LAXPC instrument onboard AstroSat: Five exciting years of new scientific results specially on X-ray Binaries
J. S. Yadav, P. C. Agrawal, Ranjeev Misra, Jayashree Roy, Mayukh Pahari, R. K. Manchanda
aa r X i v : . [ a s t r o - ph . H E ] F e b J. Astrophys. Astr. (0000) :
LAXPC instrument onboard AstroSat: Five exciting years of new scien-tific results specially on X-ray Binaries
J. S. Yadav , P. C. Agrawal , Ranjeev Misra , Jayashree Roy , Mayukh Pahari and R. KManchanda Department of Physics, Indian Institute of Technology, Kanpur 208016, India DAA (retd), Tata Institute of Fundamental Research, Homi Bhabha Road, Mumbai 400005, India Inter-University Centre for Astronomy & Astrophysics, Ganeshkhind, Pune-411007, India School of Physics and Astronomy, University of Southampton, Highfield Campus, Southampton SO17 1BJ, UK Department of Physics, Indian Institute of Technology, Hyderabad 502285, India Centre for Astrophysics, University of Southern Queensland, QLD 4300, Australia * Corresponding author. E-mail: [email protected] received November 7, 2020; accepted January 6, 2021
Abstract.
With its large e ff ective area at hard X-rays, high time resolution and having co-aligned other instru-ments, AstroSat / LAXPC was designed to usher in a new era in rapid variability studies and wide spectral bandmeasurements of the X-ray binaries. Over the last five years, the instrument has successfully achieved to a signifi-cant extent these Science goals. In the coming years, it is poised to make more important discoveries. This paperhighlights the primary achievements of AstroSat / LAXPC in unraveling the behavior of black hole and neutron starsystems and discusses the exciting possibility of the instrument’s contribution to future science.
Keywords.
Black hole—Neutron star—accretion—X-ray—radio jet—Space instrument.
1. Introduction
X-ray astronomy was born in 1962 with the chance dis-covery of a bright X-ray source Scorpius X-1 (Sco X-1) in a rocket flight experiment. It is still the brightestX-ray source in the sky radiating energy in X-rays ata prodigious rate of ∼ ergs / sec. What processgenerates X-rays at such high rate, remained a mysteryfor many years. In 1933 F.Zwicky and W.Baade madea prophetic suggestion that when cores of high massstars at the end of their life collapse, it becomes a Neu-tron Star in a massive explosion called supernova. Theequation of state of these dense stars were investigatedby J. R. Oppenheimer and G. M. Volko ff in 1939. TheNeutron stars remained a theoretical curiosity till 1967when A. Hewish and Jocelynn-Bell discovered a radiosource now called PSR B1919 +
21. producing periodicsignal every 1.3373 sec. These pulsating objects cameto be known as Pulsars, were interpreted by T. Goldas spinning Neutron stars. About a few thousand Pul-sars are now known. X-rays with 33 msec pulsation pe-riod of were discovered from a source (named as PSRB0531 +
21) in the Crab Nebula in 1969 in a rocket ex-periment. The Crab pulsar was found to be spinningdown in all the wavebands monotonically. This sug- gests that these objects are rotation powered pulsars.However, these discoveries did not contribute signifi-cantly to explain the nature of Sco X-1. Soon after thelaunch of the first X-ray satellite UHURU in 1970, itdiscovered a bright X-ray source Cen X-3, whose in-tensity was oscillating with 4.8 sec period. The objectwas found to be a Binary since its light curve showed,periodic X-ray eclipses every 2.09 days. This discoveryled to the conclusion that the bright X-ray sources arecompact objects, either a neutron star or a black hole, ina binary with a companion star. The strong gravity ofthe compact objects pulls matter from the companionstar. This matter forms an accretion disk from whichthe matter accretes to the compact object.If the compact object is a neutron star, matter ac-cretes either on the surface of the neutron star in case ofthe neutron star with low magnetic field ( gauss),or accretes at the magnetic poles of a highly magnetizedneutron star. If the magnetic axis is o ff set from the ro-tation axis, the X-ray intensity oscillates as the neutronstar spins. Several hundred X-ray binaries have beendiscovered in the Galaxy, majority of them with neu-tron star as the X-ray source. The in falling matter atthe poles under strong gravity of the neutron star getsheated up as it settles down on the surface of the neu- © Indian Academy of Sciences 1
J. Astrophys. Astr. (0000) : tron star or at its magnetic poles. The X-ray luminosityof the binary is due to release of gravitational energyand depends on the accretion rate.1.1
Classification of X-ray Binaries and their SalientFeatures :
The X-ray binaries are classified in two categories; theLow Mass X-ray Binaries (LMXBs) and the High MassX-ray Binaries (HMXBs). The LMXBs usually have alow mass ( ≤ > gauss) but a small numberof the neutron star LMXBs have higher magnetic fields > Gauss ( ∼ Gauss). The well known binaryHer X-1 is an example of the LMXBs in which theneutron star has a high magnetic field as inferred fromthe presence of the cyclotron feature in its energy spec-trum. These systems also exhibit regular X-ray pul-sations similar to those in the HMXBs. The LMXBswith low magnetic field neutron stars, are thought tohave evolved from high magnetic field neutron star bi-naries in which the accretion has spun up the neutronstars over a long period during which the magnetic fieldhas significantly decayed. The accreting millisec pul-sars (spin periods ∼ msecs) are an example of the spunup neutron stars from their progenitor LMXBs. TheLMXBs with weak magnetic field, are characterised byshort orbital periods and sporadic or regular short X-raybursts (strictly not periodic), that have a typical dura-tion of about a few seconds to a minute or in rare caseslonger.The matter accretes from the companion star on theneutron star via Roche Lobe overflow leading to theformation of an accretion disk around the neutron star.From the inner part of the accretion disk, the matterfalls on the surface of the neutron star. The accretedmatter piles up on the surface of the neutron star form-ing a layer with increasing thickness. As the accretionrate is variable, thickness of the layer in not uniform.When the accumulating matter on the neutron star sur-face develops density and temperature conditions in alocalized region that ignite thermonuclear reactions apowerful X-ray Burst occurs. These are called ther-monuclear bursts. Initially the flame is localized in aspot but it spreads quickly over the neutron star sur- face. Due to uneven spread of the nuclear burning, theX-ray in-tensity is modulated due to rapid spin of theneutron star. The intensity oscillations with frequencyof several hundred to about a kHz are detected in theLMXB bursts. This is one of the ways to measure thespin periods of the neutron stars in the LMXBs.Using the RXTE / PCA data, kHz QPOs were dis-covered from about a dozen LMXBs (usually occur-ring in pairs) (van der Klis, 2001). The precise originof kHz QPOS is still not well understood. The LMXBshave multi-component continuum energy spectra con-sisting of a disk blackbody component and a ThermalCompton power-law. The thermal Compton part of thespectrum is most likely due to the Compton up scat-tering of X-rays by hot electrons in a halo of accretedmaterial surrounding the disk.A precise modeling of the continuum is essentialto detect weak absorption features known as CyclotronResonant Scattering Features (CRSFs) or commonlyreferred as Cyclotron lines, that originate in the accre-tion column above the magnetic poles. The CRSFs aredetected in the strong magnetic field HMXBs as well asin the spectra of the strong field LMXBs. In the neutronstars with strong magnetic fields, the accreting matteris guided from the disk by the magnetic field lines ofthe neutron star to its magnetic poles. The X-ray spec-trum emitted by the plasma in the accretion column isa ff ected by the magnetic field. This interaction givesrise to the CRSFs. The energy of Cyclotron lines pro-vides a direct measure of the dipole magnetic field ofthe neutron star in the accretion column region.The HMXBs usually have a pulsating X-ray source.The pulsation periods have a range from ∼ +
65 has a spin pe-riod of 2.7 hours. The spin rate of the pulsars is variableand depends on the accretion rate. A study of the evo-lution of the pulsation periods of many pulsars, showsthat most pulsars exhibit both spin up as well as spindown episodes. Depending on the accretion rate, a pul-sar may make transition from spin up to spin down andvice versa. In a class of X-ray Binaries, the optical staris a Be star with a shell or disk of matter ejected bythe companion. Whenever the neutron star crosses theshell or disk during the course of its orbital motion, theaccretion rate shoots up resulting in dramatic increasein brightness known as X-ray outbursts that occur onceor twice in an orbital period. Some times there is a sud-den catastrophic release of matter from the disk or shellof the Be star. This may be due to thermal instabilitythat has still not been well understood. This leads toa massive spike in the accretion rate resulting in a gi-ant outburst which is non-periodic and unpredictable.This phenomenon also manifests in many X-ray tran-sients that remain in a dormant state with an exceed- . Astrophys. Astr. (0000) : ingly low accretion rate rendering them invisible. Thishibernation may last for years and even decades. Thendue to reasons still not understood, they spring back tolife with giant X-ray outbursts lasting for tens of daysto months. QPOs of ∼ mHz to tens of Hz have beendetected in many HMXBs. Study of these QPOs pro-vides insight into the radiation environment closest tothe neutron star.In black hole X-ray binaries (BHXBs), accre-tion disk and relativistic radio jets are integral parton all black hole mass scales and they provide sim-ple scaling of time and length with the mass of theblack hole from supermassive black holes in activegalactic nuclei (AGNs) to stellar mass black holes inBHXBs in our Galaxy. When accretion rate from thecompanion star is not su ffi cient to support continu-ous accretion flow to the black hole, matter fills theouter disk until a critical surface density is reachedand an outburst is triggered. An outburst in theBHXBs may last from ∼
