Optical and Ultraviolet Monitoring of the Black Hole X-ray Binary MAXI J1820+070/ASASSN-18ey for 18 Months

Low-mass X-ray binaries (LMXBs) are close binary systems consisting of a compact object and a low-mass companion star having filled its Roche lobe. The mass-donor companion can be a main-sequence star, a subgiant, or a white dwarf (WD), while the compact object can be a black hole (BH) or a neutron star (NS). Materials of the donor star transfer to the compact object through the $L1$ Lagrangian point, forming a surrounding accretion disc (Charles & Coe, 2006; Gu et al., 2019). LMXBs often stay for years or decades in a state of quiescence with X-ray luminosities of $\sim 10^{29}-10^{33.5}$erg s^-1 before turning into outbursts (Remillard & McClintock, 2006). The mass of a BH in an X-ray binary can be determined based on the periodic variations inferred from the spectroscopic and photometric observations in quiescence (e.g.,Casares & Jonker 2014; Wu et al. 2015; Wu et al. 2016). During the outburst, the luminosity of these LMXBs can increase by many orders of magnitudes relative to its quiescent state (Chen et al., 1997). The outbursts can be explained by instability of accretion discs surrounding the compact object (Osaki, 1995; Lasota, 2001; Dubus et al., 2001), which is also called the “thermal-viscous disc instability model" (DIM). Some LMXB transients can experience several outbursts, and those with higher mass transfer rate tend to have higher outburst frequencies (Lin et al., 2019).

According to the irradiated disk model, the optical and ultraviolet (UV) -band emissions are believed to arise from reprocessing of the X-ray emission (Rykoff et al., 2007; Gierliński et al., 2008, 2009). In this scenario, the temperature structure of the outer disc will be altered by the irradiated inner disk and Compton tail, leading to production of the observed optical/UV emission (Gierliński et al., 2009). The optical spectra of LMXBs often contain emission lines, such as the Balmer series. The emission lines are believed to originate from rotating accretion flow of the disc, which is ionized by irradiation from a central high-luminosity X-ray source. Note, however, that there are other arguments about the origin of the optical radiation of LMXB transients, such as the jet/corona (Eikenberry et al., 1998; Markoff et al., 2005; Russell et al., 2006) or the atmosphere of an optically thick disc in bright/soft X-ray states (Wu et al., 2001, 2002). Moreover, the H_α emission line could also originate in a dense outflow in low/hard X-ray states (Wu et al., 2001). At low-luminosity state, the ionizing source is usually considered to be viscous heating of the disc (van Paradijs & McClintock, 1994; Fender et al., 2009).

The outburst of the LMXB system, MAXI J1820+070 (ASASSN-18ey), was initially discovered as an optical transient ASASSN-18ey on UT Mar.06.58 2018 by the All Sky Automated Survey for SuperNovae (ASAS-SN; Shappee et al. 2014; Kochanek et al. 2017) at R.A. = $18^{h}20^{m}21^{s}.9$, dec. = $+07^{\circ}11^{\prime}07^{\prime\prime}.3$ (J2000) ¹¹1http://www.astronomy.ohio-state.edu/asassn/transients.html. About one week later, the Monitor of All-sky X-ray Image (MAXI; Matsuoka et al. 2009) Gas Slit Camera (Mihara et al., 2011) nova alert system detected a bright X-ray transient at the same location with a flux of 32 $\pm$ 9 mCrab in 4 - 10 keV (Kawamuro et al., 2018), which was designated as MAXI J1820+070. Multiwavelength follow-up observations suggested that MAXI J1820+070 is a BH LMXB (e.g., Baglio et al., 2018; Homan et al., 2018; Tucker et al., 2018). The distance to this system is estimated as $3.46^{+2.18}_{-1.03}$ kpc according to the Gaia DR2 parallax data (Gandhi et al., 2019). Atri et al. (2020) provided a consistent and more precise measurement on the distance to MAXI J1820+070 as 2.96$\pm$0.33 kpc using the parallax obtained from radio interferometry.

