left: Insight–HXMT only (LE: 2--10 keV; ME: 8--30 keV; HE: 28--100 keV), covering 2--100 keV in fitting, with luminosities evaluated over 2 -- 150 keV; Middle: Joint fits in which NICER (0.7--10 keV) replaces the LE data when observations are within 24 hr (ME: 8--30 keV; HE: 28--100 keV); luminosities are also evaluated over 2 -- 150 keV; Right: NICER-only narrowband fits (0.7--10 keV), with luminosities computed in the same energy band. The grey dashed line marks the adopted transition luminosity $L_{\rm t}$ (0.7--10 keV). The purple shaded band in the left and middle panels spans the 90% credible range for each data set.
• Figure 3: Broken--linear fit to $Radius$ versus normalized luminosity $x = L / (1.8 \times 10^{38})$ with $x \le 0.49$. Data points are color--coded by year; the solid curve represents the maximum--a--posteriori (MAP) broken--line model, and the vertical dashed line marks the break location. The gray and purple shaded bands show, respectively, the 90% credible intervals for $x_b$ with and without the inclusion of distance uncertainty. Left panel (Insight--HXMT, frozen $N_{\rm H}$): the break corresponds to $L_x = (4.46^{+0.53}_{-0.53}) \times 10^{37}\ {\rm erg\ s^{-1}}$. Right panel (Insight--HXMT, thawed $N_{\rm H}$): the break is $L_x = (4.42^{+0.54}_{-0.54}) \times 10^{37}\ {\rm erg\ s^{-1}}$.
• Figure 4: Panel (a): Comparison of spectral fits at similar luminosities from three observations in different outbursts, using Insight--HXMT data only. Panels (b)–(d): Spectral fits for three representative observations at luminosities below, around, and above $L_{\rm t}$, respectively, obtained with different continuum models; the corresponding best-fit parameters are listed in Table 2. Panel (e): Joint spectral fitting of the 2018 February 02 observation over 0.7--100 keV with $N_{\rm H}$ free, comparing three cases: (1) a fully free joint fit; (2) a joint fit with the continuum fixed to the Insight--HXMT-only best fit in 2--100 keV; and (3) a joint fit with the continuum fixed to the NICER-only best fit in 0.7--10 keV. In cases (2) and (3), $N_{\rm H}$, the normalization, and the cross-calibration constant are allowed to vary. Panel (f): Fits to the 2018 February 02 NICER and Insight--HXMT LE data with $N_{\rm H}$ free, comparing: (1) a free NICER-only fit in 0.7--10 keV; (2) a free Insight--HXMT LE-only fit in 2--10 keV with parameters initialized at the NICER best-fit values; and (3) a NICER-only fit initialized at the best-fit parameters from the fully free joint fit.