20 days to several months.There are over 20 confirmed BHXBs and many morecandidates (Remillard & McClintock, 2006). Out of20 BHXBs, 17 are transient X-ray sources (mostlyLMXBs) and the rest three are persistent bright X-ray sources which have a massive O / B type stars astheir companion (HMXBs) (Remillard & McClintock,2006). Many new BHXBs have been discovered withtime (Mudambi et al ., 2020a; Bhargava et al ., 2019),and MAXI shows now lightcurves of around 40BHXBs.The BHXBs show di ff erent X-ray states at di ff er-ent stages of an outburst. Initially, luminosity was thesole criteria to determine X-ray states (van der Klis,1994; Tanaka et al ., 1995). With the availability oflarger data sample during RXTE / PCA era, it be-came clear that X-ray states are not simple func-tion of the luminosity (Lewin & Van der Klis, 2006;Remillard & McClintock, 2006). It is suggested thatbeside the mass accretion rate, there must be at leastone additional parameter which drives X-ray state tran-sition. The hardness-intensity diagram (HID) whichshows a q-shaped track, has been used to study the evo-lution of outbursts and associated X-ray states in theBHXBs systems. The four distinct X-ray states iden-tified in the HID are 1. Low Hard state (LS), 2. HighSoft state (HS), 3. Hard Intermediate State (HIMS),and 4. Soft Intermediate state (SIMS) (Belloni, 2006).The LS state is associated with relatively low accre-tion rate (lower than other bright X-rays states) but canbe observed at all luminosity. It is a radio loud state.The energy spectrum is dominated by a hard power-law while power spectrum shows a strong band-limitednoise. The HS state is strongly dominated by thermaldisk component. Radio quenching is seen during this
Figure 1 . Astrosat observations during observing cyclesA2-A9 showing percentage of time devoted to di ff erentinstruments as primary instrument. state. The intermediate states (HIMS & SIMS) havesignificant contributions from both the hard power lawand the thermal disk components and show the mostcomplex variability characteristics including most ofthe quasi-periodic oscillations(QPOs). Transient ra-dio jets are seen during HIMS to SIMS transition.Remillard & McClintock (2006) provided alternativescheme of X-rays states in BHXBs in terms of HardState (HS), Thermal State (TS) and Steep Power Law(SPL) state. Here major di ff erence is the SPL statewhich has significant contributions from Thermal aswell as non-thermal emissions.1.2 LAXPC Instrument and its features to study Tem-poral and Spectral properties :
Large Area X-ray Proportional counter (LAXPC)instrument is one of the major instruments on-board Astrosat which is India‘s first space sci-ence observatory (Agrawal, 2006). There are threeX-ray instruments on-board AstroSat which covera wide energy band. These X-ray instrumentsare: (i) LAXPC instrument (Yadav et al ., 2016b;Agrawal et al ., 2017; Yadav et al ., 2017) covering 3-80 keV region, (ii) a Cadmium-Zinc-Telluride Im-ager (CZTI; Bhalerao et al . (2017)) array covering 30-100 keV, and (iii) a Soft X-ray Imaging Telescope(SXT; Singh et al . (2016)) covering 0.3 - 8 keV. Be-side these X-ray instruments, there is an Ultra-VioletImaging telescope (UVIT: Tandon et al . (2017)). Allthe above instruments are co-aligned to provide si-
J. Astrophys. Astr. (0000) : multaneous multi wavelength observations from op-tical to hard X-ray. There is also a Scanning SkyMonitor (SSM) onboard Astrosat. The LAXPC in-strument uses three co-aligned identical LAXPC detec-tors (LAXPC10, LAXPC20 and LAXPC30) to achievelarge area. The performance of LAXPC detectors arediscussed by Antia et al . (2020) in the same issue. Ahigh detection e ffi ciency in the entire 3-80 keV energyregion is achieved by having a 15 cm deep detectionvolume divided in 5 identical Anode Layers and fill-ing the Xe + CH gas at two atmosphere pressure (1520torr).The LAXPC instrument was designed to meet theobjectives of studying the intensity variations withtime, have a high detection e ffi ciency in 3-80 keV band,provide a large e ff ective area over the broad energyrange to be able to study weak ( ∼ a few milliCrab)sources and have a good spectral resolution to measurethe X-ray continuum spectra of X-ray binaries with pre-cision for deciphering the presence of weak cyclotronlines in the NSXBs. A unique feature of the LAXPCis that every detected photon is time tagged to an accu-racy of 10 µ sec to enable investigation of rapid inten-sity variations over even sub-millisec scale. This fea-ture provides LAXPC with the capability to measurehigh frequency QPOs of even a few kHz. Good energyresolution (FWHM ∼
15% at 60 keV) was achievedby using an onboard gas purifier system and contin-uously monitoring of the resolution of the Veto layerby shining 60 keV X-rays on it from an Am source.We have done extensive lab as well as in orbit calibra-tion of LAXPC detectors (Antia et al ., 2017). Furtherimprovements and updates on LAXPC detector cali-bration and background are done (Antia et al ., 2020;Misra et al ., 2020b) (this issue).Since October 2016, AstroSat observatory has op-erated on the basis of targets selection through scien-tific proposals. during the observing cycles A02-A09in the last 5 years. A comparison chart for the observa-tions with di ff erent instruments as primary instrumentis shown in Figure 1. Primary instrument is definedin the Astrosat science proposals (as science proposalrequirement). It may be noted that simultaneous datafrom all the instruments will be available for a givensource pointing if all the instruments are on during theobservation.
2. Highlight of Results on Neutron Star X-ray Bi-naries (NSXBs):
Large number of X-ray binaries with an accreting Neu-tron star as the X-ray source, have been observed withthe LAXPC instrument to investigate their timing and spectral characteristics. Due to various reasons, brieflydiscussed in Antia et al( 2020), less than 50% data ofthe sources have been analyzed so far. Nevertheless,several new and interesting results have emerged fromthese studies. Here we summarize highlights of someof these results.
Figure 2 . Evolution of the Pulsar period of 4U 1909 + Timing Properties : Spin Periods and their evolu-tion, Low and High Frequency QPOs.
The spin periods of the neutron stars are a ff ected bythe accretion rate and the resulting accretion torque thatvaries with times. Usually the low and high frequencyQPOs are found in the LMXB pulsars while the low fre-quency QPOs are more common in the HMXB pulsars.Astrosat has studied several neutron star binaries to in-vestigate their spin rates and their evolution with time.LAXPC has observed a large number of HMXBs to in-vestigate their periodic and aperiodic variations like de-termination of the spin periods and their time evolution,QPOs and their origin We summarize some of the sig-nificant results derived from Astrosat.The LMXB 3A 1822-371 has an orbital pe-riod of 5.2 hours and pulsates with 0.59 sec period(Jonker & van der Klis, 2001). The spin period from1996 and 1998 data leads to a spin-up rate of -2.85 × − s s − . Subsequently measurements of the spin periodfrom several satellite observations have shown that thepulsar is monotonically spinning up implying that theaccretion rate has been constant over ∼
20 years. As-trosat LAXPC observations of 3A 1822-371 in Septem-ber 2016 yielded a barycenter corrected pulsar spin pe-riod to be Pspin = ± et al ., 2020) . Astrophys. Astr. (0000) : Making a linear fit to all the measured spin periods,gives a spin up rate = (-2.62 ± × − s s − givinga spin up timescale (P / P) = et al . (2010).Astrosat also studied accretion powered 2.26 mil-lisecond pulsar SAX J1748.9-2021, when a short andfaint outburst occurred in it in 2017, with the SXTand LAXPC instruments. From the spectral and tim-ing analysis of the LAXPC data,,the best-fitting orbitalsolution for the 2017 outburst was derived. Using thisan average local spin frequency of 442.361098(3) Hzwas obtained. The pulse profile obtained in 3-7 keVand 7-20 keV gave a constant fractional amplitude ∼ + LAXPC spectrum in 1-50keV showed the source to be in a hard spectral state.This spectrum is best fitted with a single temperatureblackbody and thermal Comptonization model. Timeresolved analysis of the bursts revealed complex evo-lution in emission radius of blackbody for the secondburst suggestive of a mild expansion of photosphericradius.The HMXB Pulsar 4U 1909 +
07 studied with theAstrosat-LAXPC observation of 2017 July, showedpresence of 604 sec X-ray pulsations. Using the spinperiod measurements obtained earlier with the variousX-ray satellites and the period deduced from the presentwork, a plot of Spin Period versus Time is shown in Fig2 (Jaisawal et al ., 2020b). Despite a brief episodes ofspin-down, there is a clear long term trend of spin-up.The pulse period changed from 604.7 s to 603.6 s be-tween 2001-2017 resulting in an average spin-up rate of1.71 ± − ss − . The pulse profiles show strong energydependence evolving from a complex broad structure insoft X-rays into a profile with a narrow emission peakfollowed by a plateau in energy ranges above 20 keV.The change in the pulse profile suggests a change in thebeaming pattern.Another HMXB 4U 1907 +
09 with a spin period of ∼
404 sec was studied by by using the LAXPC obser-vations of 4th and 5th June (2020). Timing analysis ofthe LAXPC data yielded a period of 442.33 ± +
09 shownin Fig 3 (left panel), exhibited di ff erent energy depen-dence, the pulsed fraction of the main peak increasedtill about 40 keV and decreased after that while the sec-ondary peak disappeared at energy above about 20 keV.Energy resolved pulse profiles created from combineddata of the three LAXPCs are shown in Fig 3 (rightpanel). The pulsar spin-down was found to continue(Varun et al ., 2019b).The Be binary with 194 sec pulsar GRO 12058 + ± ± × − Hz s − Mukerjee et al .(2020). Pulse profiles in 3-80 keV were found to be en-ergy dependent. The Power Density Spectrum (PDS) ofthe source revealed a 0.090 Hz QPO and its higher har-monics. Another HMXB pulsar OAO 1657-415 with10.4 day orbital period, was studied by Jaisawal et al .(2020a) with two LAXPC observations in March andJuly 2019. The observations covered orbital phases of0.681-0.818 and 0.808-0.968. Despite being outsidethe eclipsing regime, the PDS from the first data did notshow any clear signature of pulsation or quasi-periodicoscillations. However, during the second observation,the X-ray pulsations at a period of 37.0375 s. wereclearly detected in the orbital phase range 0.808-0.92.The pulse profile of the pulsar from the second obser-vation consisted of a broad single peak with a dip-likestructure in the middle all across the observed energyrange. The evolution of the emission geometry wasprobed by constructing the pulse profile in narrow timeand energy segments. The energy spectrum of OAO1657-415 is approximated by an absorbed power-lawmodel with an iron fluorescent emission line. These re-sults are explained in the frame of stellar wind accretionmodel.Another HMXB X-ray binary pulsar 3A 0726-260(4U 0728-25) was observed on 2016 May 6-7 with theLAXPC and SXT, after a gap of almost 20 years. Thelight curves of the binary show Strong X-ray pulsationswith a period of 103.144 ± ± Figure 3 . The pulse profile of 4U 1907 +
09 has double peakshape shown in the left panel. The energy resolved pulseprofiles created from combined data of the three LAXPCsare shown in the right panel.