The X-ray and optical monitoring of this system showed that there is a time lag of 7.20$\pm$0.97 days between the outbursts detected in optical and X-ray bands for MAXI J1820+070 in March, 2018 (Tucker et al., 2018). This delay suggests that the thermal instability initially triggered an outburst in the thermal-viscous disc. The X-ray and optical monitoring of this system suggests that the state transition is not only related to the mass accretion rate (Shidatsu et al., 2019). They proposed that the jet contributed to the optical emission in the low/hard state, whereas the outer disc emission dominated the optical flux in high/soft state. Polarization observations also favored for the existence of jet or hot flow for MAXI J1820+070 (Veledina et al., 2019). Dynamical modeling of MAXI J1820+070 system was carried out by Torres et al. (2019). They confirmed the property of stellar-mass BH for the compact primary of MAXI J1820+070, with its mass function as $f(M)=5.18\pm 0.15M_{\sun}$ and an orbital period of $0.68549\pm 0.00001$ days. Muñoz-Darias et al. (2019) detected P-Cygni profiles and broad wings in He i $\lambda$5876 and H_α emission lines in the high-resolution optical spectra of MAXI J1820+070 taken during low/hard states, suggesting the existence of accretion disc wind.

In this paper, we report extensive follow-up observations of MAXI J1820+070 in optical, UV, and X-ray bands for about 18 months, and present its observational properties. The relations between optical/UV and X-ray features have been analyzed, with an attempt to better constrain the radiation physics for this BH binary system. The paper is organized as follows: in Section 2, the observations and data reductions are described. Section 3 presents the light/color curves, and Section 4 presents the spectral evolution. In Section 5, we discuss the properties of MAXI J1820+070, including spectral energy distribution (SED) and the correlation between optical and X-ray properties. We summarize our results in Section 6.

After the discovery of MAXI J1820+070, we started a long-term spectroscopic and photometric monitoring campaign in optical bands. The follow-up photometric observations started on March 20, 2018, using the 0.8-m Tsinghua-NAOC Telescope (TNT) (Huang et al. 2012; see Figure 1 for a sample image) and the Yaoan High Precision Telescope in China, as well as the AZT-22 1.5-m telescope (hereafter AZT) at the Maidanak Astronomical Observatory in Uzbekistan (Ehgamberdiev, 2018). We use standard IRAF routines to pre-process all CCD images, which include bias subtraction, flat fielding, and the removal of cosmic rays. Point-spread-function (PSF) photometry was performed for both the object and the reference stars using the pipeline $Zuruphot$ developed for automatic photometry of TNT (Mo et al. in prep.). The instrumental magnitudes were then converted into those of the Johnsons $BV$ (Johnson et al., 1966) and Sloan Digital Sky Survey (SDSS) $gri$-band system (Fukugita et al., 1996). Table 1 lists the standard BV- and $gri$-band magnitudes of the comparison stars, which are also marked in Figure 1. MAXI J1820+070 was also observed by the Ultraviolet/Optical Telescope (UVOT; Roming et al. 2005) onboard the Neil Gehrels Swift Observatory (Swift; Gehrels et al. 2004) in three UV ($uvw2$, $uvm2$ and $uvw1$) and three optical filters ($u$, $b$ and $v$). The final calibrated magnitudes in the Swift filters are presented in Table 2 and Table 4, respectively. The Swift images of MAXI J1820+070 were reduced using the HEASOFT ²²2HEASOFT, the High Energy Astrophysics Software https://www.swift.ac.uk/analysis/software.php with the latest Swift calibration database ³³3https://heasarc.gsfc.nasa.gov/docs/heasarc/caldb/swift/.