J. Astrophys. Astr. (0000) : C O UN T S / SE C O ND LAXPC 205000 10 Figure 4 . The background subtracted light curves from the3 LAXPCs showing clear intensity oscillations of ∼ ∼ + energy spectrum of the source is derived from the com-bined analysis of the SXT and LAXPC spectral data in0.4-20 keV. The best spectral fit is obtained by a powerlaw model with a photon index (1.7 ± ff at 12.9 ± ∼ et al ., 2020).Detection of ∼ ∼ +
63 were reported by Roy et al .(2019) using the Astrosat-LAXPC observation of theHMXB. The QPOs were detected on 2015 October 24during the peak of a giant type II outburst. Prominentintensity oscillations seen at ∼ ∼ ∼ mHz oscillations are examined. These oscillationsbear resemblance to the intensity oscillations observedfrom some other neutron star and black hole sourcesand may have a common origin. Current models to ex-plain the instability in the inner accretion disk causingthe intense oscillations were examined and found to beinadequate.2.2 Energy Spectra, Detection and Study of CRSFs:
LAXPC has observed a large number of HMXBs toinvestigate their periodic and aperiodic variations andmeasure their continuum spectra and their characteriza-tion to reveal the presence of Cyclotron lines (CRSFs).Thus far Astrosat-LAXPC has discovered Cyclotronlines in a few pulsars in which earlier there was no re-port of the presence of CRSF. Study of cyclotron lineenergy and its profile enable determination of the mag-netic field of the neutron star and at the site of theirorigin in the accretion columns from the following re-
Figure 5 . Shift in the energy of the Cyclotron line in HerX-1. The 2018 observation with the LAXPC gave a value of37.5 keV. lation E c = . B / G )(1 + z ) − keV (1)E = Energy of the Cyclotron line, B = Magneticfield of the neutron star in unit of 10 gauss and z isgravitational red shift due to neutron star.There are reports of shift of the line energy withtime and correlation of this with the X-ray luminosity.The results on this are conflicting and reality of varia-tion of line energy is still an open question.Her X-1 was the first source in whichTruemper et al . (1978) discovered a CRSF usu-ally termed as Cyclotron line at ∼
40 keV. The CRSFenergy was found to vary with pulse phase, X-rayluminosity, the phase of 35-day precession cycle andwith time (Staubert et al ., 2014). Using data acquiredfrom several X-ray satellites over a 20 year period,the CRSF energy was inferred to decline by 4.5 keVfrom 1996 till 2012 (Staubert et al ., 2014) as seen inFig 5 from Bala et al . (2020). After 2015 the trendwas reversed and the CRSF energy began to recoverreaching a value of 37.4 keV (Staubert et al ., 2019).Recent analysis of Her X-1 observations in 2018 withthe LAXPC shows that since then the CRSF energyappears to be constant around 37.5 keV (Bala et al .,2020).Amin et al . (2020) constructed the energy spec-trum of the pulsar 3A 1822-371 using LAXPC10 andLAXPC20 data and fitted it using 3 di ff erent contin-uum models. In all the cases an absorption feature at23 keV was present which is suggested to be a CRSFand if correct this implies a magnetic field strength of B = (2.7-3.4) × Gauss for the neutron star. The en-ergy spectrum of HMXB 4U 1907 +
09 (spin period of ∼
404 sec) was studied by by (Varun et al ., 2019b) usingthe LAXPC observations of 4th and 5th June (2020). A . Astrophys. Astr. (0000) :
CRSF was detected in the spectrum at 18.5 ± +
42 was observed withthe LAXPC. Its energy spectrum extracted fromthe LAXPC data revealed presence of 3 CRSFs(Mukerjee et al ., 2020). The cyclotron absorption fea-tures were detected in (9.7-14.4) keV, (19.3-23.8) keVand (37.8-43.1) keV, one of which is the fundamen-tal line and the other 2 are harmonics The pulse phaseresolved spectroscopy of the source showed phase de-pendent variation in line energy and relative strengthof these features. An important finding from Astrosatis the discovery of a Cyclotron line at ∼
22 keV inthe LAXPC energy spectrum by of the HMXB pulsar4U 1538-52 which has an orbital period of 3.75 daysand the spin period is currently ∼
527 sec Varun et al .(2019a).The Pulse profile is double peaked at low en-ergy and has a single peak in high energy range, thetransition taking place around the cyclotron line energyof the source.The CRSF is detected with a very high significancein the phase averaged spectrum shown in Fig 6. It is oneof the highest signal to noise ratio detection of CRSFfor this source. A detailed pulse phase resolved spec-tral analysis with 10 independent phase bins was per-formed and parameters of the continuum spectrum andCRSF parameters were derived. Theses show pulsephase dependence over the entire phase with a CRSFenergy variation of ∼
13% in agreement with previousstudies. An increase in the centroid energy of the CRSFobserved between the 1996-2004 (RXTE) and the 2012(Suzaku) observations, is confirmed a ffi rming that theincrease in the line energy was a long-term change.AstroSat observed the Be / X-ray binary pulsar SXP15.3 in the Small Magellanic Cloud (SMC) during itsoutburst in late 2017, when the source reached a lu-minosity level of ∼ erg s − . Timing and spectralanalysis between 3 and 80 keV lead to the pulse pro-file that exhibits a significant energy dependence. Thepulse shape changes from a double peaked profile toa single broad pulse at energies >
15 keV. The energyspectrum suggests presence of a Cyclotron ResonanceScattering Feature (CRSF) at ∼ et al .,2019). To the best of our knowledge this is the lowestenergy Cyclotron line confirmed in any pulsar. This in-dicates a magnetic field strength of 6 × G for theneutron star (Maitra et al ., 2018).
Figure 6 . The energy spectrum of the pulsar 4U 1538-52derived from the LAXPC data. Top panel shows theobserved spectrum by the Black points and the best fittedspectrum by a Red line. The middle panel shows Residualswhen one fits only the continuum spectrum and the bottompanel shows fit with continuum and a Cyclotron line at 22keV.
Thermonuclear Bursts and QPOs in LMXBs:
The thermonuclear bursts (or commonly called asType-1 bursts) occur in the LMXBs having a weaklymagnetized ( < G) neutron star. Detailed timing andspectral studies of these bursts provides informationabout the spin period, temperature of the thermonuclearburning surface and radius of the neutron star. Theevent mode data in which each photon is time taggedto an accuracy of about 10 microsec, enables studiesof high frequency phenomenon like kHz oscillations,Some significant results on the Type-1 bursts observedwith the LAXPC are summarized below.LAXPC capability for high time resolution stud-ies of phenomenon like detection of Coherent Oscil-lations , kHz QPOs etc has been demonstrated byVerdhan Chauhan et al . (2017). They analyzed theLAXPC observations of 8th March 2016 for the LMXBsource 4U 1728-24. A 3 ks data stretch revealed occur-rence of a typical Type-1 burst of about 20 sec dura-tion. Dynamical power spectrum of the data in the 3-20keV band, shows presence of Coherent burst oscillationwhose frequency increased from 361.5 to 363.5Hz con-sistent with an earlier result showing the same spin fre-quency of the neutron star. Our knowledge of the spinperiods of the neutron stars in LMXBs can also be de-rived from detection of the Coherent Burst Oscillationsin the initial phase of the bursts. A kHz QPO, whosefrequency drifted from ∼
815 Hz to ∼
850 Hz, was de-tected just before the burst. The QPO was detected be-low 10 keV as well as in 10-20 keV band, which was
J. Astrophys. Astr. (0000) : not possible with the RXTE. Even for a short obser-vation with a drifting QPO frequency, the time-lag be-tween the 5-10 and 10-20 keV bands can be constrainedto be less than 100 microseconds.Beri et al . (2019) studied the LMXB 4U 1636-536using 65 ks LAXPC observation over 2 days and de-tected seven thermonuclear X-ray bursts including arare triplet of X-ray bursts. Time resolved spectroscopyperformed during these seven X-ray bursts suggestedthe presence of Photospheric Radius Expansion in threeof these X-ray bursts. A transient QPO at 5 Hz was alsodetected. No evidence of kilo-hertz QPOs or coherentburst oscillations, was found in the bursts and may per-haps be due to the hard spectral state of the source.E ff ects of Thermonuclear X-ray burst on non-burstemissions in the soft state of the LMXB 4U 1728-34was investigated by Bhattacharyya et al . (2018) to un-derstand if a significant fraction of the burst emission,which is reprocessed, contributes to the changes in thepersistent emission during the burst. This is importantsince it can introduce significant systematics in the neu-tron star radius measurement using burst spectra. An-alyzing the bursts data for 4U 1728-34 in the soft stateit was concluded that the burst emission is not signif-icantly reprocessed by a corona covering the neutronstar.Devasia et al . (2021) detected 5 Type-1 thermonu-clear X-ray bursts and one burst-like event in the neu-tron star LMXB source Cygnus X-2 using LAXPC X-ray data. An energy resolved burst profile analysis andtime resolved spectral analysis for each of the burstswas performed to characterize the burst properties. Anevolution of the blackbody temperature and radius isalso observed during each burst. A search for CoherentBurst Oscillations gave only upper limits. A study ofthe hardness-intensity and color-color diagrams showthat during the 2016 LAXPC observation, Cygnus X-2was in the early Flaring Branch (FB).The LMXB 4U 1323-62 is an interesting sourcefrom which periodic X-ray dips were first detected byEXOSAT van der Klis & Jansen (1985); Parmar et al .(1989). It was extensively studied by PCA / RXTE anda study of 40 Type 1 bursts showed a recurrence timeof 2.45-2.59 hr. which is probably the orbital periodof the binary. The LMXB 4U 1323-62 was observedwith the LAXPC and detected 6 Type-1 thermonuclearX-ray bursts in ∼
50 ks exposure. The time gap be-tween the successive thermonuclear bursts was foundto be consistent with the orbital period. Using the lin-ear and orbital quadratic ephemerides, the value of theOrbital period is estimated to be 2.65 to 2.70 h. Thegap observed between the bursts in two case, is nearlydouble the wait time of consecutive bursts as 2 burstswere missed in the data gaps. The nearly fixed time gap is the time required to accumulate the accreted matterto reach the level at which the thermonuclear reactionsset in producing a burst. The light curve of 4U 1323-62 also revealed the presence of two dips. A knownlow Frequency QPO (LFQPO) was detected at ∼ et al ., 2020).The spectral and timing properties of the atollsource 4U 1705-44 were studied by Agrawal et al .(2018) using 100 ks data from the LAXPC instrument.The source was in the high-soft state during the LAXPCobservations and traced out a banana track in the Hard-ness Intensity Diagram (HID). From the Power DensitySpectra (PDS) a broad Lorentzian feature centered at1-13 Hz and a very low frequency noise (VLFN) is de-tected. The energy spectra are well described by sum ofa thermal Comptonized component, a power-law anda broad (FWHM ∼ ∼ + et al . (2020) in 0.8-50 keV using the Soft X-ray Telescope (SXT) and theLAXPC data. For the first time, Cross-correlation stud-ies were performed using SXT soft and LAXPC hardlight curves and they exhibited correlated and anti-correlated lags of the order of a hundred seconds. Spec-tral modeling gave disk radius of ∼ + et al . (2020) investigated evolution of tim-ing and spectral properties of the bright Z-source GX5-1 using February, 2017 observations with the SXTand LAXPC instruments. The 0.8-20 keV spectra fromsimultaneous SXT and LAXPC data at di ff erent loca-tions of the hardness-intensity plot is well describedby a disk emission and a thermal Comptonized com-ponent. The disk flux ratio (ratio of the disk flux tothe total flux) increases monotonically along the hori- . Astrophys. Astr. (0000) : zontal branch to the normal one. Thus in the normalbranch, the disk dominates the flux while in the hori-zontal branch the Comptonized component dominates.The disk flux scales with the inner disk temperature asT . and not as T suggesting that either the inner radiichanges dramatically or that the disk is irradiated by thethermal component changing its hardness factor. ThePDS reveal a QPO whose frequency changes from ∼ / E0.8),while the harder X-ray seems to lag the soft ones witha time-delay of a milliseconds. The results suggest thatthe spectral properties of the source are characterizedby the disk flux ratio and that the QPO has its origin inthe corona producing the thermal Comptonized compo-nent.The results inferred, mainly from the LAXPC data,on various timing and spectral properties of LMXBssummarized here, demonstrate the capability of theLAXPC instrument for probing the bright as well asfaint sources It is hoped that in the coming years manymore LMXBs will be studied and new results willemerge from these studies.2.4
Studies of New Ultraluminous X-ray (ULX) Pul-sars with Astrosat:
The Ultraluminous X-ray (ULXs) sources detected inthe Galaxy and other galaxies were believed for a longtime to be Intermediate Mass Black Holes accretingmatter from the companion star. This changed whenPulsations with 1.37 sec period were discovered fromthe ULX in M82 . Since then more ULX Pulsars havebeen discovered and at present 8 ULX Pulsars havebeen discovered and their properties are summarized inTable 1 taken from Chandra et al . (2020). Astrosat dis-covered ULX Pulsar nature of a transient X-ray sourceRX J0209.6-7427 located in the Magellan Bridge whenit became active after a gap of 26 years in 2019 andwas studied by Chandra et al . (2020) with the SXT andLAXPC instruments on Astrosat. The transient first de-tected by ROSAT during its 1993 outburst, went into adeep hibernation for 26 years and suddenly sprang backto life in 2019 with a giant outburst in 2019.Using the SXT and LAXPC observations,Chandra et al . (2020) detected strong pulsations inRX J0209.6-7427 with 9.29 s periodicity over a broadenergy band covering 0.3-80 keV (first reported by theNICER mission in the 0.2-10 keV energy band). Thepulsar exhibited a rapid spin-up during the outburstas can be seen from Fig 7. Energy resolved foldedpulse profiles were generated in several energy bandsin 3-80 keV. This is the first report of the timing andspectral characteristics of this Be binary pulsar in . . . . Sp i np e r i o d ( s ) . . . . . . . . Days since MJD 58806 . − . . . R e s i du a l s Figure 7 . Evolution of the Pulsar period during the outburstof RX J0209.6-7427 in 2019 is shown in this plot. Spin-upof the pulsar during the outburst in clearly seen. hard X-rays. There is suggestion of evolution of thepulse profile with energy. The energy spectrum of thepulsar is determined and from the best-fitting spectralvalues, the X-ray luminosity of RX J0209.6-7427 isinferred to be 1.6 × erg s − . These timing andspectral studies suggest that this source has features ofan ultraluminous X-ray pulsar in the Magellan Bridge.A second ULX Pulsar Swift J0243.6 + x ∼ × and 6 × erg s − .A detailed Broadband timing and spectral study byBeri et al . (2020) show that X-ray pulsations at ∼ Figure 8 . The pulse profiles derived from SXT (0.5-7.0keV), LAXPC (3-80 keV) and CZTI (30-150 keV) areshown in the figure for the ULX Pulsar Swift J0243.6 + J. Astrophys. Astr. (0000) :
Table 1.
Summary of the 8 known ultraluminous X-ray pulsars.
Name of ULX
Host Galaxy Spin period (s) Orbital period (days) Spin-up / down L X (10 ergs s − )M82 X-2 M82 1.37 ∼ . ∼ .
42 64 Spin-up ∼ ∼ .
13 5.3 Spin-up ∼ ∼ . Swift
J0243.6 + ∼ . ∼ . ∼ ∼ . ∼ ∼ ∼ . ∼ ∼ ∼ . ff powerlaw ( α ∼ ∼ ff ective area enables studies of relatively faintersources like ULX Pulsars.