We also analyzed the Swift/XRT observations of MAXI J1820+070 obtained during the period from MJD 58189 to MJD 58591. These observations were obtained in both window timing (WT) and photon count (PC) modes. All of the XRT data were first reduced by running the $xrtpipeline$ task of HEASOFT (version 6.19). Since the count rate of MAXI J1820+070 can reach beyond 500 cps, the source region suffers serious pile-up effect. To avoid this effect, we extract the source photons using an annular region. The outer radius is set as 40 pixels, and the inner radius is adjustable according to the count rates. In the WT mode observations, the inner radius is set as RATE/100 pixels (RATE stands for the count rate). In the PC mode observations, the inner radius is fixed as 2 pixels if the count rate is higher than 0.5 cps; otherwise, the inner radius is set to zero. The background photons were estimated from an annular region with an outer radius of 120 pixels and an inner radius of 60 pixels. The energy spectra were extracted by the XSELECT package and grouped into bins with a minimum of 20 photons. We adopted response matrix files (RMFs) from the Swift/XRT calibration database (CALDB) and created ancillary response files (ARFs) from exposure maps using the $xrtmkarf$ task. The XRT spectra between 0.6 and 10 keV were fitted with a model composed of a power-law component (the $powerlaw$ model) plus a multi-black body component (the $diskbb$ model). The power-law component accounts for the comptonized X-ray emission, and the multi-black body component represents the thermal X-ray emission from the surface of the accretion disc, while the X-ray absorption is characterized by the $wabs$ model. During the phase from 336 to 371 days after the initial detection, the X-ray count rate is quite low so we use a single power-law model to estimate the flux (Kirsch et al., 2005).

A total of 66 spectra were obtained for MAXI J1820+070 with the Lijiang 2.4-m telescope (+YFOSC) and Xinglong 2.16-m telescope (+BFOSC and +OMR) (Zhang et al., 2016), covering the phases from +11.9 days to +385.3 days since the discovery. A log of the spectra is given in Table 6. We reduced all the spectra using standard IRAF routine. Flux calibration of the spectra was performed with spectrophotometric standard stars taken on the same nights. The spectra were corrected for atmospheric extinction using the extinction curves of local observatories; and the telluric lines were removed from the spectra.

Figure 2 shows the optical and UV light curves of MAXI J1820+070, covering the phase from +6.92 to +536.67 days relative to the first detection of optical outburst. For the TNT observations, the first gap appeared between July and August, 2018, covering the phase from +114 to +185 days, is due to the maintenance of the telescope in the summer season, while the second gap, seen during the phase from t$\sim$ +258 to +372 days, is because that the source moved close to the sun. To complement the sampling of the light curves, we also collected the $B$-band data from the American Association of Variable Star Observers (AAVSO).

At t $\sim$ +100 days and 210 days after outburst, MAXI J1820+070 is found to exhibit two rebrightening behaviors. However, it did not show quiescence between these two epochs. The first rebrightening occurred before the transition from low/hard to high/soft states, while the second one occurred shortly after the transition from the high/soft to low/hard states. By t$\sim$ +360 days, this system faded gradually towards its quiescent state (Russell et al., 2019b). At t $\sim$ +370 days, however, this system is found to experience a post-outburst “rebrightening" in both optical and X-ray bands (Ulowetz et al., 2019; Bahramian et al., 2019), and reached a B-band peak of 14.17 mag at t$\sim$+380 days. This corresponds to an absolute magnitude of 1.47 mag in B band using the Gaia distance (Gandhi et al., 2019). After that, MAXI J1820+070 tended to decrease in both X-ray and optical bands and entered into the quiescent state based on the X-ray flux reported by Swift and NuSTAR (Vozza et al., 2019; Tomsick & Homan, 2019). At t$\sim$+507 days, MAXI J1820+070 was found to rebrighten again in opitcal (see Figure 2) and other bands including radio, near-infrared and X-ray bands (see Hambsch et al. 2019, Xu et al. 2019, Bright et al. 2019 and Hankins et al. 2019).

In this work, we present optical, UV and X-ray observations and studies for the BH binary system MAXI J1820+070. Several outbursts/rebrightenings were recorded during our monitoring campaign spanning from its initial optical outburst to $\sim$550 days after. The main results from our analysis of the data are summarized as follows.