3. Study of Black Hole X-ray Binaries (BHXBs)with LAXPC / Astrosat
As has been mentioned earlier, the advantages that As-troSat / LAXPC has over RXTE are (i) an higher e ff ec-tive area at energies >
30 keV, (ii) event mode data al-lowing for temporal analysis in arbitrary user definedenergy bins and (iii) simultaneous observations fromother instruments on board. These have primarily de-fined and driven the LAXPC study of Black hole bina-ries.It has been known that BHXBs are variable on dif-ferent time-scales and exhibit both broad band con-tinuum noise and narrow features termed as Quasi-periodic oscillations. While there have been severalmodels to explain the dynamic origin of these variabil-ties, a consensus and conclusive explanation for themremains illusive. One of the possible reasons for thisimpasse is that the radiative processes that are involved in these variations have not been clearly identified. Thetime-averaged spectra of black hole binaries are usuallyrepresented by physically motivated radiative compo-nents such as disk emission, thermal comptonization,and refection components.The broad band spectrum from AstroSat allows forfitting of complex models such as General Relativis-tic disk emission and blurred reflection components,which can be used to constrain physical parameterssuch as the spin of the black hole and the accretion rate.It is also imperative to identify which of these compo-nents are responsible for the variability and in partic-ular to identify the spectral parameters which may bechanging that cause the observed variability. It is likelythat more than one of these spectral parameters are in-volved and estimating any time di ff erence between thevariations of these parameters can provide critical in-formation regarding the illusive dynamic origin of phe-nomena. AstroSat observations of BHXBs can providethese estimates paving the way for an emerging fieldwhere variability is quantified and understood in termsof the active radiative processes of the system. Here wepresent some of the scientific results obtained so far. Itshould be emphasised that these results were possibledue to the unique capabilities of LAXPC (wide bandspectroscopy, higher e ffi ciency in the hard X-rays andfine time resolution) and the simultaneous spectra avail-able from SXT.After decommissioning of RXTE in 2012,
As-troSat / LAXPC is the only X-ray timing instrument bestsuited for studying spectral-temporal characteristics ofblack hole X-ray binaries (BHXBs). During last fiveyears, LAXPC successfully observed many BHXBsand very interesting results have come out. Here wediscuss some of BHXBs results which were observedon several occasions with LAXPC instrument and in-depth analysis were performed. In §3.1-3.5 we presentthe primary results from LAXPC grouped on the ba-sis of broad topics which highlight the advantage of theinstrument. In §3.6, we enumerate these results in adi ff erent degree of detail, for the individual black hole . Astrophys. Astr. (0000) : sources that have been observed by LAXPC.3.1 Energy dependent temporal properties:
Typically the variability is quantified by the powerspectrum which is the square of the Fourier transformof a lightcurve obtained over a certain energy range.If normalized in a certain way, the integration of thepower spectrum (over some frequency range) gives thevariance of the variation in that energy and frequencyrange. Another useful quantity is the fractional rootmean square, (frms) which is the square root of thevariance normalized the mean of the lightcurve. It isinsightful to consider this as a Data Cube where thepower is provided as a function of frequency and en-ergy. Collapsing the Data Cube over energy (or a givenrange of energy) provides the power spectrum in thatenergy range. On the other hand collapsing the DataCube over a given range frequency, provides the frmsas a function of energy for that frequency range. More-over, cross-frequency analysis between energy bandsprovides time-lag as a function of energy.LAXPC can e ffi ciently provide frms and time-lagas a function of energy for di ff erent frequency bandsas shown in the early analysis of the black hole sys-tems GRS 1915 +
105 (Yadav et al ., 2016a) and CygnusX-1 (Misra et al ., 2017) Encouraged by these unprece-dented results, attempts were made to develop for-malisms which could obtain physically meaningfulradiative parameter variations and time-lag betweenthem. In the hard state, the spectra of Cygnus X-1above 3 keV, is dominated by thermal Comptonization.Maqbool et al . (2019). developed a model to predictthe variability of such a spectrum when the seed pho-ton flux and the heating rate of the corona varies with atime-lag between them. Fitting the predictions to theobserved time-lag and frms as a function of energy,led them to quantify the variability in the seed pho-ton flux and the coronal heating rate as a function offrequency. Mudambi et al . (2020a) used the same anal-ysis to constrain the variability of these quantities forthe bright transient X-ray binary, MAXI J1820 +
070 inits hard state observation by AstroSat / LAXPC allow-ing for a comparison with Cygnus X-1. While theseanalysis are for the continuum variability, Jithesh et al .(2019) applied the model to a QPO observed by As-troSat LAXPC in the hard state of the black hole sys-tem, SWIFT J1658.2-4242.The model used in these works mentioned above,is only applicable to the thermal Comptonization com-ponent, while a thermal component is often observed inthe spectra of black hole systems which are not in a purehard state. A generalized method which can predict en-ergy dependent variability properties for arbitrary spec-tral components is challenging since (i) the response of the spectrum to parameter variation has to be donenumerically and (ii) the spectral parameters have to becast into physical ones rather than empirical ones (fore.g. heating rate of corona instead of say the asymptoticpower-law index for Comptonization models). An ini-tial step in this direction has been taken by Garg et al .(2020) who have fitted the energy dependent frms andtime-lag of the QPO observed in GRS 1915 +
105 byAstroSat / LAXPC.3.2
Variations of timing and spectral properties andtheir correlation :
An important and promising use of AstroSat / LAXPCdata is the study of the variation of timing features andtheir correlation with spectral parameters. The sensitiv-ity of LAXPC to detect small variation in rapid timingproperties was proved by the first time detection of thesmall ∼
7% variation of the high frequency ( ∼
70 Hz)QPO in GRS 1915 +
105 (Belloni et al ., 2019), whichalthough small has the potential to di ff erentiate betweengeneric models. The QPO has also been reported bySreehari et al . (2019) who showed that its presence de-pends on the strength of the high energy spectral com-ponent.The correlation with spectral parameters wasdemonstrated by Bhargava et al . (2019) where a longmonitoring of the black hole transient, MAXI J1535-571, showed a tight correlation between the QPO fre-quency with the high energy spectral index rather thanthe flux. An important expected correlation is that be-tween QPO frequency and the inner radius of a trun-cated disk, since a physical characteristic time-scaleassociated with the radius maybe responsible for theQPO. Using LAXPC and SXT observations of theblack hole system GRS 1915 + et al . (2020a)showed that there is such a correlation and identi-fied the QPO frequency with the dynamical time-scalecorrected by General Relativity as predicted decadesago. This important result was possible because of thebroad band spectral coverage by SXT and LAXPC andthe simultaneous rapid timing information provided byLAXPC.3.3 Long term variability :
Long observations of BHXBs by LAXPC, which couldbe continuous exposure for more than a day, or moni-toring of sources on weeks / months time-scale have pro-vided extensive information regarding the long termbehaviour of these sources. A good example is theenigmatic black hole system GRS 1915 +
105 and As-troSat was fortunate to observe it during a transitionfrom a non-variable class to a structured large ampli-tude one (Rawat et al ., 2019). This allowed for tracking
J. Astrophys. Astr. (0000) : of the QPO and its energy dependent property rms andtime-lag as the source made the transition. The tran-sient Swift J1658.2-4242 shows ’flip-flop’ state transi-tions on time-scale of minutes, which is reflected bothin the spectra and power density spectrum was studiedby LAXPC and other instruments (Bogensberger et al .,2020).Monitoring of an outburst from a black hole binarycan be used to track the spectral and timing evolutionas has been done for two outbursts of 4U 1630-472(Baby et al ., 2020). The study inferred the appearanceof the standard disk after a few hours of the burst and itspersistence as the source evolved to the soft state. Mon-itoring observations of Cygnus X-3 by LAXPC has pro-vided important clues on the formation of the radio jetand its connection to the accretion disk (Pahari et al .,2018b). The observations of Cyg X-3 provides a mea-surement of the orbital period and the discovery of lowfrequency mHz QPO whose energy dependent rms andtime-lag could be quantified (Pahari et al ., 2017).3.4
Broadband Spectral Fitting :
Although LAXPC spectral resolution is lower than thatof other instruments such as Nustar, however it doesprovide a wide energy band especially when combinedwith SXT data. Hence AstroSat observations have beenused to fit complex spectral models to constrain sys-tem parameters. For example, after fitting standardmodels to the AstroSat spectra of the black hole tran-sient MAXI J1535-571, Sreehari et al . (2019), fitted thetwo component accretion flow model to the data andfound that the accretion rate to be nearly at the Edding-ton value and could further constrain the mass of theblack to be around ∼ et al . (2019), used the relativistic diskand blurred reflection model to constrain the spin, dis-tance to the source and mass of the black hole to be ∼
10 solar masses. The di ff erent mass estimate are dueto di ff erences in the model and other assumptions, butwhat is perhaps important is the ability of these dif-ferent models to make quantitative estimates given As-troSat data. Black hole mass has also been estimatedusing two component flow for the black hole binary4U 1630-472, where the utility of long term monitor-ing of a source as has been demonstrated. The extragalactic black hole systems LMC X-1 has a well con-strained distance and black hole mass and hence As-troSat data has proved useful to constrain the black holespin (Mudambi et al ., 2020b). Relativistic disk fittingof well studied sources like GRS 1915 + et al ., 2020). 3.5 Synergy with other observatories:
Perhaps the most promising use of AstroSat data iswhen they are analyzed in conjunction with data fromother instruments of observatories. The complementarynature of the di ff erent instruments provides a multi-faceted and unprecedented view of the systems. Sinceblack hole binaries are variable simultaneous (or quasi-simultaneous) data would be optimal.A good example of using the di ff erent capabilitiesof instruments was the study of the black hole system4U 1630-47 using the high energy grating spectra fromChandra along with AstroSat’s LAXPC and SXT data(Pahari et al ., 2018a). While the latter provided evi-dence for a highly ionized wind, relativistic disk modelfit to the continuum using AstroSat data showed thatthe black hole is highly spinning, thus making an in-teresting connection between wind outflow and blackhole spin. Another example is when the combination ofhigh resolution Nustar data (3-80 keV) with AstroSat(SXT, LAXPC and CZTI) (0.7-200 keV) of MAXIJ1820 +
070 allowed for the use of more complex andsophisticated models as compared to only AstroSat data(Mudambi et al ., 2020a) or just Nustar data.Even when the observations are not simultaneous,the use of multiple instruments has proved extremelyworthwhile as in the spectral and timing study of theflip-flop state transitions of Swift J1658.2-4242 us-ing a host of instruments, XMM-Newton, NuSTAR,Swift, Insight-HXMT, INTEGRAL, ATCA and As-troSat (Bogensberger et al ., 2020).It is fortunate that at present along with AstroSat,there are two other observatories capable of high timeresolution operating namely NICER and INSIGHT-HXMT. The three observatories together can provideunprecedented timing information in an energy range(0.2-200 keV) which is about three order of magnitudeThe improvement in our understanding that would beobtained by such analysis is illustrated by Xiao et al .(2019) where they performed timing analysis of SwiftJ1658.2-4242 outburst in 2018 with Insight-HXMT,NICER and AstroSat. They found a range of QPO ac-tivities detected by the di ff erent instruments and quanti-fied their dependence. The potential of such analysis isclear, especially when one notes that the observationsused by them were often not strictly simultaneous. Itis expected that soon, joint analysis of AstroSat datawith one or both of these instruments, will be provid-ing deeper insights into the nature of Black hole sys-tems. The importance having coordinated observationswith all three observatories cannot be over emphasised. . Astrophys. Astr. (0000) : . . . . . ( . − . ) k e V / ( . − . ) k e V HR HR1(5.0−13.0) keV/(3.0−5.0) keV
Non−Variable stateIntermediate stateFlaring stateRise Fall
Figure 9 . Color-color diagram during fast transition amongthree X-ray states.
Important results of some of BHXBs with LAXPCinstrument:
GRS 1915 +
105 :
GRS 1915 +
105 is a highlyvariable BHXBs (Yadav et al ., 1999; Belloni et al .,2000) with frequent radio emission (often referred asMicroquasar) (Fender et al ., 2004; Yadav, 2006). Wehave monitored this source regularly. Yadav et al .(2016a) have studied spectra and timing properties ofGRS 1915 +
105 when the source had the characteristicsof being in Radio-quiet during March 5-7, 2016.The en-ergy dependent power spectra reveal a strong low fre-quency (2 – 7 Hz) Quasi-periodic oscillation (LFQPO)and its harmonic along with broad band noise. TheQPO frequency changes rapidly with flux. At the QPOfrequencies, the time-lag as a function of energy hasa non-monotonic behavior such that the lags decreasewith energy till about 15–20 keV and then increase forhigher energies.Rawat et al . (2019) have studied GRS 1915 + χ class) to a periodic flaring state (similar to ρ class). It is shown that such transition takes place via an Q P O f r e qu e n cy ( H z ) / M do t ( g / s ) Radius (r g ) χ classIMSHSa=0.97, N=0.22a=0.91, N=0.35a=0.99, N=0.17 Figure 10 . Color-color diagram during fast transition amongthree X-ray states. intermediate state when the large-amplitude, irregularvariability of the order of thousands of seconds turnedinto a 100-150 sec nearly periodic flares similar to ρ class / heartbeat oscillations. Figure 9 shows the color-color diagram during this fast transition. It is interest-ing to note that HR2 remains same when source tran-sits to the intermediate state but when source finally at-tains the flaring state, it regains the same HR1 but HR2has changed. Belloni et al . (2019) have analysed 92 ksof data obtained with the LAXPC instrument and theyhave detected around seven percent variation in HighFrequency QPOs in GRS 1915 + et al .(2020) have studied temporal and spectral propertiesof GRS 1915 +
105 during θ class (Belloni et al ., 2000).Sreehari et al . (2020) have studied this source duringthe soft X-ray state and have detected HFQPs in therange of 67.96 – 70.62 Hz with rms ∼ et al . (2020a) showed that there is a correla-tion between QPO frequency and disk radii, and identi-fied the QPO frequency with the dynamical time-scalecorrected by General Relativity as predicted decadesago, using LAXPC and SXT simultaneous data of GRS1915 + + Cygnus X-3 and relativistic large radio jets :
Cygnus X-3 is an extraordinary HMXBs (a close binarywith ∼ +
105 and XTE J1550-564 favours the binary har-bouring a black hole. The source is bright and persis-tent in X-rays, and the X-ray emission originates as aresult of wind accretion from a Wolf-Rayet compan-ion star onto the central compact object. In contrastto the canonical BHXBs, six di ff erent spectral stateshave been reported in Cygnus X-3 by Koljonen et al .(2010) using simultaneous X-ray and radio observa-tions. Spectral states of Cygnus X-3, as observed byLAXPC are shown in the bottom panel of Figure 11where unfolded, best-fit spectra from SXT and LAXPCjoint fitting are shown. Details of spectral modellingare discussed in Koljonen et al . (2010). With LAXPC,we observed four di ff erent spectral states: soft, hard,hypersoft and very high state. The presence of the veryhigh state is anticipated before but observed for the firsttime with LAXPC. Dramatic changes in flux in bothsoft and hard bands in di ff erent states are visible. Veryhigh energetic photons, of the order of GeV, have beendetected during the hypersoft state when the radio emis- J. Astrophys. Astr. (0000) : . . . C oun t s / s / c m Time (MJD)15−50 keV Swift/BAT lightcurve of Cyg X−3 − . . X − r a y f l u x : k e V ( P ho t on s s − c m − k e V − ) Energy (keV)Spectral states of Cyg X−3 as observed by SXT+LAXPC
Hypersoft (01 April, 2017)
Very high state (02 April, 2017)
Soft (12 August, 2016)
Hard (20 November, 2016)
Figure 11 . Top panel shows a typical 15-50 keV
Swift / BATlightcurve of Cygnus X-3 over a timescale of 1.5 year.
AstroSat / LAXPC observations during this period are shownby stars and filled circles. Bottom panel shows unfoldedspectra using SXT + LAXPC joint fitting in the energy range0.5-60 keV. Data sets used for spectral fiting are shown withfilled circle in respective colours. sion is entirely quenched.
Astrosat / LAXPC performed over 20 observationsof Cygnus X-3 during last five years. The top panelof the Figure 11 shows 15-50 keV swift / BAT lightcurveover 1.5 years and LAXPC pointing observations dur-ing this period are shown by stars and filled circles. Us-ing three long observations during 2015-2016 spanningover several orbits, Pahari et al . (2017) determined thebinary orbital period of 17253.56 ± ∼ ∼ ∼ ∼ / PCAdespite its most extensive archival data. Interest-ingly, 7-15 mHz QPOs from Cygnus X-3 was reportedwith
Exosat / ME observations (van der Klis & Jansen, 1985). Detailed analysis with LAXPC observationsshowed that QPOs were observed only during the flar-ing hard X-ray state and when the source is brightest(the peak phase of the binary orbital period). It is ex-plained in terms of increased mass accretion rate whenthe compact object passes through the denser wind sec-tion. An enhanced supply of material may temporarilyboost the temporal variability features.Until now, investigating the connection between ac-cretion disk and the radio jets in Cygnus X-3 has beenpartially successful because of the lack of truly simul-taneous X-ray and radio data. X-ray and radio monitor-ing program of Cygnus X-3 with LAXPC provided arare opportunity to explore the radio / X-ray connectionin this X-ray binary.Using long-term X-ray / radio monitoring campaignwith Swift / BAT and 11.2 GHz RATAN-600 telescope, ithas been observed that Cygnus X-3 used to move intothe hard X-ray quenched state where hard X-ray fluxdecreased by order of magnitude, and subsequently itshowed major radio flare ejection event when the ra-dio flux density increases dramatically from few tens ofmJy up to 20 Jy. Such conjunction was detected withLAXPC and RATAN-600 telescopes on 1-2 April 2017.Using detailed X-ray analysis, Pahari et al . (2018b)found that Cygnus X-3 undergo spectral state transi-tion from the hypersoft state (HPS) to a harder, moreluminous state which was never observed before. Weterm it as the very high state (VHS). Such a transi-tion occurred within a few hours when the radio fluxdensity increases from ∼
100 mJy to ∼
478 mJy. Us-ing SXT + LAXPC joint spectral analysis, they observedno hard X-rays above 17 keV during the HPS state.Within hours timescale, a flat power law appeared inthe spectra with the power-law index of 1.49 + . − . andextended up to 70 keV. Such an observation provideddirect evidence of synchrotron emission that originatesin the radio-emitting blob which was caught in the actof decoupling from the accretion disk. Such a detailedradio / X-ray coupling event was observed for the firsttime and made possible due to the LAXPC’s capabil-ities of higher e ffi ciency for hard X-ray above 30 keVand high time resolution.3.6.3 Cygnus X-1 :
Probably one of the best stud-ied black hole systems, the bright black hole system,Cygnus X-1, presents an excellent example to under-line LAXPC capabilities and the new understandingthat LAXPC data can bring. The higher e ff ective area athigh energies has allowed LAXPC to study in detail theenergy dependent rapid variability of Cygnus X-1 in thehard state. LAXPC data analysis by Misra et al . (2017),revealed the rms and time-lag variation as a function ofFourier frequencies in the broad energy range of 4 – 80 . Astrophys. Astr. (0000) : keV, extending the earlier results which were limited to30 keV. They also showed that the event mode data ofLAXPC allows for flux resolved spectroscopy in rapid( ∼
1) second time-scales, which for Cygnus X-1 re-vealed a correlation between photon index and flux. Amore extensive analysis of six observations of CygnusX-1 by Maqbool et al . (2019), showed that the energyand frequency dependent rapid variability is di ff erentfor observations which are in the hard state. More im-portantly, they could now quantitatively fit and explainthis variability in terms of a single zone stochastic fluc-tuation model.3.6.4 MAXI J1535-571 :
The bright outburst of thisrecently discovered black hole binary has been ob-served by a large number of observatories includingAstroSat and has provided one of the most compre-hensive view of any such systems. LAXPC detectedprominent QPOs at ∼ et al ., 2019) when the source wasin the hard intermediate state. Spectral modelling usingstandard Comptonization models and physically moti-vated ones like the two-component accretion flow re-vealed that source was nearly at Eddington limit andits black hole mass is ∼ M ⊙ . A detailed study ofthe broad band spectrum from SXT and LAXPC us-ing relativistic reflection models provided an estimateof the black hole spin parameter ( a ∼ .