1. 1.

   The spectra of MAXI J1820+070 are characterized by blue continuum and the superimposed Balmer series, helium emission lines, and Bowen blend features, similar to those seen in several other well-known BH LMXBs like V404 Cygni (Mata Sánchez et al., 2018). The H_α emission becomes progressively strong within $\sim$300 days from the first detection of outburst, followed by a subsequent weakening during 350$\sim$400 days. A similar trend can be also found in the H_β feature. The He i $\lambda$5876 line is also found to become progressively stronger during the phase from the first detection to about 400 days after that. Such an evolution trend can be explained by the change in optical depth of continuum, as a result of the temperature change of the outer disc during the outburst.

2. 2.

   Based on extensive optical spectroscopy and the X-ray observations of MAXI J18020+070, we analyzed the relationship between the X-ray luminosity and the pEW of the H_α emission, and we found an anticorrelation as $\textup{pEW}_{H_{\alpha}}=10^{4.3\pm 1.1}L_{x}^{-0.092\pm 0.030}$, confirming previous results.

3. 3.

   After removing the effect of continuum, the absolute flux of H_α, H_β and He ii $\lambda$4686 showed positive correlations with the X-ray flux obtained at similar phases. This confirms that the high-energy irradiation can be regarded the pumping source for the emission of MAXI1820+070. In addition, the FWHM of H_β and He ii $\lambda$4686 is found to become smaller with the increase of X-ray flux but they tend to stabilize at line forming region of $1.25-1.71R_{\sun}$, consistent with the typical radius of accretion disk around a stellar-size black hole.

4. 4.

   The time lag found for the change in emission between optical and X-ray bands is $\Delta t=t_{0.01}^{XRT}-t_{0.01}^{B}=20.80\pm 2.85$ ($\Delta t\pm 2\sigma$) days, which means that during the rebrightening process, the start of the rising in optical preceded that in the X-ray.

5. 5.

   At t$\sim$200-300 days, the X-ray flux is found to show a sudden drop, while the flux variation in optical/UV flux is much less significant. This discrepancy suggests that the viscous energy of the accretion disk can contribute significantly to the optical/UV flux when irradiation diminished.

We thank the anonymous referee for suggestive comments to help improve this work. We acknowledge the support of the staff of the Lijiang 2.4 m and Xinglong 2.16 m telescopes. Funding for the LJT has been provided by Chinese Academy of Sciences and the People’s Government of Yunnan Province. The LJT is jointly operated and administrated by Yunnan Observatories and the Center for Astronomical Mega-Science, CAS. We acknowledge the support of the staff of the Xinglong 80cm telescope. It was partially supported by the Open Project Program of the Key Laboratory of Optical Astronomy, National Astronomical Observatories, Chinese Academy of Sciences. This work is supported by the National Natural Science Foundation of China (NSFC grants 12033003, 11761141001 and 11633002), the National Program on Key Research and Development Project (grant no. 2016YFA0400803). X.W. is also supported by the Scholar Program of Beijing Academy of Science and Technology (BS2020002). J.W. acknowledges the support of the National Natural Science Foundation of China (grant No. U1938105) and the President Fund of Xiamen University (No. 20720190051). Lingzhi Wang is sponsored (in part) by the Chinese Academy of Sciences (CAS), through a grant to the CAS South America Center for Astronomy (CASSACA) in Santiago, Chile. CYW is supported by the National Natural Science Foundation of China (NSFC grants 12003013). This research has made use of data provided by the Yaoan High Precision Telescope. We thank the staff of AZT for their observations and allowance of the use of the data. We acknowledge with thanks the variable star observations from the AAVSO International Database contributed by observers worldwide and used in this research.

software: ZrutyPhot (Mo et al. in prep.), IRAF (Tody, 1986, 1993), SExtractor (Bertin & Arnouts, 1996)

The data underlying this article are available in the article. Our photometric data has been attached in Table 1, 2 and 4 in the appendix. The log table of spectral data is also shown in Table 6.