7) and a massof ∼ M ⊙ (Sridhar et al ., 2019). Correlation betweenthe QPO frequency and spectral parameters was studiedby Bhargava et al . (2019) who found that the frequencytightly correlates with the photon index rather than theflux.3.6.5
4U 1630-472 :
Among recurrent black hole X-ray transients, 4U 1630 −
472 is unique as it shows fre-quent outbursts at about 600 days interval. The sourcewas also at the centre of focus due to its ‘superoutburst’ . × − f X P ( f ) Frequency (Hz)
Figure 12 . The frequency times the power spectra of CygnusX-1 for two energy bands; 3-10 keV band (black points) and20-40 keV band (red points). which typically lasts for two years. Because no dy-namical mass measurements have been performed, thesource has been categorised as the black hole candidatedue to its similarity of X-ray spectra-timing character-istics with other confirmed black hole X-ray binaries. Adust scattering halo analysis placed the source between4.7-11 kpc Kalemci et al . (2018) while the spectral in-dex vs mass accretion rate correlation was used to esti-mate the black hole mass of 10 ± ⊙ (Seifina et al .,2014).Two major monitoring campaigns with As-troSat / LAXPC took place between 27 August − −
17 September 2018 to mon-itor ‘mini’ outbursts from 4U 1630 −
472 which typi-cally lasts 5-6 months. Using both outburst observa-tions, Baby et al . (2020) performed detailed spectro-timing analysis. Using broadband (0.7-20.0 keV) spec-tral modelling, they detected no disk component dur-ing the onset of the 2016 outburst while later a geo-metrically thin accretion disc with the temperature of ∼ ∼ ± ± et al . (2018a) determined the presence of a fast-spinning black hole in 4U 1630-472. Using relativis-tic continuum spectral modeling on three independentmeasurements with Chandra, SXT + LAXPC and Chan-dra + SXT + LAXPC and applying Markov Chain Monte
J. Astrophys. Astr. (0000) :
Carlo simulations on fitted spectral parameters, theyconstrained the spin of the black hole to be 0.92 ± MAXI J1820 +
070 :
At a distance of 3.46 + . − . kpc (Gandhi et al ., 2020), MAXI J1820 +
070 was dis-covered as one of the closest and the brightest GalacticBHXBs known till date (Corral-Santana et al ., 2016).With the Swift / BAT peak flux of ∼ et al ., 2018). However, with theMAXI data, ‘q’ shaped track is clearly visible in theHID (Chakraborty et al ., 2020) strengthening the factthat the source harbours a black hole. AstroSat ob-served the source on two occasions during the peak ofthe outburst.Using the two days long observation with LAXPCon 30 March 2018, Mudambi et al . (2020a) observed47.7 mHz QPO in the power density spectra whichis similar to what observed from two other BHXBs:GRS 1915 +
105 and IGR J17091-3624. Using anSXT + LAXPC joint spectral analysis in the energyrange 0.7-30 keV, they observed a flat spectral indexof ∼ ∼ et al ., 2019). Such a detailed temporal anal-ysis brings out the importance of LAXPC observations.Chakraborty et al . (2020) studied the broadbandspectra which include SXT, LAXPC and CZTI in theenergy range of 0.3 −
120 keV and the best fit spectralmodelling shows the presence of a soft excess compo-nent in the form of thermal disk blackbody which issignificantly below three keV, a thermal comptonisationmodel with a very flat spectral index close to 1.4 andtwo relativistic reflection models to fit broad and nar-row iron line complex and a Compton hump. Two rela-tivistic models di ff er significantly in their coronal elec-tron temperature by a factor of ∼
7. Interestingly, therelativistic reflection component dominates the broad-band spectra. Comparing with the detailed NuSTARspectral study by Buisson et al . (2019), they inferredthat the corona in MAXI J1820 +
070 is inhomogeneous and residing close to the black hole ( < g ).
4. Active Galactic Nuclei
AstroSat / LAXPC has the potential to monitor thevariability of Active Galactic Nuclei (AGN) and inconjunction with SXT and UVIT to study the theirbroad band spectra. Blazars are jet dominated AGNwhich show high amplitude variability. Using AstroSatLAXPC and SXT observations Banerjee et al . (2020)constructed the power spectrum for the lighcurve of theblazar, Mrk 421, and detected a break or a character-istic time-scale. Such characteristic time-scale havebeen detected before for X-ray binaries and regularAGN and are believed to originate in the accretion disk.Hence this result suggests that the jet variability alsohas an disk origin. For the blazar, RGB J0710 + et al . (2020) reported a significant deviationfrom a power-law shape for the X-ray spectrum ob-tained from LAXPC and SXT. Such deviation or curva-ture reflect the underlying shape of the particle energyspectrum that produces the emission.The AstroSat results suggested that particle distri-bution has a maximum energy cuto ff as predicted bymodels where the particles are shock accelerated andradiatively cooled. Goswami et al . (2020) studied thebroad band spectral energy distribution for the blazar4C + ff erent times and movedaway from the central region. For the regular AGN, REJ1034 + et al . (2018) fitted the LAXPCand SXT spectra to reveal the presence of a soft excess,consistent with earlier results.Studies of faint sources like AGN by LAXPC islimited by the accuracy of the instrument’s backgroundestimation. It should be noted that most of the re-sults of AGN data analysis using the RXTE / PCA ob-servations were undertaken in the later times of thesatellite operations when the response and backgroundwere better characterised. With improvements and dif-ferent estimation techniques for LAXPC background(Antia et al ., 2020; Misra et al ., 2020b) this issue, it isexpected that in the near future there will be significantspectral and variability studies of AGN using LAXPCdata.
5. Summary
LAXPC instrument has carried on the legacy of RXTEPCA and HEXTE by being one of the primary in-struments to study rapid variability of X-ray systems. . Astrophys. Astr. (0000) :
As compared to RXTE / PCA, it has enhanced featureswhich are (i) an higher e ffi ciency at hard X-rays (ener-gies ≥
30 keV), (ii) event mode data allowing for tem-poral analysis in arbitrary user defined energy bins and(iii) simultaneous SXT data at the soft X-ray band (0.5–8 keV). These advantages have been demonstrated by anumber of publications reporting important results dis-cussed in this paper for more than 30 sources. More-over, the increasing reservoir of archived and futureLAXPC data is expected to provide an unprecedentedview of known sources as well as those that are yet tobe discovered. Improvements in the calibration, espe-cially in the background estimation, would lead to moredetailed spectral and timing analysis of the bright X-raysources and will enable analysis of fainter sources suchas a larger number of active galactic nuclei.It is fortunate that at present there are several X-ray missions in operation, which have complimen-tary capabilities to the instruments on-board AstroSat.Several works have already demonstrated the dra-matic advantage of having simultaneous (or even quasi-simultaneous) observations of a source with multipleX-ray missions and with those in other wavebands.There is now an unprecedented opportunity of detailedtiming studies for a wide energy range using simulta-neous observations of AstroSat / LAXPC and NASA’sNicer. Despite the challenges of organizing and ad-ministrating such coordinated observations by multiplemissions (Nicer, Nustar, chandra and other), it is imper-ative for break through results, that a large number ofthem should be undertaken.
Acknowledgements
We acknowledge the strong support from Indian SpaceResearch Organization (ISRO) in various aspects ofinstrument development, space qualification, softwaredevelopment, mission operation and data dissemina-tion. We specially acknowledge ISAC support forelectronics development and during space qualificationtests. We thank IISU tean for providing us bellow pumpalong with space qualified pump driver. We acknowl-edge support of the scientific and technical sta ff of theLAXPC instrument team for their excellent team workas well as sta ff of the TIFR Workshop who helped us atvarious level of LAXPC instrument development. References
Agrawal, P. C. 2006, Advances in Space Research, 38,2989Agrawal, P. C., Yadav, J. S., Antia, H. M., et al . 2017,Journal of Astrophysics and Astronomy, 38, 30 Agrawal, V. K., Nandi, A., Girish, V., & Ramadevi,M. C. 2018, MNRAS, 477, 5437Amin, N., Roy, J., Chakroborty, S., et al . 2020, privateCommunicationAntia, H. M., Agrawal, P. C., Dedhia, D., et al . 2020,JApA, this volumeAntia, H. M., Yadav, J. S., Agrawal, P. C., et al . 2017,ApJ Supp, 231, 10Baby, B. E., Agrawal, V. K., Ramadevi, M. C., et al .2020, MNRAS, 497, 1197Bala, S., Bhattacharya, D., Staubert, R., & Maitra, C.2020, MNRAS, 497, 1029Banerjee, A., Bhattacharjee, A., Chatterjee, D., et al .2020, arXiv e-prints, arXiv:2007.05273Belloni, T. 2006, Advances in Space Research, 38,2801Belloni, T., Klein-Wolt, M., M´endez, M., Van Der Klis,M., & Van Paradijs, J. 2000, arXiv preprintastro-ph / et al .2019, MNRAS, 489, 1037Beri, A., Paul, B., Yadav, J. S., et al . 2019, MNRAS,482, 4397Beri, A., Naik, S., Singh, K. P., et al . 2020, MNRAS,arXiv:2010.08334Bhalerao, V., Bhattacharya, D., Vibhute, A., et al . 2017,Journal of Astrophysics and Astronomy, 38, 31Bhargava, Y., Belloni, T., Bhattacharya, D., & Misra,R. 2019, MNRAS, 488, 720Bhattacharyya, S., Yadav, J. S., Sridhar, N., et al . 2018,ApJ, 860, 88Bhulla, Y., Roy, J., & Jaa ff rey, S. N. A. 2020, Researchin Astronomy and Astrophysics, 20, 098Bogensberger, D., Ponti, G., Jin, C., et al . 2020, A&A,641, A101Buisson, D. J. K., Fabian, A. C., Barret, D., et al . 2019,MNRAS, 490, 1350Chakraborty, S., Navale, N., Ratheesh, A., & Bhat-tacharyya, S. 2020, MNRAS, 498, 5873Chandra, A. D., Roy, J., Agrawal, P. C., & Choudhury,M. 2020, MNRAS, 495, 2664 J. Astrophys. Astr. (0000) :
Chaudhury, K., Chitnis, V. R., Rao, A. R., et al . 2018,MNRAS, 478, 4830Corral-Santana, J. M., Casares, J., Mu ˜noz-Darias, T., et al . 2016, A&A, 587, A61Devasia, J., Raman, G., & Paul, B. 2021, NA, 83,101479Fender, R. P., Belloni, T. M., & Gallo, E. 2004,Monthly Notices of the Royal Astronomical Society,355, 1105Gandhi, P., Rao, A., Charles, P. A., et al . 2020, MN-RAS, 496, L22Garg, A., Misra, R., & Sen, S. 2020, MNRAS, 498,2757Goswami, P., Sinha, A., Chandra, S., et al . 2020, MN-RAS, 492, 796Jain, C., Paul, B., & Dutta, A. 2010, MNRAS, 409, 755Jaisawal, G. K., Naik, S., Epili, P. R., et al . 2020a,JApA, this volumeJaisawal, G. K., Naik, S., Ho, W. C. G., et al . 2020b,MNRAS, 498, 4830Jithesh, V., Maqbool, B., Misra, R., et al . 2019, ApJ,887, 101Jonker, P. G., & van der Klis, M. 2001, ApJL, 553, L43Kalemci, E., Maccarone, T. J., & Tomsick, J. A. 2018,ApJ, 859, 88Koljonen, K. I. I., Hannikainen, D. C., McCollough,M. L., Pooley, G. G., & Trushkin, S. A. 2010, MN-RAS, 406, 307Lewin, W., & Van der Klis, M. 2006, Compact stellarX-ray sources, Vol. 39 (Cambridge University Press)Maitra, C. 2017, Journal of Astrophysics and Astron-omy, 38, 50Maitra, C., Paul, B., Haberl, F., & Vasilopoulos, G.2018, MNRAS, 480, L136Malu, S., Sriram, K., & Agrawal, V. K. 2020, MNRAS,doi:10.1093 / mnras / staa2939Maqbool, B., Mudambi, S. P., Misra, R., et al . 2019,MNRAS, 486, 2964Misra, R., Rawat, D., Yadav, J. S., & Jain, P. 2020a,ApJL, 889, L36 Misra, R., Roy, J., & Yadav, J. S. 2020b, JApA, thisvolumeMisra, R., Yadav, J. S., Verdhan Chauhan, J., et al .2017, ApJ, 835, 195Mudambi, S. P., Maqbool, B., Misra, R., et al . 2020a,ApJL, 889, L17Mudambi, S. P., Rao, A., Gudennavar, S. B., Misra, R.,& Bubbly, S. G. 2020b, MNRAS, 498, 4404Mukerjee, K., Antia, H. M., & Katoch, T. 2020, ApJ,897, 73Pahari, M., Antia, H. M., Yadav, J. S., et al . 2017, ApJ,849, 16Pahari, M., Bhattacharyya, S., Rao, A. R., et al . 2018a,ApJ, 867, 86Pahari, M., Yadav, J. S., Verdhan Chauhan, J., et al .2018b, ApJL, 853, L11Parmar, A. N., Gottwald, M., van der Klis, M., & vanParadijs, J. 1989, ApJ, 338, 1024Rawat, D., Pahari, M., Yadav, J. S., et al . 2019, ApJ,870, 4Remillard, R. A., & McClintock, J. E. 2006, Annu. Rev.Astron. Astrophys., 44, 49Roy, J., Agrawal, P. C., Singari, B., & Misra, R. 2020,Research in Astronomy and Astrophysics, 20, 155Roy, J., Agrawal, P. C., Iyer, N. K., et al . 2019, ApJ,872, 33Seifina, E., Titarchuk, L., & Shaposhnikov, N. 2014,ApJ, 789, 57Singh, K. P., Stewart, G. C., Chandra, S., et al . 2016,in Society of Photo-Optical Instrumentation Engi-neers (SPIE) Conference Series, Vol. 9905, SpaceTelescopes and Instrumentation 2016: Ultraviolet toGamma Ray, ed. J.-W. A. den Herder, T. Takahashi,& M. Bautz, 99051ESreehari, H., Nandi, A., Das, S., et al . 2020, arXiv e-prints, arXivSreehari, H., Ravishankar, B. T., Iyer, N., et al . 2019,MNRAS, 487, 928Sridhar, N., Bhattacharyya, S., Chandra, S., & Antia,H. M. 2019, MNRAS, 487, 4221Staubert, R., Klochkov, D., Wilms, J., et al . 2014,A&A, 572, A119 . Astrophys. Astr. (0000) : Staubert, R., Tr¨umper, J., Kendziorra, E., et al . 2019,A&A, 622, A61Tanaka, Y., Lewin, W., Van Paradijs, J., & VanDen Heuvel, E. 1995, Cambridge, UK: Cam, 126Tandon, S. N., Hutchings, J. B., Ghosh, S. K., et al .2017, Journal of Astrophysics and Astronomy, 38,28Truemper, J., Pietsch, W., Reppin, C., et al . 1978,ApJL, 219, L105Uttley, P., Gendreau, K., Markwardt, C., et al . 2018,The Astronomer’s Telegram, 11423, 1van der Klis, M. 1994, SSRv, 69, 208van der Klis, M. 2001, in American Institute of PhysicsConference Series, Vol. 599, X-ray Astronomy: Stel-lar Endpoints, AGN, and the Di ff use X-ray Back-ground, ed. N. E. White, G. Malaguti, & G. G. C.Palumbo, 406–415van der Klis, M., & Jansen, F. A. 1985, NATURE, 313,768Varun, Maitra, C., Pragati, P., Harsha, R., & Biswajit,P. 2019a, MNRAS, 484, L1Varun, Pradhan, P., Maitra, C., Raichur, H., & Paul, B.2019b, ApJ, 880, 61Verdhan Chauhan, J., Yadav, J. S., Misra, R., et al .2017, ApJ, 841, 41Xiao, G. C., Lu, Y., Ma, X., et al . 2019, Journal of HighEnergy Astrophysics, 24, 30Yadav, J. 2006, The Astrophysical Journal, 646, 385Yadav, J., Rao, A., Agrawal, P., et al . 1999, The Astro-physical Journal, 517, 935Yadav, J. S., Agrawal, P. C., Antia, H. M., et al . 2017,Current Science, 113, 591Yadav, J. S., Misra, R., Verdhan Chauhan, J., et al .2016a, ApJ, 833, 27Yadav, J. S., Agrawal, P. C., Antia, H. M., et alet al